Ebook PDF

All rights reserved.
May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.
Copyright @ 2017. ASHP.
EBSCO : eBook Collection (EBSCOhost) - printed on 2/20/2019 8:52 PM via UNIVERSITY TEKNOLOGI MARA
AN: 1523639 ; Zdanowicz, Martin M., American Society of Health-System Pharmacists.; Concepts in Pharmacogenomics
Account: s6790720.main.ehost
All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.
Pharmacogenomics
Fundamentals and Therapeutic Applications in Personalized Medicine
Any correspondence regarding this publication should be sent to the publisher, Ameri-
can Society of Health-System Pharmacists, 4500 East-West Highway, Suite 900,
Bethesda, MD 20814, attention: Special Publishing.
The information presented herein reflects the opinions of the contributors and advi-
sors. It should not be interpreted as an official policy of ASHP or as an endorsement
of any product.
Because of ongoing research and improvements in technology, the information and

its applications contained in this text are constantly evolving and are subject to the
professional judgment and interpretation of the practitioner due to the uniqueness of
a clinical situation. The editors and ASHP have made reasonable efforts to ensure the
accuracy and appropriateness of the information presented in this document. However,
any user of this information is advised that the editors and ASHP are not responsible
for the continued currency of the information, for any errors or omissions, and/or for
any consequences arising from the use of the information in the document in any and
all practice settings. Any reader of this document is cautioned that ASHP makes no
representation, guarantee, or warranty, express or implied, as to the accuracy and ap-
propriateness of the information contained in this document and specifically disclaims
any liability to any party for the accuracy and/or completeness of the material or for
any damages arising out of the use or non-use of any of the information contained in
this document.
Editorial Project Manager: Ruth Bloom

Production Manager: Johnna Hershey
Cover & Page Design: David Wade
Page Production: Carol Barrer
Library of Congress Cataloging-in-Publication Data
Names: Zdanowicz, Martin M., editor. | American Society of Health-System Pharmacists.

Title: Concepts in pharmacogenomics / [edited by] Martin M. Zdanowicz.
Description: Second edition. | Bethesda, MD : American Society of Health-System
Pharmacists, [2017] | Includes bibliographical references and index.
Identifiers: LCCN 2016026095 | ISBN 9781585285167
Subjects: | MESH: Pharmacogenetics
Classification: LCC RM301.3.G45 | NLM QV 38.5 | DDC 615/.7--dc23
LC record available at https://lccn.loc.gov/2016026095
© 2017, American Society of Health-System Pharmacists, Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any

means, electronic or mechanical, including photocopying, microfilming, and recording,
or by any information storage and retrieval system, without written permission from the
American Society of Health-System Pharmacists.
ASHP is a service mark of the American Society of Health-System Pharmacists, Inc.;

registered in the U.S. Patent and Trademark Office.
ISBN: 978-1-58528-516-7
10 9 8 7 6 5 4 3 2 1
Dedication
To my wife, Christine—my
soulmate, for her boundless
love and support.
To my children, Alex and Olivia,

for the love and joy they bring
into my life.
Contents
Contributors ................................................................................................................................................................................................................vi
Preface .............................................................................................................................................................................................................. viii
PART I—Fundamentals of Pharmacogenomics ....................................................................................................... 1

Chapter 1: Pharmacogenomics: Current and Future Promise..............................................................................................................3
Martin M. Zdanowicz, PhD, MEd, MA
Chapter 2: The Genetic Basis of Pharmacogenomics................................................................................................................................19
Samit Shah, PhD, RPh, MBA; Kathy D. Webster, PharmD, PhD
Chapter 3: Methodologies in Pharmacogenomics........................................................................................................................................ 59
Amy-Joan L. Ham, PhD
Chapter 4: The Pharmacogenetics of Drug Metabolism.......................................................................................................................107
Ali Roberts, PharmD, BCPS; Landry K. Kamdem, PharmD, PhD; G. Scott Weston, RPh, PhD
Chapter 5: Pharmacogenomics of Transport Proteins.............................................................................................................................. 151
Arthur G. Cox, PhD
Chapter 6: Pharmacodynamics and Pharmacogenomics....................................................................................................................183
Kathy D. Webster, PharmD, PhD; Samit Shah, PhD, RPh, MBA
PART II—Applications of Pharmacogenomics in Therapeutics...............................................................213

Chapter 7: Cardiovascular Disease.................................................................................................................................................................................215
Kathryn M. Momary, PharmD, BCPS
Chapter 8: Hematology/Oncology Pharmacogenomics.....................................................................................................................257
Todd A. Thompson, PhD
Chapter 9: Central Nervous System.......................................................................................................................................................................... 283
Megan J. Ehret, PharmD, MS, BCPP
Chapter 10: Infectious Diseases........................................................................................................................................................................................ 309
Edward Grace, PharmD, BCPS (AQ-ID), AAHIVP; Christopher Farrell, PhD
Chapter 11: Respiratory Diseases.....................................................................................................................................................................................327
Samit Shah, PhD, RPh, MBA; David Ha, PharmD
Chapter 12: Toxicogenomics...................................................................................................................................................................................................357
Helen E. Smith, MS, PhD, RPh
Chapter 13: The Pharmacogenetics of Addiction.......................................................................................................................................... 375
Patti W. Adams, PhD; Erika E. Tillery, PharmD, BCPP, CGP
Chapter 14: Pharmacogenomics and Diabetes Mellitus....................................................................................................................... 405
Launa M. J. Lynch, PhD; Kimberly L. Barefield, PharmD, BCPS, CDE
PART III—Important Issues in Pharmacogenomics..........................................................................................435

Chapter 15: Pharmacogenomic Testing and Drug Labeling...............................................................................................................437
Kimberly L. Barefield, PharmD, BCPS, CDE; James W. Fetterman, Jr., PharmD;
Andria Fetterman, PharmD
Chapter 16: Pharmacogenomics in Practice: The Role of the Pharmacist........................................................................453
Keri C. Anderson, PharmD, BCPS; Kenric B. Ware, PharmD, MBA, AAHIVP
Chapter 17: Ethics and Pharmacogenomics.........................................................................................................................................................467
Sally A. Huston, PhD
Chapter 18: Future of Pharmacogenomics in Clinical Practice....................................................................................................... 491

Rocco J. Rotello, PhD; Miriam A. Ansong, PharmD, EMBA
INDEX .................................................................................................................................................................................... 499
v
Contributors
Patti W. Adams, PhD James W. Fetterman, Jr., PharmD

Associate Professor Professor Pharmacy Practice and
Assistant Dean of Operations, Savannah Experiential Education Coordinator
South University School of Pharmacy South University School of Pharmacy
Savannah, Georgia Savannah, Georgia
Keri C. Anderson, PharmD, BCPS Edward Grace, PharmD, BCPS (AQ-ID),

Clinical Pharmacist AAHIVP
Commercial & Specialty Pharmacy Associate Professor and Vice-Chair
St. Simons Island, Georgia Infectious Diseases Specialist
Department of Clinical and Administrative
Miriam A. Ansong, PharmD, EMBA Sciences
Notre Dame of Maryland University
Associate Professor and Director
Baltimore, Maryland
Drug-Health Information and Informatics
Department of Pharmacy Practice
Cedarville University School of Pharmacy David Ha, PharmD
Cedarville, Ohio Assistant Professor, Clinical Sciences
Keck Graduate Institute: School of Pharmacy
Claremont, California
Kimberly L. Barefield, PharmD, BCPS, CDE
Associate Professor of Pharmacy Practice,
Internal Medicine Amy-Joan L. Ham, PhD
South University School of Pharmacy Assistant Professor
Savannah, Georgia Department of Pharmaceutical, Social, and
Administrative Sciences
Belmont University College of Pharmacy
Arthur G. Cox, PhD
Nashville, Tennessee
Assistant Dean and Chair
Department of Pharmaceutical Sciences
Sullivan University College of Pharmacy Sally A. Huston, PhD
Louisville, Kentucky Associate Professor, Social and Administrative
Sciences
Keck Graduate Institute: School of Pharmacy
Claremont, California
Behavioral Health Clinical Pharmacy Specialist
Fort Belvoir Community Hospital
Fort Belvoir, Virginia Landry K. Kamdem, PharmD, PhD
Associate Professor
Harding University College of Pharmacy
Christopher Farrell, PhD
Searcy, Arkansas
Associate Professor, Pharmaceutical and
Administrative Sciences
Presbyterian College School of Pharmacy Launa M. J. Lynch, PhD
Clinton, South Carolina Assistant Professor, Pharmacology
South University School of Pharmacy
Savannah, Georgia
Andria Fetterman, PharmD

Assistant Professor, Pharmacy Practice
South University School of Pharmacy
Savannah, Georgia
vi
Kathryn M. Momary, PharmD, BCPS Erika E. Tillery, PharmD, BCPP, CGP
Associate Professor, Pharmacy Practice and Associate Professor of Pharmacy Practice
Pharmaceutical Sciences Presbyterian College School of Pharmacy
Mercer University College of Pharmacy and Clinton, South Carolina
Health Sciences Center
Atlanta, Georgia
Kenric B. Ware, PharmD, MBA, AAHIVP

Assistant Professor of Pharmacy Practice
Ali Roberts, PharmD, BCPS South University School of Pharmacy
Clinical Scientist, Healthcare Services Columbia, South Carolina
Aegis Sciences Corporation
Nashville, Tennessee
Kathy D. Webster, PharmD, PhD
Founding Dean
Rocco J. Rotello, PhD Professor
Associate Professor, Pharmaceutical Sciences Keck Graduate Institute: School of Pharmacy
Cedarville University Claremont, California
Cedarville, Ohio
G. Scott Weston, RPh, PhD
Samit Shah, PhD, RPh, MBA Associate Dean, Assessment and Academic
Associate Dean of Academic Affairs Affairs
Chair, Biopharmaceutical Sciences Professor of Pharmaceutical Sciences
Keck Graduate Institute: School of Pharmacy Belmont University College of Pharmacy
Claremont, California Nashville, Tennessee
Helen E. Smith, MS, PhD, RPh Martin M. Zdanowicz, PhD, MEd, MA

Associate Professor, Department of Associate Dean, Health Studies
Pharmaceutical Sciences University of Miami School of Nursing & Health
Feik School of Pharmacy Studies
University of the Incarnate Word Coral Gables, Florida
San Antonio, Texas

Associate Professor, Pharmaceutical Sciences
University of New Mexico College of Pharmacy
Albuquerque, New Mexico
vii
Preface
It was quite gratifying to hear then President Obama outline his precision medicine initia-
tive during the 2016 State of the Union address. This initiative devotes $215 million to
“pioneer a new model of patient-powered research that promises to accelerate biomedical
discoveries and provide clinicians with new tools, knowledge, and therapies to select which
treatments will work best for which patients.”1 Several key objectives were outlined in this
initiative including improved treatments for cancer that are genetically based; creation of
a voluntary national research cohort for precision medicine; strengthening of patient data
privacy and security in the context of precision medicine; modernization of the current
regulatory environment as it relates to the development of new genetic and genomic tech-
nologies; and enhanced public−private partnerships focused on precision medicine.
The release of our second edition is particularly timely given the renewed focus on
precision or individualized medicine. As with the first edition, it was our goal to ensure
that the chapters were written in a clear, simple, and organized manner. The material is
presented at a level that would be easy to understand for students, residents, fellows,
and practitioners looking to better comprehend the fundamentals of pharmacogenomics.
Relevant clinical applications were also emphasized throughout the chapters to illustrate
to our target audience how the science of pharmacogenomics is currently impacting their
clinical practice and how it might help shape their practice in the future.
The new edition contains updates on current and next-generation genomic technolo-
gies that will be used to identify new drug targets and improve overall drug safety. Empha-
sis is placed on the role that pharmacogenomics plays in the individualization of cancer
chemotherapy and the future development of new cancer drug targets. Two new chapters
cover the pharmacogenomics of drug addiction and antidiabetic drugs.
The text is organized into three parts. The first focuses on the basic science involved
in pharmacogenomics with an emphasis on methodologies as well as the general effects
of genetic variability on the pharmacokinetics and pharmacodynamics of drug therapy. The
second part presents a systems-based review of current pharmacogenomics applications
to clinical practice. The third part focuses on relevant topics designed to give students and
practitioners a richer background of the many aspects that impact the full understanding
and application of pharmacogenomics.
The text incorporates numerous features that are designed to enhance its readability
and usefulness as a teaching aid. Key definitions are included at the beginning of each
chapter to help readers understand the terminology they will encounter as they review the
chapters. Practical case studies are also embedded within each chapter to stimulate critical
thinking and facilitate the reader’s clinical application. Clinical pearls, which are sprinkled
throughout, highlight key concepts and applications. Numerous figures and summary
tables are included within the text to enhance presentation of the material and facilitate
the readers’ understanding.
viii
I hope that you enjoy reading this text and that the material contained within enhances
your understanding and appreciation for the impact pharmacogenomics is having and will
continue to have on safe and effective pharmacotherapy.
Martin M. Zdanowicz
February 2017
1. Obama BH II. Remarks of President Barack Obama—State of the Union Address. Oral address
presented at: State of the Union Address; January 12, 2016; Washington, DC. https://www.whitehouse.
gov/the-press-office/2016/01/12/remarks-president-barack-obama-%E2%80%93-prepared-delivery-
state-union-address. [July, 11, 2016].
ix
Chapter 1 • Pharmacogenomics: Current and Future Promise 1
PART I
Fundamentals of
Pharmacogenomics
CHAPTER 1: Pharmacogenomics: Current and Future Promise...................3
CHAPTER 2: The Genetic Basis of Pharmacogenomics..................................19
CHAPTER 3: Methodologies in Pharmacogenomics......................................... 59
CHAPTER 4: The Pharmacogenetics of Drug Metabolism.........................107
CHAPTER 5: Pharmacogenomics of Transport Proteins................................151
CHAPTER 6: Pharmacodynamics and Pharmacogenomics......................183

1
CHAPTER
1
Pharmacogenomics:
Current and Future Promise
Martin M. Zdanowicz, PhD, MEd, MA
LEARNING OBJECTIVES KEY DEFINITIONS

After completing this chapter, the reader ALLELE—one of a pair of genes on a specific
should be able to location of a chromosome that controls
the same trait.
Explain the meaning of key terms used in
pharmacogenomics. BIOMARKER—a substance, which can be
measured in the body, that may be
Describe the effects of genetic
indicative of a particular disease or
polymorphisms on drug pharmacokinetic
process.
and pharmacodynamics.
DEOXYRIBONUCLEIC ACID (DNA)—a self-
Discuss the potential benefits of
replicating material present in nearly all
pharmacogenomics on healthcare and
living organisms as the main constituent
the various roles pharmacists might play
of chromosomes.
in its application.
PHARMACODYNAMICS—the study of how a
List major obstacles to the expansion of
drug exerts its effects in the body.
pharmacogenomics in clinical practice.
PHARMACOGENETICS—the study of
Discuss specific ways in which
pharmacists might take the lead in the inherited differences or variations in drug
application of pharmacogenomics to metabolism and response.
enhance their practice. PHARMACOGENOMICS—the study of the
role of inheritance in individual variation
in drug response. This term refers to
the general study of the many different
genes that determine drug behavior.
PHARMACOKINETICS—a study of
absorption, distribution, metabolism, and
excretion of drugs.
POINT MUTATION—a change in a single
nucleotide of the genome that occurs in
≤1% of the population.
POLYMORPHISM—the presence of two or
more alleles for a gene or DNA sequence

in a population.
PROTEOMICS—the large-scale study of
proteins that focus on their structure and
function.
3
4 CONCEPTS IN PHARMACOGENOMICS
SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—a point mutation occurring in >1% of the

population.
WHOLE GENOME SEQUENCING—a laboratory technique in which an individual’s entire
sequence of DNA is determined at one time.
INTRODUCTION
For many years, healthcare professionals have been taught the classic approach to deter-
mining drug doses in their patients. The vast majority of pharmacologic agents used today
are dosed based only on the patient’s body weight. Even in pediatric patients where it
has long been recognized that “children are not miniature adults,” specialized dosage
formulations and detailed pharmacokinetic and pharmacodynamic studies are still lacking.
Pharmacists have always been at the forefront in emphasizing the importance of individu-
alizing drug therapy. In pharmacokinetics they are taught that many factors can influence
drug disposition in their patients including organ function, blood flow, age, and body fat.1
However, a key component of what makes a patient an individual, namely his or her unique
genetic makeup, is often overlooked.
The science of pharmacogenomics is a dynamic and evolving field of study; its goal
is to understand how an individual’s unique genetic composition can alter his or her phar-
macokinetic and pharmacodynamic responses to a specific drug or class of drugs. Beyond
tailoring a drug to an individual based on size, weight, or age, the science of pharmaco-
genomics strives to tailor drug therapy to individual patients based on their own unique
molecular characteristics (e.g., individual differences in drug-metabolizing enzymes, drug
transporter activity, receptor sensitivity). Ultimately, such an approach may augment our
current knowledge of pharmacotherapy to enhance the efficacy of drugs while reducing
their unwanted side effects and potential toxicities.
The earliest clinical observations related to the impact of genetics on drug disposition
and effects were made in the 1950s. Kalow published a landmark paper in 1956 in which he
described several patients who did not exhibit a typical response to succinylcholine as a
result of genetic “variations” in the activity of their plasma cholinesterase enzymes.2 During
the same year, another study by Carson reported that a genetic deficiency in glucose-6
phosphate dehydrogenase enzyme was responsible for the excess hemolysis observed in
primaquine-sensitive individuals.3 Several years later, Evans detailed how genetic differ-
ences in metabolizing enzymes gave rise to patients who were “slow inactivators” or “rapid
inactivators” of the antituberculosis drug isoniazid.4 Fredrich Vogel first used the term
pharmacogenetics in 1959.5
Although studies in the new discipline of pharmacogenomics continued to expand
throughout the 1960s, the early 1970s saw the next major expansion of this field occurring
with the documentation of several genetic variations in key drug-metabolizing enzymes
of the liver. Mahgoub published a 1977 study that quantitated measurable differences
in elimination of the antihypertensive drug debrisoquine in 94 volunteers.6 Variability in
drug disposition was due to genetic variation in the individuals’ debrisoquine hydroxylase,
an enzyme later identified as cytochrome p450, 2D6 (CYP2D6). Since this initial finding,
numerous other drugs have also been shown to be substrates for CYP2D6 and thus poten-
tially affected by genetic variation in this enzyme.
In the 1980s, genetic variability in other drug-metabolizing enzymes such as thiopurine
methyltransferase (involved in 6-mercaptopurine methylation) and CYP2C19 were also

identified.7,8 As the emerging science of pharmacogenetics continued to evolve, it eventual-
ly made its way into mainstream clinical pharmacy journals in 1992.9,10 At this same time, an
explosion in molecular biology and widespread availability of new genetic biotechnology
would directly impact the pace at which pharmacogenetics progressed in the next decade.
In 1988, Congress commissioned formation of the Human Genome Project. Jointly run
by the National Institutes of Health (NIH) and the Department of Energy, the goal of the
Human Genome Project was to sequence the entire human genome. Information obtained
from this project was to be made freely available to all interested parties via a database
set up by the National Center for Biotechnology Information. The final results of the
Human Genome Project were presented in 2003—a full 2 years ahead of the anticipated
completion date.11 In 1999, 10 large pharmaceutical companies and the U.K. Wellcome Trust
philanthropy formed the SNP Consortium. The purpose of this collaboration was to find
and map 300,000 common single nucleotide polymorphisms (SNPs). The ultimate goal of
the consortium was to generate a widely accepted, high-quality, extensive, publicly avail-
able map using SNPs as markers evenly distributed throughout the human genome. When
this group completed its initial work, a total of 1.8 million SNPs were actually identified.
In 2000, the NIH established the Pharmacogenetics Research Network. This nationwide
collaboration of scientists had three main goals: (1) to examine the relationship between
genetic variation and drug response; (2) to become a resource through which researchers in
the field of pharmacogenomics could interact and share knowledge, tools, and data; and (3)
to create a publicly available database through which researchers and other professionals
could freely access information regarding the link between various phenotypes and geno-
types (https://www.pharmgkb.org/).
With the advent of new technologies that can rapidly and accurately analyze an entire
human genome for variations and the completion of Human Genome and HapMap projects,
it is now possible to conduct genome-wide association studies (GWAS). Using an SNP-
chip, such studies are able to analyze an individual’s entire deoxyribonucleic acid (DNA) to
identify various SNPs. Two groups of individuals, those with a particular disease and those
without the disease, are analyzed in parallel. Genetic polymorphisms are identified in the
affected individuals who are not found in the healthy subjects and, therefore, may be poly-
morphisms that are associated with a particular disease process. GWAS have thus far been
able to identify SNPs associated with conditions such as Parkinson’s disease, type 2 diabe-
tes, Crohn’s disease, and Alzheimer’s disease. GWAS can also be used to identify genetic
variants that may affect drug toxicity or efficacy. In the future, such GWAS may provide
valuable insights regarding the role of genetics in disease and medication response.
CASE STUDY
Azathioprine is an immunosuppressant used to treat autoimmune disease and prevent
rejection of transplant organs. Azathioprine is converted to 6-mercaptopurine. In 2014,
Manzi et al. reported on two cases of boys who had undergone renal transplantation and
developed significant neutropenia following treatment with azathioprine.12
Questions
1. What enzyme is responsible for the inactivation of 6-mercaptopurine? Describe the metabo-
lism of azathioprine.
2. How can the administration of azathioprine lead to neutropenia?
3. How would you describe the genetic variant that is associated with azathioprine-induced
neutropenia?
4. How would the dose of azathioprine need to be adjusted in patients at risk for neutropenia?
Discuss how genetic testing might be used to predict patients who would be at the greatest
risk for neutropenia when receiving azathioprine.
SOURCES OF GENETIC VARIABILITY IN DRUG RESPONSE

Presently, the human genome is estimated to contain 30,000 to 40,000 distinct genes.
When one factors in alternative splicing and post-translational modifications, the human
genome may code for in excess of 100,000 individual proteins. Although genetic variation
may arise due to a rare mutation, it most commonly occurs as random variation between
the nucleotide sequences of different individuals. The genomes of any two given individu-
als are estimated to differ by approximately one nucleotide in every thousand, or a differ-
ence of approximately 3 million base pairs in total.
Single base-pair substitutions that occur with a frequency of ≥1% in a population are
referred to as SNPs. SNPs are the most common type of variation found in human DNA.
These substitutions can occur anywhere in the DNA. To date, over 1.4 million SNPs have
been identified with more than 60,000 occurring in the coding regions for proteins. The
genes that code for the CYP enzymes 2A6, 2C9, 2C19, 2D6, and 3A4, for example, have been
shown to be polymorphic with functional variations in a significant percentage of certain
ethnic groups. A second, less common form of genetic variation stems from insertion,
deletion, or duplication of bases. These types of mutation rarely occur within the coding
regions of genes.
The most common and best studied genetic polymorphisms to date are those that
affect drug pharmacokinetics; however, an increasing number of polymorphisms are being
identified that can affect drug pharmacodynamics too. Because the overall effect of a drug
in the body is based on both pharmacokinetic and pharmacodynamic interactions, pharma-
cogenomics will need to expand its future focus to include polygenic variations that can
affect both facets of a drug’s action.
GENETIC VARIABILITY AND DRUG PHARMACOKINETICS

The pharmacokinetic profile of a particular drug may be determined by four main factors:
1. The extent to which a drug is absorbed from its site of administration.
2. How the drug is distributed in various body compartments after it is absorbed.
3. The extent and means by which a drug is metabolized within the body.
4. How a drug is excreted from the body (e.g., kidney, liver, gastrointestinal [GI]).
The single greatest source of pharmacogenetic variability identified thus far is vari-
ability that occurs in drug-metabolizing enzymes. Phase I metabolism generally involves
oxidation and reduction reactions carried out by the CYP450 system of the liver. Phase II
reactions are conjugation reactions designed to make the product of the reaction more
polar and water soluble to facilitate elimination by the kidneys. Clinically important poly-
morphisms have been identified in most of the major enzymes involved in both phase I and
phase II drug metabolism (Table 1-1).13 The majority of the genetic variations in CYP450
genes are due to single amino acid substitutions.
Although numerous genetic variants have not been reported to significantly alter
activity of the enzyme they code for in vivo, several variants have clearly been associated
with altered activity of the enzyme to the point where it impacts drug disposition. The
Table 1-1
Examples of Pharmacogenomics Variants that Affect Drug
Pharmacokinetics
Genetic Variants Effect on Activity Drug(s) Affected
Phase I Enzymes
CYP2C9*2, *3, *8, *11 Decreased Warfarin, phenytoin

CYP2C19*2 Decreased Diazepam, omeprazole, clopidogrel
CYP2C19*17 Increased
CYP2D6*3-*6, *10, *17 Decreased Codeine, anti-psychotics, SSRIs
CYP2D6*1xN, 2xN Increased
CYP3A4*22 Decreased Tacrolimus, paclitaxel, midazolam
Phase II Enzymes
N-acetyltransferase
NAT2*5, *6, *7, *14 Decreased Isoniazid, rifampin
NAT2*12, *13 Increased
Thiopurine methyltransferase
TPMT*2, *3 Decreased 6-Mercaptopurine, azathioprine
UDP-glucuronosyltransferase
UGT1A1*28 Decreased Irinotecan
Drug Transporters
P-glycoprotein Decreased Numerous substrates including

3435C>T digoxin, anti-fungals, antivirals, and
immunosuppressants
Breast cancer resistance protein Decreased Numerous substrates including
BRCP/ABCG2 chemotherapy agents, antibiotics,
and immunosuppressants
Organic anion-transporting Decreased Numerous substrates including
proteins statins, chemotherapy agents,
OAT1B1 diuretics, and metformin
SSRIs, selective serotonin reuptake inhibitors.
CYP2D6 family of enzymes, for example, exhibits a number of polymorphisms that results
in diminished activity of numerous enzymes within this family (e.g., CYP2D6*4, CYP2D6*5);
other polymorphisms (mainly duplicated or amplified alleles) result in enzymes that
metabolize their substrates more rapidly or extensively (e.g., CYP2D6*2).
Enzymes from the CYP2D6 family are involved in the metabolism of many important
and widely utilized drugs such as codeine, fluoxetine, haloperidol, and propranolol. Approx-
imately 5% to 10% of the Caucasian population may be classified as “poor metabolizers”
of agents metabolized by CYP2D6 due to the presence of a polymorphism in genes for
this particular CYP family. In patients of Asian and African heritage, a higher prevalence of
certain CYP subtypes (CYP2D6*10 and CYP2D6*17, respectively) have been identified that
are associated with reduced rates of drug metabolism for certain substrates. Other poly-
morphisms in CYP2D6 can lead to the phenotypic presentation of patients that are rapid
or ultra-rapid metabolizers (UM). In some populations (e.g., Ethiopian), the prevalence of
the UM phenotype can be as high as 29%.14
The actual clinical effect of CYP2D6 polymorphism depends on the specific drug being
metabolized. In cases where the drug being acted on by CYP2D6 enzymes is a prodrug or
less potent compound (e.g., the conversion of codeine to more potent morphine), poor
metabolizers might require higher doses of drug to obtain the required therapeutic effect.
If the drug is inactivated by CYP2D6 enzymes, then individuals who are poor metabolizers
would require lower doses to yield the desired clinical effect. Clinical effects that might
be partly attributable to CYP2D6 polymorphisms include altered codeine efficacy, the risk
of tardive dyskinesia from antipsychotics, and the overall efficacy of certain b-blockers.
The CYP2C9 and CYP2C19 families of liver enzymes also contain a number of clini-
cally significant polymorphisms. The 2C9 family is involved in the metabolism of many
clinically important drugs, including several with narrow therapeutic indices such as
phenytoin and warfarin. Two of these genetic variants involve amino acid substitutions at
the active site of the enzyme, significantly reducing the overall activity of the enzyme. The
2C19 family of enzymes is involved in the metabolism of several commonly used proton-
pump inhibitors as well as certain benzodiazepines such as diazepam. Approximately 1% to
3% of Caucasians are poor metabolizers for warfarin and phenytoin. It puts these patients
at an increased risk of bleeding and phenytoin toxicity, respectively, at therapeutic doses.
Poor metabolizers of proton-pump inhibitors might actually have a therapeutic advan-
tage because reduced inactivation of these agents can lead to higher levels in the gut.
Furata’s study of genetically poor metabolizers of omeprazole had significantly higher
cure rates for H. pylori than did patients taking similar doses who were normal metaboliz-
ers of the drug.15 A higher prevalence of polymorphisms for these enzymes among Asian
populations has also resulted in the prescribing of lower doses of diazepam for patients
of Asian descent.16
Although the CYP3A4 family of liver enzymes accounts for more than half of all
hepatic metabolism, few significant polymorphisms in this family have been reported.
The Human CYP Allele Nomenclature Database (http://www.cypalleles.ki.se/) lists only

three loss-of-function CYP alleles, but they are rare (well below 1%) in most populations.
However, a recent study found that one loss-of-function CYP3A4 allele, CYP3A4*20, was
present in 1.2% of individuals from Spain and even higher (up to 3.8%) in certain regions
of Spain.17 Although still rare, such loss of function variants could predispose patients to
adverse drug events. The activity of this particular group of enzymes can also be greatly
influenced by numerous drugs that act as inducers (e.g., carbamazepine, phenobarbital,

phenytoin) or inhibitors (e.g., cimetidine, erythromycin, ketoconazole) of 3A4.
A number of clinically significant polymorphisms have been reported in phase II
metabolizing enzymes such as N-acetyltransferase, glutathione transferases, and thio-
purine methyltransferase (TPMT).13 One particularly relevant clinical example involves
the enzyme TPMT that metabolizes anticancer drugs such as 6-mercaptopurine and
6-thioguanine as well as azathioprine, a purine antimetabolite used to treat Crohn’s
disease and rheumatoid arthritis. Several SNPs have been identified for TPMT, which can
greatly alter its activity. Since methylation is involved in both activation and metabolism
of mercaptopurine, altered enzyme activity will affect the concentration of both active
and toxic metabolites. The therapeutic index for the thiopurine agents is very narrow with
life-threatening myelosuppression being the major concern. Patients with reduced activity
TPMT will require significant dose reduction to avoid toxicity.
As a result of the demonstrated clinical impact of polymorphisms in CYP2C9 and TPMT,
the U.S. Food and Drug Administration (FDA) recommends that at-risk patients be tested
for the presence of the variants before receiving warfarin or azathioprine, respectively. The
availability of rapid and simple genetic testing can greatly facilitate genotypic testing of
patients for metabolizing enzyme polymorphisms prior to drug administration. In addition
to reducing adverse effects, studies have shown that such testing is clearly cost effective.18
A second factor that can significantly impact drug pharmacokinetics is polymorphisms
in drug transporters. A number of transmembrane transport proteins are present in the
walls of the GI tract, hepatocytes, kidney tubules, and blood-brain barrier. These trans-
porters are responsible for selectively transporting substance across biologic membranes.
Two superfamilies of transport proteins have been identified with important effects on the
absorption, distribution, and excretion of drugs, the ATP-binding cassette and the solute
carrier superfamilies. Perhaps the best studied group of transport proteins are the p-glyco-
proteins (P-gp), which function as energy-dependent, multidrug efflux pumps.19
P-gp is widely distributed in normal cells and involved in the efflux of numerous drugs
including digoxin and immunosuppressants, such as tacrolimus and human immunodefi-
ciency virus protease inhibitors. The expression of P-gp appears to differ significantly from
individual to individual. Various degrees of P-gp expression also occur in cancer cells and
bacteria where they play a key role in anticancer drug and antibiotic resistance. Numerous
polymorphisms have been identified in the multidrug resistance gene, which codes for
P-gp. One of these, an SNP in exon 26, is associated with altered intestinal expression of
P-gp and thus significantly altered absorption of drugs like digoxin or protease inhibitors.
GENETIC VARIABILITY AND DRUG PHARMACODYNAMICS

Pharmacodynamic variations in drug response are those that occur due to genetic differ-
ences in drug targets (e.g., receptors, enzymes). A significant number of drug target poly-
morphisms have been identified thus far (Table 1-2). One clinically relevant example involves
a reduced response to asthma medications due to variants in β2-adrenergic receptors and

5-lipoxygenase. At least 11 SNPs have been identified in the β2-receptor gene. Numerous
polymorphisms appear to alter receptor expression, down-regulation, or second-messenger
coupling and therefore patient response.20 Polymorphisms in the β-receptor gene have
also been identified, which may alter the responsiveness of certain patients to the cardio-
vascular effects of b-blockers too. Serotonin receptor and transporter polymorphisms have
Table 1-2
Examples of the Impact of Pharmacogenomics on Drug
Pharmacodynamics
Gene Product Affected Drug(s) Clinical Impact
Angiotensin-converting ACE inhibitors Reduced efficacy

enzyme (ACE)
ß2-Adrenergic receptors ß-blockers Variable efficacy
Serotonin transporter SSRIs, SNRIs Altered efficacy
HMG-CoA reductase Statin drugs Altered efficacy
Vitamin K epoxide reductase Warfarin Altered efficacy and increased
(VKORC1) risk for adverse effects
5-Lipoxygenase Zileuton Altered efficacy
SNRIs, selective norepinephrine reuptake inhibitors; SSRIs, selective serotonin reuptake inhibitors.
likewise been reported, which can alter the efficacy of certain antidepressant and anti-
psychotic agents. Other drug-target polymorphisms with potential clinical significance
include those for various enzymes such as 5-lipoxygenase (altered Zileuton response),
HMG-CoA reductase (altered statin response), and angiotensin-converting enzyme.21
The anticoagulant warfarin is a particularly interesting case because it is affected by
both drug-metabolizing enzyme (CYP2C9) polymorphisms and drug target polymorphisms.
CLINICAL PEARL
The response of a patient to a drug may be polygenic in that it can be
influenced by many different and possibly unrelated genes, as in the case
of warfarin.
The anticoagulant actions of warfarin are due to its inhibition of vitamin K reductase,
an enzyme involved in the regeneration of reduced vitamin K. This is a necessary cofactor
for synthesis of clotting factors by the liver. Several SNPs have been identified in the vita-
min K epoxide reductase gene, which can reduce its susceptibility to blockade and thus
lead to patients who are “warfarin resistant.”22 In 2007, the FDA concluded evidence was
sufficient to warrant a warfarin labeling change to include information about the potential
impact of genetic variation on dosing.
Identification of genetic variability in drug targets has also been of great utility in
cancer chemotherapy. For example, lung tumor cells with activation mutations in the
tyrosine kinase portion of their epidermal growth factor receptors are known to be particu-
larly sensitive to drugs (gefitinib) that inhibit this receptor. A second enzyme, thymidylate
synthase, is expressed in both normal and cancer cells. Anticancer drugs, such as metho-
trexate and fluorouracil, inhibit this enzyme in cancer cells in order to interfere with cancer
cell nucleic acid metabolism. Polymorphisms in this enzyme have been identified that can
affect the efficacy of these agents in cancer cells as well as their toxicity in normal cells.23
Several interesting polymorphisms have also been documented in genes coding for
various ion channels in the heart.24 These polymorphisms can alter the flux of ions like
potassium, which in turn may affect cardiac conduction. Some conduction changes such
as prolonged QT intervals can predispose an individual to dangerous arrhythmias such as
torsades des pointes.
Despite significant expansion of pharmacogenomic research in the past decade, the

actual application of it in daily clinical practice is still relatively limited. A number of
challenges need to be considered for a comprehensive application of pharmacogenom-
ics to therapeutics. First is the polygenic nature of many drug responses. The example of
warfarin highlights just how complex the effects of genetic variation can be with regard to
a patient’s overall drug response. The effect of polymorphisms in a particular drug dispo-
sition pathway may be heightened or blunted by other variations in separate but related
pathways. This finding is further complicated by the fact that a large number of SNPs are
likely yet to be identified and characterized.
Second, therapeutic options for a number of diseases and conditions may be relatively
limited. If pharmacogenomic testing eliminates one or more of the drug options for a patient,
there may not be any effective alternative therapies for treatment. Third is the lack of
economic incentive for developing drugs that may be useful only in a limited patient popu-
lation. Because the cost of bringing a new drug to market can be several hundred million
dollars, it would not be worthwhile for a drug company to spend a large amount of money
on a new medication if it cannot recoup its initial investment and eventually turn a profit.
Finally, healthcare providers will need to receive an extensive education with regard
to the fundamentals of pharmacogenomics including various types of pharmacogenomic
testing, interpretation of these tests, and application of their results in daily practice.
It is perhaps in this last area of clinician and patient education that pharmacists might
have the most immediate impact. When compared to other healthcare specialists, phar-
macists have the most training with regard to the pharmacology, pharmacokinetics, and
pharmacodynamics of various drugs. Clinical pharmacists currently practice a form of
individualized medication therapy when they consider the many factors that can alter
drug pharmacokinetics and effect from patient to patient. Pharmacists tailor medications
to individuals based on factors such as their liver function, renal function, age, size, and
concomitant drug use, so the concept of using another specific trait (i.e., genetic makeup)
should not seem unusual.
CLINICAL PEARL
Pharmacogenomics may be used to optimize drug dose and choice, thus
improving therapeutic efficacy and reducing side effects.
In addition to their role as healthcare providers, pharmacists currently serve as educa-

tors and clinical consultants—places where their knowledge and understanding of phar-
macogenomics can be invaluable. Pharmacists need not become experts in molecular
biology or genetics to be competent and comfortable with the applications of pharma-
cogenomics. They do, however, need to become familiar with the basic terminology that
defines this area of study. Pharmacists should understand how genetic data from patients
is obtained, along with the significance of such data and how it may be applied clinically.
From the authors’ own research, the amount of pharmacogenomic information that
pharmacists and pharmacy students currently receive is insufficient.25-27 Although educa-
tional resources for pharmacists in the area of pharmacogenomics have increased signifi-
cantly, there is still a great need for continuing education and training of pharmacists.
These educational needs may be met in part through an expanded number of specialized
CE programs, workshops, and journal articles dedicated to training current and future phar-
macists in the fundamentals and applications of pharmacogenomics. Likewise, the amount
and type of pharmacogenomics education that pharmacy students receive must be evalu-
ated now and continually reassessed in the future.
CLINICAL PEARL
As a result of their training in drug pharmacokinetics and
pharmacodynamics, pharmacists are ideally situated to advance the
frontiers of pharmacogenomics and facilitate its entry into the mainstream
of clinical practice.
In 1990, Speedie predicted that the newly emerging biotechnology related to phar-
macogenomics would have great impact on both pharmacy education and practice in the
coming years.28 The availability of pharmacogenomics information greatly enhances the
pharmacist’s decision-making ability. There are currently 137 medications, of which 166
different pharmacogenomics biomarkers have been identified with data included in the
FDA-approved label.29 ASHP recently published a statement on the pharmacist’s role in
clinical pharmacogenomics describing how this area of research and drug treatment can
improve medication-related outcomes across the continuum of care in all health-system
practice settings.30
Although the current role of the pharmacist in pharmacogenomics is still evolving, a
number of areas especially need pharmacists who are knowledgeable. Pharmacists, for
example, can take the lead in educating physicians and patients about pharmacogenomics.
In the near future, it is estimated that nearly one quarter of all prescriptions may contain
some pharmacogenomic information as part of their package insert. Inclusion of such infor-
mation will undoubtedly elicit many questions from both patients and healthcare providers
that the trained pharmacist would be ideally suited to answer. Likewise, as pharmaco-
genomic testing of patients expands, pharmacists will be asked many questions by both
patients and healthcare providers about interpreting and applying the results of such tests.
Pharmacists in various institutional clinical settings can become the main consultants on
pharmacogenomic issues and applications related to drug therapy.
Another important area where pharmacists might take the lead in applying pharma-
cogenomics is in drug efficacy and safety. Pharmacogenomics has the potential to identify
patient populations that will be most likely to experience specific benefits or adverse
effects from a particular drug. Such information can be used by pharmacists to guide drug
therapy to ensure its maximal efficacy and safety. Examples include the analysis of drug-
metabolizing enzymes as biomarkers to identify patients who might be “slow metabolizers”

for a certain drug or the use of pharmacogenomic data to optimize the dose of warfarin a
patient receives.
Pharmacists might also utilize pharmacogenomic information to predict (and thus
prevent) potential drug interactions in a specific patient population. Given their training
and experience, pharmacists may serve in many different roles by planning, executing, and
maintaining a successful pharmacogenomics program.
IMPACT OF PHARMACOGENOMICS ON DRUG DEVELOPMENT

Despite the fact that money spent by the pharmaceutical industry on new drug develop-
ment has essentially doubled in the last decade, the number of new drugs brought to
market in this same period has remained the same. Likewise, the success rate of drugs in
various phases of clinical development has declined significantly. Starting back in 2003, the
FDA made a concerted effort to promote the use of pharmacogenomics in drug develop-
ment. In addition to hosting numerous workshops, the FDA also released written guidelines
for the submission of genomic data and published a table of valid genomic biomarkers
to include in FDA-approved drug labels.29,30 In their publication Guidance for Industry—
Pharmacogenomic Data Submissions, the FDA encouraged pharmacogenomic testing in
new drug development and in a follow-up white paper they published in 2004; the use
of pharmacogenomics was identified as a key opportunity for future drug development.31
CLINICAL PEARL
Pharmacogenomics may facilitate the identification of new drug therapies
and enhance the drug development process.
Pharmacogenomics and pharmacogenetics can provide investigators with a tool for

identifying new drug targets. As the use of gene-wide microarrays expands, research-
ers will likely identify numerous genetic variants associated with specific disease states
or conditions. These variations in gene expression might serve as the starting point for
the development of drugs targeting the product of these genes or ones that can directly
modify expression of the variant genes themselves. However, one main drawback to the
genome-wide approach of identifying new candidate genes is that it does not account for
the potential interaction of other genes impacting the overall phenotype being studied.
An interesting approach to identifying new potential pharmacogenomic targets
involves the concept of examining “disease” or “drug response” pathways. For example, if
one looks at the disease of asthma, there are a number of documented polymorphisms
affecting the response to both β2-adrenergic agents and corticosteroids. The polymor-
phisms occur in genes not only coding the actual drug targets (receptors) but also coding
for signaling proteins, second messengers, and other factors that modulate the response of
these agents. Polymorphisms in the 5-lipoxygenase pathway have also been identified that
can alter response of asthma medications directed at inhibition of this enzyme pathway.
By focusing on the specific pathways of drug response in asthma, one might be more likely
to identify and characterize not just polymorphisms but interactions between polymorphic
genes in this pathway.
Pharmacogenomics can also assist in developing drugs designed for specific patient
populations or subsets of patients who might be most responsive to a particular medica-

tion. Such new drugs could potentially have enhanced efficacy in this target population.
Furthermore, this selectivity might reduce the overall occurrence of adverse effects that
could occur if the drug were widely used in a population not likely to respond to it. To some
extent, the application of pharmacogenomics might shift the focus of pharmaceutical
drug development away from broad-market drugs to those targeted more toward specific
subpopulations. Targeted drug development may present some economic challenges to
drug companies because the profitability of new compounds for select markets may be
limited. However, the potential wealth of new drug targets identified by pharmacogenomics
could offset this drawback.
Pharmacogenomics might also aid new drug development by allowing investigators

to identify patient populations more likely to exhibit adverse effects from a new drug and,
thus, eliminate this population from clinical trials. Some drugs removed from the market
in recent years (due to adverse reactions) might not have been if potentially unidentified
genetic susceptibilities had been identified in a percentage of the population and those
patients were excluded from taking the drug. Likewise, pharmacogenomics might allow
for earlier detection of specific adverse effects before a new drug goes further on in the
development process.
An interesting example is the cholesterol-lowering drug cerivastatin. This agent
was removed from the market in 2001 due to the occurrence of rhabdomyolysis (muscle
destruction). Although all of the statins marketed today have a similar mechanism of
action, they differ in their potential for causing rhabdomyolysis. Some evidence suggests
that the risk for this potentially serious adverse effect correlates with the extent by which
the individual statins are metabolized by various CYP450 enzymes.32 In the case of ceriva-
statin, the greatest risk for rhabdomyolysis occurred in patients taking gemfibrozil, a second
cholesterol-lowering agent that can inhibit key CYP isoenzymes involved in the metabo-
lism of cerivastatin. Could some of these instances of rhabdomyolysis been avoided if
pharmacogenomic testing were used to identify specific patients who were “slow metabo-
lizers” and avoid use of this agent in this population?
Pharmacists may also play an important role in new drug development as part of a
multidisciplinary, collaborative drug development team. Their expertise in drug formulation,
pharmacokinetics, and clinical pharmacotherapy can make them important contributors to
the drug development process at many different phases. Some pharmacist-researchers are
currently at the forefront of pharmacogenomics research and even leading major research
projects in this area.
FUTURE PROMISE OF PHARMACOGENOMICS

Pharmacogenomics is currently impacting several important aspects of clinical practice on
a daily basis, and the list of polymorphisms with potential clinical significance is steadily
growing. With the completion of the Human Genome Project, increased pharmacogenom-
ics research, and greater application of pharmacogenomic testing, the next decade will
likely be one of significant growth in terms of the amount of clinical practice and drug
development impacted by pharmacogenomics. Pharmacogenomics clearly has significant
potential for aiding in the development of drugs that are highly specific and efficacious due
to their targeting of specific enzymes, proteins, or other cellular targets.
Pharmacogenomics might also provide an additional and highly specific means of

determining drug dose for an individual, one that goes beyond just weight and age but
considers perhaps a patient’s ability to metabolize a drug or respond to it. This individual
dose maximization would also improve drug efficacy and reduce side effects. Overall drug
safety could likewise be improved by the use of pharmacogenomics because it might allow
practitioners to utilize the best possible drug or dose for a patient the first time.
The drug discovery and development process can also benefit greatly from pharma-
cogenomics. A wide array of potential genomic targets can be readily identified as start-
ing places for new drug development. Multiple genes in pathways involved in a particular
disease or drug effect can be studied simultaneously with respect to their relationship and
role in the disease process or drug response. The likelihood of a new drug failing in clini-
cal trials might be significantly reduced if studies are conducted in populations identified
as more likely to respond to the drug, or less likely to exhibit adverse effects to the drug
based on genetic profile.
Finally, the potential economic benefit of pharmacogenomics cannot be overlooked.
By improving drug efficacy, reducing adverse drug effects, decreasing drug trial failure, and
speeding new drug development, pharmacogenomics can have an unequivocal impact on
the overall cost of healthcare (Table 1-3).
CLINICAL PEARL
The money spent on pharmacogenomic research and testing could reduce
overall healthcare costs by improving drug efficacy, reducing adverse drug
effects, decreasing drug trial failure, and speeding new drug development.
If the use of pharmacogenomics is to become fully incorporated into daily clinical

practice, pharmacists will need to be an integral part of the process. One day patients may
carry around a “super card” that contains all of their medical history and data, including a
detailed genetic profile. Pharmacists would be able to access this data from the patient
instantly at their practice sites and use it to ensure that drug choice and dose was optimal.
Pharmacists may also one day play a vital role in performing key genetic tests on patients
at their point of contact. Both of the above scenarios would require pharmacists to be
Table 1-3
Pharmacogenomics: Benefits and Challenges
Benefits Challenges
Optimization of drug choice and dose to Expanding the application of

minimize side effects and maximize efficacy pharmacogenomics in clinical practice
Targeted drug development Existing legal and ethical framework
Reduced failure of drugs in clinical trials if Patient and provider education regarding
participants are selected based on their pharmacogenomic testing
pharmacogenomic profile
Identification of polygenic drug effects Informed consent and the use of genetic
information
More rapid drug development Cost effectiveness

Identification of new drug targets Attitudes toward genetic testing
Reduction in healthcare costs Complexity of polygenic drug responses
Enhanced disease screening Pharmacogenomics could lead to potentially
smaller and more specialized drug markets
well trained in the fundamentals of pharmacogenomics. They need to be able to not only
interpret the results of pharmacogenomic tests but also be able to apply it to the specific
pharmacotherapy of their patient. Pharmacists are currently called on by physicians to
optimize patient drug therapy; pharmacogenomic data would simply be another variable or
tool that pharmacists could use to do so.
In the future, pharmacists will probably have to become familiar with other disciplines
that likewise may utilize genomics to impact clinical care. The “omics” revolution includes
other fields of study such as proteomics, which explores changes in protein expression of
individuals under various conditions; nutrigenomics, which studies the potential interaction
between dietary nutrients and genotypic expression; metabonomics, which investigates the
alteration of metabolites and metabolic pathways under various conditions; and toxico-
genomics, which studies how genomes are affected by toxins or other environmental
factors.
SUMMARY
Although a number of challenges must still be overcome, the promise of pharmaco-
genomics remains bright. Pharmacists must be willing to embrace not only this emerging
discipline but to take the lead in its future implementation and direction in order to make
this promise a reality.
REFERENCES
1. Burton ME, Shaw LM, Schentag JJ, et al. Applied Pharmacokinetics & Pharmacodynamics. Baltimore,
MD: Lippincott Williams & Wilkins; 2006.
2. Kalow W. Familial incidence of low pseudocholinesterase level. Lancet. 1956;2:576-577.
3. Carson PE, Flanagan CL, Ickes CE, et al. Enzymatic deficiency in primaquine-sensitive erythrocytes.
Science. 1956;124:484-485.
4. Evans DAP, Manley KA, McKusick VA. Genetic control of isoniazid metabolism in man. Br Med J.
1960;2:485-491.
5. Vogel F. Moderne probleme der humangenetik. Ergeb Inn Med Kinderheilkd. 1959;12:52-125.
6. Mahgoub A, Idle JR, Lancaster R, et al. Polymorphic hydroxylation of debrisoquine in man. Lancet.
1977;17:584-586.
7. Weinshilboum RM, Sladek SL. Mercaptopurine pharmacogenetics: Monogenic inheritance of erythro-
cyte thiopurine methyltransferase activity. Am J Hum Genet. 1980;32:651-662.
8. Wedlund PJ, Aslanian WS, McAllister CB. Mephenytoin hydroxylation deficiency in Caucasians:
Frequency of a new oxidative drug metabolism polymorphism. Clin Pharmacol Ther. 1984;36:773-780.
9. Gibaldi M. Pharmacogenetics: part I. Ann Pharmacother. 1992;26:121-126.
10. Gibaldi M. Pharmacogenetics: part II. Ann Pharmacother. 1992;26:255-261.
11. Collins FS, Green ED, Guttmacher AE, et al. A vision for the future of genomics research. Nature.
2003;422:835-847.
12. Manzi S. Toce M, Burns M. Pharmacogenomic testing to mitigate azathioprine adverse drug effects.
Clin Toxicol. 2014;52:328.
13. Wilkinson GR. Drug metabolism and variability among patients in drug response. N Engl J Med.
2005;352:2211-2221.
14. Akullu E, Persson I, Bertilsson L, et al. Frequent distribution of ultra-rapid metabolizers of debriso-
quine in an Ethiopian population carrying duplicated and multi-duplicated functional CYP2D6 alleles.
J Pharmacol Exp Ther. 1996;278:441-446.
15. Furata T, Ohashi K, Kamata T, et al. Effect of genetic differences in omeprazole metabolism on cure
rates for Helicobacter pylori infection and peptic ulcer. Ann Intern Med. 1998;129:1027-1030.
16. Ozawa S, Soyama A, Saeki M. Ethnic differences in genetic polymorphisms CYP2D6, CYP2C19,
CYP3A4’s and MDR1/ABCB1. Drug Metab Pharmacokinet. 2004;19:83-95.
17. Apellániz-Ruiz M, Inglada-Pérez L, Narajo MEG, et al. High frequency and founder effect of the
CYP3A4*20 loss of function allele in the Spanish population classifies CYP3A4 as a polymorphic
enzyme. Pharmacogenomics J. 2015;15:288-292.
18. Van den Akker-van Marle ME, Gurwitz D, Detmer SB, et al. Cost-effectiveness of pharmacogenomics
in clinical practice: a case study of thiopurine methyltransferase genotyping in acute lymphoblastic

leukemia in Europe. Pharmacogenomics. 2006;7:783-792.
19. Kerb R. Implications of genetic polymorphisms in drug transporters for pharmacotherapy. Cancer Lett.
2006;234:4-33.
20. Brodde OE, Leineweber K. Beta-2-adrenoceptor gene polymorphisms. Pharmacogenet Genomics.
2005;15:267-275.
21. Evans WE, McLeod HL. Pharmacogenomics-drug disposition, drug targets and side effects. N Engl J
Med. 2003;348:538-549.
22. Hall AM, Wilkins MR. Warfarin: a case history in pharmacogenetics. Heart. 2005;91:563-564.
23. Lee W, Lockhart AC, Kim RB, et al. Cancer pharmacogenomics: powerful tools in cancer chemotherapy
and drug development. Oncologist. 2005;10:104-111.
24. Pfeufer A, Jalilzadeh S, Siegfried P. Common variants in myocardial ion channel genes modify the QT
interval in the general population. Circ Res. 2005;96:693-701.
25. Latif DA. Pharmacogenetics and pharmacogenomics instruction in schools of pharmacy in the USA: is
it adequate? Pharmacogenomics. 2005;6:317-319.
26. Zdanowicz MM, Huston SA, Weston GS. Pharmacogenomics in the professional pharmacy curriculum:
content, presentation and importance. Int J Pharm Ed. 2006;2:1-12.
27. Koomer A, Dutta AP, Tran HT. Current state of pharmacogenomics/pharmacogenetics information in
the schools and colleges of US, Canada and UK. Paper presented at: Annual Meeting of the American
Association for Colleges of Pharmacy. July 19-23, 2008; Chicago, IL.
28. Speedie MK. The impact of biotechnology upon pharmacy education. Am J Pharm Ed. 1990;54:55-60.
29. U.S. Food and Drug Administration. Genomics. May 2015. Available at: http://www.fda.gov/drugs/scien-
ceresearch/researchareas/pharmacogenetics/ucm083378.htm
30. ASHP Statement on the Pharmacist’s Role in Clinical Pharmacogenomics. Available at: http://www.
ashp.org/DocLibrary/BestPractices/FormularyPositions.aspx. Accessed October 29, 2015.
31. U.S. Food and Drug Administration. Innovation, stagnation. Challenge and opportunity on the critical
path to new medical products. Available at: http://www.fda.gov/oc/initiatives/criticalpath/whitepaper.
html. Accessed April 30, 2009.
32. Schreiber DH, Anderson T. Statin-induced rhabdomyolysis. J Emerg Med. 2006;31:177-180.
CHAPTER
2
The Genetic Basis of
Pharmacogenomics
Samit Shah, PhD, RPh, MBA and Kathy D. Webster, PharmD, PhD

After completing this chapter, the reader ALLELE—alternative forms of a gene at a
should be able to given locus.
Describe how the “message” in DNA is AUTOSOME (AUTOSOMAL)—any
used to make proteins. chromosome other than the sex
chromosomes or the mitochondrial DNA.
Describe human gene structure and its
regulatory sequences. CHROMATID—the two parallel identical
strands, connected at the centromere,
Discuss the importance of studying
of a doubled chromosome after
genetic variation.
chromosomal replication but before
Differentiate between a polymorphism anaphase.
and a mutation.
CHROMOSOME—organized structures
Describe the different types of that carry the DNA. Humans have 22
polymorphisms and the nomenclature pairs of autosomes and 1 pair of sex
used to describe them. chromosomes.
List examples for the different types of COPY NUMBER VARIANT
polymorphisms. POLYMORPHISMS—a polymorphism that
Describe the Hardy-Weinberg Equilibrium, occurs when a segment of genome (such
linkage disequilibrium, haplotype, and as an entire gene) is doubled during the
genome-wide association study. crossing-over process in meiosis.
Calculate frequencies of genotypes and DEOXYRIBONUCLEIC ACID (DNA)—a self-
alleles from genotype data. replicating material present in nearly all
Explain the role of racial/ethnic genetic of chromosomes.
differences in pharmacogenomics
studies. EXON—sequence found in pre-mRNA that
codes for the peptide.
Note: Significant contributions from Dr. Taimour Y. Langaee and Dr. Jaekyu Shin, authors of the chapter
published in the previous edition, are gratefully acknowledged.
19
GENETIC CODE—set of rules that define how information in the DNA or RNA is used to
create chains of amino acids (proteins).
GENOME—the total genetic material contained within the chromosomes of an organism. In
humans 46 chromosomes make up the genome, and almost all cells contain an entire
copy of the genome.

GENOME-WIDE ASSOCIATION STUDY (GWAS)—a survey of several genetic variants in
different individuals to study if there is an association with traits such as disease
susceptibility or how individuals respond to certain drugs.
GENOTYPE—genetic constitution of an organism used to refer to the set of two alleles
inherited for a particular gene or locus. Genotype also refers to an individual’s genetic
makeup or sum total of genes.
HAPLOTYPE—a set of alleles or polymorphisms on a chromosome often inherited together.
HETEROZYGOTE (HETEROZYGOUS)—an individual who has two different alleles at a given
locus on a pair of homologous chromosomes.
HISTONES—proteins that organize DNA into units termed nucleosomes.
HOMOZYGOTE (HOMOZYGOUS)—an individual possessing a pair of identical alleles at a
given locus on a pair of homologous chromosomes.
INSERTION/DELETION POLYMORPHISMS—polymorphisms that result from addition or
deletion of nucleotide(s) or base pair(s) from the DNA sequence of a gene.
INTRON—sequence found in pre-mRNA that is a noncoding region.
LINKAGE DISEQUILIBRIUM—non-random association of alleles at different sites.
MESSENGER RIBONUCLEIC ACID (mRNA)—large family of RNA molecules that convey
genetic information from DNA to the ribosome.
NONHOMOLOGOUS—not pertaining to a pair of chromosomes.
NONSYNONYMOUS SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—single nucleotide
changes in the coding regions of DNA resulting in a codon that codes for a different
amino acid.
PHENOTYPE—physical manifestation of a genetic trait or a general constitutional
manifestation of health or disease in an individual.
RIBONUCLEIC ACID (RNA)—a nucleic acid present in all living cells.
SHORT TANDEM REPEAT (STR) POLYMORPHISMS—short sequences of DNA (2−5 base
pairs), which are repeated multiple times in a sequence; also called microsatellites.
SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—a DNA sequence variation occurring when a
single nucleotide in the genome differs between paired chromosomes in an individual.
SYNONYMOUS SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—single nucleotide changes in
the coding regions of DNA that does not result in a change in the protein sequence.
WILD TYPE—phenotype or gene of the typical form found in natural conditions.
Chapter 2 • The Genetic Basis of Pharmacogenomics 21
INTRODUCTION
The field of molecular biology and genetics has advanced greatly in the past 10 years. New
technologies, assays, and methods are being developed at a fast pace. Although it is chal-
lenging to collect and present all the genetic information related to pharmacogenomics in
a single chapter, we focused on key areas that would be of value to practicing pharmacists.
The first section provides a brief introduction to molecular biology and includes informa-
tion about the structure of deoxyribonucleic acid (DNA), ribonucleic acid (RNA), human
genes, and a short description of transcription and translation. Transcription is the transfer
of genetic information from DNA to messenger ribonucleic acid (mRNA) and translation is
the synthesis of protein based on the mRNA template.
In the second section, genetic variation (DNA mutations and polymorphisms) geno-
types and phenotypes are discussed. In the third section, we cover the Hardy-Weinberg
Equilibrium (HWE), linkage disequilibrium, haplotype, and the genetic differences among
racial and ethnic groups. The goal of this chapter is to provide readers with a background
in genetics that allows them to better understand and apply the fundamental concepts
of pharmacogenomics.
CASE STUDY—WARFARIN AND ATENOLOL

A.J. is a 56-year-old African-American man who presented to an emergency department
with signs of a TIA (transient ischemic attack). He is diagnosed with atrial fibrillation
and hypertension. He does not smoke or have any other comorbidity. A.J. comes to your
pharmacy with prescriptions for warfarin and atenolol. His geneticist friend told him that
his genes may affect the way he responds to his prescribed medications.
Questions
1. Atenolol is a b1-adrenergic receptor (ADRB1) antagonist. Polymorphisms in ADRB1 can influ-
ence how patients respond to atenolol. Gly389Arg (glycine to arginine change at position
389 of the protein) is a commonly found polymorphism in African Americans. Is this an
example of a synonymous SNP, nonsynonymous SNP, STR, or an insertion/deletion poly-
morphism?
2. Thirty-four percent of African Americans are homozygous for the arginine allele at codon 389
in ADRB1 gene. What is the frequency of the arginine allele in African Americans?
3. The enzyme CYP2C9 is important for metabolizing warfarin, and individuals who have lower
CYP2C9 activity may require lower doses of warfarin. Individuals with a CYP2C9*3 allele
have decreased enzymatic activity compared to individuals with a CYP2C9*1 allele. A lower
daily dose of warfarin is required in individuals with which of the following genotypes—
CYP2C9*1/*1 or CYP2C9*3/*3?
4. The CYP2C9*15 (9100 C>A) allele involves a change from a C to A, resulting in a codon for ser-
ine replaced with a stop codon. What would be the functional consequence of the nucleotide
change from C to A?
DEOXYRIBONUCLEIC AND RIBONUCLEIC ACIDS

DNA is the hereditary material that is passed from one generation to the next and contains
information to make RNA, which then is used to make proteins. DNA and RNA are linear
polymers composed of long strands of similar units (nucleotides) connected to each other
via covalent phosphodiester bonds.
The basic unit of DNA is a deoxyribonucleotide. Each deoxyribonucleotide contains

three parts: a 5-carbon deoxyribose sugar, a phosphate group, and a nitrogenous base. The
nitrogenous bases are either pyrimidines or purines. In DNA, the purine bases are adenine
and guanine, and the pyrimidine bases are cytosine and thymine (Figures 2-1 and 2-2).
DNA exists as a double-stranded helix with the two strands of DNA running anti-
parallel to each other. A 5′OH of one deoxyribonucleotide unit binds to the 3′OH of anoth-
er unit via a phosphodiester bond (Figure 2-2). When multiple deoxyribonucleotide units
are connected, 5′ and 3′designations are used to distinguish the two ends of a DNA strand.
The strands are considered anti-parallel because one strand of DNA runs in the 5′ to 3′
direction, and the other strand that binds to it runs in the 3′ to 5′ direction. In DNA, adenine
(A) always base pairs with thymine (T), and cytosine (C) base pairs with guanine (G). Because
of this specific base pairing, the complementary strand sequence of DNA can be predicted
if the sequence information for the other DNA strand is known.
The specific base pairing between DNA bases also helps explain how DNA is replicated
in a semi-conservative manner. The two strands of DNA are separated, and each strand
can then serve as a template to aid synthesis of the complementary strand. The reason
is because each strand contains information necessary to make its complementary strand.
Several proteins and enzymes participate in DNA replication, including DNA polymerases
that are equipped with proof reading and repair systems. DNA polymerases catalyze syn-
thesis of new DNA strands by adding a complementary nucleotide to the hydroxyl group on
the 3′ end of the growing DNA strand, and they also correct errors that occur during DNA
synthesis. DNA replication is termed semi-conservative as each newly synthesized double-
stranded DNA is made of one old and one new strand of DNA.
RNA consists of a long chain of ribonucleotide subunits. Each ribonucleotide, like a
deoxyribonucleotide, contains three parts: a nitrogenous base, a phosphate, and a 5-carbon
sugar (ribose). The principal differences between RNA and DNA are that RNA nucleotides
contain a 5-carbon ribose instead of a deoxyribose sugar, and RNA contains the nitrog-
enous base uracil (U) instead of thymine (T). DNA is more stable than RNA; and unlike DNA,
RNA is usually single-stranded (Figures 2-1 and 2-2).1
GENE FUNCTION AND STRUCTURE

DNA is found in almost all cells within an organism. To fit the long strands of DNA within
a cell, DNA is tightly packaged into structures termed chromosomes. Most multicellular
organisms possess two copies of each chromosome. DNA sequences in chromosomes are
organized as genes. A gene is a sequence of DNA nucleotides at a specific location on
the chromosome, which contains information to make a specific sequence of RNA. Thus,
genes are the basic unit of heredity and play a role in determining most characteristics of
an organism.
The location of a gene or a DNA marker on the chromosome is termed a locus. An
allele is the alternative form of a gene at a given locus. The total of human genetic mate-
rial, termed the genome, is contained within 46 chromosomes. Most cells within an organ-
ism (with exceptions such as mature red blood cells) contain an entire copy of the genome.
All cells in an organism contain the same DNA and genes. However, expression of genes
varies between cells from different tissues and organs. Thus, while the liver and kidney
cells from an individual would have the same DNA, the proteins expressed in these differ-
ent cell types will be different.
Purines
NH2 O
N N N NH
N N N N NH2
H H
Adenine (A) Guanine (G)
Pyrimidines
NH2 O O
N NH NH
N O N O O
N
H H H
Cytosine (C) Thymine (T) Uracil (U)
(DNA) (RNA)
FIGURE 2-1 Purine and pyrimidine bases.

FIGURE 2-2 DNA and RNA structure.
In humans, the information in the protein-coding genes is used to make precursor

mRNA (pre-mRNA). Most pre-mRNA contains both coding (exons) and noncoding (introns)
sequences. Pre-mRNA processing, which involves removal of the introns and additional
modifications, gives rise to mature messenger RNA (mRNA). The mRNA is then translated to
produce proteins. In addition to the exons and introns, genes also have regulatory sequenc-
es located before or upstream (5′) of the coding sequence called 5′ untranslated region (5′
UTR), and after or downstream (3′) of the coding sequence called 3′ untranslated region
(3′ UTR). These untranslated regions are transcribed into mRNA but are not translated into
amino acids (Figure 2-3). Regulatory sequences such as enhancers, silencers, insulators,
and cis and trans-acting elements play important roles in the regulation of gene expression
at the transcription and translation level.
Gene expression is the process used in the synthesis of a functional gene product
(such as protein) from a gene, and it may be used to describe the relative levels of gene
product produced. Transcription factors bind to regulatory sequences that may be located
in close proximity or far from the promoters in the introns and cause an increase or de-
crease in gene expression. Repressors of translation (protein synthesis) can also bind to
cis and trans-acting elements on the mRNA to repress or inhibit protein synthesis. The
regulation of the iron-storage protein ferritin by the concentration of iron in the cells is an
example of translation (protein synthesis) regulation. Promoters are nucleotide sequences
in DNA that are located at 5′ or upstream of coding region and are sites where RNA poly-
merase II and other transcription factors bind to start the process of transcription of RNA
from DNA (Figure 2-3).2-4
TRANSCRIPTION
Transcription is the process by which information in DNA is transcribed or copied into an
RNA sequence. Several different types of RNA are transcribed from the DNA. One of these
is the messenger ribonucleic acid (mRNA), which conveys the message from the DNA to
the protein synthesis machinery of the cell to produce proteins. The two strands of the
DNA are temporarily separated, and RNA is transcribed in the 5′→3′ direction using the
complementary DNA strand running in the 3′→5′ direction as the template. The strand
used during transcription as the template is referred to as the template or the noncoding
or the antisense strand.
DNA-dependent RNA polymerase (RNA polymerase or RNAP) is the enzyme that pro-
duces the RNA using the DNA as a template. Promoters are regions on the DNA that define
where transcription is initiated by RNAP. Eukaryotic transcription initiation is complex; sev-
Exon Intron Exon Intron Exon

5’ 1 1 2 2 3 3’
Promoter
5’ UTR 3’ UTR
(Untranslated region) (Untranslated region)
FIGURE 2-3 Human gene structure.
eral proteins (general transcription factors) along with RNAP II bind to the promoter region
to initiate transcription. Cis-acting elements (enhancers and silencers) are non-coding DNA
sequences that regulate the transcription of a neighboring gene, and trans-acting ele-
ments (activators and repressors) are coding genes that produce RNA or proteins inhibiting
transcription of another gene. These elements help to precisely regulate gene expression.
RNAP possesses helicase activity, which separates the two DNA strands to form a
transcription bubble. Unlike DNA replication, a primer is not required for initiation of RNA
synthesis. Complementary ribonucleotides are matched to the antisense (template) strand
of DNA and added to the growing chain through the action of RNAP. The sequence of
the mRNA is exactly the same as the sense (coding or non-template) strand of the DNA,
except that in RNA the base thymine is replaced with uracil and the nucleotides contain
ribose instead of deoxyribose. A termination signal in the DNA is used to mark the point at
which RNA polymerase is separated from the DNA template, and the transcription process
is terminated.1,5-7
Extensive processing of the RNA occurs in eukaryotes prior to the RNA being utilized
by the protein machinery for synthesis of proteins. The newly synthesized RNA (precursor
mRNA or pre-mRNA) undergoes 5′-capping, 3′-polyadenylation, and splicing. A 7-methyl-
guanosine cap is added to the first nucleotide via a 5′-5′ triphosphate linkage. Multiple
adenosine monophosphates are added to the 3′ end of the chain.8 Pre-mRNA consists of
sequences used by the translational machinery (exons) to express proteins and noncod-
ing regions (introns) that are removed from the RNA through splicing. A complex of small
nuclear RNAs (snRNAs) and other proteins named a spliceosome catalyzes the removal of
introns from the pre-mRNA. Alternative splicing or processing of a single pre-mRNA can
give rise to multiple mRNAs, depending on which exons are retained. This process can help
explain how a limited number of genes result in a larger, more complex proteome (Figure
2-4).9,10
THE GENETIC CODE

The genetic code is a set of rules that defines how the information in the DNA or RNA is
used to create chains of amino acids (polypeptides). Each three nucleotides or bases des-
ignate an amino acid. Because we have four possible nucleotides and each codon consists
of three nucleotides, there are a total of 43 or 64 codons. Of the 64 codons, three codons
(UAG, UGA, and UAA) code for no amino acid but rather act as “stop codons”; the rest code
for 20 amino acids.
The first codon of the mRNA that is translated is termed the start codon. Typically, the
start codon is AUG. Because AUG codes for methionine, proteins start with this amino acid.
The genetic code is called degenerate since 61 codons code for only 20 amino acids, and
multiple codons may code for the same amino acid. The number of codons that code for
an amino acid varies from one codon (AUG for the amino acid methionine) to six codons
(CGU, CGC, CGA, CGG, AGA, and AGG for the amino acid arginine). Codons that specify
for the same amino acid are said to be synonyms. The genetic code is unambiguous—one
codon codes for a single amino acid, and the code is read continuously; no nucleotides are
skipped between codons.
The amino acids that are coded by a sequence of nucleotides varies depending on
which set of codons from the mRNA are “read” to produce the polypeptide. The way the
5’ 3’
5’ UTR Exon 1 Intron 1 Exon 2 Intron 2 Exon 3 3’ UTR
Gene
Promoter
Transcription
5’ UTR Exon 1 Intron 1 Exon 2 Intron 2 Exon 3 3’ UTR

Pre-mRNA
Splicing Introns are spliced

(removed)
Exon 1 Exon 2 Exon 3
mRNA
Nucleus Open Reading Frame (ORF)
Translation
COOH
H2N
Protein
FIGURE 2-4 Transcription and RNA processing.
codons are read depends on the first nucleotide that is read in the sequence, and each of
the three possible ways of reading a nucleotide sequence is a reading frame.
To better understand how the reading frame of a particular nucleotide sequence may
vary, we can look at an example of a short nucleotide sequence—ACCGGGAAA. If the
reading of the codon starts from the first position, the codons read will be ACC, GGG, and
AAA, which translates into the amino acids threonine (Thr), glycine (Gly), and lysine (Lys). If
the reading starts from the second position, the codons read will be CCG and GGA, which
code for proline (Pro) and glycine (Gly). If the reading starts from the third position, the
codons read will be CGG and GAA, which would then code for arginine (Arg) and glutamate
(Glu). Thus, three reading frames are possible for each strand of DNA.1,5 This example also
illustrates how deletion or insertion of a single nucleotide in a sequence can disrupt the
reading frame and result in a group of entirely different amino acids being translated, even
if the rest of the nucleotide sequence has no changes.
TRANSLATION
The processed mRNA that is formed through transcription is transported from the nucleus
to the cytoplasm, where the protein synthesis machinery is located. The message in the
language of nucleotides is then translated into the language of amino acids through the
use of transfer RNAs (tRNAs) and ribosomes. As described above, the genetic code defines
how the message in the mRNA is translated to form proteins. Transfer RNAs are short RNA
molecules that contain a three nucleotide region (the anticodon) and carry an amino acid
corresponding to the anticodon. Ribosomes are the molecular machines that are respon-
sible for protein synthesis. Found in both prokaryotes and eukaryotes, ribosomes are com-
plexes composed of both protein and RNA components.
The process of translation (protein synthesis) takes place in three steps—initiation,
elongation, and termination. Initiation of translation is a complex process that involves

binding of the initiation factors and ribosome to the 5′ end of the mRNA. Transfer RNAs
carry the amino acid corresponding to the mRNA sequence, and elongation involves ad-
dition of the correct amino acid to the growing polypeptide chain through formation of a
peptide bond. The elongation of the polypeptide chain from the amino (N-terminal) to the
carboxyl (C-terminal) corresponds to the 5′→3′ direction in the mRNA. When one of the
stop codons (UAA, UAG, or UGA) in the mRNA is reached, release factors are incorporated
into the ribosome in lieu of tRNAs. This results in release of the polypeptide chain and
termination of protein synthesis.1,5,11,12
Following translation, proteins may undergo post-translation modifications such as gly-
cosylation, methylation, and phosphorylation. Post-translational modifications expand the
functional repertoire of the proteome and play a key role in the regulation of protein activity.
REGULATION OF GENE EXPRESSION

The levels and activity of a specific protein need to be precisely regulated in each cell.
Thus, a variety of mechanisms are utilized by the cell to regulate gene expression. DNA
methylation involves addition of methyl groups on certain regions of the DNA, which allows
recruitment of other proteins and subsequent silencing of gene expression.13 Modifications
such as acetylation or methylation of histones, the proteins around which DNA is wrapped
in the nucleosomes, can change the accessibility of DNA to the transcriptional machinery
and thus allow regulation of gene expression without alteration in DNA sequences.14 Ribo-
switches are RNA molecules that may bind to specific effector molecules and regulate the
activity of these RNAs. Riboswitches can allow control of translation and, thereby, regulate
the amount of protein that is made in the cells in response to other small molecules pres-
ent in the cell.15 MicroRNAs are short RNAs that are partially complementary to multiple
mRNA sequences and down-regulate gene expression by binding to mRNA and causing
translational repression.16 Even after a protein is formed, post-translational modifications
can dictate whether a protein is active, therefore, helping to regulate its activity.
GENETIC VARIATION
Genes code for proteins and so variations in gene sequence can result in production of
proteins that may have a different amino acid sequence. The most common DNA sequence
variation is a single nucleotide polymorphism known as an SNP or “snip.” In some cases, a
variation in a single nucleotide may result in a synonymous codon and, thus, production of
the same polypeptide. However, in other cases, variation in a single nucleotide may result
in incorporation of a different amino acid and potentially a change in the activity of the
protein. Besides a single base substitution, other sequence variations are also found that
involve removal or addition of several nucleotides. Some of these variations may change
the reading frame and result in production of nonfunctional proteins.
Variations in gene sequences are known as mutations or polymorphisms, depending

on their occurrence in the population. A polymorphism is an allele that is present in 1% or
more of the population, whereas a mutation is a rare allele found in less than 1% of the
population.17 Mutations in DNA are due to environmental factors such as radiation, cigarette
smoke, drugs, or sunlight. They can also be caused by copying errors during the DNA rep-
lication process. Some of these variations in gene sequence may result in only small dif-
ferences in traits among people such as height or eye color, while others may have a more
dramatic effect and result in diseases. Polymorphisms are found in >1% of the population
and typically do not result in severe diseases. They are most likely responsible for differ-
ences in traits such as hair color or increased susceptibility for certain diseases. Proteins
also play a key role in drug response, and thus it follows that polymorphisms can affect the
way in which an individual will respond to drugs. Pharmacogenomics is the study of how
genetic variations influence an individual’s response to drugs.
GENOTYPES AND PHENOTYPES

Humans possess two copies of each chromosome, and the set of alleles for a particular
locus or gene is referred to as the genotype. Phenotype, on the other hand, refers to ob-
servable or functional traits such as eye color, height, or protein activity. In 1908, Wilhelm
Johannsen proposed the distinction between genotype and phenotype after realizing that
the hereditary and developmental pathways had separate causes. Johannsen’s distinction
between genotype and phenotype was deduced from Mendel’s work on inheritance in
garden peas in 1900. Mendel’s work clearly made the distinction between what we call the
genome and the phenome today. The genome includes all of the genes and noncoding
sequences. Humans have two different genomes: a nuclear genome (~3.2 billion base pairs)
and a mitochondrial genome (~16,500 base pairs). The phenome refers to all the pheno-
types expressed by an organism. The association between genotype and phenotype is not
always clear, and the phenotype depends on both the genotype and the environmental
factors that regulate the expression of the genotype. In some cases a single phenotype
may be associated with many genotypes, as seen in multigenic diseases.
Researchers can acquire individual-level genotype-phenotype data from the data-
base of the Genotype and Phenotype (dbGaP; http://www.ncbi.nlm.nih.gov/entrez/query.
fcgi?db=gap). The dbGaP was created and is operated by the National Library of Medicine’s
National Center for Biotechnology Information (NCBI; http://www.ncbi.nlm.nih.gov/), which
receives and archives data from studies that investigated the association between pheno-
type and genotype such as genome-wide association studies (GWAS).18
Individuals may have a homozygous (possess identical alleles) or a heterozygous
(possess different alleles) genotype for a given trait. When individuals are heterozygous,
the interaction between the two alleles produces the phenotype. One allele may be com-
pletely dominant over the other or exhibit incomplete dominance or codominance. Because
each of the two alleles contributes to the genotype, genetic sequence variations have dif-
ferent effects depending on whether they are present on one or both of the alleles. For
example, consider a genetic variation resulting in an allele that produces a dysfunctional

truncated protein. Individuals that possess both alleles with this genetic variation will have
no functional protein levels. On the other hand, individuals with one normal-function or
wild-type allele and one variant allele will produce a dysfunctional truncated protein that
will likely produce some functional protein. Individuals who possess two normal-function
or wild-type alleles will generally produce high levels of the functional protein.
GENETIC VARIATION NOMENCLATURE

A consistent nomenclature system is essential to describe variations in gene sequences.
Recommendations for the description of DNA sequence variants have been published by
the Human Genome Variation Society.19 Some key points in the recommendations include
referring to the A of the initiator codon as +1 (and the immediately preceding nucleotide
as −1, with no nucleotide 0), indicating substitutions by “>” (1997 G>T thus indicates that
nucleotide G in the reference sequence at position 1997 is changed to a T), indicating dele-
tions by “del,” and indicating insertions by “ins” (1997-1998insT thus indicates that a T was
inserted between nucleotides 1997 and 1998 of the reference sequence).
When referring to a cDNA sequence, the nucleotide number is preceded by a “c”, as
in c.187C>G. On the other hand, when a variation results in a change in the amino acid se-
quence, the change is denoted by the original amino acid followed by the position in the
peptide chain followed by the amino acid found in the mutation, as in p.His63Asp. A “p”
may be placed in front of the amino acid to indicate that this is a change in the protein se-
quence. Letters and numbers are used to identify a particular gene, thus VKORC1 indicates
the gene that makes the protein vitamin K oxidoreductase complex 1.
Variations in the cytochrome P450 (CYP) enzyme system are a common source of vari-
ability in an individual’s response to drugs. A star (*) nomenclature system has been used
to describe variations in the CYP450 genes.20 The wild-type or normal function allele is
generally referred to as the *1 allele. Thus a *1 allele for CYP2D6 is indicated as CYP2D6*1.
Higher numbers (*2, *3) refer to polymorphisms that may have altered activity. The allele
numbers may indicate different functional effects in different genes. Thus a *2 allele for
one gene may lack activity, while a *2 allele for another gene may have reduced activity.
When extra copies of a gene are present, they are indicated with a *2Xn, where n refers to
the number of copies of the gene on the same chromosome. When using the star nomen-
clature to indicate the genotype, both alleles are listed. In the case of the CYP2D6 *1/*3
genotype, one allele is a *1 allele and the other is a *3 allele.
A SNP database (dbSNP) has also been developed and maintained by the NCBI to
serve as the central repository for all genetic sequence variants, including both polymor-
phisms and mutations. A unique number, a reference SNP (rs), is assigned to each genetic
variation to serve as an identifier for the location and kind of variation.21 The dbSNP data-
base is available online through the NCBI Entrez system.
TYPES OF POLYMORPHISMS
Several different types of polymorphisms have been found. These range from variations
involving substitution of a single nucleotide to others involving hundreds or thousands of
nucleotides. The frequency of polymorphisms varies among different racial/ethnic groups.
The different types of polymorphisms are described below.
Single Nucleotide Polymorphisms

This is the most common form of polymorphism, where one nucleotide or base is substi-
tuted by another. Millions of SNPs have been identified, and they are responsible for more
than 90% of variations in the human genome. SNPs are found in the coding, noncoding,
and regulatory regions of genes. Based on the location of SNPs, they have different effects
on gene expression and function.22-25 SNPs located in the coding, promoter, or regulatory
regions of a gene can alter the amino acid sequence of the expressed protein or level of
gene expression.
Some SNPs are located outside the coding and regulatory regions of a gene and have
no effect on gene expression (the amino acid sequence or expression levels of the gene).
However, these SNPs may still be linked to disease susceptibility or drug response be-
cause several SNPs can travel together in a population. SNPs not directly responsible for
a phenotype may also be found together with other SNPs that contribute to alterations in
protein sequence or protein levels. SNPs that do not reside in regulatory or coding regions
of genes but are associated with a certain trait are referred to as linked or indicative SNPs.
Coding Region SNPs
There are an estimated 20,000–25,000 protein-coding genes in the human genome.26 The
human genome comprises about 180,000 exons comprised of 30 million nucleotides.27
Single nucleotide changes in the coding regions of a DNA sequence may result in a codon
that codes for the same amino acid, a codon that codes for another amino acid, or a stop
codon. Because of the redundancy of the genetic code, some SNPs do not lead to changes
in amino acids despite replacement of a nucleotide by another nucleotide. These SNPs
are termed synonymous SNPs or silent polymorphisms. Although synonymous polymor-
phisms do not result in a change of amino acids, they may affect RNA secondary structure
and could thereby alter gene expression. The synonymous SNP in the DNA excision repair
gene (ERCC1) associated with altered 5-fluorouracil/oxaliplatin therapy in colorectal can-
cer is an example of this type of SNP.28-30A change from a C to a T at codon 118 results
in conversion of a commonly used AAC codon to a less commonly used AAT codon. Both
of these code for asparagine, but the change of codon has been shown to reduce ERCC1
gene expression and repair activity, resulting in a higher response rate to 5-fluorouracil/
oxaliplatin combination chemotherapy.30
SNPs resulting in codons that code for a different amino acid can change protein
activity. These types of polymorphisms are referred to as nonsynonymous SNPs or non-
synonymous polymorphisms. The nonsynonymous polymorphisms in the thiopurine meth-
yltransferase (TPMT) gene result in several variant alleles called TPMT*2, TPMT*3A, and
TPMT*3C.31,32 These TPMT variants produce proteins that have reduced enzymatic activity
and are associated with increased toxicity of TPMT-metabolized thiopurine drugs such as
azathioprine and mercaptopurine.
When one nucleotide in a codon is replaced with another, the replacement can also
result in a stop codon (TAA, TGA, or TAG), which leads to premature termination of protein
synthesis. This kind of polymorphism results in expression of truncated proteins, which
generally lack normal function. An example of a nonsense mutation is the CYP2C19*3 al-
lele. A nucleotide change from G to A results in the codon being read as UAG instead of
UGG, leading to production of a truncated protein that lacks drug metabolism activity.33
Noncoding Region SNPs
The noncoding region that includes introns, regulatory sequences, and most intergenic
DNA comprises the largest fraction of the human genome. SNPs in regulatory sequences
or elements such as promoters and enhancers can alter levels of gene expression. Poly-
morphisms in noncoding regions can also affect splicing, mRNA stability, and degradation
resulting in different proteins or altered expression. SNPs that occur in the noncoding,
regulatory regions of genes are referred to as regulatory SNPs (rSNPs).34-36
A rSNP in the promoter region of tumor necrosis factor α (TNF-α) gene creates a new
binding site for the oct-1 transcription factor that results in over expression of TNF-α in
monocytes and increased susceptibility to cerebral malaria in affected Africans.37 Altera-
tion of gene expression caused by rSNPs in the 3′untranslated region (3′UTR) of genes can
affect mRNA stability, half-life, and degradation.38,39 A rSNP in the 3′UTR region of the
human dihydrofolate reductase (DHFR) gene results in increased expression of the DHFR
protein. This rSNP could alter the effectiveness of drugs such as methotrexate, which exert
their action through inhibition of the DHFR enzyme.40
Polymorphisms in splice sites can result in alternative splicing, abnormal protein pro-
duction, and clinical consequences. Alteration of the relative levels of the two main mRNA
splice isoforms of the dopamine receptor D2 (DRD2) is associated with two intronic SNPs.
The differences in D2 mRNA splicing are strongly associated with increased susceptibility
toward cocaine abuse.41
Insertion/Deletion (INDEL) Polymorphisms

When nucleotide(s) or base pair(s) are added or deleted from the DNA sequence of a
gene, they are called insertions and deletions, respectively. The insertion/deletion (I/D)
polymorphisms (indels), which are the insertion or deletion of one or more nucleotide(s),
are widely distributed in the human genome and can occur anywhere in the DNA se-
quence of genes. Many of these insertion or deletion polymorphisms can disrupt the
reading frame and thereby lead to production of nonfunctional or truncated proteins. The
following example illustrates how insertion of one nucleotide can shift the reading frame:
If a sequence starts as GAA/TTC/AAG/GTT, it is coding for glutamic acid/

phenylalanine/lysine/valine. If there is an addition (insertion) of a C
nucleotide at the beginning of the sequence [C]GAA/TTC/AAG/GTT, it
changes to CGA/ATT/CAA/GGT, which now codes for arginine/isoleu-
cine/glutamine/glycine. Frameshift mutations often result in stop codons,
causing production of truncated proteins. If the indels are three nucleo-
tides or multiples of three nucleotides, they may not change the reading
frame but can still alter the activity of proteins.
CYP2D6*3 is a polymorphism in which deletion of a single nucleotide A at position

2549 causes a frameshift that abolishes the activity of the enzyme. The clinical conse-
quence of this polymorphism is that affected patients are unable to activate prodrugs or
metabolize drugs requiring CYP2D6. CYP3A5*7 is a variant in which insertion of a single
nucleotide T between positions 27131 and 27132 results in a frameshift and production of a
nonfunctional protein.42 Examples of indel polymorphisms with clinical importance include
insertions or deletions in the angiotensin-I converting enzyme (ACE) gene, the alpha2β-
adrenoceptor (ADRA2B) gene, and the nucleotide-binding oligomerization domain-con-
taining protein 1/caspase recruitment domain 4 (NOD1/CARD4), a gene associated with
susceptibility to inflammatory bowel disease.43
Copy Number Variant or Gene Deletion Polymorphisms

Copy number variants occur when a segment of the genome is doubled during the
crossing-over process in meiosis. This usually happens when the two sister chromatids
are not aligned properly, and the breaking and rejoining of DNA is not 100% accurate. This
results in one chromosome having more DNA than the other of the pair. Duplicated genes
(paralogs) can result in gene loss because redundant genes are usually removed from the
genome. If duplicated genes persist, they display a different function or result in increased
gene expression and thus increased protein levels. An example of a copy number variant is
the CYP2D6*2Xn allele, in which a single chromosome has multiple copies of the CYP2D6
gene. CYP2D6 gene duplication with increased enzyme levels results in ultra-rapid me-
tabolism of drugs that are substrate for this enzyme. The ultra-rapid genotype occurs in 3%
of the northern European white population, 5% to 10% of southern European and Arabian
populations, and 10% to 30% of Ethiopian and Saudi Arabian populations.44,45
The CYP2D6*5 allele is an example of a gene deletion in which the entire CYP2D6*5
gene is missing.46 CYP2D6 gene deletion results in poor metabolism of drugs that are sub-
strates. Variation of CYP2D6 levels due to copy number variants or gene deletion polymor-
phisms can affect the activation or metabolism of a number of drugs and prodrugs such as
amitriptyline, codeine, and tamoxifen.47
Short Tandem Repeat Polymorphisms

Short tandem repeat (STR) or variable number of tandem repeat (VNTR) polymorphisms
are short sequences of DNA, which are repeated in tandem and occur in a variable copy
number. The two main types of tandem repeats are microsatellites and mini satellites.
Microsatellites typically contain multiple repeats of 10–100 base pairs, whereas mini satel-
lites typically refer to tandem repeats of 2–4 base pairs.48
One example of a STR polymorphism is a dinucleotide TA repeat in the promoter
region of the UDP-glucuronosyltransferase 1A1 (UGT1A1) gene. Instead of the six repeats
found in the wild-type allele (UGT1A1*1), seven TA repeats are found in the variant al-
lele UGT1A1*28. This polymorphism (UGT1A1*28) decreases expression of UGT1A1 protein,
which is involved in the metabolism of irinotecan, a topoisomerase I inhibitor used in the
treatment of colorectal cancer. Irinotecan is a prodrug that is converted to the active me-
tabolite SN-38 by carboxylesterases. Low levels of UGT1A1 have been shown to result in
accumulation of SN-38 with an increase in irinotecan-related toxicity, including diarrhea
and neutropenia.49
Another example of this type of polymorphism is tandem repeats of the multiple
nucleotide sequence of 5′-GGCGGG-3′ in the promoter region of 5-lipoxygenase (ALOX5)
gene, which has been implicated as an important marker for inflammatory diseases.
The most commonly occurring allele has five repeats (5′-GGCGGG GGCGGG GGCGGG
GGCGGG GGCGGG -3′), and the three most common variant alleles have four STRs with
deletion of one repeat (5′-GGCGGG GGCGGG GGCGGG GGCGGG -3′), three STRs with
deletion of two repeats (5′-GGCGGG GGCGGG GGCGGG-3′), and six STRs with addition
or insertion of one repeat (5′-GGCGGG GGCGGG GGCGGG GGCGGG GGCGGG GGCGGG
-3′). It has been shown that the variant alleles have decreased promoter activity and, con-
sequently, decreased transcription of the ALOX5 gene. Reduced ALOX5 protein production
may in turn affect the clinical response of patients with asthma to drugs such as zileuton,
which target the 5-lipoxygenase pathway.50
TYPES OF MUTATIONS
A genetic variation that occurs in <1% of the population is termed a mutation. Mutations in
DNA can be caused by environmental factors such as radiation, cigarette smoking, chemical
agents, ultraviolet light, or by copying error during the DNA replication process. Different
types of mutations are described below with some examples of the diseases they cause.
Substitutions (Single Base-Pair Changes)

Substitutions involve a single nucleotide change and may be a missense, nonsense, or
silent mutation. A missense mutation nucleotide substitution changes a codon so it now

codes for another amino acid. In sickle-cell anemia, the substitution of A in codon GAG
with T (GTG) at the nucleotide position 17 of beta chain of the hemoglobin gene results
in the change of the amino acid from glutamic acid (GAG) to valine (GTG).51 A nonsense
mutation occurs when one nucleotide in a codon is replaced with another that results in
a stop codon (TAA, TAG, and TGA), which terminates the process of translation or protein
synthesis prematurely. An example of a nonsense mutation (TGG codon of tryptophan
amino acid at peptide position 1282, and 1316 to stop codon) occurs in the cystic fibrosis
transmembrane conductance regulator (CFTR) gene, which results in a premature stop in
the protein synthesis of this transporter.52 When substitution of one nucleotide by another
results in no change in the coded amino acid, it is termed a silent mutation.
Insertions and Deletions

Insertion and deletion mutations involve insertion or deletion of a nucleotide(s) or base
pair(s) from the DNA sequence of a gene. If indels are one or two nucleotides, they change
the reading frame and cause a frameshift mutation. Even indels of three nucleotides or
multiples of three nucleotides can cause human genetic disorders. In Huntington’s disease,
the insertion of CAG that codes for amino acid glutamine creates the Huntington protein,
which in turn results in abnormal increases of p53 protein levels in the brain cells that
causes cell death by apoptosis. The genetic mutation in Huntington’s disease is also called
guanine expansion. Although the general population may have from six to 26 of the CAG
repeats, patients with Huntington’s disease may have from 40 to >100 CAG repeats.53
Fragile X syndrome is another example of an indel mutation, which is caused by
insertion and expansion of trinucleotide repeats (CGG) in 5′UTR of the fragile X mental-
retardation-1 gene (FMR-1) on the human X chromosome. The number CGG repeats in
normal people may range from six to 60 repeats, but in individuals with this disorder the
number of repeats exceeds 200 and may grow up to 1,000 repeats.54 Intragenic deletions
and duplications have also been reported and are responsible for more than two-thirds of
the mutations that lead to Duchenne muscular dystrophy and Becker muscular dystrophy.55
Inversion Mutations
Inversion mutations occur when a short (a few bases) fragment or an entire section of DNA
is reversed. Inversions in some cases may involve several genes from a large region of the
chromosome. Inversion of DNA can result in genomic disorders such as Hunter’s syndrome
(an X-linked recessive disorder with progressive damage of various tissues and organs) and
some forms of hemophilia.56
Translocations
When a segment of one chromosome is transferred (translocated) to another non-
homologous chromosome, it is called a translocation. The joining of two unlinked pieces
of the genome can lead to diseases such as leukemia. Translocations can also make a gene
nonfunctional if the break takes place within a gene, or it can result in synthesis of a hybrid
or fusion gene. If the hybrid genes are translated, they may yield proteins that have an N-
terminal of one protein coupled to the C-terminal of another protein. The altered chromo-
some 22 (also known as Philadelphia chromosome) is the result of translocation that is
created by reciprocal fusion of two segments of chromosomes 9 and 22. This translocation
produces a hybrid or fusion gene that includes two genes called BCR and ABL-1, which in
humans may cause hematologic malignancies.

The Philadelphia chromosome is often seen in the cancer cells of patient with chronic
myelogenous leukemia (CML). In most B-cell tumors, a translocation occurs between chro-
mosome 8 and three other chromosomes (2, 14, and 22). Most patients with the Burkitt’s
lymphoma (a high grade B-cell neoplasm and a childhood tumor that is also seen in adults)
carry translocation of c-myc oncogene from chromosome 8 to either immunoglobulin (Ig)
heavy chain region on chromosome 14 or on light chain loci of chromosomes 2 or 22. In this
translocation no fusion gene or protein is produced, and the oncogene from chromosome
8 is under transcriptional control of an Ig gene promoter, which causes overexpression of
the oncogenic protein involved in Burkitt’s tumor.57,58
Chromosomal Aberrations (Monosomy and Trisomy)

Chromosomal aberrations may also be referred to as genome mutations. In trisomy, there is
an extra copy of a chromosome. An example is in trisomy 21 (Down syndrome, having three
chromosome 21 instead of two), which occurs in 1 of 660 newborns. In monosomy, a whole
chromosome is absent. An example of monosomy is seen in Turner syndrome (monosomy
X, only has one X sex chromosome), which occurs in 1 of 2,500 females.59,60
GENETIC ASSOCIATION STUDIES

As previously discussed, polymorphisms can influence susceptibility to certain diseases
or an individual’s response to certain drugs. Genetic association is when one or more
genotypes are associated with a genetic trait beyond what would be expected by chance.
Genetic association studies are conducted to identify relationships between patterns of
polymorphisms and traits. Pharmacogenomic association studies compare frequencies of
genotypes and alleles for a drug response phenotype. Therefore, it is important to under-
stand how frequencies of genotypes and alleles are calculated.
Calculation of Genotype and Allele Frequencies from Genotype Data

If nAA, nAa, and naa are the number of subjects with corresponding AA, Aa, and aa genotypes
in a pharmacogenomic study, then the total number of subjects (N) in the study will be N
= nAA + nAa + naa. Let PAA, PAa, and Paa be the frequencies of the AA, Aa, and aa genotypes
in the study. Then, PAA = nAA/N, PAa = nAa/N, and Paa = naa/N. In addition, PAA + PAa + Paa = 1.
If the alleles A and a are located on an autosomal (nonsex) chromosome, the fre-
quency of each allele can be calculated by the following formula:
PA = (2nAA + nAa)/2N
Pa = (2naa + nAa)/2N
Note that the denominator should be 2N because an individual has a pair of autosomal
chromosomes. Table 2-1 shows genotype data obtained from 293 subjects. The frequen-
Table 2-1
A Genotype Data from 293 Subjects
Genotype
AA Aa Aa Total
Count 220 65 8 293

Frequency 0.75 0.22 0.03 1.0
(= 220/293) (= 65/293) (= 8/293) (= 0.75 + 0.22 + 0.03)
cies of AA, Aa, and aa genotypes are 0.75 (= 220/293), 0.22 (= 65/293), and 0.03 (= 8/293),
respectively. The A and a alleles occur in 86% (= ([2 × 220] + 65) × 100/[2 × 293]) and 14%
(= ([2 × 8] + 65) × 100 /[2 × 293]) of the population.
Hardy-Weinberg Equilibrium
Hardy-Weinberg Equilibrium (HWE) is a fundamental principle in population genetics. In
1908, an English mathematician G.H. Hardy and a German physician W. Weinberg individu-
ally described the principle to explain why dominant traits do not automatically replace
recessive traits in a population.61,62 HWE states that the genotype and allele frequencies
of a large, randomly mating population remain constant from generation to generation un-
less factors that disrupt the equilibrium have occurred. Because pharmacogenomic studies
often involve a large number of subjects with a certain phenotype who are genotyped for a
particular locus, understanding the concept of HWE and the potential causes of departure
from HWE is very important.
Suppose we start with a single genetic locus with two alleles represented by A and a.
Let the frequencies of the A and a alleles in a population be p and q, respectively. Because
there are only two alleles in the population, the sum of their frequencies is 1 (p + q = 1).
In addition, suppose mating occurs in the population independent of genotypes (random
mating). Then, three genotypes will be in the second generation (Table 2-2): AA, Aa, and
aa. The frequencies of each of the genotypes in the second generation will be p2, 2pq, and
q2 and p2 + 2pq + q2 =1.
If random mating occurs among the AA, Aa, and aa genotypes in the second genera-
tion (Table 2-3), the resultant genotypes and their frequencies (in parenthesis) in the next
generation will be as follows:
AA × AA = AA (p4)
AA × Aa = AA (2p3q) and Aa (2p3q)
AA × aa = Aa (2p2q2)
Aa × Aa = AA (p2q2), Aa (2p2q2) and aa (p2q2)

Aa × aa= Aa (2pq3) and aa (2pq3)
aa × aa = aa (q4)
Table 2-2
Genotype Frequencies in the Second Generationa,b
Maternal gametes
A (p) a (q)
Paternal gametes A (p) AA (p2) Aa (pq)

a (q) Aa (pq) aa (q2)
a
Parentheses are the frequencies of alleles or genotypes.
b
Genotype frequency: AA = p2; Aa = 2pq; aa = q2.
Table 2-3
Frequency of Type of Mating in the Third Generationa
Maternal genotypes
AA (p2) Aa (2pq) aa (q2)
Paternal genotypes AA (p2) AA × AA AA × Aa AA × aa

(p4) (2p3q) (p2q2)
Aa (2pq) Aa × AA Aa × Aa Aa × aa
(2p3q) (4p2q2) (2pq3)
aa (q2) aa × AA aa × Aa aa × aa
(p2q2) (2pq3) (q4)
a
Parentheses are the frequencies of the genotypes.
As a result, the frequencies of the AA, Aa, and aa genotypes in the third generation are as
noted below:
AA = p4 + 2p3q + p2q2 = p2 (p2 + 2pq + q2) = p2 (p + q)2 = p2 × 1 = p2

Aa = 2p3q + 4p2p2 + 2pq3 = 2pq (p2 + 2pq + q2) = 2pq (p + q)2 = 2pq
aa = p2q2 + 2pq3 + q4 = q2 (p2 + 2pq + q2) = q2 (p + q)2 = q2
From our original equation p2 + 2pq + q2 = 1, we know the frequency of each genotype
remains constant and stable over successive generations. A population with a stable geno-
type frequency is said to be in HWE.
Calculation of Genotype and Allele Frequencies Using HWE

HWE can be used to determine genotype frequencies from allele frequency data and/or
allele frequencies from genotype frequency data in a study population. For example, cyto-
chrome P450 2C9*2 (CYP2C9*2) is a C to T change at nucleotide position 430 in CYP2C9
gene. The frequencies of the C and T alleles in Caucasians are 87% and 13%, respectively.63
According to HWE, it is estimated that 76% (= 0.87 × 0.87 × 100) of Caucasians carry CC; 22%
(= 2 × 0.87 × 0.13 × 100) are CT; and 2% (= 0.13 × 0.13 × 100) have TT. β1-adrenergic receptor
gene (ADRB1) contains an arginine to glycine change at codon 389. Thirty-four percent of
the African Americans are homozygous for the arginine allele at codon 389 in ADRB1.64
What are the frequencies of the arginine and glycine alleles in this population? Let the
frequencies of the arginine and glycine alleles be p and q, respectively. According to HWE,
p2 = 0.34. Therefore, p = √0.34 = 0.58. Since p + q = 1, q = 1 – 0.58 = 0.42. Thus, 58% and
42% of African Americans carry arginine and glycine alleles, respectively. In addition, 49%
(= 2 × 0.58 × 0.42 × 100) are arginine/glycine heterozygotes, and 17% (= 0.42 × 0.42 × 100) are
glycine/glycine homozygotes.
HWE in Pharmacogenomic Studies
If pharmacogenomic studies assess multiple genetic polymorphisms, HWE should be test-
ed for each polymorphism. Departure from HWE indicates that experimental errors and/or
factors that disrupt HWE may have influenced the study (see below). When applied to a
cohort study, HWE should be tested in an entire study population. On the other hand, HWE
should be tested only in the control population in a case-control study because departure
from HWE is expected among cases if a genetic polymorphism is associated with a phe-
notype.65 Since HWE assumes random mating independent of genotypes, Pearson’s chi-
squared (X2) test for independence is commonly used to evaluate the equilibrium.66 This
test compares the observed genotype count in the study population with the expected
count under HWE. In the example in Table 2-4, X2 test statistics = Σ (observed – expected)2/
expected = 1.194 < 3.841, the critical value of X2degree of freedom = 1 at α = 0.05.
As a result, the genotype frequency in the study population is in HWE. The X2 test can
produce a false-positive result when the study has a small size and/or a low minor allele
frequency since the test assumes an asymptotic distribution of genotypes in the popula-
tion.66,67 Thus, an exact test should be used in studies with a small sample size and/or a low
minor allele frequency. Alternative exact tests of HWE have been developed, and computer
software codes for an exact test of HWE are freely available (http://www.sph.umich.edu/
csg/abecasis/Exact/index.html).68
Table 2-4
Pearson’s C 2-Test of the Genotype Data for Hardy-Weinberg
Equilibriuma
AA Aa aa Total
Observed Count 220 65 8 293

Genotype 0.75 0.22 0.03
Allele frequency A: 0.86
a: 0.14
Expected Genotype 0.74 0.24 0.02

(= 0.86 × 0.86) (= 2 × 0.86 × 0.14) (= 0.14 × 0.14)
Count 216.8 70.3 5.9
(= 0.74 × 293) (= 0.24 × 293) (= 0.02 × 293)
a
∑ (observed-expected)2/expected = (220 - 216.8)2/216.8 + (65 - 70.3)2/70.3 + (8 - 5.9)2/5.9 = 0.047 + 0.400 + 0.747 = 1.194.
Factors Disrupting HWE

HWE can be disrupted by biological (or evolutionary) and experimental factors. Biologi-
cal factors include nonrandom mating, migration, genetic drift, founder effect, mutation,
and natural selection.69 Nonrandom genotyping error and missing genotype data are the
main experimental causes.65 In human population, members of a particular subpopulation
(social, ethnic) commonly mate with each other (nonrandom mating).69 Inbreeding or mat-
ing between close relatives increases homozygosity for all genes. Assortative mating or
mating between individuals who have a similar or a dissimilar phenotype can also change
frequencies of homozygotes.69 Mating with a similar phenotype increases homozygosity for
genes involved in its expression, whereas mating with a dissimilar phenotype decreases the
homozygosity. Migration also influences genotype frequencies in a population.69,70
When a small population with a certain recessive genotype moves to a geographical
region and becomes an isolate, it will have a higher frequency of the genotype. Genetic
drift or allelic drift is a process in which genotype frequencies change from one generation
to the next due to chance.70 The effect is greater in small populations, which are also more
susceptible to founder effect. If founders of a population fail to pass a genetic allele on
to the next generation, only the alternative alleles will be found in the successive genera-
tions.69,70 If the founders pass a rare genetic allele to the next generation, its frequency
increases in the successive generations. As a result, the founder effect changes the fre-
quency of a certain rare genotype in a genetic isolate. Mutation, a change in the genetic
material, occurs at a rate of 10-6 to 10-4 mutations per locus per gamete per generation.69,70
At equilibrium, the genotype frequency is a balance between the rates of the introduction
of new alleles by mutation and of the removal of the mutated alleles by negative selection.
Environmental factors such as radiation and chemicals that change the mutation rate
can influence the balance. Biological fitness may differ by phenotype (hence genotype). A
phenotype with low biological fitness has negative selection pressure, which will reduce
the frequency of the genotype in the successive generations.68,69 In contrast, a positive se-
lection pressure will increase the frequency of the phenotype with high biological fitness.
As a result, natural selection influences genotype frequencies in the population.
Nonrandom genotyping errors involve a systematic genotyping error, which disrupts
HWE by misclassifying a particular genotype.65 A preferentially missing genotype or al-
lele can also break HWE. In addition, nonrandom genotyping error and missing data may
lead to a spurious genotype−phenotype association; therefore, pharmacogenomic stud-
ies should have certain measures of quality control of genotyping assay to minimize the
experimental errors.
Testing Genotype Data for Deviations from HWE
Testing of genotype data for deviations from HWE is conducted to identify potential geno-
typing errors. In general, genotypes that are not in HWE are excluded from genotype–phe-
notype association analyses.65 Often it is difficult to distinguish whether the departure from
HWE is due to systematic experimental errors, violations of the assumptions of HWE, or a
true association with a phenotype. A thorough examination of data is a key to identifying
the cause. For example, data indicating heterozygotes are disproportionately missing relative
to homozygotes may suggest a systematic experimental error. One genetic marker—among
the set of markers in strong linkage disequilibrium—that does not follow HWE may also
imply a systematic experimental error. On the other hand, a consistent pattern of departure
from HWE among the set of genetic markers in strong linkage disequilibrium may suggest
violations of the HWE assumptions. Departure from HWE due to violations of the assump-
tions may be replicated in an independent cohort, while that due to chance may not.
HAPLOTYPE
Haplotype is defined as a set of alleles or polymorphisms on a chromosome that tend

to be inherited together.69 Just as a genotype refers to a set of two alleles inherited for a
particular gene, a diplotype refers to a set of two haplotypes. Suppose that a pair of ho-
mologous chromosomes has three polymorphic loci (C or A at locus 1, T or G at locus 2, and
C or G at locus 3) (Figure 2-5). Because sets of polymorphisms can be tightly linked and
inherited together, fewer than the theoretically possible 8 (23) combinations may be found
to exist. Thus instead of each of the combinations (CTC, CGC, CTG, CGG, ATC, AGC, ATG,
AGG), only two (CTC and AGG) may be found to exist in the human population.
In Figure 2-5, suppose that the dark gray and the light gray chromosomes are passed
down from the person’s mother and father, respectively. The genotypes in Figure 2-5 are
CA at locus 1, TG at locus 2, and CG at locus 3. On the other hand, the haplotypes—the
multiple alleles on the same chromosome—are CTC (on the dark gray chromosome) and
AGG (on the light gray chromosome). Individuals who are homozygous for C at locus 1, T at
locus 2, and C at locus 3 carry two copies of CTC haplotype and 0 copy of AGG haplotype.
Therefore, the maximum number of copies of a particular haplotype an individual can carry
is 2 and the minimum number is 0.
Haplotypes can allow us to obtain information on local linkage disequilibrium and
historical recombination events (see below).71,72 Haplotype information can be obtained in
two ways. Molecular haplotyping involves direct sequencing of multiple polymorphic loci
on a chromosome. This method produces the most accurate information about haplotypes;
however, it is often laborious and expensive as well as technically challenging, particularly
if the genetic loci are far apart.66 Statistical haplotyping, such as an accelerated expecta-
tion maximization method, is more widely used to infer haplotypes in a study population.66
It is based on the fact that the human genome consists of many genomic regions with a rel-
Gene loci
C T C
A G G
Locus 1 Locus 2 Locus 3

Genotype: CA, TG, CG
Haplotype: CTC/AGG
FIGURE 2-5 The concept of haplotype.
atively small number of haplotypes due to strong linkage disequilibrium (see below). Various
computer software programs such as PHASE, SNPHAP, and FASTPHASE have been devel-
oped to infer haplotypes. These programs are accurate in inferring haplotypes, particularly
when they are used in data with high genomic marker density and few missing genotypes.73
CLINICAL PEARL
Haplotypes can allow us to obtain information on past recombination
events and local linkage disequilibrium.
LINKAGE DISEQUILIBRIUM
Linkage disequilibrium is the nonrandom association of alleles at different sites.69 Alleles
in linkage disequilibrium are tightly linked together; as a result, they are more likely found
together in a population. Linkage disequilibrium is a key concept in pharmacogenomic as-
sociation studies, and it is used to select not only genomic markers but also to analyze
genomic data in the association studies. It is also important in GWAS, which survey up to
a million SNPs to associate a phenotype. As a basis of linkage disequilibrium, we will first
review the recombination process during meiosis.
Recombination Process
During meiosis, homologous chromosomes are paired together; subsequently, some of
the genetic materials are exchanged between the pair. This process is called crossover
or recombination.76 Alleles close to each other on a chromosome are more likely to be
passed along together, whereas alleles far apart are more likely to be regrouped during
this process. Consider two bi-allelic loci on two different chromosomes (Figures 2-6; A/a
and B/b on chromosome 1, and C/c and D/d on chromosome 2). Because the two loci in
Figure 2-6 (A) are close, the next generation will have only two haplotypes: AB and ab.
(Recall that a haplotype is a group of alleles on a chromosome.) In contrast, the two loci in
Figure 2-6 (B) may be regrouped during meiosis because they are far apart; as a result, the
next generation will have four haplotypes (CD, cD, Cd, and cd). The recombination process
is a basis of linkage disequilibrium and genetic diversity in human population. On average,
30–40 recombination events occur per chromosome during meiosis.76
Mathematical Expression of Linkage Disequilibrium

Mathematical formulas are helpful to understand the concept of linkage disequilibrium.
Consider two bi-allelic loci (A or a allele at one locus, and B or b allele at the other locus)
in a population. Assume the population is in HWE at both loci. Let PAB be the frequency of
AB haplotype in the population. In addition, let PA and PB be the frequencies of A and B
alleles. If the two loci are independent of each other, then PAB = PA × PB. If the two loci are
correlated with each other or in linkage disequilibrium, PAB ≠ PA × PB.
Tables 2-5 and 2-6 illustrate this concept. Both tables contain two bi-allelic loci with
four haplotypes (AB, Ab, aB, and ab). In Table 2-5, the frequency of the AB haplotype is 0.1
(PAB = 0.1), and it is the product of the frequencies of the two alleles, PA and PB (PA × PB = 0.2
× 0.5 = 0.1): PAB = PA × PB. In addition, PAb = PA × Pb, PaB = Pa × PB and Pab = Pa × Pb; as a result,
the two loci in Table 2-5 are not in linkage disequilibrium. In contrast, PAB ≠ PA × PB in Table
C c
A a
(A) B b (B)
D d
a a
A A C C c c
B B b b
D D d d
a C c
A
B b
D
d
a c c
A A a C C
B B b
b
d D d
D
FIGURE 2-6 Recombination process.
2-6 since PAB = 0.2, while PA × PB = 0.6 × 0.5 = 0.3. Also, PAb ≠ PA × Pb, PaB ≠ Pa × PB and Pab ≠
Pa × Pb; as a result, the two loci in Table 2-6 are in linkage disequilibrium.
Measures of Linkage Disequilibrium

Three parameters are commonly used to measure linkage disequilibrium in pharmaco-
genomic studies: D, D′, and r2. D is how much observed frequency of a haplotype differs
from its expected frequency, and D′ is a D value adjusted for allele frequency. An alterna-
tive parameter, r2, is a measure of how strongly the two variables are correlated.
Suppose we have two bi-allelic loci, A or a allele at locus 1 and B or b allele at locus 2.
PAB is the observed frequency of the haplotype AB formed by the two alleles A and B. D is
defined as D = PAB – PA × PB = PAb – PA × Pb = PaB – Pa × PB = Pab – Pa × Pb.77 If the two loci are
not in linkage equilibrium, D = 0 because PAB = PA × PB. In Table 2-5, the two loci are not in
linkage disequilibrium because D = 0 (PAB – [PA × PB] = 0.1 – (0.5 × 0.2) = 0). In Table 2-6, the
two loci are in linkage disequilibrium because D ≠ 0 (PAB – [PA × PB] = 0.2 – [0.6 × 0.5] = –0.1
or PAb – [PA × Pb] = 0.4 – [0.6 × 0.5] = 0.1).
Table 2-5
Linkage Disequilibrium Example 1a
Locus 2
Allele
B b Total
Locus 1 Allele A 0.1 0.4 0.5

a 0.1 0.4 0.5
Total 0.2 0.8 1.0
The numbers are the frequencies of haploytpes and alleles. For example, the frequencies of AB and Ab haplotypes are
a
0.1 and 0.4, respectively. The frequencies of A and B alleles are 0.5 and 0.2, respectively.
Table 2-6
Locus 2
Allele
B b Total
Locus 1 Allele A 0.2 0.4 0.6

a 0.3 0.1 0.4
Total 0.5 0.5 1.0
The numbers are the frequencies of haplotypes and alleles. For example, the frequencies of AB and Ab haplotypes are
a
0.2 and 0.4, respectively. The frequencies of A and B alleles are 0.6 and 0.5, respectively.
D intuitively explains the concept of linkage disequilibrium. However, its numerical

value does not quantify the strength of linkage disequilibrium since the value varies as al-
lele frequency changes. D′ is defined as D′ = D/❘D❘ max where ❘D❘max = min (PA × Pb, Pa × PB), if
D > 0 or min (PA × PB, Pa × Pb), if D < 0.77
D′ accounts for allele frequency because it is a normalized D value relative to the
maximum D value achievable given allele frequency.78 D′ ranges between −1 and 1. ❘D′❘ = 1
suggests the two SNPs have not been separated by recombination. In other words, ❘D′❘ = 1
indicates perfect linkage disequilibrium between the loci. ❘D′❘ <1 suggests the perfect linkage
disequilibrium has been disrupted by recombination. The two loci in Table 2-5 are not in
linkage disequilibrium because its D′ = 0. In Table 2-6, D′ = -0.1/min (0.3, 0.2) = −0.1/0.2 = −0.5;
as a result, the two loci in Table 2-6 have some degree of linkage disequilibrium. The use
of D′ has limitations. First, it is not clear how to interpret values of D′ between 0 and 1 (e.g.,
we are not sure how different the two D′ values, 0.4 and 0.6, are). In addition, sample size
of a study can influence the D′ value because a small sample size tends to overestimate
the value.66 As a result, D′ may not be a good measure for comparison between studies
with different sample sizes.
The parameter r2 is a statistical coefficient of determination. It is defined as r2 = D2/

(PA × Pa× PB × Pb).77 It measures the extent of correlation between a pair of variables. Values
of r2 ranges between 0 and 1, and a higher value indicates a higher degree of correlation
between the pair. A perfect correlation of r2 = 1 indicates a perfect linkage disequilibrium
between the loci. Because two bi-allelic loci can be perfectly correlated only if minor al-
lele frequencies are identical at both loci, r2 does not give information on recombination
between the loci. In Table 2-5, r2 = 0. In Table 2-6, r2 = (−0.1)2/(0.6 × 0.4 × 0.5 × 0.5) = 0.17. Al-
though D value in Table 2-6 suggests linkage disequilibrium, the r2 value indicates that the
degree of correlation between the two loci in the population is weak. The r2 value is used
to calculate the sample size required for a pharmacogenomic association study because
the sample size is inversely related to an r2 value given a fixed effect size.79
Suppose three polymorphic loci (locus 1, 2, and 3) have different r2 values: r2 = 0.5
between loci 1 and 2, and r2 = 1.0 between 1 and 3. If a study selects the locus 1 to cover
the locus 2, it requires twice as many subjects as a study that genotypes for the locus 1 to
represent the locus 3. As with D, the r2 value is influenced by allele frequency because its
calculation does not require normalization of allele frequency. Finally, the r2 value is more
useful for selecting the tagging SNP from a set of the potential SNPs because r2 represents
a degree of correlation between the SNPs.80
Which linkage disequilibrium measure should be used? Choice of a measure depends
primarily on the purpose of the study. In a pharmacogenomics association study, the r2 val-
ue is more useful because it helps calculate the sample size and selection of the tagging
SNP. In general, an r2 value between the two loci ≥0.8 is considered a strong correlation.80
CLINICAL PEARL
The r2 value is more often used to obtain tagging SNPs than other
measures of linkage disequilibrium.
Linkage Disequilibrium in Pharmacogenomic Association Studies

How is linkage disequilibrium used in a pharmacogenomic association study? Let’s consider
Figure 2-7. Suppose that the locus A in the gene harbors an SNP that causes variable drug
response by changing an amino acid in a protein. If the loci B and C are in strong linkage
disequilibrium with the locus A, then the SNPs at either the locus B or C can be chosen to
represent the locus A. A SNP that represents the other SNPs, such as the SNPs at the loci
B and C, is called a tagging SNP. The tagging SNP may not cause phenotype variability
among individuals. It is often hard to pinpoint a causal SNP that changes phenotype. Be-
cause tagging SNPs are highly correlated with a causal SNP, they can be used to identify
the causal SNP or to narrow down the genomic region where the causal SNPs might be
located.
For example, the population in Table 2-7 has only two haplotypes: AB and ab. If a
person carries the A allele, then the individual never carries the b allele but always the B
allele on the same chromosome. As a result, genotyping for either locus 1 or 2 instead of
both of the loci will obtain the allelic information of the other locus. Either one of the SNPs
can be used to tag the other SNP. On the other hand, neither of the SNPs in Table 2-6 can
be used as a tagging SNP because its population has four haplotypes (AB, Ab, aB, and ab).
FIGURE 2-7 Locus A is located in a gene while loci B and C are in nongenetic region. The
loci B and C are in linkage disequilibrium with locus A. Because of linkage disequilibrium,
genotyping at either locus B or C can give the genotype information on locus A. SNP refers to
single nucleotide polymorphism.
Table 2-7
Locus 2
Allele
B b Total
Locus 1 Allele A 0.7 0 0.7

a 0 0.3 0.3
Total 0.7 0.3 1.0
The numbers are the frequencies of haplotypes and alleles. For example, the frequencies of AB and Ab haplotypes are
a
0.7 and 0, respectively. The frequencies of A and B alleles are 0.7 and 0.7, respectively.
Note the SNPs in Table 2-6 are in weak linkage disequilibrium since D = 0.1, D′ = 0.5, and r2
= 0.17. In contrast, the SNPs in Table 2-7 are in strong linkage disequilibrium:
D = PAB – (PA × PB) = 0.7 – (0.7 × 0.7) = 0.21

D′ = D/ ❘D❘max = 0.21/min ([0.7 × 0.3], [0.3 × 0.7]) = 0.21/0.21 = 1
r2 = D2/(PA × Pa × PB × Pb) = (0.21)2/(0.7 × 0.3 × 0.7 × 0.3) = 1
The r2 value is ≥0.8 in Table 2-7, suggesting a strong correlation. These examples show that
a pharmacogenomic association study can be more efficiently conducted if the SNPs in the
study are in strong linkage disequilibrium.
CLINICAL PEARL
Linkage disequilibrium provides a tool that can be used for genetic
association studies, including GWAS.
USE OF LINKAGE DISEQUILIBRIUM AND HAPLOTYPES IN

CLINICAL PRACTICE
Because many polymorphisms are linked to each other, it may not always be necessary
to obtain information about every single polymorphism when conducting clinical tests
or GWAS. When multiple polymorphisms are inherited together, only one or few of all
polymorphisms inherited together need to be analyzed for clinical or research studies.
Moreover, the haplotype rather than the multiple genotypes may be used to describe the
polymorphisms. The vitamin K epoxide reductase complex subunit 1 (VKORC1) gene, for
example, encodes for a target protein of the anticoagulant warfarin. Although the gene
contains several nonsynonymous SNPs, their frequencies are low (minor allele frequency
<5%) in the population.74 The gene also includes about 30 SNPs in the noncoding regions.
Ten noncoding SNPs with a minor allele frequency ≥5% have been studied for an as-
sociation with interindividual variability in warfarin dose requirements.74 SNPs with a minor
allele frequency <5% are often excluded in a pharmacogenomic study because they usu-
ally give inadequate power to the study and may be clinically unimportant.75 Five of the
10 SNPs are tightly linked together and form two common haplotypes that have been
identified as a genetic factor, which influences the interindividual variability in warfarin
dose requirements (Table 2-8): C at −4931 always occurs with A at −1639, T at 1173, C at
1542, and T at 2255.74 Likewise, T is always associated with G at −1639, C at 1173, G at 1542,
and C at 2255. As such, all of the five SNPs need not be genotyped to predict a patient’s
warfarin dose. Instead, only one locus can be chosen to infer alleles at the other four loci.
For example, if a patient has C at 1173, the other alleles should be T at −4931, G at −1639, G
at 1542, and C at 2255. Thus, the use of haplotypes can decrease the number of SNPs to
be genotyped for clinical practice.
CLINICAL PEARL
Tagging a single SNP within the haplotype can be used to predict drug
therapy outcomes: VKORC1 −1639G/A is a tag SNP to predict warfarin
dose.
Table 2-8
Five Noncoding SNPs in VKORC1 Gene Associated with Interindividual
Variability in Warfarin Dose Requirementsa
Locus –4931 –1639 1173 1542 2255
SNPs C A T C T
T G C G C
a
Because of strong linkage disequilibrium, SNPs in the first row (CATCT) occur together and SNPs in the second row
(TGCGC) occur together.
INTERNATIONAL HAPMAP AND 1000 GENOMES PROJECTS

International HapMap Project was a multinational study to identify and catalog human ge-
netic diversities.81 In this project, which started in 2002, DNA samples from four representa-
tive populations (Yoruba in Ibadan, Nigeria; Caucasians in Utah, United States; Han Chinese
in Beijing, China; and Japanese in Tokyo, Japan) were genotyped to create a haplotype map
of the human genome. One million SNPs from 270 samples were genotyped in the first
phase of the project, and 3 million SNPs from 270 samples were genotyped in the project’s
second phase. The project’s third phase involved genotyping of 1.4 million SNPs from 1,301
samples obtained from 11 different populations.82 The data from this project are publicly
available at www.hapmap.org.
According to the HapMap data, most recombination events have occurred at short ge-
nomic regions called hotspots.81 Linkage disequilibrium often breaks down around hotspots
due to their high recombination rate; as a result, the human genome has discontinuous
linkage disequilibrium patterns and can be considered a block-like structure of linkage
disequilibrium. Each block has limited numbers of common haplotypes due to its strong
linkage disequilibrium. In fact, the average number of the common haplotypes in a block
is four to five, even if each block contains 30–70 SNPs.81 In addition, only a limited number
of tagging SNPs are required to obtain information about genetic variations in a block. The
strong linkage disequilibrium within a block can allow us to select only a few of the several
SNPs in a block for carrying out genetic association studies.
The information obtained from the haplotype map was critical for the design and
success of GWAS (see below). However, due to technological limitations, the International
HapMap project focused on relatively more common DNA variants. Due to advances in
sequencing technologies, the 1000 Genomes Project was initiated in 2008. The project’s
goal is to sequence around 2,500 samples and obtain information for most rare variants
with minor allele frequencies (MAF) >1%. The International HapMap and 1000 Genomes
Projects have helped in advancing genotyping and next-generation sequencing platforms,
in evaluating studies involving population structure, and in choosing tagging SNPs for con-
ducting GWAS.82
GENOME-WIDE ASSOCIATION STUDIES

Candidate gene and GWAS are utilized to correlate alleles that are associated with a spe-
cific trait. In the candidate gene approach, a gene that is known to be related to a particular
trait in some manner is selected and studied. Following selection of genes that may be as-
sociated with the trait, polymorphisms in these genes that may be responsible for the phe-
notype are identified. A study is then conducted between subjects who exhibit a particular
trait and a control group that does not exhibit the trait to test if a particular polymorphism
is indeed associated with the phenotype.83
Unlike candidate gene association studies, GWAS do not require selection of genes
that may be related to a specific trait prior to the study. The human genome is estimated
to have about 12 million common SNPs (MAF >5%). GWAS survey these SNPs to correlate
a drug response (phenotype) to a genetic variant. Genotyping of all of these SNPs is tech-
nically feasible but laborious and time consuming. GWAS genotype 100,000 to 1,000,000
SNPs in the entire human genome.75 Even if these numbers are smaller than the total
number of the common SNPs, the information obtained from the tagging SNPs would be
similar to that from all the common SNPs; as a result, GWAS are efficient.
GWAS can provide new biological information by allowing the identification of genes
involved in drug response phenotypes. For example, one study identified a genomic marker
that is associated with statin-induced myopathy (phenotype).84 The study, which enrolled
85 cases of statin-induced myopathy and 80 controls (no statin-induced myopathy), geno-
typed over 300,000 SNPs to associate a genetic marker with the risk of the phenotype.
An SNP, rs4363657, located in the solute carrier organic anion transporter family member
1B1 gene (SLCO1B1), was found to be strongly associated with the phenotype. The SLCO1B1
gene encodes for organic anion transporting polypeptide, which mediates cellular uptake
of various drugs, including statins. Because the SNP was located in the intron, the exons
of the SLCO1B1 gene were re-sequenced to discover SNPs, which were in strong linkage
disequilibrium with rs4363657, and thus might influence the functions of the protein by
changing an amino acid.
The rs4363657 SNP was found to be in strong linkage disequilibrium (r2 >0.95) with a
nonsynonymous SNP, rs4149056, which changes valine to alanine at codon 174. This dis-
covery may lead to better understanding of pathogenesis of the statin-induced myopathy.
CLINICAL PEARL
GWAS can help establish relations between genetic variants and specific
traits, without any prior knowledge of genes that may be related to the trait.
This can also help discover new functions for a gene.
LINKAGE DISEQUILIBRIUM PLOT

A linkage disequilibrium plot is often used to visualize pairwise linkage disequilibrium be-
tween SNPs. In addition, it helps identify haplotype blocks in a genomic region. Computer
software such as Haploview are employed to construct a linkage disequilibrium plot from
genotype data.85 Figures 2-8A-C are linkage disequilibrium plots in β2-adrenergic receptor
(ADRB2) gene region in Caucasians, Yoruba Africans, and Chinese/Japanese. The Cauca-
sians have seven common SNPs with minor allele frequency >5% and two haplotype blocks
in the region (Figure 2-8A). Haplotype block 1 includes rs2400707, rs12654778, rs1042713,
rs1042714, and rs1042717; block 2 consists of rs1042718 and rs1042719. Linkage disequilib-
rium plots can display values of the various measures of linkage disequilibrium. In Figures
2-8A through 2-8C, r2 values (shown as r2) are used to select tagging SNPs. The two
SNPs (rs2400707 and rs1042714) have a strong correlation because their r2 value is ≥80%.
As a result of this strong correlation (one, between rs2400707 and rs1042714; and two,
between rs12654778 and rs1042713) in the haplotype block 1, only three SNPs (rs1042717,
either rs2400707 or rs1042714, and either rs12654778 or rs1042713) are required to cover
the block. Because the two SNPs (rs1042718 and rs1042719) in the haplotype block 2 have
r2 <80%, both of them may need to be genotyped. However, due to a strong correlation
between rs1042717 and rs1042718 (r2 = 80%), only one of them needs to be selected. As a
result, the total number of the tagging SNPs in the Caucasian population is four.
In contrast to Caucasians, Yoruba Africans have 12 common SNPs with three haplotype
blocks in the region (Figure 2-8B). Because of strong correlations between some of the
SNPs, information on the ADRB2 variations can be obtained with only six SNPs (rs2053044,
rs12654778, rs1042711, rs1042713, rs1042718, and rs3729943).
rs12654778
rs1042713
rs1042714
rs1042717
rs1042718
rs2400707
rs1042719
Block 1 (1 kb) Block 2 (0 kb)

1 3 4 5 6 8 10
42 92 37 23 80 57
43 36 12 16 63
96 14 11 27
22 12 0
15 0
26
FIGURE 2-8A The linkage disequilibrium plot is obtained from single nucleotide
polymorphisms (SNPs) genotyped in the HapMap Project. The plot is generated by
Haploview. The bars in the long white box in the upper part of the figure indicate the
locations of the seven SNPs with minor allele frequency >5% in the Caucasian population.
Note the region contains two haplotype blocks: block 1 (SNPs 1, 3, 4, 5, and 6) and block 2
(SNPs 8 and 10). The diamond boxes at the intersection of two SNPs indicate the r2 values for
that pair. The darker the diamond, the higher the r2 value.
The Chinese/Japanese populations have two haplotype blocks with 11 common SNPs
in ADRB2 (Figure 2-8C). Linkage disequilibrium data suggest the region requires eight tag-
ging SNPs (rs2053044, rs17108803, rs12654778, rs1042711, rs1042713, rs1042714, rs1042717,
and rs1042719).
GENETIC DIFFERENCES AMONG RACIAL AND ETHNIC GROUPS

Study populations with different ancestral history should be considered when pharma-
cogenomic studies are designed and interpreted as evidenced by the linkage disequilibrium
plots of the three distinct populations for ADRB2. First, the number of the common SNPs
in a genomic region may differ by study population. In ADRB2, there are 7, 12, and 11 SNPs
in Caucasians, Yoruba Africans, and Chinese/Japanese, respectively. The number of genetic
variations depends on when a population was formed. A population with a longer history
has more opportunities to develop new genetic variations through mutation and recombi-
nation. Since African-descent populations were formed prior to the establishment of Euro-
pean or Asian-descent populations, they have the largest number of genetic variations.70 In
addition, African-descent populations have shorter haplotype blocks and more haplotypes
rs2400707
rs2053044
rs12654778
rs11168070
rs1042711
rs11168070
rs1042713
rs1042714
rs1042717
rs1042718
rs3729943
rs1042719
Block 1 (0 kb) Block 2 (0 kb) Block 3 (1 kb)

1 2 4 5 6 7 8 9 10 12 13 14
81 18 5 91 100 14 19 10 100 9 9
21 6 4 86 13 80 49 11 9 96
24 8 4 19 84 8 48 0 96
18 7 28 100 8 8 4 11
18 21 5 10 8 0 45
5 4 13 10 0 8
25 33 13 0 8
44 32 1 11
45 3 14
4 28
41
FIGURE 2-8B The linkage disequilibrium plot is obtained from single nucleotide poly-
morphisms (SNPs) genotyped in the HapMap Project. The plot is generated by Haploview.
The bars in the long white box in the upper part of the figure indicate the locations of the 12
SNPs with minor allele frequency >5% in the Yoruba population. Note the region contains
three haplotype blocks: block 1 (SNPs 1, 2, and 4), block 2 (SNPs 5, 6, and 7), and block 3
(SNPs 8, 9, 10, 12, 13, and 14). The diamond boxes at the intersection of two SNPs indicate the
r2 values for that pair. The darker the diamond, the higher the r2 value.
within a block than the other populations because of their higher genetic variability.71,81
Populations living in close geographic regions have a similar pattern of genetic varia-
tions compared with those living far away from each other.70 For example, Japanese and
Chinese populations are genetically closer than Japanese and Europeans. Japanese and
Chinese have similar frequencies of the minor allele (G) of rs1042714 (8 and 12%), while
those in Caucasians and Yoruba Africans are 47% and 18%, respectively. Because study
populations with different ancestral history may have differences in allele frequency, it
should be checked for each population studied.
Linkage disequilibrium patterns may also differ by study population. In ADRB2, for
example, Yoruba Africans have three haplotype blocks while Caucasians and Chinese/
Japanese have two blocks. In addition, the structures of the haplotype blocks are different
between the Caucasian and the Chinese/Japanese populations despite having the same
number of the blocks. Finally, study populations may have distinct degrees of pairwise
linkage disequilibrium. For example, Caucasians have strong linkage disequilibrium between
rs1042714 and rs2400707 (r2 = 96%); in contrast, it is weak in Yoruba Africans and Chinese/
Japanese (r2 = 25% and 35%).
rs2400707
rs2053044
rs17108803
rs12654778
rs11168070
rs1042711
rs1042713
rs1042714
rs1042717
rs1042718
rs1042719
Block 1 (0 kb) Block 2 (1 kb)

1 2 1 4 5 6 8 9 10 12 14
80 2 4 6 57 5 9 8 93 72
2 13 0 3 9 57 60 8 71
18 8 0 49 100 4 61 8
35 23 7 6 8 4 43
19 6 0 34 8 5
3 8 11 32 8
35 19 11 15
22 21 8
24 25
27
FIGURE 2-8C The linkage disequilibrium plot is obtained from single nucleotide
polymorphisms (SNPs) genotyped in the HapMap Project. The plot is generated by
Haploview. The bars in the long white box in the upper part of the figure indicate the
locations of the 11 SNPs with minor allele frequency >5% in the Chinese/Japanese
population. Note the region contains two haplotype blocks: block 1 (SNPs 1 and 2) and block
2 (SNPs 4, 5, 6, 8, 9, 10, and 12). The diamond boxes at the intersection of two SNPs indicate
the r2 values for that pair. The darker the diamond, the higher the r2 value.
The frequencies of common ADRB2 haplotypes, defined as those with a frequency

>5% in a given study population, can be calculated using an accelerated expectation maxi-
mization algorithm (Tables 2-9A-C).85 Caucasians have four common haplotypes (Table
2-9A). Because of the eight common SNPs, Caucasians can have up to 256 (= 28) potential
haplotypes; however, the four common haplotypes account for about 93% of the total
haplotype variability in Caucasians. Significant portions of the total variability in Yoruba
Africans and Chinese/Japanese are also explained by the small number of the common
haplotypes (Table 2-9B and 2-9C). The frequencies of the haplotype that occur most
often are 0.442, 0.291, and 0.337 in Caucasians, Yoruba Africans, and Chinese/Japanese,
respectively. Because African-descent populations have higher genetic variability than the
other populations, Yoruba Africans have the lowest frequency of the haplotype (0.291). The
higher genetic variability in African-descent population also explains the lowest contribu-
tion (87%) of the total haplotype variability by the common haplotypes in Yoruba Africans.
Table 2-9A
Common ADRB2 Haplotypes in Caucasiansa,b

Haplotype block # 1 2
SNP # 1 3 4 5 6 8 10 Frequency
Haplotype A G G G G C G 0.442
G A A C G C G 0.240
G G G C A A C 0.175
G A A C G C C 0.069
a
SNP #1, rs2400707; SNP #3, rs12654778; SNP #4, rs1042713; SNP #5, rs1042714; SNP #6, rs1042717; SNP #8,
rs1042718; SNP #10, rs1042719.
b
rs numbers are reference SNP accession IDs in dbSNP (single nucleotide polymorphism database in National Center
for Biotechnology Information). Genotype data are obtained from www.HapMap.org. AGGGGCG is a haplotype with
a frequency of 0.442. Haplotypes are inferred by an accelerated expectation maximization method on Haploview
software.
Table 2-9B
Common ADRB2 Haplotypes in Yoruba Africansa,b
Haplotype block # 1 2 3 Frequency
SNP # 1 2 4 5 6 7 8 9 10 12 13 14
Haplotype G G G C T T G C A A C C 0.291
A A G C T T A C G C C G 0.267
G G A C T T A C G C C G 0.192
A A G G C C G G G C C G 0.122
a
SNP #1, rs2400707; SNP #2, rs2053044; SNP #4, rs12654778; SNP #5, rs11168070; SNP #6, rs1042711; SNP
#7, rs1801704; SNP #8, rs1042713; SNP #9, rs1042714; SNP #10, rs1042717; SNP #12, rs1042718; SNP #13,
rs3729943; SNP #14, rs1042719.
b
for Biotechnology Information). Genotype data are obtained from www.HapMap.org. Haplotypes are inferred by an
accelerated expectation maximization method on Haploview software.
Table 2-9C
Common ADRB2 Haplotypes in Chinese/Japanesea,b
Haplotype block # 1 2
SNP # 1 2 3 4 5 6 8 9 10 12 14 Frequency
Haplotype G G T G C T G C A A C 0.337
G G T A C T A C G C G 0.253
A A T G C T A C G C G 0.139
G G G G C T G C A A C 0.079
G G T A C T A C G C C 0.057
A A T G G C G G G C G 0.056
a
SNP #1, rs2400707; SNP #2, rs2053044; SNP #3, rs17108803; SNP #4, rs12654778; SNP #5, rs11168070; SNP
#, rs1042711; SNP #8, rs1042713; SNP #9, rs1042714; SNP #10, rs1042717; SNP #12, rs1042718; SNP #14,
rs1042719.
b
for Biotechnology Information). Genotype data are obtained from www.HapMap.org. Haplotypes are inferred by an
accelerated expectation maximization method on Haploview software.
CLINICAL PEARL
African-descent populations are genetically more diverse than European-
descent and Asian-descent populations; African-descent populations
may need more SNPs to predict drug response compared with the other
populations.
Population Stratification and Admixture

Pharmacogenomic association studies are often carried out as a case-control study. Cases
are those that exhibit a particular drug response phenotype, and controls are the patients
who do not exhibit the particular phenotype. This design aims to discover the differences
in allele frequency between cases and controls. Population stratification and admixture
should be considered as potential confounders of these studies.86
A study involving two or more unevenly distributed populations with differences in
allele frequency between cases and controls can produce a false association. There is
population stratification when differences in the allele frequency between cases and
controls are not related to the phenotype but are attributable to the uneven distribution
of the background populations.87 For example, a study found that a certain haplotype in
human leukocyte antigen gene was associated with the risk of type II diabetes mellitus in
Pima Indians. Because the haplotype has been more common in Caucasians, the data were
reanalyzed according to Pima Indian ancestry. After adjustment for Pima Indian ancestry, it
was found not to be associated with an increased risk for the disease.88
Population admixture can also confound pharmacogenomic studies. When two popu-
lations are mixed, allele frequency in an admixed population will be changed. Comparing
such populations for a phenotype difference may lead to a spurious association. Hispanics,
for example, are an admixture of Caucasians, Africans, and Native Americans. Ancestral
history in Hispanics differs by geographic region. Puerto Ricans are mainly an admixture
of Caucasians and Africans, whereas Mexicans are of Caucasians and Native Americans. As
a result, comparison of Puerto Ricans with Mexicans may produce a false association be-
tween an allele and a phenotype. A spurious association can also be found in a seemingly
homogenous population if the population has genetically distinct subpopulations due to

nonrandom mating.86
Careful matching strategies and sampling can minimize the confounding effects of
population stratification and admixture in a pharmacogenomic case-control study.87 In ad-
dition, the difference in population structure can be adjusted by using anonymous genetic
markers and/or ancestral information markers.87,88 A set of anonymous genetic markers
scattered throughout the genome can be selected as an indicator of the amount of diver-
sity in cases and controls. Ancestral information markers are genetic markers, often SNPs,
which help identify a distinct population (e.g., Europeans, Africans, Asians) because these
markers have differences in their frequencies among populations.89 The ancestral informa-
tion markers may be useful in distinguishing complex admixed populations, such as His-
panic populations, in pharmacogenomics studies.
CLINICAL PEARL
Population structure and admixture can produce false associations
between genes and drug responses.
SUMMARY
The field of pharmacogenomics has burgeoned in recent years due to rapid advances in
molecular biology and expanded knowledge of human genomics. As pharmacogenomics
begins to impact clinical practice, a fundamental understanding of human genetics will
be essential for applying pharmacogenomics and interpreting the results of pharmaco-
genomic testing data on patients. The frequency of alleles and population studies in ge-
netic association studies must be analyzed to validate the results. The fact that multiple
genetic variations can impact the actions of a single drug, as in the case of warfarin, adds
a layer of complexity to our understanding of the impact of pharmacogenomics in clini-
cal practice. Although pharmacists and clinicians do not need to be geneticists to benefit
from the application of pharmacogenomics, their knowledge of human genomics and the
fundamental principles of pharmacogenomics will need to be enhanced and updated on
a regular basis.
REFERENCES
1. Strachan T, Read AP. DNA structure and gene expression. In: Strachan T, Read AP, eds. Human Molec-
ular Genetics. 3rd ed. Oxford: Garland Science; 2003:4-32.
2. Kornberg RD. Eukaryotic transcriptional control. Trends Cell Biol. 1999;9:M46-M49.

3. Burgess-Beusse B, Farrell C, Gaszner M, et al. The insulation of genes from external enhancers and
silencing chromatin. Proc Natl Acad Sci USA. 2002;99(Suppl 4):16433-16437.
4. Goessling LS, Daniels-McQueen S, Bhattacharyya-Pakrasi M, et al. Enhanced degradation of the ferri-
tin repressor protein during induction of ferritin messenger RNA translation. Science. 1992;256:670-
673.
5. Langaee TY, Zineh I. Applied molecular and cellular biology. In: Pharmacogenomics: Applications to
Patient Care. Lenexa, KS: American College of Clinical Pharmacy (ACCP); 2004:53-116.
6. Dvir A, Conaway JW, Conaway RC. Mechanism of transcription initiation and promoter escape by RNA
polymerase II. Curr Opin Genet Dev. 2001;11:209-214.
7. Korzheva N, Mustaev A. Transcription elongation complex: structure and function. Curr Opin Microbiol.
2001;4:119-125.
8. Herbert A, Rich A. RNA processing and the evolution of eukaryotes. Nat Genet. 1999;21:265-269.
9. Kornblihtt AR. Chromatin, transcript elongation and alternative splicing. Nat Struct Mol Biol. 2006;13:5-
7.
10. Lander ES, Linton LM, Birren B, et al. Initial sequencing and analysis of the human genome. Nature.
2001;409:860-921.
11. Sachs AB, Sarnow P, Hentze MW. Starting at the beginning, middle, and end: translation initiation in
eukaryotes. Cell. 1997;89:831-838.
12. Nakamura Y, Ito K, Isaksson LA. Emerging understanding of translation termination. Cell. 1996;87:147-
150.
13. Breiling A, Lyko F. Epigenetic regulatory functions of DNA modifications: 5-methylcytosine and beyond.
Epigenetics Chromatin. 2015;8:24.
14. Rice JC, Allis CD. Histone methylation versus histone acetylation: new insights into epigenetic regula-
tion. Curr Opin Cell Biol. 2001;13(3):263-273.
15. Breaker RR. Riboswitches and the RNA world. Cold Spring Harb Perspect Biol. 2012 Feb 1;4(2).
pii: a003566.
16. He L, Hannon GJ. MicroRNAs: small RNAs with a big role in gene regulation. Nat Rev Genet.
2004;5(7):522-531.
17. Brookes AJ. The essence of SNPs. Gene. 1999;234:177-186.
18. Rasmuson M. The genotype-phenotype link. Hereditas. 2002;136:1-6.
19. den Dunnen JT, Antonarakis SE. Mutation nomenclature extensions and suggestions to describe
complex mutations: a discussion. Hum Mutat. 2000;15(1):7-12.
20. Sim SC, Ingelman-Sundberg M. The Human Cytochrome P450 (CYP) Allele Nomenclature website: a
peer-reviewed database of CYP variants and their associated effects. Hum Genomics. 2010;4(4):278-
281.
21. Kitts A, Phan L, Ward M, et al. The Database of Short Genetic Variation (dbSNP) 2013 Jun 30 [Updated
2014 Apr 3]. In: The NCBI Handbook [Internet]. 2nd ed. Bethesda, MD: National Center for Biotechnol-
ogy Information (US); 2013.
22. Sherry ST, Ward MH, Kholodov M, et al. dbSNP: the NCBI database of genetic variation. Nucleic Acids
Res. 2001;29:308-311.
23. Marsh S, Kwok P, McLeod HL. SNP databases and pharmacogenetics: great start, but a long way to go.
Hum Mutat. 2002;20:174-179.
24. Prokunina L, Alarcón-Riquelme ME. Regulatory SNPs in complex diseases: their identification and
functional validation. Expert Rev Mol Med. 2004;6:1-15.
25. Vignal A, Milan D, SanCristobal M, et al. A review on SNP and other types of molecular markers and
their use in animal genetics. Genet Sel Evol. 2002;34:275-305.
26. International Human Genome Sequencing C. Finishing the euchromatic sequence of the human
genome. Nature. 2004;431(7011):931-945.
27. Marian AJ. Sequencing your genome: what does it mean? Methodist Debakey Cardiovasc J.
2014;10(1):3-6.
28. Nackley AG, Shabalina SA, Tchivileva IE, et al. Human catechol-O-methyltransferase haplotypes
modulate protein expression by altering mRNA secondary structure. Science. 2006;314:1930-1933.
29. Wang D, Sadée W. Searching for polymorphisms that affect gene expression and mRNA processing:
example ABCB1 (MDR1). AAPS J. 2006;8:E515-E520.
30. Viguier J, Boige V, Miquel C, et al. ERCC1 codon 118 polymorphism is a predictive factor for the tumor
response to oxaliplatin/5-fluorouracil combination chemotherapy in patients with advanced colorec-
tal cancer. Clin Cancer Res. 2005;11:6212-6217.
31. Relling MV, Hancock ML, Rivera GK, et al. Mercaptopurine therapy intolerance and heterozygosity at
the thiopurine S-methyltransferase gene locus. J Natl Cancer Inst. 1999;91:2001-2008.
32. McLeod HL, Siva C. The thiopurine S-methyltransferase gene locus—implications for clinical pharma-
cogenomics. Pharmacogenomics. 2002;3:89-98.
33. Fuchshuber-Moraes M, Carvalho RS, Rimmbach C, et al. Aminoglycoside-induced suppression of

CYP2C19*3 premature stop codon. Pharmacogenet Genomics. 2011;21(11):694-700.
34. Prokunina L, Alarcón-Riquelme ME. Regulatory SNPs in complex diseases: their identification and
functional validation. Expert Rev Mol Med. 2004;6:1-15.
35. Marsh S. Pharmacogenomics. Ann Oncol. 2007;18(Suppl 9):ix24-ix28.
36. Roden DM, Altman RB, Benowitz NL, et al. Pharmacogenetics Research Network—pharmacogenomics:
challenges and opportunities. Ann Intern Med. 2006;145:749-757.

37. Knight JC, Udalova I, Hill AV, et al A polymorphism that affects OCT-1 binding to the TNF promoter
region is associated with severe malaria. Nat Genet. 1999;22:145-150.
38. Jacobson A, Peltz SW. Interrelationships of the pathways of mRNA decay and translation in eukary-
otic cells. Annu Rev Biochem. 1996;65:693-739.
39. Decker CJ, Parker R. Mechanisms of mRNA degradation in eukaryotes. Trends Biochem Sci. 1994;
19:336-340.
40. Goto Y, Yue L, Yokoi A, et al. A novel single-nucleotide polymorphism in the 3′-untranslated region of
the human dihydrofolate reductase gene with enhanced expression. Clin Cancer Res. 2001;7:1952-1956.
41. Moyer RA, Wang D, Papp AC, et al. Intronic polymorphisms affecting alternative splicing of human
dopamine D2 receptor are associated with cocaine abuse. Neuropsychopharmacology. 2011;36(4):753-
762.
42. Kuehl, P, Zhang J, Lin Y, et al. Sequence diversity in CYP3A promoters and characterization of the
genetic basis of polymorphic CYP3A5 expression. Nat Genet. 2001;27:383-391.
43. McGovern DP, Hysi P, Ahmad T, et al. Association between a complex insertion/deletion poly-
morphism in NOD1 (CARD4) and susceptibility to inflammatory bowel disease. Hum Mol Genet.
2005;14:1245-1250.
44. Kirchheiner J, Schmidt H, Tzvetkov M, et al. Pharmacokinetics of codeine and its metabolite morphine
in ultra-rapid metabolizers due to CYP2D6 duplication. Pharmacogenomics J. 2007;7:257-265.
45. Ingelman-Sundberg M. Genetic polymorphisms of cytochrome P450 2D6 (CYP2D6): clinical conse-
quences, evolutionary aspects and functional diversity. Pharmacogenomics J. 2005; 5:6–13.
46. Fukuda T, Maune H, Ikenaga Y, et al. Novel structure of the CYP2D6 gene that confuses genotyping
for the CYP2D6*5 allele. Drug Metab Pharmacokinet. 2005;20(5):345-350.
47. Samer CF, Lorenzini KI, Rollason V, et al. Applications of CYP450 testing in the clinical setting. Mol
Diagn Ther. 2013;17(3):165-184.
48. Ramel C. Mini- and microsatellites. Environ Health Perspect. 1997;105(suppl 4):781-789.
49. Liu X, Xu W. UGT1A1*28 polymorphisms: a potential pharmacological biomarker of irinotecan-based
chemotherapies in colorectal cancer. Pharmacogenomics. 2014;15(9):1171-1174.
50. Drazen JM, Yandava CN, Dubé L, et al. Pharmacogenetic association between ALOX5 promoter geno-
type and the response to anti-asthma treatment. Nat Genet. 1999 Jun;22:168-170.
51. Ingram VM. Sickle-cell anemia hemoglobin: the molecular biology of the first “molecular disease”—
the crucial importance of serendipity. Genetics. 2004;167:1-7.
52. Rolfini R, Cabrini G. Nonsense mutation R1162X of the cystic fibrosis transmembrane conductance
regulator gene does not reduce messenger RNA expression in nasal epithelial tissue. J Clin Invest.
1993;92:2683-2687.
53. Snell RG, MacMillan JC, Cheadle JP, et al. Relationship between trinucleotide repeat expansion and
phenotypic variation in Huntington′s disease. Nat Genet. 1993;4:393-397.
54. Crawford DC, Schwartz CE, Meadows KL, et al. Survey of the fragile X syndrome CGG repeat and the
short-tandem-repeat and single-nucleotide-polymorphism haplotypes in an African American popu-
lation. Am J Hum Genet. 2000;66:480-493.
55. Tayeb MT. Deletion mutations in Duchenne muscular dystrophy (DMD) in Western Saudi children.
Saudi J Biol Sci. 2010;17(3):237-240.
56. Korneev S, O’Shea M. Evolution of nitric oxide synthase regulatory genes by DNA inversion. Mol Biol
Evol. 2002;19:1228-1233.
57. Mitelman F, Johansson B, Mertens F. The impact of translocations and gene fusions on cancer causa-
tion. Nat Rev Cancer. 2007;7:233-245.
58. Neri A, Barriga F, Knowles DM, et al. Different regions of the immunoglobulin heavy-chain locus are
involved in chromosomal translocations in distinct pathogenetic forms of Burkitt lymphoma. Proc Natl
Acad Sci USA. 1988;85:2748-2752.
59. Korenberg JR, Chen XN, Schipper R, et al. Down syndrome phenotypes: the consequences of chromo-
somal imbalance. Proc Natl Acad Sci USA. 1994;91:4997-5001.
60. Hook EB, Warburton D. The distribution of chromosomal genotypes associated with Turner’s
syndrome: livebirth prevalence rates and evidence for diminished fetal mortality and severity in geno-
types associated with structural X abnormalities or mosaicism. Hum Genet. 1983;64:24-27.
61. Hardy GH. Mendelian proportions in a mixed population. Science. 1908;28:49-50.
62. Weinberg W. Über den nachweis der vererbung beim menschen. Jahreshefte des Vereins für vaterlän-
dische Naturkunde in Württemberg. 1908;64:368-382.
63. Gage BF, Lesko LJ. Pharmacogenetics of warfarin: Regulatory, scientific, and clinical issues. J Thromb
Thrombolysis. 2008;25:45-51.
64. Shin J, Johnson JA. Pharmacogenetics of beta-blockers. Pharmacotherapy. 2007;27:874-887.
65. Lunetta KL. Genetic association studies. Circulation. 2008;118:96-101.
66. Balding DJ. A tutorial on statistical methods for population association studies. Nat Rev Genet.
2006;7:781-791.
67. Emigh TH. A comparison of tests for Hardy-Weinberg equilibrium. Biometrics. 1980;36:627-642.
68. Wigginton JE, Cutler DJ, Abecasis GR. A note on exact tests of Hardy-Weinberg equilibrium. Am J
Hum Genet. 2005;76:887-893.
69. Strachan T, Read AP. Human Molecular Genetics. 3rd ed. Oxford: Garland Science/Taylor & Francis
Group; 2003.
70. Cavalli-Sforza LL. Human evolution and its relevance for genetic epidemiology. Annu Rev Genomics
Hum Genet. 2007;8:1-15.
71. Gabriel SB, Schaffner SF, Nguyen H, et al. The structure of haplotype blocks in the human genome.
Science. 2002;296:2225-2229.
72. de Bakker PI, Yelensky R, Pe′er I, et al. Efficiency and power in genetic association studies. Nat Genet.
2005;37:1217-1223.
73. Marchini J, Cutler D, Patterson N, et al. A comparison of phasing algorithms for trios and unrelated
individuals. Am J Hum Genet. 2006;78:437-450.
74. Rieder MJ, Reiner AP, Gage BF, et al. Effect of VKORC1 haplotypes on transcriptional regulation and
warfarin dose. N Engl J Med. 2005;352:2285-2293.
75. Hirschhorn JN, Daly MJ. Genome-wide association studies for common diseases and complex traits.
Nat Rev Genet. 2005;6:95-108.
76. Gelehrter TD, Collins FS, Ginsburg DG. Principles of Medical Genetics. 2nd ed. Baltimore, MD: Williams
& Wilkins; 1998.
77. Devlin B, Risch N. A comparison of linkage disequilibrium measures for fine-scale mapping. Genomics.
1995;29:311-322.
78. Lewontin RC. The interaction of selection and linkage. II. Optimum models. Genetics. 1964;50:757-782.
79. Pritchard JK, Przeworski M. Linkage disequilibrium in humans: Models and data. Am J Hum Genet.
2001;69:1-14.
80. Carlson CS, Eberle MA, Rieder MJ, et al. Selecting a maximally informative set of single-nucleotide
polymorphisms for association analyses using linkage disequilibrium. Am J Hum Genet. 2004;74:106-
120.
81. International HapMap Consortium. A haplotype map of the human genome. Nature. 2005;437:1299-
1320.
82. International HapMap Consortium. The International HapMap Project. Nature. 2003 Dec 18;
426(6968):789-796.
83. Patnala R, Clements J, Batra J. Candidate gene association studies: a comprehensive guide to useful
in silico tools. BMC Genet. 2013;14:39.
84. SEARCH Collaborative Group, Link E, Parish S, et al. SLCO1B1 variants and statin-induced myopathy—a
genome-wide study. N Engl J Med. 2008;359:789-799.
85. Barrett JC, Fry B, Maller J, et al. Haploview: Analysis and visualization of LD and haplotype maps.
Bioinformatics. 2005;21:263-265.
86. Hu D, Ziv E. Confounding in genetic association studies and its solutions. In: Qing Y, ed. Pharmaco-
genomics in Drug Discovery and Development. New York: Humana Press; 2008:31-39.
87. Cardon LR, Palmer LJ. Population stratification and spurious allelic association. Lancet. 2003;361:598-
604.
88. Knowler WC, Williams RC, Pettitt DJ, et al. Gm3;5,13,14 and type 2 diabetes mellitus: an association in
American Indians with genetic admixture. Am J Hum Genet. 1988;43:520-526.
89. Halder I, Shriver MD. Measuring and using admixture to study the genetics of complex diseases. Hum
Genomics. 2003;1:52-62.
CHAPTER
3
Methodologies in
Pharmacogenomics
Amy-Joan L. Ham, PhD
LEARNING OBJECTIVES
After completing this chapter, the reader Illustrate how Food and Drug
should be able to Administration−approved clinical
genotyping tests are used to determine
Describe the processes of DNA collection
an individual’s genetic makeup.
and DNA isolation.
Describe the chemistry used in the
polymerase chain reaction and how it has
been adapted to streamline genotyping
KEY DEFINITIONS
methods and to allow for multiplexing
capacities. ALLELE—one of a number of alternative DNA
sequences at a particular site in the DNA.
Explain the difference between the
allele discrimination reaction, the allele ALLELE DETECTION—method used in a
detection platform, and allele detection genotyping reaction to capture the
methods. information from the allele discrimination
step.
Specify factors that may influence the
selection of a particular genotyping ALLELE DISCRIMINATION—method used in
method. a genotyping reaction to differentiate
between various alleles (i.e., wild-type
Compare and contrast assay detection versus polymorphic allele).
platforms and detection methods with
respect to their relative cost, throughput, ALLELE-SPECIFIC OLIGONUCLEOTIDE (ASO)
and multiplexing capacities. HYBRIDIZATION—a common allele
discrimination method used in
Compare and contrast the genotype genotyping procedures whereby probes
discrimination reactions used in are designed to bind complementary to
common genotyping methods specific DNA sequences and are used to
including next-generation sequencing interrogate the polymorphism of interest.
and enzyme-based methods, allele-
specific hybridization, allele-specific ASSAY—an analytical procedure used for
amplification, and single-base extension qualitative or quantitative measurement
methods. of an analyte (molecule). In the setting of
pharmacogenomics, it is a procedure or
Discuss how genotyping tests are
test used to determine a person’s genetic

regulated by the federal government makeup or genotype.
and which agencies are responsible for
oversight.
59
DEOXYRIBONUCLEIC ACID (DNA)—a self-replicating material present in nearly all living

organisms as the main constituent of chromosomes.
GENOTYPE—an individual collection of genes. Generally refers to the two alleles inherited
for a particular gene.
GENOTYPING—determination of a person’s genetic makeup at a particular site in the DNA.

MASSIVELY PARALLEL SEQUENCING—a high-throughput sequencing approach in which many
sequencing reactions (100s–1,000s) can be run at the same time using the same platform
(in parallel). This approach is used in next-generation sequencing technologies.
MULTIPLEXING—analysis of multiple polymorphisms in the same genotyping reaction.
OLIGONUCLEOTIDES—short fragments of single-stranded DNA that are designed to bind to
specific DNA sequences. Oligonucleotides are often referred to as primers or probes.
POLYMORPHISM—the presence of two or more alleles for a gene or DNA sequence that
exists in more than 1% of the population.
PRIMER EXTENSION—a common allele discrimination method used in genotyping procedures
where a single-stranded oligonucleotide is bound complementary to the sequence of
interest, and the strand is extended over the polymorphism of interest.
RESTRICTION ENZYMES—enzymes that recognize a specific short sequence of nucleotides in
the DNA and cut the DNA at that site.
RESTRICTION SITES—short nucleotide sequences (generally four to eight base pairs in
length) that are present throughout the genome that are sites of cleavage for restriction
enzymes.
THROUGHPUT—the number of samples or polymorphisms that can be genotyped with a
particular genotyping method.
WILD TYPE—the most prevalent (“normal”) DNA sequence for a given gene and/or allele
within a species.
INTRODUCTION
The field of pharmacogenomics is aimed at understanding the influence of interindividual
genetic variability on drug disposition and response.1 Pharmacogenomic studies have heavily
populated the scientific literature in recent years, and this pharmacogenomic information
is starting to migrate into clinical practice.2-5 One of the key factors driving pharmaco-
genomic research, and the subsequent incorporation of this information into clinical prac-
tice, is the increased number of sophisticated technologies that are available to determine
a person’s genetic makeup (i.e., genotyping).6 Genotyping technologies have rapidly evolved
over the past two decades and continue to evolve at a rapid pace. As a result of these
advances, the accurate and timely determination of a person’s genetic makeup is no longer
the rate-limiting step in the field of pharmacogenomics.
Although clinicians will rarely be asked to perform laboratory-based genetic tests, an

understanding of how genetic information is obtained, processed, and analyzed is crucial
for the successful clinical application of pharmacogenomic information.7 Moreover, an
understanding of the principles underlying common genotyping methods will allow for
the appropriate analysis and interpretation of pharmacogenomic studies that are published
Chapter 3 • Methodologies in Pharmacogenomics 61
in the literature. The goal of this chapter is to highlight the most common methodologi-
cal procedures used in the field of pharmacogenomics. From deoxyribonucleic acid (DNA)
sample collection to genotype determination, the basic laboratory steps involved in the
pharmacogenomic process will be discussed. The discussion will include
1. DNA preparation including DNA sample collection, DNA isolation, and DNA target sequence
amplification using the polymerase chain reaction (PCR) or variations thereof.

2. The components of genotyping assays (allelic discrimination reactions, assay detection
platforms, and detection methods) including examples such as restriction fragment length
polymorphisms (RFLP) analysis, pyrosequencing, and TaqMan®.
3. Data interpretation utilizing algorithms and bioinformatics analysis.
These steps and their primary components are listed in Figure 3-1. Although data inter-
pretation and bioinformatics are integral to the overall process, this chapter will focus
primarily on the genotyping assays and the chemistries involved. Genotyping assays can
be further broken down into allelic discrimination reactions, assay detection platforms,
and detection methods. Much of the chapter will be devoted to the explanation of these
various approaches for genotyping (Table 3-1). In addition, these assays can be directly
DNA Preparation Genotyping Assay Data Interpretation
DNA Sample Allelic Discrimination Data Interpretation

Collection Reaction Algorithm/Software
Assay Detection Bioinformatic

DNA Isolation
Platform Analysis
Template
Detection Method
Preparation/
Amplification
FIGURE 3-1 Laboratory components of genotyping analysis for pharmacogenomics.
Table 3-1
Components of a Genotyping Assay
Reaction Product
Allelic Discrimination Reaction Assay Detection Platform Detection
DNA sequencing Microarray Radiography

Enzyme-based assays Microsphere Fluorescence
Hybridization Mass spectrometry Mass spectrometry
Allele-specific amplification Electrophoresis FRET
Single-base extension Microtiter plate Chemiluminescence
Electric current
disruption by
semi-conductor or
biomolecule (Ion
Torrent, Oxford
Nanopore)
FRET, fluorescence resonance energy transfer.
combined with various methods for amplification based on the PCR that helps to stream-
line the assays and allows for multiplexing for large-scale analyses. By understanding the
various approaches for genotyping assays and how the components have been “mixed
and matched” in different methods, one can better comprehend new approaches as they
continue to be developed.
CASE STUDY
Cytochrome P450 (CYP) 2D6 is an oxidative enzyme that is responsible for the hepatic
metabolism of a wide variety of clinically important drugs such as antidepressants,
antipsychotics, antiarrhythmics, opiates, antiemetics, and ß-adrenergic receptor
blockers.8,9 CYP2C19 is also an oxidative enzyme that is responsible for the hepatic
metabolism of drugs such as anticonvulsants, proton pump inhibitors, anticoagulants,
benzodiazepines, and antimalarials.8,9 The CYP2D6 gene is highly polymorphic, with
over 100 polymorphisms identified in the gene thus far (http://www.cypalleles.ki.se/
cyp2d6.htm). Functional polymorphisms in the CYP2D6 gene are associated with
altered enzyme activity and result in different CYP2D6 metabolizing enzyme phenotypes
such as poor metabolizers, intermediate metabolizers, normal (extensive) metabolizers,
and ultrarapid metabolizers. Two common polymorphisms in the CYP2C19 gene are
associated with a poor metabolizer phenotype. Knowledge of an individual’s CYP2D6 and
CYP2C19 genotype and phenotype may help predict their CYP2D6 and CYP2C19 drug-
metabolizing enzyme activity and may help clinicians select the safest and most effective
medications for treating a particular disease in that individual.8,9
The U.S. Food and Drug Administration (FDA) has approved several tests for CYP2D6
and CYP2C19 polymorphisms, including the Roche AmpliChip CYP450 test. This test is
intended to identify a patient’s CYP2D6 and CYP2C19 genotype. The package insert for
the Roche AmpliChip CYP450 test states that information about CYP2D6 and CYP2C19
genotypes may be used as an aid to clinicians in determining therapeutic strategy
and treatment dose for therapeutics that are metabolized by the CYP2D6 or CYP2C19
gene product.10 The Roche AmpliChip CYP450 test can be prescribed by a physician
and processed at a participating laboratory. The CYP2D6 and CYP2C19 genotype and
phenotype results are sent back to the provider usually within a week.
Questions
1. In clinical practice, what are the most common ways to collect human genomic DNA?
2. How is DNA isolated from nucleated cells? Explain the process.
3. What role does PCR play in the genotyping process?
4. What factors influence the choice of a genotyping method for a particular laboratory or phar-
macogenomic application?
5. Can you explain how genotyping by a microarray is different than genotyping with other
methods such as PCR-RFLP, pyrosequencing, or other next-generation sequencing (NGS)
techniques (TaqMan®)? The Affymetrix DMET™ Plus array is a DNA microarray.
DNA SAMPLE COLLECTION

Pharmacogenomics is based on the accurate determination of an individual’s genetic
makeup at a particular site in his or her genomic DNA. Although nonhuman genetic
information (e.g., tumor, bacteria, virus) can be used to guide pharmacologic therapy, the
processes described hereafter are focused on the collection, isolation, and interrogation of
human genomic DNA. Human genomic DNA can be obtained from any cell in the body that
contains a nucleus, most commonly blood lymphocytes or buccal (cheek) cells. Genomic
DNA cannot be obtained from non-nucleated cells, such as red blood cells or platelets. The
gold standard method to obtain a DNA sample is through collection of peripheral whole
blood.11 Whole blood collection is the preferable method because it yields a large amount
of DNA. For example, a 5-mL whole blood sample yields enough DNA (approximately
150 mcg) to perform thousands of genotyping reactions.12 Although clinical genetic tests
usually require only a few genotyping reactions, whole blood collection is advantageous
because it allows for the storage of an ample amount of DNA for future pharmacogenomic
investigations. Furthermore, some studies suggest that the DNA obtained from whole blood
is of superior quality than DNA obtained from other sources, such as buccal cells or saliva.13
The disadvantage of whole blood collection is that it involves an intravenous blood draw.
The venipuncture procedure may be resource-, time-, or cost-prohibitive in certain patient
care settings, and it may pose challenges or discomfort in certain patient populations
(e.g., pediatric patients).
Another common method for DNA sample collection is through the collection of
buccal epithelial cells using a cheek swab, cheek brush, or oral rinse.11,14,15 The cheek swab
and brush methods involve rubbing a foam-tipped swab or cytobrush against the inside
of the cheek for approximately 30 seconds.16 The swab or brush is then placed in a sterile
container. For the oral rinse method, patients are asked to vigorously swish approximately
10-30 mL of commercially available mouthwash for 30-60 seconds and then expectorate
into a sterile collection container.14,15,17 Some studies have shown that the oral rinse method
produces higher DNA yields than the swab or brush methods (e.g., 55 mcg of DNA oral rinse
method versus 12 mcg of DNA brush or swab methods).15,17,18 In terms of advantages, buccal
cell methods are noninvasive, easy to perform, and relatively painless. However, buccal
cell methods may be disadvantageous because they result in lower DNA yields than whole
blood collection, often have nonhuman DNA contamination (e.g., bacteria), and may not
be ideal for certain patient groups. For example, in children, cytobrushes may be painful,
and the oral rinse technique may be difficult for children to execute without swallowing or
aspirating the liquid.19 Additionally, some data suggest that DNA obtained from buccal cell
methods may be of lower quality than DNA obtained from whole blood, and thus may not
perform as well in subsequent genotyping reactions.20
The newest DNA collection method involves obtaining a whole saliva sample, which
includes both buccal epithelial cells and white blood cells found in the mouth. In this
method, patients are asked to expectorate approximately 2 mL of saliva into a sterile
container. The whole saliva method yields more DNA (i.e., approximately 35 mcg per 2-mL
saliva sample) than other buccal cell collection methods; however, the amount of non-
human DNA is high.21 A commercial whole saliva collection kit is available (Oragene® DNA
self-collection kit, DNA Genotek Inc.); after a patient expectorates into the container and
closes the cap, the contents in the container initialize the beginning phases of DNA isola-
tion and stabilization. This type of collection kit lends itself well to situations where DNA
needs to be stored for long periods of time or shipped at room temperature.16
TEMPLATE PREPARATION
DNA Isolation
After DNA collection, the next step in the process is to extract, isolate, and purify the DNA
from the cells in the sample. This process is referred to as DNA isolation. There are many
commercially available kits to aid in the DNA isolation process, and the methods used in
these kits differ based on the type of sample that is collected, the chemicals used in the
DNA isolation process, and the quantity of DNA that needs to be isolated. In general, the
process of genomic DNA isolation involves the following steps: (1) disruption and lysis of
cells to release genomic DNA, (2) removal of proteins and cellular debris, and (3) recovery
of purified DNA.22
With DNA isolation from whole blood, the first step in the process is to lyse the red
blood cells (which do not contain genomic DNA) while keeping the white blood cells
(which contain genomic DNA) intact. The white blood cells are then collected in a pellet
through centrifugation, and the cellular membranes lysed with a detergent. During the
disruption and lysis process, protease is added to the sample to digest proteins contained
in the cells.22 In some DNA isolation protocols, organic solvents (e.g., phenol, chloroform,
isoamyl alcohol) or high concentrations of salts (e.g., potassium acetate, ammonium
acetate) are used to extract the proteins from the lysed cells.22 These methods have some
limitations in that they are time consuming and often require the use of toxic substances
(e.g., organic solvents). In most commercially available DNA isolation kits, the use of prote-
ase is preferred because it is easy, reliable, and relatively nontoxic. Following treatment
with protease, the digested cellular proteins are collected in a pellet by centrifugation
and the supernatant, which contains the genomic DNA, is collected. The genomic DNA in
the supernatant is then recovered from the sample by an alcohol precipitation step using
ethanol or isopropanol.22 The precipitated DNA is collected in a pellet by centrifugation
and resuspended in an appropriate buffer for long-term storage.22
The process of isolating DNA from buccal cells varies slightly from the isolation
process of whole blood samples because it does not require a red blood cell lysis step.
Instead, buccal cell samples are centrifuged to concentrate the cells in a pellet. This pellet
is then resuspended, and the processes of cell lysis, protein removal, DNA precipitation,
and DNA recovery are performed.18 Following the DNA isolation process, the purity and
concentration of DNA can be measured by spectrophotometry. The concentration of DNA
is determined by measuring the absorbance at 260 nm. The purity of DNA is determined
by measuring the ratio of absorbances at 260 nm and 280 nm. A ratio of 1.7 to 1.9 indicates
that the DNA sample is pure and free of protein contaminants.22
DNA Amplification
PCR
The human genome contains approximately 3 billion base pairs. Most clinical pharma-
cogenomic tests require that a specific region of DNA within the genome be targeted and
amplified. A specific region of DNA that contains the gene or polymorphism of interest is
often referred to as a target sequence. Target sequences can vary in size depending on the
subsequent genotyping method to be used; but, in general, target sequences are usually a
few hundred base pairs in length. PCR is the method that is used to target and amplify a
specific sequence of DNA within the genome.23,24
PCR was developed based on the inherent physiochemical properties of DNA, which
consists of two strands bound together in antiparallel form (5′ to 3′ and 3′ to 5′). The
nucleotide bases (adenine, thymine, cytosine, and guanine) in the two strands are bound
complementary to each other by hydrogen bonds such that adenine binds with thymine
and cytosine binds with guanine.25 The hallmark of PCR is the cycling of different tempera-
tures, in the presence of key reaction components, to target and exponentially amplify a
specific DNA target sequence. A PCR mixture generally contains the following reaction
components: genomic DNA, deoxynucleoside triphosphates (dATP, dCTP, dGTP, and dTTP),
buffer, cations (e.g., magnesium, potassium), primers, and DNA polymerase.26 The major
steps of the PCR process (Figure 3-2) are (A) denaturation (i.e., strand separation), (B) primer
annealing, and (C) strand extension.7,23
During the process of denaturation, the hydrogen bonds holding the double-stranded
DNA molecule together are broken, and the double-stranded DNA molecule is separated
into two single-stranded molecules (Figure 3-2). This typically occurs at a temperature
of approximately 95°C. Following denaturation, two single-stranded primers (also known
as oligonucleotides) are annealed to the single-stranded DNA molecules. A primer is a
short sequence of nucleotides (generally 17 to 30 base pairs in length) that is designed
to bind complementary to a specific sequence of nucleotides in the single-stranded DNA
molecule.24 One primer is designed to bind complementary to one strand of the DNA
molecule in the forward direction (5′ to 3′), while the second primer is designed to bind
complementary to the other strand of the DNA molecule in the reverse direction (3′ to 5′).24
The temperature required for the primers to anneal to the single-stranded DNA molecule
is highly dependent on the sequence of nucleotides in the template DNA. However, in
general, annealing temperatures range from 40° to 70°C.
After the two primers are annealed to the single-stranded DNA molecules, the process
of extension can occur. The extension step of the PCR reaction is typically carried out at
72°C and is catalyzed by an enzyme called DNA polymerase, which promotes the synthesis
of a complementary strand of DNA in the 5′ to 3′ direction.22 Specifically, DNA polymerase
functions to add deoxynucleoside triphosphates (dATP, dCTP, dGTP, and dTTP) to the 3′
end of each primer that is annealed to the single-stranded DNA molecule. In this way, each
single-stranded DNA template strand is built up to form a double-stranded DNA replicate.
There are many different types of DNA polymerases that may be used in the DNA process;
however, the most common is Taq polymerase, a heat-stable DNA polymerase that comes
from the bacterium Thermus aquaticus.23 The cycle of denaturation, annealing, and exten-
sion is repeated 30 to 40 times resulting in an exponential increase in the number of DNA
replicates. At the end of the PCR process, millions of copies of the DNA target sequence
are present in the reaction mixture.
After PCR is complete, a small amount of the PCR product is often subjected to gel
electrophoresis to verify that the PCR worked correctly and that the amplified target
sequence (also referred to as an amplicon) is the correct size (Figure 3-3).7,26 In gel electro-
phoresis, a portion of the PCR product is mixed with a tracking dye and injected onto an
agarose gel, along with a DNA molecular weight marker (also called a ladder). The agarose
gel is placed in an electrophoresis unit, where an electric current is then passed through
the unit. The electric current prompts the negatively charged DNA to pass through the gel
toward the positive electrode. The agarose gel is porous, and the speed at which the PCR
product moves through the gel is dependent on the size of the PCR product.27 For example,
Polymerase Chain Reaction (PCR) DNA

Copies
5’ 3’ 1
3’ 5’
(A)
5’ 3’
+ 3’ 5’
5’ 3’ (B) 3’ 5’
5’ 3’
(C) 3’ 5’
5’ 3’
5’ 3’
2
3’ 5’ 3’ 5’
+
(D) 4
+
8
+
16
FIGURE 3-2 PCR. (A) Denaturation: The first step in the PCR process is to separate the
double-stranded DNA into two single-stranded template molecules. Denaturation is typically
accomplished at 95°C. (B) Annealing: Following strand separation, the reaction is cooled to
allow the primers present in the reaction mixture to hybridize to the single-stranded DNA
templates. One primer binds complementary to one strand in the forward direction, while
the second primer binds complementary to the other strand in the reverse direction. (C)
Extension: DNA polymerase catalyzes the addition of deoxynucleotides to the 3’ end of each
hybridized primer. Extension is typically accomplished at 72°C. Following extension, the
complementary strand of each of the single-stranded products is built up to form two double-
stranded replicates. (D) Repeat cycles: The replicates then serve as templates in the next
series of temperature cycles. This series of cycles (denaturation, annealing, and extension)
is repeated 30–40 times, and the number of double-stranded replicates increases in an
exponential fashion.
large PCR products move slowly through the gel, whereas small PCR products move quick-
ly through the gel.7 The agarose gel is typically stained with ethidium bromide, a substance
that intercalates into DNA and can be visualized under ultraviolet light. After completion of
electrophoresis, the gel can be viewed under ultraviolet light, and the position of the PCR
product (i.e., band) is compared to the position of the bands of the DNA molecular weight
marker. This allows the size of the PCR product to be estimated.7
Lane
Amplicon size (bp)

FIGURE 3-3 Gel electrophoresis. A 106 base pair target sequence in human genomic DNA
was amplified by PCR. The PCR product underwent electrophoresis on a 1.5% agarose gel
and was stained with ethidium bromide to allow for visualization under ultraviolet light.
Lane 1 shows the DNA molecular weight marker (“ladder”). Lanes 2 through 5 show the PCR
product (highlighted bands) from four patient samples. The expected size of the PCR product
(106 base pairs) can be verified by comparing the band position on the gel in relation to the
band positions of the DNA ladder (118 bp and 72 bp).
Originally published in Aquilante CL, Zineh I, Beitelshees AL, et al. Common laboratory
methods in pharmacogenomics studies. Am J Health-Syst Pharm. 2006;63:2101-2110.
© 2006, American Society of Health-System Pharmacists, Inc. All rights reserved. Reprinted
with permission (R0932).
CLINICAL PEARL
Most amplification methods produce amplicons 0.1 to 10 kilo base pairs
(kbp) but may produce fragments up to 40 kbp in size. The proper buffer
solutions and carefully designed primers are necessary to reduce errors
during amplification and to ensure the proper region of DNA is amplified.
Newer Methods for DNA Amplification

Newer, higher throughput PCR-based methods for DNA amplification such as emulsion PCR,
solid-phase PCR (bridge amplification), rolling circle amplification (RCA), and digital droplet
PCR (ddPCR) have also been developed that allow for massively parallel sequencing or
genotyping.28-37 These methods increase the speed of amplification and/or the number
of sequences that can be amplified at the same time. The advantage of these methods is
that they can be directly coupled to high-throughput NGS technologies (discussed below)
or other types of genotyping methods, which eliminates the amplification as a completely
separate step and allows for the amplification and sequencing/genotyping on the same
platform. This not only increases the throughput of sequencing and genotyping, but has
also driven the cost down to a point that routine sequencing is becoming far more plau-
sible. Some of these newer methods are depicted in Figure 3-4.
Emulsion PCR utilizes beads with primers bound to the surface that are complemen-
tary to adaptor sequences coupled to the target DNA (Figure 3-4A).33,34,36-38 The beads
are then added to an oil emulsion along with the reactants necessary for PCR. Each bead
represents a single PCR product, but with the reaction occurring in emulsion, millions of
beads can be processed in the same sample tube. The beads are then separated for indi-
vidual sequencing reactions but on a platform that allows for multiplexing or massively
parallel genotyping such as NGS (discussed later in this chapter).
Digital droplet PCR or ddPCR is a similar quantitative PCR approach (qPCR) to emul-
sion PCR but does not use beads, and the PCR reaction takes place in the presence of a
detection probe.28 With ddPCR, the PCR reaction is randomly partitioned into nanoliter-
sized aqueous droplets using water-oil emulsion and microfluidics to generate the drop-
lets. The droplets are separated (i.e., into wells on a 96-well plate) and serve as reaction
chambers for the PCR reaction. After the PCR reaction is allowed to proceed, final products
are then separated using flow cells and detected with fluorescently labeled probes.
Solid-phase bridge amplification is achieved by binding the target DNA sequence to
adaptor sequences on a solid surface (Figure 3-4B). Two separate adaptor sequences are
coupled to the target DNA, and the primers for both adaptor sequences are attached to the
same area of the solid surface.36-39 When the prepared DNA sample is added, the adaptor
sequence on one end of the DNA sequence anneals with its primer. Because the primer for
the adaptor sequence on the opposite end of the target sequence is also on the surface,
the DNA forms a bridge by annealing to the second primer while still bound to the first
primer. The polymerase then extends the sequence though to the primer, which results
in the covalent attachment of the target sequence to the surface.36,39 This continues with
both adaptor primers producing clusters of DNA that have been amplified on the surface.
Rolling circle amplification or RCA is an isothermal amplification technique that utilizes
DNA polymerases, most frequently Phi29, to generate long single strands of DNA (ssDNA)
from a circularized piece of target DNA (Figure 3-4C). The advantage of polymerases used
for RCA, such as Phi29, is that they have the ability to displace the complementary strand
on DNA allowing them to continue progression despite encountering double-stranded DNA
that has not been denatured into a single strand. This allows the synthesis of the ssDNA,
from a circular template, to continue extension through the original start site to which the
primer binds, without interruption.40 As a result, one long, continuous piece of ssDNA is
generated with multiple copies of the target DNA in tandem.30,40,41 Another advantage of
this technique is that Phi29 is thermostable so, unlike PCR, the amplification procedure
can take place at a constant temperature and can be used either in solution or on a solid
surface.40 There are many modifications to this approach in which multiple primers are
used to bind multiple sites on the circularized DNA template so that multiple sequences
can be amplified at the same time.40 For example, additional primers can be added to
the circular template to allow for branched amplification from the original amplification
product. These approaches have been coupled to either NGS or other allele discrimination
reactions for genotyping analysis.40 One use for this approach is to form the DNA nanoballs
used for the sequencing technology called Revolocity™, developed by Complete Genom-
ics, discussed below.42,43
A. Emulsion PCR
Template
Emulsion
PCR
Primer
attached to
bead Bead
Amplified PCR
product on
bead
B. Solid-phase “bridge” amplification
C. Rolling circle amplification

Circularized
Target DNA
Target DNA Add primer
& polymerase
Polymerase-
displacement and
Multiple copies of DNA Continuous First pass
as continuous strand extension extension
FIGURE 3-4 Amplification methods used for NGS. (A) Emulsion PCR: DNA is fragmented,
denatured and adapter sequences are added to either end of the DNA. A sequence
complementary to one of the adapter sequences is attached to beads, and a sequence
complementary to the other adapter sequence serves as a primer for amplification. The
beads are added to an oil emulsion along with reactants for PCR, and PCR is performed.
The original sequences are then removed through denaturation, leaving the beads with
multiple copies of the DNA on a single bead. (B) Solid-phase “bridge” amplification: DNA is
fragmented, denatured and adapter sequences are added to either end of the DNA. Sequences
complementary to the adapter sequences are attached to a solid surface. The prepared DNA
is then added to the solid phase and binds to one of the complementary adapter sequences
attached to the surface. The DNA then forms a bridge by binding to the other adapter sequence
on the solid surface, which serves as the primer for PCR. PCR is preformed, and clusters of
the sequences are amplified on the solid surface. (C) Rolling circle amplification: Target DNA
is circularized with a sequence complementary to a primer inserted into the sequence. The
primer and a polymerase, which has the ability to displace a strand of DNA from double-
stranded DNA, are added and amplification proceeds. After the first pass of DNA extension,
the polymerase continues to amplify DNA by displacement of the original strand and
produces multiple copies of the target DNA in a single continuous strand.
CLINICAL PEARL
DNA amplification is necessary for many genotyping assays to ensure
the quantity of DNA is adequate for detection, but this can also introduce
error due to spurious mutations during amplification that can result in an
incorrect genotypic assignment.
PRINCIPLES OF GENOTYPE DETERMINATION

After the DNA sequence of interest is targeted and amplified, the next step in the pharma-
cogenomic process is to determine a person’s genetic makeup (i.e., genotype) at a particular
site within that DNA sequence. Many genotyping methods and technologies are avail-
able to accomplish this goal (Table 3-2).7,29,35,41,44-52 At each site in DNA, a person has two
alleles—one from their mother and one from their father. Methods for genotyping can be
broken down into three separate parts: (1) the allele discrimination reaction, (2) the allele
detection platform, and (3) the reaction product detection method. The allele discrimina-
tion reaction refers to the chemistry that is used to distinguish between polymorphic (i.e.,
variant) and nonpolymorphic (i.e., wild type) alleles that are present at a particular locus
in a person’s DNA sample. The allele detection platform refers to the physical platform
on which the allele discrimination reaction is performed (i.e., microarray, beads, mass spec-
trometry). The reaction product detection method refers to the instrumentation and chemi-
cal principle on which the product is ultimately detected from the allele discrimination
reaction (i.e., fluorescence, molecular mass).
Beyond differences in allele discrimination reaction, platform, and detection, the choice
of genotyping method can be influenced by other factors such as the number of samples
and polymorphisms that can be genotyped at one time (also referred to as throughput);
the type of polymorphism that can be genotyped using the method (e.g., single nucleotide
polymorphisms [SNPs] versus insertion/deletion polymorphisms); equipment acquisition
and genotyping costs (i.e., labor, consumables); turnaround time; required technical exper-
tise; and the ability of the method to genotype more than one polymorphism at one time
(i.e., multiplex genotyping).7,45,53-58 Ideally, the chosen method should be able to determine
a genotype in one attempt; if a second attempt is necessary, it should be fast and inexpen-
sive.7,44 Given the millions of polymorphisms present in the human genome, the genotyp-
ing method should also allow for fast and easy assay development and validation.59 The
ability to set up a new genotyping assay is a key component for productivity in the field
of pharmacogenomics. The genotyping method should also provide genetic results in a
manner that is software-driven and easy to interpret.60 This decreases the chance for errors
in human interpretation and ambiguous genotype determinations.
As pharmacogenomic tests begin to make their way into the clinical arena, addi-
tional analytical parameters such as specificity, sensitivity, reproducibility, and accuracy
are also important considerations in the genotyping process.60,61 The challenge associated
with evaluating genetic test performance characteristics is that many tests arise out of
academic laboratories and are not FDA-approved diagnostic tests.61 As such, analytical test
performance characteristics may be unpublished or may differ between or within a labora-
tory.61 Analytical test characteristics that should be considered in the genotyping method
selection process include analytical sensitivity, analytical specificity, reproducibility, and
Table 3-2
Summary of Selected Genotyping Methods
Allele
Genotyping Discrimination Allele Detection Cost Estimates

Method Method Method/Platform Throughput per SNPa
DNA microarray Allele-specific Fluorescence/ High Low

hybridization solid surface or
beads
Invader® assay Cleavage- Fluorescence/96- Medium to high Low
base with or 354-well
endonuclease PCR plates
enzyme (e.g.,
Cleavase®)
Ligation Ligation-based Fluorescence/96- Medium to high Low
using allele- or 354-well
specific PCR plates
hybridization
and ligase
enzyme
Mass Various (e.g., Molecular Medium to high Low
spectrometry allele-specific weight/mass
hybridization, spectrometry
primer (e.g., MALDI-
extension) TOF)
NGS reversible Reversible chain Fluorescence/ Medium Medium to lowb
chain termination solid surface
termination chemistry
PCR-RFLP Restriction Fluorescence or Low High
endonucleases radiation/gel
electrophoresis
Pyrosequencing Primer extension Light/beads Medium Medium
Sanger DNA Chain termination Fluorescence/ Low Medium
sequencing chemistry capillary gel
electrophoresis
SNaPShot® Primer extension Fluorescence/ Medium to high Low
capillary gel
electrophoresis
TaqMan® Allele-specific Fluorescence/96- Medium to high Low
hybridization or 354-well
PCR plates
DNA, deoxyribonucleic acid; MALDI-TOF, Matrix Assisted Laser Desorption Ionization-Time of Flight; NGS, next-
generation sequencing; PCR, polymerase chain reaction; RFLP, restriction fragment length polymorphism; SNP,
single nucleotide polymorphism.
a
Cost estimates: high, greater than $4.00 per SNP; medium , $1.00 to $4.00 per SNP; low, less than $1.00 per SNP.
b
Depends on the specific instrument platform.
accuracy.60,62 In the context of genetic testing, analytical sensitivity refers to the probability
that a test will be positive when a particular DNA sequence is present (true positive), while
minimizing the detection of a genotype that is not present (false positive). The analyti-
cal specificity refers to the probability that a test will be negative when a particular DNA
sequence is absent (true negative), while not missing to the detection of a genotype (false
negative).63 Reproducibility refers to the probability of the test repeatedly producing the
same results in the same person.62,63 Accuracy is the degree to which the observed geno-
type matches the true genotype.60
Oversite and regulation of clinical diagnostic laboratories is the responsibility of the
Centers for Medicare & Medicaid Services (CMS) under the Clinical Laboratory Improvement
Amendment (CLIA) of 1988.64,65 Regulation of genotyping tests by the FDA, which falls
under the regulation of medical devices, is increasing. The FDA is working to regulate these
types of tests to improve the reliability and to ensure accuracy and validity.
Many different methods are available to determine a person’s genetic makeup, and
the list grows substantially each year as genomic technologies become more sophisti-
cated and less expensive. This chapter will first focus on the four primary components of
an assay, so that it is easier to understand new methods as they develop. Each method
to determine genotype can be broken down into four primary categories: (1) template
preparation and purification (discussed above), (2) allelic discrimination reaction, (3) assay
detection platform, and (4) reaction product detection (Table 3-1).53,54,66-69 In addition, a
discussion of pertinent factors (e.g., throughput, cost) that might influence the choice of
that particular genotyping method will be included. An abbreviated summary of three
different parts (following template preparation) of some of most widely used genotyping
methods is presented in Table 3-1. A summary of some common methods discussed in this
chapter are presented in Table 3-2. In addition, the primary advantages and disadvantages
of some popular genotyping methodologies are presented in Table 3-3. The discussion of
genotyping methods is focused on their ability to genotype SNPs. However, it is important
to point out that the genotyping method can be engineered in many cases to detect other
polymorphisms such as insertions/deletions, copy number variants, and nucleotide repeats.
Assay Detection Platforms and Reaction Product Detection

To better understand the various allele discrimination reactions, it is helpful to first under-
stand the detection platforms and the chemistries behind the reaction product detection.
There are a number of methods of product detection and platforms on which the reactions
and detection can take place. Platforms include gel electrophoresis and capillary electro-
phoresis, array-based platforms, flow-cell analysis, and others. The platform on which the
assay is performed frequently determines how many samples can be processed, if they
can be achieved in parallel, and the amount of labor versus automation that is possible.
Platforms such as microarray analysis, which is a fairly high-throughput method, allow for
multiple genetic mutations to be analyzed at once. NGS platforms also increase throughput
by allowing massively parallel sequencing but at the cost of accuracy. Continued improve-
ments in these methods will decrease the errors that we see with these technologies as
the speed of sequencing continues to improve in throughput and speed. These platforms
will be discussed in more detail in the discussion of allele discrimination reactions below.
Reaction products can be analyzed by a number of product detection methods. The
oldest method of detection is probably the use of radiography for the detection of radio-
labeled probes of nucleic acids. This required the modification of molecules to incorporate
a radioactive atom. This radiation could then be detected using an x-ray film. The most
common methodology currently used for detection is fluorescence, which not only elimi-
nated the risk associated with handling radioactivity but also allowed for methods in which
more than one molecule is labeled using different fluorescent dyes; these dyes fluoresce
Table 3-3
Notable Advantages and Disadvantages of Some Popular Genotyping
Methods
Genotyping Method Advantages Disadvantages
DNA microarray Capable of genotyping hundreds High cost per chip may be cost
of thousands of polymorphisms prohibitive for studies involving
Utilized in genome-wide thousands of patients
association studies
Invader® assay Does not require a PCR Requires large quantities of

amplification step genomic DNA
Highly accurate Limited multiplexing capability
Software assigns genotype call
Mass spectrometry High sensitivity Requires high equipment
High throughput acquisition costs
Higher order multiplexing Requires technical expertise
capability Requires labor-intensive
Software assigns genotype calls purification procedures
NGS reversible Reversible chain termination High instrument acquisition costs
chain termination Solid-surface array format Large-scale runs
Excellent multiplexing capability High computational needs
High throughput Long run time
High sensitivity and specificity, Short read length (50–150 bp)
low error rate
Lowest cost per Mb
Software assigns genotype calls
PCR-RFLP Low equipment acquisition costs Low throughput
Lengthy sample processing times
High per SNP cost
Limited multiplexing capability
Requires user to assign genotype
calls, which can introduce error
Pyrosequencing Primer extension Higher per SNP cost
Provides sequence information Difficulty interrogating regions
for the region surrounding the that contain long stretches
SNP, thus providing a specificity of the same nucleotide (i.e.,
measure in each reaction homopolymeric regions)
High sensitivity and specificity, Roche/454 platform will be
low error rate discontinued
Medium read length Lower order multiplexing
(400–1,000 bp) capability
Software assigns genotype calls
TaqMan® Combines PCR and allele Less amenable to multiplexing

discrimination in a single Fluorescent-labeled probes can
reaction, which results in time be expensive
and cost savings Lower order multiplexing
Software assigns genotype calls capability
DNA, deoxyribonucleic acid; NGS, next-generation sequencing; PCR, polymerase chain reaction; RFLP, restriction
fragment length polymorphism; SNP, single nucleotide polymorphism.
at different wavelengths, allowing for multiple reactions at the same time. Fluorescence
is generated when a molecule is excited by photons of a specific wavelength and releases
the energy by emitting photons of a different wavelength. The fluorescence is typically the
result of the incorporation of a fluorescently labeled deoxy nucleotide during the addition
of nucleotides to the reaction or by the annealing of a fluorescently labeled allele-specific
probe. A variation of this methodology is the use of quenching agents that suppress the
fluorescence of a reporter dye when in close proximity, but the fluorescence is released
and allowed to produce its emission after the reporter dye is separated from the quencher
(i.e., TaqMan, Invader®).7,49,70-74
The release of light (chemiluminescence) is the method of detection for pyrosequenc-
ing in which light is emitted through the catalysis of an enzymatic reaction of luciferase by
adenosine triphosphate (ATP) as a cofactor.57 A more recently applied method of detection
for DNA sequencing analysis is by measuring changes in electrical current. Oxford nano-
pore sequencing uses a protein that forms a pore inserted into a membrane to sense the
differing electrochemical properties of the nucleotides as they pass through the pore.75
Ion torrent (Thermo Scientific) sequencing uses an ion semiconductor to measure the
hydrogen ion released during the incorporation of a nucleotide by polymerase (Ion Torrent,
Thermo Scientific).76,77
Mass spectrometry has become a useful allele detection method in the genotyp-
ing process. The technique of matrix-assisted laser desorption-ionization time-of-flight
(MALDI-TOF) mass spectrometry is used to measure the molecular mass of oligonucle-
otides.51,52,78 While other allele detection methods rely on light, electrophoresis, or fluores-
cence, mass spectrometry differentiates among alleles based on mass alone. As such, the
technology eliminates the need for expensive primer labeling or lengthy gel electrophore-
sis runs.79,80 Additionally, mass spectrometry has excellent precision and high-throughput
capabilities. The primary drawbacks of mass spectrometry as an allele detection method
are the high equipment acquisition costs, high level of required technical expertise, and
rigorous sample purification procedures.
Because mass spectrometry is used primarily as an allele detection method, it is typi-
cally coupled with allele discrimination methods, such as single primer extension or allele-
specific hybridization,81 that allow for the analysis of differences in mass for detection. For
example, the MassARRAY® system (Agena Bioscience) uses primer extension whereby prim-
ers are annealed upstream of the polymorphism of interest and are extended to yield prod-
ucts that differ in molecular weight.7,35,82-87 The mass of the different extension products is
measured by MALDI-TOF mass spectrometry, and genotypes are assigned accordingly.
Allelic Discrimination Reaction

The allelic discrimination reaction refers to the method by which the individual allele can
be distinguished from other alleles. These reactions can be classified into six different cate-
gories (Table 3-1): (1) DNA sequencing, (2) next-generation sequencing (NGS), (3) enzyme-
based methods, (4) hybridization, (5) allele-specific (AS) amplification, and (6) single base
extension.69 Although these classifications are useful in understanding the various types
of discrimination that are used to distinguish alleles, there is also some crossover; combi-
nations of these methods are often used for genotyping. Because adaptations of these
methods are continually developing, this section will describe the basis of these different
types of discrimination reactions so that as methods develop and adaptations are made,
the information presented can serve as a guide to understanding these new developments.
DNA Sequencing
Before the recent advent of sophisticated high-throughput genotyping technologies, direct
DNA sequencing (i.e., determination of the sequence of nucleotides along a DNA strand)
was one of the most popular methods used to determine a person’s genetic makeup.
Although other genotyping modalities are often used today, direct DNA sequencing is still
considered a gold standard genotyping method. Furthermore, many advances in sequenc-

ing technology have resulted in faster, less cumbersome methods that are more amenable
to high-throughput genomic analyses.36,39
CHAIN TERMINATION SEQUENCING (SANGER SEQUENCING). The origins of direct DNA

sequencing are largely based on a method developed by Dr. Fred Sanger in 1975.88,89 The
Sanger method is often referred to as the chain termination method because this method’s
foundation is the use of dideoxynucleotide chain terminators. Dideoxynucleotides are simi-
lar to nucleotides, except they do not have a hydroxyl group at the 3′ carbon position of
the molecule (Figure 3-5A).90 To elongate a DNA chain, a phosphodiester bond must form
between the 3′ carbon of the nucleotide already incorporated into the DNA sequence
and the 5′ carbon of the nucleotide that is added to the DNA sequence. When added to a
sequencing reaction, dideoxynucleotides can incorporate into the DNA sequence; however,
because they lack a 3′ hydroxyl group, no additional nucleotides can be added to the
growing DNA chain (i.e., no phosphodiester bond can be formed). Thus, the DNA chain is
terminated.88-91
Sanger used the chain-terminating chemistry of dideoxynucleotides to develop a novel
DNA sequencing method (Figure 3-5B).88,89 In this method, DNA is separated into single-
stranded molecules, and a primer is annealed to the strand. The mixture is divided into four
separate sequencing reactions that contain DNA polymerase, all four deoxynucleotides
(dATP, dCTP, dGTP, and dTTP), and one dideoxynucleotide (e.g., ddATP in reaction mixture
1, ddCTP in reaction mixture 2).39,90 In each reaction mixture, DNA polymerase catalyzes the
incorporation of deoxynucleotides or dideoxynucleotides into the growing complementary
DNA chain. In the reaction mixture, deoxynucleotides are present at a higher concentration
than dideoxynucleotides, thus the deoxynucleotides and the dideoxynucleotides compete
for incorporation into the growing DNA chain.
If a dideoxynucleotide gets randomly incorporated into the complementary DNA
sequence and the chain elongation reaction is terminated, this results in a mixture of
DNA fragments of different lengths.39,91 These fragments all have a common 5′ end (due
to the primer) but, because of variable chain termination, will have different 3′ ends and
thus different sizes.39,91 The size of the fragments in each of the four sequencing reaction
mixtures can then be assessed using a denaturing gel or capillary gel electrophoresis,
and the DNA sequence can be determined from the position of bands on the gel (for a
denaturing gel electrophoresis) or the order of elution (capillary gel electrophoresis).39,91 In
many automated DNA sequencing systems, the primers or dideoxynucleotides are labeled
with a fluorescent dye. As the DNA fragments move through the gel and pass a laser, the
fragments fluoresce and the output is recorded as peaks on a chromatogram. The chro-
matogram is recorded and serves as the basis from which the DNA sequence can be recon-
structed.39,91 When used for genotyping purposes, the direct DNA sequencing output can be
used to determine if an individual has polymorphic or wild-type alleles at a particular site
in the DNA sequence (Figure 3-5B).
A. Di-deoxy-nucleotides
Base
(A, C, G, T)
B. Sanger Sequencing
ddGT
ddCT
ddAT
ddTT
ddATP
ddCTP
C
T
ddGTP
T Denaturing gel
ddTTP G
C
A
5’ 3’
C
T
T
Capillary gel
G
C
A
FIGURE 3-5 Principles of the Sanger method of DNA sequencing. (A) Dideoxynucleotides
(also known as chain terminators) are similar to deoxynucleotides, except they lack a
hydroxyl group at the 3′ carbon position. (B) Chain-termination method for DNA sequencing.
A primer is annealed to a single-stranded DNA molecule in four separate reactions, one
for each dideoxynucleotide. In each reaction mixture, DNA polymerase catalyzes the
incorporation of deoxynucleotides or dideoxynucleotides into the growing complementary
DNA chains. Deoxynucleotides are present at a higher concentration than dideoxynucleotides
so that the incorporation of the dideoxynucleotide does not dominate the reaction. If a
dideoxynucleotide gets randomly incorporated into the DNA sequence, the chain elongation
reaction is terminated because the dideoxynucleotide lacks a 3′ hydroxyl group (i.e., a
phosphodiester bond cannot form with the 5′carbon of the next incoming nucleotide). The
random incorporation of the dideoxynucleotides into the growing DNA chains results in
chain termination and produces different sized fragments.
Each individual reaction is separated by size using electrophoresis, traditionally on
a denaturing gel electrophoresis and more recently by capillary gel electrophoresis.
The separation of the chain-terminated sequences in the four separate mixtures by

electrophoresis determines which dideoxynucleotide was incorporated. Because the size
of the fragments differs by the addition of the dideoxynucleotides, the DNA sequence
can be reconstructed from the size of the gel fragments. In part B of this figure, the first
dideoxynucleotide incorporated is ddATP, followed by ddCTP, then ddGTP, etc., resulting in
the final sequence ACGTTC.
Next-Generation Sequencing
Like many of the genotyping technologies discussed in this chapter, DNA sequencing has
become high throughput in nature. Many researchers are moving away from the traditional
Sanger method of DNA sequencing and are opting to use newer, more sophisticated instru-
ments.36,39,92 In general, these newer instruments (e.g., Illumina’s HiSeq, MiSeq, and NovaSeq
sequencers; and Complete Genomics Revolocity sequencer) have circumvented the chal-
lenges associated with older sequencing instruments by eliminating the need for individual
DNA sample preparation and electrophoretic gel separation.36,39,93 Instead, many of these
NGS technologies are carried out on solid phase support structures (e.g., beads, or a solid
surface) with nongel detection systems that allow several hundred thousand sequenc-
ing reactions to be carried out in parallel. This compares to only 96 to 384 sequencing
reactions with traditional Sanger sequencing.36,39,92-95 As a result, NGS systems are being
used in whole genome resequencing efforts because of their high-throughput capabilities,
improved depth of sequencing coverage, low cost, and high resolution.36,37,39,65,92-94
Although multiple platforms are used for NGS, there are essentially three primary
approaches for sequencing: (1) sequencing by synthesis, (2) sequencing by ligation, and (3)
single molecule sequencing. Sequencing by synthesis includes pyrosequencing96-99 and
sequencing with reversible chain termination.36-39,100 In addition, these approaches utilize
various detection methods including fluorescence, photon (light) detection, ion sensing
by semiconductors,39,58 and ion current sensing by biological Nanopore.75,101-104 Table 3-4
summarizes these various sequencing technologies, and a more detailed description of
some of these methods follows.
Table 3-4
Comparison of NGS Platforms
NGS Instrument
Platform Amplification Method Sequencing Method Detection
Complete Genomics Rolling circle PCR Unchained ligation Fluorescence
Revolocity™ (DNA nanoball)
Illumina MiSeq™, Solid-phase bridge Reversible chain Fluorescence
NextSeq™, amplification termination
HiSeq™, HiSeqX™,
NovaSeq™
Pacific Biosciences None Sequencing by Fluorescence
SMRT® synthesis
Roche/454™ emPCR Pyrosequencing Light (photon)
sequencers GS detection (chemi-
Junior,® GS Junior+® luminescence)
and GS-FLX+®
Thermo Ion Torrent™ emPCR Sequencing by H+ ion sensing by
synthesis semi-conductor
Thermo SOLiD® emPCR Chained ligation Fluorescence

Oxford Nanopore ®
None Depolymerization Biological electrical
and cleavage of sensing
nucleotides
DNA, deoxyribonucleic acid; NGS, next-generation sequencing; PCR, polymerase chain reaction.
SEQUENCING BY SYNTHESIS. The basic principle behind sequencing by synthesis NGS tech-
niques is that by adding nucleotides that are complementary to a DNA template of interest
using a polymerase, as happens during replication, you can then determine the order of the
sequence of the DNA template. This is the same basic premise on which Sanger sequencing
is based, but the techniques have evolved using various approaches toward the addition and
detection of the added nucleotides. The two primary approaches use either pyrosequencing
or cyclic reversible chain termination, but others that measure the synthesis more directly,
such as Ion Torrent™ (Thermo Scientific), are becoming more popular.
Pyrosequencing—Uses primer extension/polymerase-mediated synthesis to generate
sequence complementary to DNA using the capture of light to detect the nucleotides that
are added sequentially. Pyrosequencing is sometimes referred to as real-time sequenc-
ing by synthesis because it uses some of the concepts of traditional DNA sequencing and
provides information about the polymorphism and the surrounding sequence in the geno-
typing reaction.96,97,99,105
Prior to performing pyrosequencing, the target sequence of DNA that contains the
polymorphism of interest is amplified, typically by emulsion PCR. The DNA target library
(attached to beads, Figure 3-4) is then denatured to form single-stranded PCR molecules,
and the beads are loaded into a reactor that contains wells (one bead per well) where the
pyrosequencing reaction takes place. An illustration of the chemistry used in the pyro-
sequencing method is summarized in Figure 3-6.57,96,99,106
The first step in the pyrosequencing method is a primer extension reaction whereby
a sequencing primer is hybridized to the single-stranded PCR product either next to or a
few bases upstream of the polymorphism of interest. Next the reaction mixture is placed in
an automated instrument, and DNA polymerase and nucleotides are added to initiate the
sequencing reaction. With the pyrosequencing method, nucleotides are added in a sequen-
tial fashion based on the known DNA template sequence.57,99,107,108 If an added nucleotide is
complementary to the template sequence, it is incorporated by DNA polymerase into the
complementary DNA strand, starting from the 3′ end of the hybridized sequencing primer.
When a nucleotide is incorporated into the DNA sequence, a pyrophosphate (PPi) molecule
is released. This release of PPi drives a series of enzymatic reactions that result in the
production of visible light.57,96,99,109 Specifically, the pyrophosphate molecule reacts with
ATP sulfurylase in the reaction mixture and produces ATP. The ATP then reacts with lucif-
erase enzyme and D-luciferin, and light is produced. This light is captured on a camera and
appears as a peak on the pyrosequencing output (also called a pyrogram). If the nucleotide
that was added does not bind complementary to the template sequence, it is degraded by
another enzyme in the reaction mixture called apyrase. Since a molecule of PPi is released
for each nucleotide incorporated, the amount of light produced during the pyrosequencing
reaction is proportional to the number of nucleotides that have been incorporated into the
DNA strand. The process of nucleotide addition and degradation is repeated in an itera-
tive fashion, culminating in the synthesis of a DNA strand that is complementary to the
template sequence of interest.98 As such, the pyrogram provides real-time sequencing data
regarding the polymorphism of interest and the surrounding DNA sequence. An example of
a patient pyrogram is shown in Figure 3-7.7

The pyrosequencing method relies on pattern recognition software to automati-
cally assign a genotype for the polymorphic site of interest.98 Specifically, the software
compares the light pattern and peak heights produced on the pyrogram to the light
pattern and peak heights of a theoretical histogram (i.e., the expected light pattern and
Pyrosequencing
Excess Apyrase
Polymerase dNMP, AMP + Pi
5’ 3’ dNTP, ATP
3’G C T T A T A C C T A A T C T A 5’
+dATP
Luci
ATP feras
5’ e
3’ G C T T A T A C C T A A T C T A 5’
+dTTP PPi PPi
5’ 3’
3’ 5’
+dCTP Light
5’ 3’
3’ 5’
No light
+dGTP
PPi
Light
5’ 3’
3’ 5’
+dATP
PPi
5’ 3’
3’ 5’
… T A - CC T A…
FIGURE 3-6 Principles of pyrosequencing chemistry. In pyrosequencing chemistry,

a sequencing primer is bound complementary to a single-stranded PCR product. In
the presence of DNA polymerase, nucleotides are added in a sequential fashion based
on the known DNA sequence. In this example, dATP was added to the reaction mixture
first, followed by dTTP, dCTP, and dGTP, respectively. As each individual nucleotide is
incorporated, a pyrophosphate (PPi) molecule is released into the reaction mixture. The PPi
reacts with the enzyme ATP-sulfurylase to produce ATP, and the ATP reacts with the enzyme
luciferase to produce light. Light emission and intensity, which is proportional to the number
of pyrophosphate molecules release, are then captured on a camera and appears as a peak on
a pyrogram. If a nucleotide does not bind complementary to the single-stranded PCR product,
it is degraded by the enzyme apyrase and the cycle is repeated with the addition of the next
nucleotide. If no nucleotide is incorporated, then no signal is generated such as the dCTP
reaction shown in the figure. The sequence can be determined based on the complementarity
of the nucleotide added to the mixture depending on the light pattern and the order in which
the nucleotides are added to the mixture. In the figure, because no light was generated when
dCTP was added, the dCTP was not incorporated so the sequence determined is TACCTA.
peak heights based on the known dispensation of nucleotides and the interrogated DNA
sequence).98,109 Based on the match between the observed and the theoretical pyrograms,
the software then scores the quality of the pyrosequencing reaction and assigns a geno-
type for the polymorphism of interest. Thus, the automated pyrosequencing recognition
software removes observer bias, which has often plagued older genotyping methods such
as PCR-RFLP (discussed below).
FIGURE 3-7 Pyrosequencing output. The ATP-binding cassette subfamily G, member 2

gene (ABCG2) contains an SNP (G→A) at nucleotide 34, resulting in an amino acid change
at codon 12 (Val→Met). A patient sample was genotyped for the Val12Met polymorphism in
the ABCG2 gene using pyrosequencing. The resulting pyrosequencing output is shown. The
top panel represents a theoretical histogram. The bottom panel represents a pyrogram from a
patient sample. The nucleotide dispensation order is indicated on the X axis. The peak height,
denoted on the Y axis, indicates the number of incorporated nucleotides. On the top panel,
peak height values of 0.5 suggest heterozygosity, while values of 1 indicate the incorporation
of 1 base (no polymorphism). Letters E and S on the pyrogram denote the addition of enzyme
and substrate to the reaction mixture. The C nucleotides, located before and after the G and A
nucleotides (in the shaded region), have no peaks because they are negative controls assigned
by the software. The sequence analyzed in this example was G/ATGTC. The patient’s genotype
was G/A (Val/Met).
Val = valine; Met = methionine.
Pyrosequencing is considered a medium-throughput genotyping method, with some

users estimating that approximately 1,000 to 2,000 genotypes can be performed per
day.45,105 The turnaround time associated with this method is also relatively fast. For a
96-sample plate, the post-PCR processing step takes between 30 minutes to 1 hour, and
the automated genotyping portion of the assay takes between 10 to 20 minutes.107 The
technology is typically used in the simplex format (i.e., the interrogation of one polymor-
phism in a single well of the plate). However, the technology does allow the user to run
different simplex assays for different polymorphisms on the same plate. Thus, a number
of different polymorphisms can be genotyped in a single pyrosequencing run. Furthermore,
the technology is capable of genotyping multiple polymorphisms in the same reaction
mixture (i.e., multiplex genotyping). As such, pyrosequencing is a relatively versatile tech-
nology for many clinical pharmacogenomic applications.56-58,99,110
Cost is one of this technology’s limitations. The equipment acquisition costs for this
system are expensive and vary significantly, ranging from $75,000 to $500,000.7,55‑58
Furthermore, the labor costs associated with the post-PCR processing step and the
reagent costs for the enzymatic reaction mixture may be prohibitive for some laboratories.7
The technology can also be problematic for sequences with long stretches of the same
nucleotide (i.e., homopolymeric regions). Although pyrosequencing is diminishing in its use

for whole genome sequencing, it continues to be useful for sequencing of short, targeted
sequences (i.e. gene-specific genotyping analysis) and for methylation analysis.112 As such,
pyrosequencing will likely remain a useful genotyping method for medium-throughput
laboratories that genotype a moderate number of polymorphisms.
Cyclic Reversible Chain Termination—Is a sequencing by synthesis technique that

utilizes reversible chain terminators. This chemistry is essentially the same as traditional
Sanger sequencing except that instead of utilizing ddNTPs, the dNTP used terminates
the reaction by covering up the 3′ hydroxl group, but the termination can be reversed by
uncovering the 3′ hydroxyl group so subsequent reactions can proceed. There are two
approaches to utilizing reversible chain terminators: amplification on a solid surface using
bridge amplification (Illumina HiSeq, MiSeq) or single molecule sequencing without ampli-
fication (discussed below under Single Molecule Sequencing).36-39,103,113,114
Bridge amplification (described earlier) is utilized to generate clusters of amplified
target DNA on a solid surface (Illumina). After amplification of the DNA template, a primer is
annealed to the template followed by the addition of DNA polymerase and all four nucleo-
tides, each labeled with a reversible terminator and four different fluorescent dyes. The
nucleotides are allowed to be incorporated through extension of the primer, but since the
3′OH group is covered, the reaction can’t proceed past the one added nucleotide.
After the nucleotides have been incorporated, the excess nucleotides are washed
away, and the fluorescence is imaged to determine which nucleotide has been incorpo-
rated for each individual spot (cluster). The fluorescent label is detected and then cleaved,
followed by the cleavage and reversal of the termination reaction by uncovering the 3′OH
group (Figure 3-8A).36,37,39,103 The cleaved fluorescent label and terminating group are then
washed away, and the cycle is repeated. The readout is automated for each cycle and
analyzed utilizing software to determine the sequences of each spot over multiple cycles
(Figure 3-8B).36,37,39,103 The advantage of this approach over traditional Sanger sequencing
is that separate reactions for each individual nucleotide are unnecessary, and multiple
reactions can be measured at the same time. The entire process can be fully automated,
which significantly increases throughput and decreases the total time to achieve sequenc-
ing.36,37,39,103 Of all of the NGS technologies discussed in this chapter, this one is the most
frequently used, particularly for whole genome sequencing.
Other Polymerase-Mediated Sequencing by Synthesis—Platforms have been devel-
oped that directly detect nucleotide incorporation without the use of fluorescent tags. One
such platform, Ion Torrent™ (Thermo Scientific), detects the incorporation of nucleotides
by measuring the change in pH due to the release of hydrogen ion (H+) with the incorpora-
tion of a nucleotide.39,58,76,77,115 The pH change is measured by semiconductor sensors or Ion
Chip. This technology utilizes emulsion PCR for amplification. Once amplified, the enriched
beads containing the amplified sequence and a sequencing primer are placed into wells on
the Ion Chip. The reactants required for sequencing are added to the well using microfluid-
ics. The bottom of the well directly interfaces with a hydrogen ion detector, and the hydro-
gen ion released with the nucleotide incorporation is detected through a pH change.77 The
sequence is determined as the nucleotides are added separately and sequentially through
the microfluidics that flow across the well. The advantage of this system is that the native
nucleotides are incorporated, and the efficiency of the reaction is not impeded by the
addition of a fluorescent tag to the nucleotide’s structure.58,76,77,115 Similar approaches are
A. Sequencing by reversible chain termination (NGS sequencing)

Cycle 1
Primer
1. Incorporate 3’-OH-protected,
fluorescently-labeled nucleotide (four
Polymeras
dCTP
Repeat
different fluorescent dyes)
dATP
1 2. Wash away unused nucleotides and
e
+ dTTP image fluorescence
Template
dGTP 3. Reverse termination and cleave dye;
wash
Cycle 2
C TG GA
GA
A
Repeat cycles
2 3 1 2 3
B. Output from sequencing by reversible chain termination
Image output
Cycle 1 Cycle 2 Cycle 3 Cycle 4 Cycle 5
Key:
A G Top spot: AGGTC
blue yellow
Bottom spot: GTGCT
C T
green red
FIGURE 3-8 Sequencing by synthesis using reversible chain termination. Reversible

chain termination sequencing utilizes the same basic approach as Sanger sequencing, but
instead of dideoxynucleotides the nucleotides have a protected 3′-hydroxyl group that can
be cleaved to reveal the 3′-hydroxyl group to allow for extension of the chain. (1) With the
amplified DNA template bound to a solid surface, a primer and the four 3′-hydroxy-protected
nucleotides (ddNTP), each with a different fluorescent dye attached (represented by circle,
triangle, star, and diamond in the figure), are added to the reaction mixture along with DNA
polymerase to allow for the extension of the DNA. (2) After the reaction is allowed to proceed,
the unbound nucleotides are washed away and the fluorescence (which is proportional to the
number of nucleotides incorporated) is measured. (3) The fluorescent label is then cleaved off
of the incorporated nucleotide, followed by the cleavage of the 3′-hydroxy-protected group,
and they are washed away. This is followed by the addition of a new batch of 3′-hydroxy-
protected nucleotides, and the cycle is repeated. After several cycles, the sequence is
determined to be the sequence that is complementary to the nucleotides incorporated. In
this figure, the nucleotides incorporated were A, followed by 2xG, T and C, respectively, so
the sequence is determined to be TCCAG (which is complementary to AGGTC). Part A of this

figure depicts the process of incorporation of the nucleotides. Part B of this figure shows the
output of this process and demonstrates how this process is multiplexed by monitoring the
sequencing of multiple DNA sequences at the same time. Each fluorescent spot on the solid
surface represents a separate DNA molecule with a different sequence, so the sequencing can
be highly multiplexed.
under development in which scanning electron microscopy or transistors are utilized for
detection.116,117
SEQUENCING BY LIGATION. Two instrument platforms are currently available that are used
to perform sequencing by ligation. These methods utilize either chained ligation (Thermo
SOLiD) or unchained ligation (Complete Genomics Revolocity™).36,37,58

The SOLiD™ (Sequencing by Oligonucleotide Ligation and Detection) system is
considered a chained ligation in which two bases are coded for each round of analysis. For
this system, the fragmented DNA is amplified using adapter-ligation and emulsion PCR
with magnetic beads.36,37,58 After the DNA sequence fragments have been amplified, the
beads are attached to a solid surface in a flow cell. The unique aspect of this platform
is that two bases are called for with each cycle, and each nucleotide base is analyzed at
least twice, which ultimately increases the accuracy of the analysis.36,37,58 For each cycle,
a primer complementary to the adapter sequence is added, followed by the addition of
octomers. The first two nucleotides at the 3′ end of the octomer are known followed by
three degenerate nucleotides (that can bind multiple, but not all nucleotides) and three
universal nucleotides (that can bind to any nucleotide). A fluorescent dye is attached to
the 5′ end, and a cleavage site is incorporated between the degenerate and universal
nucleotides. A total of 16 octomers, accounting for all possible two nucleotide sequences,
are used. If the first two bases are perfectly aligned, then the ligase will join the octomer
to the primer and the fluorescence is detected. This is then followed by chemical cleav-
age of the fluorescence by cleavage of the last three universal nucleotides. The addition
of octomers and cleavage of the fluorescence is repeated for a number of cycles (usually
10).36,37,58 After the completion of the cycles, the extended product is removed and the
entire procedure is repeated again, but this time using a primer that binds to the n-1 posi-
tion followed by n-2, n-3, etc. Each cycle encodes for every fifth nucleotide and, with the
use of the various primers, the entire sequence ultimately has each nucleotide sequenced
a minimum of two times. The sequence is then decoded by aligning the primer rounds to
determine the sequence.36,37,58
Revolocity™ sequencing (Complete Genomics, a division of BGI, Mountain View, CA) is
considered an unchained ligation-based method.42,43 Unlike the SOLiD system, this system
interrogates one nucleotide at a time and utilizes what has been termed combinatorial
probe-anchor ligation (cPAL) technology. The DNA to be amplified is first fragmented,
four adaptor sequences are inserted, and the DNA is amplified using RCA.42 This process
ultimately produces multiple repeated copies of the target DNA, including four adaptor
sequences, as a long piece of single-stranded DNA with multiple copies in tandem. The
adaptor sequences are palindromic, which allows them to hybridize to each other. As a
result, the long ssDNA with the copies of the target DNA and adapters forms a ball of DNA
termed DNA nanoballs.42 These DNA nanoballs are attached to a solid surface in a flow
cell and sequenced using anchor probes, which bind to the inserted anchor sequences, and
nonamer probes, which consist of nine nucleotides and a dye. Each interrogation probe
contains one known nucleotide at a known position of the nonamer with each nucleotide
encoded by one of four dyes, and remaining nucleotides consist of degenerate nucleotides.
A different nonamer probe is used for each round of sequencing by varying the position of
the known nucleotide.
As each nonamer is added, the probe can be ligated only if the known nucleotide is
complementary to the base being interrogated. The fluorescence is imaged, the ancho-
probe complex is removed, and another nonamer with the known nucleotide in a differ-
ent position is added for the next cycle of hybridization/ligation.42 The advantage of this
system is that each nucleotide call is completely independent of the others and, therefore,
is not affected by the previous cycle.42 The entire procedure is repeated for each adaptor
site until the sequence between each adaptor has been completed and the sequences are
computationally merged by excluding the adaptor sequences. Given the challenges that
this technology has in mapping and assembling the sequence, Complete Genomics does
not provide the instrumentation to laboratories but rather provides the technology as a
sequencing service with all sequencing contained in-house.118
SINGLE MOLECULE SEQUENCING. The disadvantage of utilizing various methods to amplify

the DNA template prior to sequencing is two-fold. First, a bias toward the amplification
of AT-rich and GC-rich sequences can cause important sequences in which variants may
occur to be missed. Second, with PCR amplification, there is a greater chance of introduc-
ing error with the incorporation of a mutation during the amplification process. Single
molecule sequencing, such as Single Molecule, Real Time Technology (Pacific Biosciences),
and Oxford Nanopore (discussed below), overcomes both of these disadvantages and also
simplifies the template preparation process.33,37-39,75,101,103,104,119-122
Pacific Biosciences uses Single Molecule, Real Time (SMRT®) technology for sequenc-
ing in which the nucleotides used for the synthesis are tagged with four different fluores-
cent tags linked to the phosphate of the nucleotides (phospholinked nucleotides).37,38,119
As these nucleotides are incorporated into the DNA, the fluorescent tag is released along
with the release of the phosphate when the nucleotide addition progresses. The unique
feature of this technology is the detection platform system. This technology utilizes a
polymerase tethered to the bottom of a well that also serves as a Zero Mode Waveguide
(ZMW).37-39,119 ZMWs only allow the illumination of light in a small space in the bottom of
the well for detection so that the fluorescence can be measured only as the nucleotide is
being incorporated.37-39,119 This occurs in milliseconds to allow for the recorded detection,
and the fluorescence dissipates when the pyrophosphate is cleaved. This system allows for
the detection of low concentrations of fluorescence because the background fluorescence
from unincorporated nucleotides is essentially nonexistent in the ZMW during detection.
The decreased background fluorescence thereby increases the sensitivity of detection to
levels of DNA that do not require amplification.37-39,119 In addition, the time for sequencing is
reduced because it measures the continuous incorporation of nucleotides. Unlike reversible
chain terminators that stop sequencing before reversing the terminator, the fluorescence
is released (as phospholinked pyrophosphate) during the incorporation of the nucleotide
by polymerase.37-39,119,56,58,76,115
Other single molecule sequencing technologies are evolving, but they are still in their
infancy with regard to their usefulness. These technologies are able to read much longer
sequences of DNA but are plagued by errors in reading the code. Nonetheless, these tech-
nologies are very promising for genome sequencing and have even been used to sequence
microbial genomes, which may be useful where antibiotic resistance is a concern.117,123 In

2012, Oxford released the first in its series of Nanopore™ instruments, MinION, PromethION,
and GridION.124 These instruments utilize protein nanopores that feed the DNA template
through the pore.75,101,103,117 The pore is set in a membrane that passes an ion current through
the protein. As the DNA is fed through the nanopore one DNA base at a time, the change in
the ion current due to the presence of the individual DNA base is measured.75,101,103,104,117 Each
base disrupts the ion current to varying extents, and it is this difference that determines the
base in the sequence. Roche (which acquired Genia Technologies in 2014) is developing a
similar technology, NanoTag, but rather than feeding the DNA strand through the nanopore,
the nanopore has an associated polymerase that forms a protein complex.125 In the Roche
NanoTag sequencing approach, each nucleotide incorporated into a DNA template carries a
unique “NanoTag” that is released and flows through the nanopore as the individual nucle-
otide is incorporated into the DNA template.125 As the NanoTag passes through the pore,
it produces a tag-specific disruption of the ion current that identifies the nucleotide incor-
porated. Other technologies, such as sequencing by expandomer (SBX) that use reporter
molecules and nonbiological nanopores (Stratos Genomics), are also under development
along with other advances in technology that continue to evolve at a rapid pace.117
CLINICAL PEARL
DNA sequencing methods have some error associated with them that
may lead to incorrect assignment of the sequence or genotype. One way to
reduce this error is to perform “deep sequencing” using NGS technologies
in which a specific genomic region is sequenced multiple times—up to
hundreds or thousands of times. This is particularly important for studies
using cell populations that are rare or in low proportion. For example,
mutated cancer cells may be contaminated with normal cells or other sub-
clones of cancer cells.
Enzyme-Based Assays
Enzyme-based methods require the use of an enzyme for discrimination between alleles.
Some of these approaches are modeled after those utilized for DNA sequencing as
discussed above, but others have unique and distinct features. These assays can be clas-
sified into three separate categories: (1) restriction enzyme-based, (2) cleavage-based, and
(3) ligation-based reactions.
PCR-RESTRICTION FRAGMENT LENGTH POLYMOPHISM (PCR-RFLP). One of the oldest

genotyping assays available is PCR coupled with RFLP analysis. The PCR-RFLP genotyping
method uses bacterial restriction enzymes to discriminate between polymorphic and wild-
type alleles.25 Subsequently, this method uses gel electrophoresis to separate the DNA
fragments and either radiolabeled or fluorescently labeled probes to detect the alleles
present in a given sample. The hallmark of RFLP is the use of restriction enzymes, which
recognize specific sequences of nucleotides in DNA, called restriction sites. Restriction
sites are short nucleotide sequences (generally four to eight base pairs in length) that are
present throughout the genome. When a specific restriction enzyme recognizes a specific
restriction site, it cuts the DNA at that site. In pharmacogenomics, investigators are inter-
ested in whether an individual carries a polymorphic or wild-type allele at a particular site
in the DNA. Oftentimes, a specific allele will either result in the presence or absence of a
restriction site. As such, a restriction enzyme can be chosen based on this knowledge.51,66
Subsequently, when the restriction enzyme is mixed with a PCR product (that contains the
target sequence of interest), the restriction enzyme will recognize either the wild-type or
polymorphic allele and cut the DNA at that site. The restriction digest reaction is then elec-
trophoretically separated on a gel, and different fragment patterns are produced based on
the size of the fragments resulting from the digest.46 An individual’s genotype can then be
visually assigned based on these restriction digest fragment patterns (Figure 3-9).7
The PCR-RFLP genotyping technique was used frequently in the pre-genomic era
because it was easy to perform, and the equipment acquisition costs (e.g., gel electro-
phoresis chamber) were low. However, since the completion of the Human Genome Proj-
ect and the emergence of more sophisticated genotyping technologies, PCR-RFLP as a

primary genotyping method has fallen out of routine use. PCR-RFLP is not well suited for
most genotyping laboratories because it is a labor-intensive and high-cost process with
long turnaround and a low throughput; it is not amenable to genotyping more than one
polymorphism at a time (i.e., it is difficult to multiplex RFLP assays). It is also associated
with observer bias during visual analysis of the gel fragment patterns.7,44 Nonetheless,
PCR-RFLP is still used today in clinical laboratories (i.e., PGXL laboratories uses it for the
detection of SLC6A4 polymorphisms), and it can be an appropriate method for laboratories
that genotype small numbers of samples for a few polymorphisms.
CLEAVAGE-BASED ASSAYS. The use of cleavage enzymes, such as flap-endonuclease, is
Lane
1 2 3 4 5
Amplicon size (bp)
603
310
FIGURE 3-9 RFLP-PCR. The Gsα gene (GNAS1) contains a SNP (T→C) at position 393
in exon 5. PCR was performed to generate a 345 base pair product. The C allele creates a
restriction site for the restriction enzyme, FokI. Subsequently, the PCR product was digested
with this enzyme. Lane 1 is the DNA ladder; lane 2 represents a patient with a T/T genotype
(uncut band; 345 bp); lanes 3 and 4 represent patients with the C/C genotype (cut band; 259
bp); and lane 5 represents a patient with a heterozygous T/C genotype (both bands; 345 and
259 bp).
the basis of another enzymatic assay used to discriminate between different alleles. The
advantage of these enzymes is that they can be applied readily to automated, high-through-
put genotyping systems.54,72 An example of one such assay is the proprietary Invader® assay
(Hologic, Inc; originally developed and patented by Third Wave Technologies).49,73,74 The
enzyme that is most commonly used for these types of assays is the proprietary enzyme
Cleavase®. The chemistry of the Invader® assay is shown in Figure 3-10.49 Nonproprietary
versions of this assay are sometimes called Cleavase®/FRET (fluorescence resonance ener-
gy transfer).72 In this technology, two allele-specific probes and one common probe are used
in the reaction mixture. The common probe (also called an invader) forms a three-dimen-
sional structure with the allele-specific probe if it has bound perfectly to the sequence of
interest. An endonuclease enzyme is then added, which recognizes and cleaves the three-
dimensional structure to release a nucleotide flap from the allele-specific probe that is
not bound to the DNA template.54 The nucleotide flap (i.e., overlapping structure) that is
cleaved in the reaction participates in a second reaction where it serves as the invader on
a reporter probe containing a fluorescence at the end of the sequence and a quencher,
bound to another part of the sequence, that suppresses the fluorescence when the probe
is intact. This reporter probe is also known as a FRET cassette.49
This reporter probe is also a sequence that binds complementary to the flap that is
cleaved from the allele-specific probe. The binding of the cleaved flap sequence with the
FRET cassette forms another three-dimensional structure that is also cleaved by endo-
nuclease enzyme (Figure 3-10). This results in the release of fluorescence due to the
cleavage by the endonuclease, which separates it from the quencher and allows it to be
detected. Since the flap was originally released from an allele-specific probe, the genotype
is determined. Notably, the Invader® assay can be used to genotype directly from genomic
DNA without a PCR amplification step.54 The Invader® assay is highly accurate; however, it is
limited by the need for large amounts of template DNA per reaction and, while throughput
is increased, it is difficult to multiplex Invader® assays.49,73,74
LIGATION-BASED ASSAYS. Single-strand breaks in DNA that are produced from damage,
replication, and/or repair of DNA are fused by DNA ligase enzymes. The DNA ligase
enzymes catalyze the reaction that facilitates the formation phosphodiester bond between
two nucleotides in the single-strand break. The DNA ligase enzyme is the basis for the
ligation-based assays, also known as oligonucleotide ligation assays or OLA.41,67,87,126-128
These methods rely on the use of a ligase enzyme to join two primers bound to the DNA
strand. The basic premise for this assay is depicted in Figure 3-11. In most ligation reactions,
two discriminating probes are used to interrogate the polymorphism of interest: one probe
that binds complementary to the sequence containing the polymorphic allele, and the
other probe that binds complementary to the sequence containing the wild-type allele. In
addition to the two allele-specific probes, a probe that is complementary to a sequence
common to both alleles is also used in the reaction.54 The common probe binds next to
the allele-specific probes, regardless of the alleles that are present in the sample. When
the ligase enzyme is added to the reaction, it joins the common probe to the allele-specific
probe that has bound perfectly to the sequence of interest (Figure 3-11).54 The xMAP®
technology (Luminex) is one such commercialized product. Although there are no current
FDA-cleared pharmacogenomic tests from Luminex, they recently received approval for a
respiratory pathogen panel called NxTag® that utilizes the xMAP® technology to determine
a number of respiratory viruses and bacteria for diagnostic purposes.129,130
Cleavase®/FRET or Invader® Assay

Reaction A Reaction B
MT WT
probe probe
A
(allele discrimination)
T T
Primary reaction
Mutant Sequence Mutant Sequence
X
MT probe No cleavase
cleaved activity
Cleaved probe
invades FRET probe
quenched
(secondary allele discrimination &
Q
Secondary reaction
detection)
Cleavase
Fluorescence
signal released
Q
FIGURE 3-10 Principles of the Cleavase®/FRET or Invader® Assay. In the primary reaction,
an allele-specific probe (MT probe or WT probe) and a common probe (invader) bind
complementary to the sequence of interest. If the sequence is complementary to the probe
(Reaction A), a three-dimensional, invasive structure forms. This three-dimensional structure
results due to the single-base overlap between the invader oligo and allele-specific probe.
An endonuclease enzyme (i.e., Cleavase®) is then added, which recognizes and cleaves the
three-dimensional structure to release the nucleotide flap from the allele-specific probe. The
nucleotide flap (i.e., overlapping structure) that is cleaved in the reaction then participates
in a secondary reaction where it serves as the invader. It binds complementary to, and
forms a three-dimensional structure with, one of two fluorescent-labeled molecules in the
reaction mixture in an allele-specific manner to the fluorescence molecule. The fluorescent
molecule is part of a FRET cassette in which the fluorescence is quenched (no fluorescence
is produced). The endonuclease enzyme recognizes and cleaves the three-dimensional
structure that has formed between the flap and the fluorescence molecule. This results in
the release of fluorescence (F), which can then be detected since it is separated from the
quencher (Q). In Reaction B, a second probe (WT probe) and a common probe (invader oligo)
are added to the mixture. However, the WT probe is not specific to the allele present in the
reaction mixture and does not bind complementary to the sequence of interest. As a result,
a three-dimensional structure does not form because there is not a single-base overlap
between the WT probe and the invader oligo. Thus no cleavage occurs when the endonuclease
is added to the mixture, a nucleotide flap is not released, and it cannot participate in the
secondary fluorescence reaction necessary for its detection.
Adapted with permission from Bender LM, Cotton SW, Willis MS. Kids in America: newborn
screening for cystic fibrosis. Labmedicine, 2011; 42(10):595-601. © 2011 American Society
for Clinical Pathology.
Oligonucleotide ligation assay (OLA)
A
Mutant allele
+
NNNN C NNNNNN T
Wild-type probe Mutant probe
NNNN NNNNNN T
A A
+
Mismatch Match → liga on
Ligase
→ NO ligation &
amplification
NNNN
NNNNNN T
A A
NO Amplification → NO signal Fluorescence
G NNNN C
Wild-type allele Wild-type probe
Common probe
A with fluorescence
NNNNNN T
Mutant allele Mutant probe
FIGURE 3-11 Principles of ligation-based assays. Ligation-based assays utilize ligase

enzymes, which connect two oligonucleotides to discriminate alleles. Allele-specific probes
(wild-type or mutant) and common probes, adjacent to the allele being discriminated, are
used that bind to target DNA. If there is a match to the allele, a ligase enzyme connects the
allele-specific probe to the common probe and amplification is able to proceed. Typically, the
common probe is tagged with a signal, such as fluorescence, that allows for detection if the
probes are ligated or a difference in mass may be detected using mass spectrometry.
OLA reactions can be detected in multiple ways. Each allele-specific probe may be
of differing size (i.e., addition of a polymer to the probe) and can be separated by elec-
trophoresis or analyzed by mass spectrometry.79,87 In addition, the common probe could
have a fluorescent tag for detection or each probe may be tagged with a different dye to
distinguish them, such as in Ilumina’s GoldenGate assay.131 Another approach would be to
add a tag with an affinity for another molecule (i.e., biotin-streptavidin) that is attached to a
color-coded bead and use flow cytometry to detect the different beads.132 Alternatively the
probes may be bound to the surface of a microarray, and the extension of bases is achieved
with fluorescently labeled ddNTPs.132 Because the allele-specific probes are of differing
molecular weights or produce different colors, they can be distinguished through separa-
tion by electrophoresis and/or by differential fluorescence. In addition, mass spectrometry
may be used to analyze the difference in the molecular weight between the different
alleles (MassARRAY®, Agena Bioscience™) due to the molecular weight difference of the
allele-specific nucleotide.
Hybridization
DNA MICROASSAYS. In recent years, DNA microarrays have gained significant ground as
a genotyping methodology in pharmacogenomic research. The move to DNA microarrays
has changed the pharmacogenomic landscape in terms of the way researchers address
hypotheses. Traditionally, pharmacogenomic hypotheses have been tested using a candi-
date-gene approach, whereby a select number of genes or polymorphisms are chosen
based on proteins known to influence the pharmacology of drug disposition and drug
response.59 For example, in the treatment of asthma, significant interindividual variability
exists in response to β2-agonists (e.g., albuterol). If one takes a candidate-gene approach
to investigate interindividual variability in albuterol response, the β2-adrenergic receptor
gene (which encodes the β2-adrenergic receptor) would be a logical starting point because
the β2-adrenergic receptor is the target of albuterol’s action. However, with the comple-
tion of the Human Genome Project, many pharmacogenomic researchers are moving away
from the candidate-gene approach and are now taking a genome-wide approach. In the
genome-wide approach, thousands of polymorphisms across the genome are interrogated
for their association with drug disposition or response.59 The move from the candidate-
gene approach to the genome-wide approach is due, in large part, to the availability of DNA
microarrays that utilize allele-specific oligonucleotide (ASO) hybridization for detection.
DNA microarrays are a collection of probes that are bound in a grid-like pattern to
a solid phase support structure (e.g., a nylon membrane, glass slide, silicon chip).133,134 As
such, microarrays are often referred to as gene chips. For genotyping purposes, the probes
that are bound to the chip are oligonucleotides—short fragments of single-stranded DNA
that are designed to bind to specific sequences in the DNA via ASO hybridization (Figure
3-12). For pharmacogenomic applications, these oligonucleotides are designed to inter-
rogate the polymorphisms of interest. Depending on the type of microarray, between
1,000 to 500,000 polymorphisms can be interrogated at one time.59 Several gene chips
are available commercially such as Affymetrix GeneChip® technology (the DMET™ Plus
array probes for nearly 2,000 different polymorphisms for ~230 genes that code for drug-
metabolizing enzymes and drug transporters).135-137 To utilize the Affymetrix DMET™ Plus
array technology, the patient’s DNA sample is digested into smaller pieces, attached to an
adaptor sequence, and then subjected to an initial multiplex PCR-amplification to remove
pseudogene bias and increase specificity using Molecular Inversion Probe (MIP) technol-
ogy.54,138-140 MIP technology utilizes linear probes with sequences at either end of the probe
that can be ligated when hybridized to the target sequence, forming a circular probe that
is then amplified using RCA.140-142 The amplified PCR product is then fluorescently labeled
and hybridized onto the chip. Each polymorphism on the chip is represented by different
ASO probe sets containing a perfect match probe and a mismatch probe.54 Thus, in the
Affymetrix DMET™ Plus array, the allele discrimination method is ASO hybridization with
enhanced specificity using MIP amplification. The binding of the patient’s DNA to a particu-
lar probe results in the release of fluorescence. The fluorescence patterns following probe
binding are analyzed, and a genotype is assigned by automated software.
Another microarray technology used in genome-wide association studies is the Illu-
mina BeadChip platform.143 This platform employs the Illumina BeadArray technology,
Allele-Specific Oligonucleotide Hybridization (ASO)

– i.e. DNA Microarray
DNA oligonucleotide probes

bound to solid surface
Allele discrimination
Fluorescently-labeled
target DNA
oligonucleotides
F F
perfect match mismatch
wash to remove mismatched

oligonucleotides
Detection
F F F F
Measure fluorescently-
F F F F labeled oligonucleotides
F F F F
F = fluorescence
FIGURE 3-12 ASO hybridization – DNA microarray. ASO hybridization techniques, such
as DNA microarray, are based upon the hybridization of target DNA sequences, which
are typically fluorescently labeled, to probes complementary to the sequences bound to a
solid surface. The target DNA is prepared by fragmentation or digestion and amplification,
followed by labeling with a fluorescent dye. The amplified, labeled target DNA sequences
are then added to the solid surface to allow for hybridization. If there is a match, the target
DNA binds tightly to the probe attached to the surface and fluorescence can be detected in
the location of the probe on the solid surface. If there is a mismatch, the DNA does not bind
as tightly and the weakly bound or unbound DNA can be washed away and no signal is
produced. Each location is mapped, so the polymorphism of interest can be identified by
detecting fluorescence at a particular location.
Adapted with permission from Bender LM, Cotton SW, Willis MS. Kids in America: newborn
screening for cystic fibrosis. Labmedicine, 2011; 42(10):595-601. © 2011 American Society
for Clinical Pathology.
whereby 3-micron silica beads are randomly assembled in microwells.54,143 Attached to each
bead are thousands of copies of a specific oligonucleotide sequence that is specific to the
sequence containing the polymorphism to be interrogated. Prior to adding the genomic
DNA to the beads, the sample is subjected to PCR and an allele discrimination reaction
that involves both ASO hybridization and ligation, which includes a fluorescent tag bound
to the allele-specific probe.25 When the processed PCR product is added to the beads, the
oligonucleotide sequences on the beads will bind complementary to the PCR product in an
allele-specific manner, and the information from this reaction is captured by fluorescence
and genotypes are assigned.54
Although the focus of this chapter has been on genotyping methodologies used to
interrogate polymorphisms in the DNA sequence, it is also important to note that micro-
arrays are available to determine the gene expression profiles (messenger ribonucleic acid
[mRNA]) of certain types of cancers (e.g., lymphoid malignancies, breast cancer prognos-
tics).144,145 This information can be used to categorize the cancer and help guide treatment.
Allele-Specific (AS) Amplification
Allele-specific (AS) amplification, also referred to as AS-PCR or AS primer extension
(ASPE), relies on the specificity of a primer to a particular allele sequence and subsequent
PCR amplification to determine if a particular allele is present or not. This is a quantitative
real-time PCR approach that can be utilized for SNPs as well as copy number variations,
because it measures the presence of the gene quantitatively by measuring the expression
of the gene during PCR. The basic premise is that if the primer is not specific, it will not
bind effectively and the DNA will not be amplified. Despite many variations of this method,
in its simplest form, a set of PCR primers are used in which the 3′ end of the primer is
specific for the allele. If the primer is specific for the sequence, a stable duplex is formed,
the extension reaction is allowed to proceed, and a PCR product is produced. If the primer
is a mismatch for the sequence, the extension of the primer is prevented and no product is
formed (Figure 3-13A).68-70,72,132
Variations of this approach include ASPE coupled with allele-specific tags that bind
to beads for flow analysis (ASPE/FLOW), amplification refractory mutation system (ARMS)
and tetra primer ARMS, AS-PCR with blocking agent (ASB-PCR), ddPCR, and isothermal
amplification techniques such as loop-mediated amplification (LAMP).28,30,132 Methods of
detection include electrophoresis, fluorescence, gold nanoparticles, and mass spectrom-
etry. Examples of commercially available assays that employ these approaches include
the GoldenGate™ (discussed above under Ligation-Based Assays) and Infinium™ assays
produced by Illumina and the TaqMan® assay, which is described below.
TAQMAN®. TaqMan® is an automated, medium-to-high throughput genotyping system that

relies on both ASO hybridization and AS amplification as the allele discrimination method.
It uses fluorescence as the allele detection method and is also considered to be a real-
time quantitative PCR (RT-qPCR) method.54,67,146,147 Figure 3-13B provides an illustration of
TaqMan® chemistry.5 The foundation of TaqMan® is the use of fluorescence-labeled probes,
which drive the ASO hybridization reaction.146 These probes are short stretches of nucleo-
tides that are designed to bind complementary to the template sequence of interest. For
biallelic polymorphisms, one probe is designed to bind complementary to the wild-type
allele and one probe is designed to bind complementary to the polymorphic allele. Each
probe is labeled with a reporter dye on the 5′ end and a quencher dye on the 3′ end. The
quencher dye neutralizes the fluorescence of the reporter dye when in close proximity; but
A. Allele-specific amplification
3’ 5’ 3’ 5’
WT MUT
Mismatch + Match
3’ 5’ 3’ 5’
X 3’ 5’
No product
Mutant Genotype
B. TaqMan® assay
R Probe with reporter
Quencher
Primer dye quenched
3’ 5’
dCTP
+ dATP
dTTP
Polymerase
dGTP
Quencher
3’ 5’
R Fluorescence
emitted
Quencher
3’ 5’
FIGURE 3-13 Principles of AS amplification and TaqMan® assays. Both AS-PCR and
TaqMan® assays are considered real-time PCR (RT-PCR) methods. (A) AS-PCR utilizes
ASOs, which serve as primers for amplification using PCR. The product can be detected
using fluorescent-labeled primers. (B) The TaqMan® assay is driven by ASO hybridization
utilizing probes that have fluorescence attached to the 5′-end of the probe and a quencher,
which suppresses the fluorescence, attached the 3′-end of the probe. The TaqMan® probe is
hybridized to the prepared single-stranded target DNA. Primer upstream of the target allele
is then added to the mixture, and PCR is allowed to advance. If the probe matches the allele,
as PCR continues, it displaces the nucleotide at the 5′-end of the probe and releases the
nucleotide with the fluorescence attached. After the nucleotide is released, it is separated
from the quencher and the molecule is allowed to fluoresce and is detected. If the TaqMan®
probe is a mismatch, the nucleotide will not be released and the fluorescence is still
suppressed by the quencher and no signal is generated.
when the reporter dye is separated from the quencher dye, the reporter dye is able to fluo-
resce.148 With this principle in mind, probes are designed such that when intact, the close
proximity of the reporter dye to the quencher dye prevents fluorescence. However, when
the probe is disrupted or cleaved, the reporter dye and quencher dye become separated
and fluorescence is emitted.148
TaqMan® offers an advantage for genotyping since the PCR amplification step and
the allele discrimination step are conducted in the same reaction, rather than in sepa-
rate steps, such as in PCR-RFLP and pyrosequencing.146 As such, TaqMan® includes a PCR
primer and two allele-specific fluorescence-labeled probes in the reaction mixture. During
the annealing step of the amplification process, both the PCR primers and fluorescently
labeled probes bind complementary to the DNA target sequence (Figure 3-13B). The probe
that binds perfectly to the target sequence (i.e., contains no nucleotide mismatches) forms
a stable duplex and the sequence can be extended, whereas the probe that contains a
mismatch does not form a stable duplex and cannot be extended.45,146 During the exten-
sion step of the amplification process, DNA polymerase extends the complementary DNA
strand from the 3′ end of the PCR primer. When the DNA polymerase encounters the probe
that is bound tightly to the strand (i.e., a perfect match), it cleaves the probe at the 5′ end.
The resulting cleavage releases the 5′ reporter dye from the probe, separating it from the
3′ quencher dye, which is still bound to the probe. As a result, fluorescence is emitted from
the reporter dye and subsequently measured. If the PCR primer is not complementary to
the sequence, the sequence will not extend and the 5′ reporter dye remains neutralized by
the 3′ quencher dye and will not fluoresce because the probe remains intact. Software is
used to process the fluorescence data and assign a genotype.
TaqMan® is considered to be a medium-to-high throughput genotyping method with
the capability of determining approximately 1,000 to 10,000 genotypes per day.45 Many
laboratories find the TaqMan® system and similar systems particularly well suited for
pharmacogenomic analyses due to advantages of performing both PCR amplification and
genotype determination into one step.148 This allows for decreased sample processing
times and reduced labor costs. The primary limitation of this method is the high cost of
the fluorescence-labeled probes, which must be designed and optimized for each poly-
morphism.71,148 Thus, for laboratories that want to genotype a large number of different
polymorphisms, the TaqMan® method may be very costly during the assay design and vali-
dation phases. The equipment acquisition costs may also be prohibitive in some settings.
The TaqMan® system uses a real-time PCR thermal cycler, which is more expensive than a
traditional thermal cycler ($30,000 to $100,000 versus $4,000 to $10,000, respectively).7
Single-Base Extension
In addition to pyrosequencing and AS amplification, a number of other primer extension
methods are used for allele discrimination. Primer extension is a versatile tool that can be
coupled with various allele detection methods (e.g., light, mass, fluorescence) to discrimi-
nate between polymorphic and wild-type alleles. In pyrosequencing, the allele detection
method is light, whereas in the MassARRAY® iPLEX system (Agena Bioscience) the allele
detection method is mass. Primer extension can also be coupled with the allele detec-
tion method of fluorescence for genotype determination.35,83,86 An example of this type of
genotyping method is the SNaPshot kit (Applied Biosystems). In the SNaPshot technology,
primers are designed to bind to a PCR product immediately upstream of the polymorphism
of interest.7,149 Fluorescently labeled dideoxynucleotides are added to the reaction mixture,
and these dideoxynucleotides bind complementary to the template to extend the DNA
strand.7 The different-sized extension products, which are fluorescently labeled, are sepa-
rated by capillary gel electrophoresis (i.e., the same type of electrophoresis that is used in
automated DNA sequencing reactions). As the DNA fragments pass through the gel and a
laser, the fragments fluoresce and the output is recorded as peaks on a chromatogram and
genotypes can be determined.

Other Genotyping Methodologies
Although this chapter has focused on the analysis of DNA for genotyping, keep in mind
there are other approaches for looking at variation in human response using molecules
downstream of DNA. These methods may help to identify other sources of variation as a
result of genome changes. RNASeq is being utilized to sequence and examine products of
transcription.94,150 The analysis of mRNA is not just a direct transcript of the genome, but it
may include splice variants that result in alternative protein forms not necessarily predict-
able from DNA sequencing. MicroRNA (miRNA) analysis is also gaining popularity as these
nucleic acid products are involved in the control transcription and translation as well as
other cellular processes.151-153 Proteomics is also a rapidly evolving field that involves the
detection of proteins, their expression levels, and their modification either as the result of
splice variants or from post-translational modification.154,155 Many of these approaches are
currently being adapted for use in the clinic too.
DATA ANALYSIS
Although this chapter will not review the types of data analysis tools available, it is impor-
tant to note that the final step of the process for genotype determination is the analysis
of the data to determine if the genotype is present. At first glance this may appear seam-
less because many technologies provide their own data analysis tools, but it can be very
complex and can be a significant source of error, particularly in the analysis of whole
genome sequencing or highly multiplexed experiments that produce enormous amounts of
data.102,109,136,156-159 Many of these tools are developed in-house as well and may use differ-
ent approaches and algorithms to analyze the final data.
FDA-APPROVED GENOTYPING TESTS IN CLINICAL PRACTICE

In the past, genotyping technologies have largely been limited to academic laboratories.
Oversight of laboratories performing clinical testing have frequently used assays that
have been developed in a kind of “home-brewed” manner, and CMS provides oversight
of clinical diagnostic laboratories provided through their program for CLIA.64 In 2010, the
FDA announced plans to regulate laboratory designed tests (LDTs) due to the complexity
of some genetic tests and the growth in commercial laboratories performing them. The
LDTs fall under the Medical Devices Amendments of 1976 and are regulated by the FDA.
In addition to stand-alone genetic tests, the FDA also regulates companion diagnostic
devices that include a number of genetic tests, particularly for companion diagnostics
for targeted cancer chemotherapies.160 The FDA regulates information about pharmaco-
genomics biomarkers in drug labeling and maintains a list of the drug labels regulated on
their website.161 In addition to CMS and FDA regulation, the Centers for Disease Control and
Prevention has the Genetic Testing Reference Materials Coordination Program for estab-
lishing the process of creating and characterizing reference materials to be used for qual-
ity control and validation of genetic tests, including those used for pharmacogenomics.162
This program characterized 107 genomic DNA reference materials for CYP2D6, CYP2C19,
CYP2C9, VKORC1, and UGT1A1 in 2010 and recently published the results for the character-
ization of reference materials for 28 genes important for pharmacogenomics testing.163,164
Many of these characterized reference materials are available for laboratories to use in
assessing and validating laboratory tests.162

The FDA has approved several pharmacogenomic diagnostic tests for use in clinical
practice. One such diagnostic test, the Roche AmpliChip CYP450 test, is intended to iden-
tify a patient’s CYP2D6 and CYP2C19 genotypes from genomic DNA extracted from a whole
blood sample.10 In the clinic, a patient’s blood sample is drawn, and the specimen is sent to
a participating laboratory with the required equipment to perform the pharmacogenomic
test. The Roche AmpliChip CYP450 test is a DNA microarray that uses Affymetrix gene
chip technology. The DNA microarray is designed to interrogate 27 alleles in the CYP2D6
gene and three alleles in the CYP2C19 gene.10 The five steps involved in the AmpliChip test
are PCR amplification of purified DNA; fragmentation and labeling of the amplified PCR
product; hybridization of the amplified product to the microarray; scanning of the micro-
array; and determination of CYP450 genotype and assignment of a predicted phenotype.10
As discussed previously, numerous oligonucleotide probes, which are designed to be
complementary to the wild-type or polymorphic sequences, are bound to the microarray.
When the patient’s DNA is added to the microarray, the pattern of hybridization to the
specific probes is analyzed and a genotype is determined.10 The genotype information is
compiled by a software program, and a report is provided that summarizes the identified
alleles and genotypes. This information is used to predict an individual’s CYP2D6 and
CYP2C19 metabolizing enzyme phenotype.10 Providers can use this phenotype informa-
tion to help guide drug therapy. In general, the Roche AmpliChip assay takes 1 or 2 days to
complete. Compared to DNA sequencing, the Roche AmpliChip is accurate, with observed
genotype call agreements of 99.2% for CYP2D6 and 99.6% for CYP2C19.10 The interlabora-
tory reproducibility of this assay is reported to be 99.9%.10 Although the Affimetrix DMET™
Plus array described earlier expands the number of genes and polymorphisms that can be
analyzed as compared to the Roche AmpliChip, it has not been approved by the FDA.
Another FDA-approved pharmacogenomic diagnostic test that is available for use in
clinical practice is the Invader® UGT1A1 Molecular Assay (Hologic). The genotyping method
used in this diagnostic test is the Invader® assay (described above). In addition, another
assay is the Infinity® UGT1A1 Assay (AutoGenomics), which utilizes hybridization with a
microarray for detection. UDP glucuronosyltransferase 1A1 (UGT1A1) is an enzyme that is
responsible for the glucuronidation of the active metabolite of the anticancer drug irinote-
can.74,165 Furthermore, UGT1A1 is responsible for the glucuronidation of bilirubin.
Several common polymorphisms exist in the UGT1A1 gene. The most widely studied
polymorphism is a TA repeat in the UGT1A1 promoter. Individuals with seven TA repeats at
this position (designated as the *28 allele) have reduced glucuronidating capacity of the
enzyme compared to wild-type individuals who possess only six TA repeats at this posi-
tion (designated as the *1 allele).74,165 In terms of irinotecan, the active metabolite, SN-38,
is inactivated by UGT1A1 through glucuronidation. Carriers of a UGT1A1*28 allele are less
able to inactivate SN-38, resulting in increased SN-38 plasma concentrations that can be
associated with side effects such as neutropenia and diarrhea.74,165 The intended use of
the Invader® UGT1A1 Molecular Assay is for the detection and genotyping of the *1 and *28
alleles in the UGT1A1 gene in genomic DNA obtained from whole blood, whereas the Infin-
ity® UGT1A1 Assay is for the detection and genotyping of *1, *28, *36, and *37.74,165,166 When
compared to bidirectional DNA sequencing, the Invader® UGT1A1 Molecular Assay has been
shown to be highly accurate, with an observed genotype call agreement of 100%.74 The
interlaboratory reproducibility of this assay is reported to be 98.1%.74 Similarly, using the

same comparator, the observed genotype call is within 98–100% agreement for the Infin-
ity® UGT1A1 Assay.165,166
In addition to the tests described above, multiple companion diagnostic tests are
approved by the FDA for pharmacogenomics and particularly for the detection of somatic
mutations in cancer to ensure the proper therapy is given if particular mutations are pres-
ent.160 An example of such a test is the therascreen® EGFR RGQ PCR Kit (Qiagen Manches-
ter, Ltd) sold as a companion diagnostic for gefitinib (Iressa). This test is a real-time PCR
test used to investigate exon 19 deletions and exon 21 (L858R) substitution mutations in
the epidermal growth factor receptor in non-small cell lung cancer tumors. The screen
detects mutations that may indicate the use of gefitinib. Other companion diagnostics
are used for other cancers and drugs used in oncology, and a list of FDA-cleared or FDA-
approved companion diagnostics can be found on the FDA website.160
CLINICAL PEARL
Genotyping is being used not only to improve outcomes by determining the
host’s genotype, but genotyping of microorganisms is increasingly being
used to determine potential resistance. This is to ensure that the patient
is receiving an effective treatment against the microorganism and also to
further prevent the development of drug resistance in the microorganism.
Many more FDA-approved tests are available for testing the variant strains
of microorganisms than tests for pharmacogenomics with respect to host-
derived pharmacokinetic/pharmacodynamic factors.
SUMMARY
Over the past few years, the field of pharmacogenomics has witnessed an explosion in
the technologies and resources available to conduct genomic analyses. Pharmacogenomic
tests are beginning to move out of the academic laboratory and into mainstream medicine.
Currently, several FDA-approved diagnostic pharmacogenomic tests are available for clini-
cal use. Undoubtedly, in the coming years, genotyping technology for pharmacogenomic
applications will become more sophisticated in nature, cost less, and have higher through-
put. Additionally, as sequencing technologies have continued to mature, the future we have
been anticipating when the cost is only $1,000 to sequence an individual’s entire genome
is coming to fruition. In fact, the National Human Genome Research Institute reported that
the cost in October 2015 was only $1245 (Figure 3-14).167 This cost is less than 5% of the
cost of whole genome sequencing just 5 years prior.167 With the growing availability of
clinical pharmacogenomic diagnostic tests, and as clinical evidence documenting improved
patient outcomes with genotype-guided pharmacotherapy mounts, pharmacogenomics
will serve as a useful tool to aid in the safe and rationale use of drug therapy.
Cost per DNA Sequencing per Genome
$100,000,000
$10,000,000
$1,000,000
Single molecule
sequencing
$100,000
$10,000
$1,000
2000 2005 2010 2015
1st Generation Sequencing Next Generation Sequencing

(Sanger sequencing) (NGS)
FIGURE 3-14 The cost of DNA sequencing per genome from September 2001 through
October 2015. Costs include labor, administration, management, utilities, reagents, and
consumables as well as sequencing instruments and equipment, informatics activities,
submission of data to a public database, and related indirect costs. Data were taken from the
from the NHGRI Genome Sequencing Program website. Available at: http://www.genome.gov/
sequencingcosts/. Accessed December 18, 2015.
REFERENCES
1. Ginsburg GS, Konstance RP, Allsbrook JS, et al. Implications of pharmacogenomics for drug develop-
ment and clinical practice. Arch Intern Med. 2005;165(20):2331-2336.
2. Collins FS, Green ED, Guttmacher AE, et al., Institute USNHGR. A vision for the future of genomics
research. Nature. 2003;422(6934):835-847.
3. Zineh I, Gerhard T, Aquilante CL, et al. Availability of pharmacogenomics-based prescribing informa-

tion in drug package inserts for currently approved drugs. Pharmacogenomics J. 2004;4(6):354-358.
4. Johnson JA. Pharmacogenetics in clinical practice: how far have we come and where are we going?
Pharmacogenomics. 2013;14(7):835-843.
5. Weitzel KW, Alexander M, Bernhardt BA, et al. The IGNITE network: a model for genomic medicine
implementation and research. BMC Med Genomics. 2016;9(1):1.
6. Tsongalis GJ, Coleman WB. Clinical genotyping: the need for interrogation of single nucleotide poly-
morphisms and mutations in the clinical laboratory. Clin Chim Acta. 2006;363(1-2):127-137.
7. Aquilante CL, Zineh I, Beitelshees AL, et al. Common laboratory methods in pharmacogenomics stud-
ies. Am J Health-Syst Pharm. 2006;63(21):2101-2110.
8. Whirl-Carrillo M, McDonagh EM, Hebert JM, et al. Pharmacogenomics knowledge for personalized
medicine. Clin Pharmacol Ther. 2012;92(4):414-417.
9. Chaudhry SR, Muhammad S, Eidens M, et al. Pharmacogenetic prediction of individual variability in
drug response based on CYP2D6, CYP2C9 and CYP2C19 genetic polymorphisms. Curr Drug Metab.
2014;15(7):711-718.
10. Roche Molecular Systems I. AmpliChip CYP450 [test prescribing information]. CE-IVD, 1/2009, Rev. 6.0
ed. Pleasanton, CA: Roche Molecular Systems, Inc.; 2009.
11. Feigelson HS, Rodriguez C, Robertson AS, et al. Determinants of DNA yield and quality from buccal
cell samples collected with mouthwash. Cancer Epidemiol Biomarkers Prev. 2001;10(9):1005-1008.
12. Lench N, Stanier P, Williamson R. Simple non-invasive method to obtain DNA for gene analysis.
Lancet. 1988 Jun 18;1(8599):1356-1358.
13. Philibert RA, Zadorozhnyaya O, Beach SR, et al. Comparison of the genotyping results using DNA
obtained from blood and saliva. Psychiatr Genet. 2008;18(6):275-281.
14. Garcia-Closas M, Egan KM, Abruzzo J, et al. Collection of genomic DNA from adults in epidemiological
studies by buccal cytobrush and mouthwash. Cancer Epidemiol Biomarkers Prev. 2001;10(6):687-696.
15. Lum A, Le Marchand L. A simple mouthwash method for obtaining genomic DNA in molecular epide-
miological studies. Cancer Epidemiol Biomarkers Prev. 1998;7(8):719-724.
16. Rogers NL, Cole SA, Lan HC, et al. New saliva DNA collection method compared to buccal cell collec-
tion techniques for epidemiological studies. Am J Hum Biol. 2007;19(3):319-326.
17. Andrisin TE, Humma LM, Johnson JA. Collection of genomic DNA by the noninvasive mouthwash
method for use in pharmacogenetic studies. Pharmacotherapy. 2002;22(8):954-960.
18. Heath EM, Morken NW, Campbell KA, et al. Use of buccal cells collected in mouthwash as a source of
DNA for clinical testing. Arch Pathol Lab Med. 2001;125(1):127-133.
19. Milne E, van Bockxmeer FM, Robertson L, et al. Buccal DNA collection: comparison of buccal swabs
with FTA cards. Cancer Epidemiol Biomarkers Prev. 2006;15(4):816-819.
20. Hansen TV, Simonsen MK, Nielsen FC, et al. Collection of blood, saliva, and buccal cell samples in a
pilot study on the Danish nurse cohort: comparison of the response rate and quality of genomic DNA.
Cancer Epidemiol Biomarkers Prev. 2007;16(10):2072-2076.
21. Ng DP, Koh D, Choo S, et al. Saliva as a viable alternative source of human genomic DNA in genetic
epidemiology. Clin Chim Acta. 2006;367(1-2):81-85.
22. QIAGEN Genomic DNA Handbook; 2001. Available at: https://www.qiagen.com/us/resources/
resourcedetail?id=97640bc9-e4fe-4c4b-83f6-ac7ca4181597&lang=en. Accessed December 15, 2015.
23. Eisenstein BI. The polymerase chain reaction. A new method of using molecular genetics for medical
diagnosis. N Engl J Med. 1990;322(3):178-183.
24. Markham AF. The polymerase chain reaction: a tool for molecular medicine. BMJ. 1993;306(6875):441-
446.
25. Strachan T, Read AP. DNA structure and gene expression. Human Molecular Genetics. 2nd ed. New
York, NY: Wiley; 1999:1-26.
26. Baumforth KR, Nelson PN, Digby JE, et al. Demystified ... the polymerase chain reaction. Mol Pathol.
1999;52(1):1-10.
27. Miesfeld RL. Applied Molecular Genetics. New York, NY: John Wiley; 1999.
28. Hindson BJ, Ness KD, Masquelier DA, et al. High-throughput droplet digital PCR system for absolute
quantitation of DNA copy number. Anal Chem. 2011;83(22):8604-8610.
29. Tanha HM, Naeini MM, Rahgozar S, et al. Modified tetra-primer ARMS PCR as a single-nucleotide poly-
morphism genotyping tool. Genet Test Mol Biomarkers. 2015;19(3):156-161.
30. Yan L, Zhou J, Zheng Y, et al. Isothermal amplified detection of DNA and RNA. Mol Biosyst.
2014;10(5):970-1003.
31. van Dijk EL, Jaszczyszyn Y, Thermes C. Library preparation methods for next-generation sequencing:
tone down the bias. Exp Cell Res. 2014;322(1):12-20.
32. Li S, Liu HN, Jia YY, et al. A novel SNPs detection method based on gold magnetic nanoparticles array
and single base extension. Theranostics. 2012;2(10):967-975.
33. Zhu Z, Jenkins G, Zhang W, et al. Single-molecule emulsion PCR in microfluidic droplets. Anal Bioanal
Chem. 2012;403(8):2127-2143.
34. Hudecova I. Digital PCR analysis of circulating nucleic acids. Clin Biochem. 2015;48(15):948-956.
35. Gabriel S, Ziaugra L, Tabbaa D. SNP genotyping using the Sequenom MassARRAY iPLEX platform. Curr
Prot Hum Genets / editorial board, Jonathan L. Haines et al. 2009;Chapter 2:Unit 2.12.
36. Mardis ER. Next-generation DNA sequencing methods. Annu Rev Genomics Hum Genet. 2008;9:387-
402.
37. Metzker ML. Sequencing technologies—the next generation. Nat Rev Genet. 2010;11(1):31-46.
38. Metzker ML. Sequencing in real time. Nat Biotechnol. 2009;27(2):150-151.
39. Mardis ER. Next-generation sequencing platforms. Annu Rev Anal Chem. 2013;6:287-303.
40. Ali MM, Li F, Zhang ZQ, et al. Rolling circle amplification: a versatile tool for chemical biology, materi-
als science and medicine. Chem Soc Rev. 2014;43(10):3324-3341.
41. Shin GW, Chung B, Jung GY, et al. Multiplex ligase-based genotyping methods combined with CE.
Electrophoresis. 2014;35(7):1004-1016.
42. Drmanac R, Sparks AB, Callow MJ, et al. Human genome sequencing using unchained base reads on
self-assembling DNA nanoarrays. Science. 2010;327(5961):78-81.
43. Peters BA, Kermani BG, Sparks AB, et al. Accurate whole-genome sequencing and haplotyping from 10
to 20 human cells. Nature. 2012;487(7406):190-195.
44. Aquilante CL, Lobmeyer MT, Langaee TY, et al. Comparison of cytochrome P450 2C9 genotyping
methods and implications for the clinical laboratory. Pharmacotherapy. 2004;24(6):720-726.
45. Chen X, Sullivan PF. Single nucleotide polymorphism genotyping: biochemistry, protocol, cost and
throughput. Pharmacogenomics J. 2003;3(2):77-96.
46. Daly AK. Development of analytical technology in pharmacogenetic research. Naunyn-Schmiedeberg’s
Arch Pharmacol. 2004;369(1):133-140.
47. Ding C, Jin S. High-throughput methods for SNP genotyping. Methods Mol Biol. 2009;578:245-254.
48. Li KC, Ding ST, Lin EC, et al. Melting analysis on microbeads in rapid temperature-gradient
inside microchannels for single nucleotide polymorphisms detection. Biomicrofluidics. 2014 Nov
26;8(6):064109.
49. Olivier M. The Invader assay for SNP genotyping. Mutat Res. 2005;573(1-2):103-110.
50. Ranade K, Chang MS, Ting CT, et al. High-throughput genotyping with single nucleotide polymor-
phisms. Genome Res. 2001;11(7):1262-1268.
51. Shi MM. Technologies for individual genotyping: detection of genetic polymorphisms in drug targets
and disease genes. Am J Pharmacogenomics. 2002;2(3):197-205.
52. Tost J, Gut IG. Genotyping single nucleotide polymorphisms by mass spectrometry. Mass Spectrom
Rev. 2002;21(6):388-418.
53. Kwok PY. Methods for genotyping single nucleotide polymorphisms. Annu Rev Genomics Hum Genet.
2001;2:235-258.
54. Kim S, Misra A. SNP genotyping: technologies and biomedical applications. Annu Rev Biomed Eng.
2007;9:289-320.
55. The AllSeq Knowledge Bank. Available at: http://allseq.com/knowledgebank. Accessed December 16,
2015.
56. NEXT GEN SEEK. Comparing price and tech. specs. of Illumina MiSeq, Ion Torrent PGM, 454 GS Junior,
and PacBio RS. 2012; http://nextgenseek.com/2012/08/comparing-price-and-tech-specs-of-
illumina-miseq-ion-torrent-pgm-454-gs-junior-and-pacbio-rs/. Accessed December 16, 2015.
57. Harrington CT, Lin EI, Olson MT, et al. Fundamentals of pyrosequencing. Arch Pathol Lab Med.
2013;137(9):1296-1303.
58. Liu L, Li Y, Li S, et al. Comparison of next-generation sequencing systems. J Biomed Biotechnol.
2012;2012:251-364.
59. Hernandez-Boussard T, Klein TE, Altman RB. Pharmacogenomics: The relevance of emerging geno-
typing technologies. MLO Med Lab Obs. 2006;38(3):24, 26-30.
60. Isler JA, Vesterqvist OE, Burczynski ME. Analytical validation of genotyping assays in the biomarker
laboratory. Pharmacogenomics. 2007;8(4):353-368.
61. Flockhart DA, O’Kane D, Williams MS, et al. Pharmacogenetic testing of CYP2C9 and VKORC1 alleles
for warfarin. Genet Med. 2008;10(2):139-150.
62. Weiss ST, McLeod HL, Flockhart DA, et al. Creating and evaluating genetic tests predictive of drug
response. Nat Rev Drug Discov. 2008;7(7):568-574.
63. Genetic Testing Task Force. Promoting safe and effective genetic testing in the United States. Final
Report of the Task Force on Genetic Testing 1997. Available at: http://www.genome.gov/10001733.
Accessed December 15, 2015.
64. Clinical Laboratories Improvement Amendment (CLIA). Available at: https://www.cms.gov/Regulations-
and-Guidance/Legislation/CLIA/index.html?redirect=/clia/. Accessed December 16, 2015.
65. Luthra R, Chen H, Roy-Chowdhuri S, et al. Next-generation sequencing in clinical molecular diagnos-
tics of cancer: advantages and challenges. Cancers. 2015;7(4):2023-2036.
66. Gut IG. Automation in genotyping of single nucleotide polymorphisms. Hum Mutat. 2001;17(6):475-492.
67. Jenkins S, Gibson N. High-throughput SNP genotyping. Compar Funct Genomics. 2002;3(1):57-66.
68. Koch WH. Technology platforms for pharmacogenomic diagnostic assays. Nat Rev Drug Discov.
2004;3(9):749-761.
69. Syvanen AC. Accessing genetic variation: genotyping single nucleotide polymorphisms. Nat Rev Genet.
2001;2(12):930-942.
70. Zhang L, Zhao J, Cui G, et al. Genotyping on ALDH2: comparison of four different technologies. PLoS
One. 2015;10(3):e0122745.
71. McGuigan FE, Ralston SH. Single nucleotide polymorphism detection: allelic discrimination using
TaqMan. Psychiatr Genet. 2002;12(3):133-136.
72. Bender LM, Cotten SW, Willis MS. Kids in America: newborn screening for cystic fibrosis. Lab Med.
2011;42(10):595-601.
73. Abou Tayoun AN, de Abreu FB, Lefferts JA, et al. A clinical PCR fragment analysis assay for TA repeat
sizing in the UGT1A1 promoter region. Clin Chim Acta. 2013;422:1-4.
74. Third Wave Technologies. Invader UGT1A1 Molecular Assay [prescribing information]. In: Technologies
TW, ed. Madison, WI; 2005.
75. Feng Y, Zhang Y, Ying C, et al. Nanopore-based fourth-generation DNA sequencing technology.
Genomics Proteomics Bioinformatics. 2015;13(1):4-16.
76. Quail MA, Smith M, Coupland P, et al. A tale of three next generation sequencing platforms: compari-
son of Ion Torrent, Pacific Biosciences and Illumina MiSeq sequencers. BMC Genomics. 2012;13:341.
77. Rothberg JM, Hinz W, Rearick TM, et al. An integrated semiconductor device enabling non-optical
genome sequencing. Nature. 2011;475(7356):348-352.
78. Sauer S, Gut IG. Genotyping single-nucleotide polymorphisms by matrix-assisted laser-desorption/
ionization time-of-flight mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci.
2002;782(1-2):73-87.
79. Gut IG. DNA analysis by MALDI-TOF mass spectrometry. Hum Mutat. 2004;23(5):437-441.
80. Storm N, Darnhofer-Patel B, van den Boom D, et al. MALDI-TOF mass spectrometry-based SNP geno-
typing. Methods Mol Biol. 2003;212:241-262.
81. Lechner D, Lathrop GM, Gut IG. Large-scale genotyping by mass spectrometry: experience, advances
and obstacles. Curr Opin Chem Biol. 2002;6(1):31-38.
82. Gilar M, Fountain KJ, Budman Y, et al. Characterization of therapeutic oligonucleotides using liquid
chromatography with on-line mass spectrometry detection. Oligonucleotides. 2003;13(4):229-243.
83. Mollinari M, Serang O. Quantitative SNP genotyping of polyploids with MassARRAY and other plat-
forms. Methods Mol Biol. 2015;1245:215-241.
84. Wang P, Williams RT, Guerrero CR, et al. Fragmentation of electrospray-produced deprotonated ions
of oligodeoxyribonucleotides containing an alkylated or oxidized thymidine. J Am Soc Mass Spectrom.
2014;25(7):1167-1176.
85. Jordana-Lluch E, Gimenez M, Quesada MD, et al. Improving the diagnosis of bloodstream infections:
PCR coupled with mass spectrometry. Biomed Res Int. 2014(2014);Article ID 501214:8.
86. Millis MP. Medium-throughput SNP genotyping using mass spectrometry: multiplex SNP genotyping
using the iPLEX(R) Gold assay. Methods Mol Biol. 2011;700:61-76.
87. Meyer K, Ueland PM. Use of matrix-assisted laser desorption/ionization time-of-flight mass spectrom-
etry for multiplex genotyping. Adv Clin Chem. 2011;53:1-29.
88. Sanger F, Coulson AR. A rapid method for determining sequences in DNA by primed synthesis with
DNA polymerase. J Mol Biol. 1975;94(3):441-448.
89. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci
U S A. 1977;74(12):5463-5467.
90. Berg JM, Tymoczko JL, Stryer L, et al. Biochemistry. 5th ed. New York, NY: W.H. Freeman; 2002.
91. Strachan T, Read AP. DNA Sequencing. Human Molecular Genetics. 2nd ed. New York, NY: Wiley;
1999:129-134.
92. Schuster SC. Next-generation sequencing transforms today’s biology. Nat Methods. 2008;5(1):16-18.
93. Morozova O, Marra MA. Applications of next-generation sequencing technologies in functional
genomics. Genomics. 2008;92(5):255-264.
94. van Dijk EL, Auger H, Jaszczyszyn Y, et al. Ten years of next-generation sequencing technology. Trends
Genet. 2014;30(9):418-426.
95. Barba M, Czosnek H, Hadidi A. Historical perspective, development and applications of next-generation
sequencing in plant virology. Viruses. 2014;6(1):106-136.
96. Ahmadian A, Ehn M, Hober S. Pyrosequencing: history, biochemistry and future. Clin Chim Acta.
2006;363(1-2):83-94.
97. Ronaghi M. Pyrosequencing sheds light on DNA sequencing. Genome Res. 2001;11(1):3-11.
98. Fakhrai-Rad H, Pourmand N, Ronaghi M. Pyrosequencing: an accurate detection platform for single
nucleotide polymorphisms. Hum Mutat. 2002;19(5):479-485.
99. Nyren P. The history of pyrosequencing®. Methods Mol Biol. 2015;1315:3-15.
100. Metzker ML, Raghavachari R, Richards S, et al. Termination of DNA synthesis by novel 3’-modified-
deoxyribonucleoside 5’-triphosphates. Nucleic Acids Res. 1994;22(20):4259-4267.
101. Bayley H. Nanopore sequencing: from imagination to reality. Clin Chem. 2015;61(1):25-31.
102. Madoui MA, Engelen S, Cruaud C, et al. Genome assembly using Nanopore-guided long and error-free
DNA reads. BMC Genomics. 2015;16:327.
103. Buermans HP, den Dunnen JT. Next generation sequencing technology: Advances and applications.
Biochim Biophys Acta. 2014;1842(10):1932-1941.
104. Haque F, Li J, Wu HC, et al. Solid-state and biological nanopore for real-time sensing of single chemi-
cal and sequencing of DNA. Nano Today. 2013;8(1):56-74.
105. Ronaghi M, Karamohamed S, Pettersson B, et al. Real-time DNA sequencing using detection of pyro-
phosphate release. Anal Biochem. 1996;242(1):84-89.
106. Langaee T, Ronaghi M. Genetic variation analyses by pyrosequencing. Mutat Res. 2005;573(1-2):96-102.
107. Eriksson S, Berg LM, Wadelius M, et al. Cytochrome p450 genotyping by multiplexed real-time DNA
sequencing with pyrosequencing technology. Assay Drug Devel Technol. 2002;1(1 Pt 1):49-59.
108. Alderborn A, Kristofferson A, Hammerling U. Determination of single-nucleotide polymorphisms by
real-time pyrophosphate DNA sequencing. Genome Res. 2000;10(8):1249-1258.
109. de Brevern AG, Meyniel JP, Fairhead C, et al. Trends in IT Innovation to build a next generation bio-
informatics solution to manage and analyse biological big data produced by NGS Technologies.
Biomed Res Int. 2015;2015:904541.
110. Langaee T, Hamadeh I, Chapman AB, et al. A novel simple method for determining CYP2D6 gene
copy number and identifying allele(s) with duplication/multiplication. PLoS One. 2015;10(1):e0113808.
111. GenomeWeb staff reporter. Roche Shutting Down 454 Sequencing Business. Available at: https://
www.genomeweb.com/sequencing/roche-shutting-down-454-sequencing-business. Accessed Decem-
ber 15, 2015.

112. Kreutz M, Schock G, Kaiser J, et al. PyroMark® instruments, chemistry, and software for Pyrosequenc-
ing® analysis. Methods Mol Biol. 2015;1315:17-27.
113. Litosh VA, Wu W, Stupi BP, et al. Improved nucleotide selectivity and termination of 3’-OH unblocked
reversible terminators by molecular tuning of 2-nitrobenzyl alkylated HOMedU triphosphates. Nucleic
Acids Res. 2011;39(6):e39.
114. Metzker ML. Emerging technologies in DNA sequencing. Genome Res. 2005;15(12):1767-1776.
115. Zhao J, Grant SF. Advances in whole genome sequencing technology. Curr Pharm Biotechnol.
2011;12(2):293-305.
116. Mankos M, Shadman K, Persson HH, et al. A novel low energy electron microscope for DNA sequenc-
ing and surface analysis. Ultramicroscopy. 2014;145:36-49.
117. Wang Y, Yang Q, Wang Z. The evolution of nanopore sequencing. Front Genet. 2014;5:449.
118. Reid C. Company profile: Complete Genomics Inc. Future Oncol. 2011;7(2):219-221.
119. Eid J, Fehr A, Gray J, et al. Real-time DNA sequencing from single polymerase molecules. Science.
2009;323(5910):133-138.
120. Mosher JJ, Bowman B, Bernberg EL, et al. Improved performance of the PacBio SMRT technology for
16S rDNA sequencing. J Microbiol Methods. 2014;104:59-60.
121. Pareek CS, Smoczynski R, Tretyn A. Sequencing technologies and genome sequencing. J Appl Genet.
2011;52(4):413-435.
122. Schadt EE, Turner S, Kasarskis A. A window into third-generation sequencing. Hum Mol Genet.
2010;19(R2):R227-240.
123. Judge K, Harris SR, Reuter S, et al. Early insights into the potential of the Oxford Nanopore MinION for
the detection of antimicrobial resistance genes. J Antimicrob Chemother. 2015;70(10):2775-2778.
124. Eisenstein M. Oxford Nanopore announcement sets sequencing sector abuzz. Nat Biotechnol.
2012;30(4):295-296.
125. Kumar S, Tao C, Chien M, et al. PEG-labeled nucleotides and nanopore detection for single molecule
DNA sequencing by synthesis. Sci Rep. 2012;2:684.
126. Homig-Holzel C, Savola S. Multiplex ligation-dependent probe amplification (MLPA) in tumor diagnos-
tics and prognostics. Diagn Mol Pathol. 2012;21(4):189-206.
127. Choi W, Shin GW, Hwang HS, et al. A multiplex single nucleotide polymorphism genotyping method
using ligase-based mismatch discrimination and CE-SSCP. Electrophoresis. 2014;35(8):1196-1203.
128. Macdonald SJ, Pastinen T, Genissel A, et al. A low-cost open-source SNP genotyping platform for
association mapping applications. Genome Biol. 2005;6(12):R105.
129. GenomeWeb staff reporter. Luminex Receives FDA Clearance for NxTag Respiratory Pathogen Panel.
Available at: https://www.genomeweb.com/molecular-diagnostics/luminex-receives-fda-clearance-
nxtag-respiratory-pathogen-panel. Accessed December 18, 2015.
130. Luminex. NxTag® Respiratory Pathogen Panel. https://www.luminexcorp.com/clinical/infectious-
disease/nxtag-respiratory-pathogen-panel/. Accessed December 18, 2015.
131. Lin CH, Yeakley JM, McDaniel TK, et al. Medium- to high-throughput SNP genotyping using VeraCode
microbeads. Methods Mol Biol. 2009;496:129-142.
132. Ye T, Tong R, Gao Z. Genotyping of single nucleotide polymorphisms. In: Volker A, Erdmann SJ,
Barciszewski J, ed. RNA and DNA Diagnostics. Switzerland: Springer International Publishing; 2015:123-
144.
133. Meloni R, Khalfallah O, Biguet NF. DNA microarrays and pharmacogenomics. Pharmacol Res.
2004;49(4):303-308.
134. Villeneuve DJ, Parissenti AM. The use of DNA microarrays to investigate the pharmacogenomics of
drug response in living systems. Curr Top Med Chem. 2004;4(13):1329-1345.
135. Fernandez CA, Smith C, Yang W, et al. Concordance of DMET plus genotyping results with those of
orthogonal genotyping methods. Clin Pharmacol Ther. 2012;92(3):360-365.
136. Guzzi PH, Agapito G, Di Martino MT, et al. DMET-Analyzer: automatic analysis of Affymetrix DMET
Data. BMC Bioinformatics. 2012;13.
137. Burmester JK, Sedova M, Shapero MH, et al. DMET microarray technology for pharmacogenomics-
based personalized medicine. Methods Mol Biol. 2010;632:99-124.
138. Grant SF, Hakonarson H. Microarray technology and applications in the arena of genome-wide asso-
ciation. Clin Chem. 2008;54(7):1116-1124.
139. Maresso K, Broeckel U. Genotyping platforms for mass-throughput genotyping with SNPs, including
human genome-wide scans. Adv Genet. 2008;60:107-139.
140. Translating pharmacogenetics. Treat everyone as an individual. DMET™ Plus Solution. In: Affymetrix
I, ed. P/N CL01351 rev. 2 ed. Available at: http://media.affymetrix.com/support/technical/brochures/
dmet_drug_metabolism_brochure.pdf. Accessed December 18, 2015.
141. Hardenbol P, Baner J, Jain M, et al. Multiplexed genotyping with sequence-tagged molecular inversion
probes. Nat Biotechnol. 2003;21(6):673-678.
142. Absalan F, Ronaghi M. Molecular inversion probe assay. Methods Mol Biol. 2007;396:315-330.
143. Illumina BeadArray Technology. Available at: http://www.illumina.com/technology/beadarray-technology.
html. Accessed December 16, 2015.
144. Gyorffy B, Hatzis C, Sanft T, et al. Multigene prognostic tests in breast cancer: past, present, future.
Breast Cancer Res. 2015;17:11.

145. Song JY, Yu J, Chan WC. Gene expression profiling in non-Hodgkin lymphomas. Cancer Treat Res.
2015;165:97-123.
146. Livak KJ. Allelic discrimination using fluorogenic probes and the 5’ nuclease assay. Genet Anal.
1999;14(5-6):143-149.
147. Livak KJ. SNP genotyping by the 5’-nuclease reaction. Methods Mol Biol. 2003;212:129-147.
148. Livak KJ, Flood SJ, Marmaro J, et al. Oligonucleotides with fluorescent dyes at opposite ends provide
a quenched probe system useful for detecting PCR product and nucleic acid hybridization. PCR Methods
Appl. 1995;4(6):357-362.
149. Sequist LV, Heist RS, Shaw AT, et al. Implementing multiplexed genotyping of non-small-cell lung
cancers into routine clinical practice. Ann Oncol. 2011;22(12):2616-2624.
150. Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet.
2009;10(1):57-63.
151. Goossens-Beumer IJ, Derr RS, Buermans HP, et al. MicroRNA classifier and nomogram for metastasis
prediction in colon cancer. Cancer Epidemiol Biomarkers Prev. 2015;24(1):187-197.
152. Xu L, Qi XL, Duan SL, et al. MicroRNAs: potential biomarkers for disease diagnosis. Biomed Mater Eng.
2014;24(6):3917-3925.
153. Buermans HP, Ariyurek Y, van Ommen G, et al. New methods for next generation sequencing based
microRNA expression profiling. BMC Genomics. 2010;11:716.
154. Zhang Y, Fonslow BR, Shan B, et al. Protein analysis by shotgun/bottom-up proteomics. Chem Rev.
2013;113(4):2343-2394.
155. Yates JR, III. The revolution and evolution of shotgun proteomics for large-scale proteome analysis. J
Am Chem Soc. 2013;135(5):1629-1640.
156. Lamy P, Grove J, Wiuf C. A review of software for microarray genotyping. Hum Genomics. 2011;5(4):
304-309.
157. Potamias G, Lakiotaki K, Katsila T, et al. Deciphering next-generation pharmacogenomics: an informa-
tion technology perspective. Open Biol. 2014;4(7).
158. Kvale MN, Hesselson S, Hoffmann TJ, et al. Genotyping informatics and quality control for 100,000
subjects in the Genetic Epidemiology Research on Adult Health and Aging (GERA) cohort. Genetics.
2015;200(4):1051-1060.
159. Lighten J, van Oosterhout C, Bentzen P. Critical review of NGS analyses for de novo genotyping multi-
gene families. Mol Ecol. 2014;23(16):3957-3972.
160. List of Cleared or Approved Companion Diagnostic Devices (In Vitro and Imaging Tools); 2015. Avail-
able at: http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/InVitroDiagnostics/
ucm301431.htm. Accessed December 16, 2015.
161. Table of Pharmacogenomic Biomarkers in Drug Labeling. Available at: http://www.fda.gov/drugs/science
research/researchareas/pharmacogenetics/ucm083378.htm. Accessed December 18, 2015.
162. Genetic Testing Reference Materials Coordination Program (GeT-RM)—Home. Available at: http://
wwwn.cdc.gov/clia/Resources/GetRM/default.aspx. Accessed December 18, 2015.
163. Pratt VM, Zehnbauer B, Wilson JA, et al. Characterization of 107 genomic DNA reference materials for
CYP2D6, CYP2C19, CYP2C9, VKORC1, and UGT1A1: a GeT-RM and Association for Molecular Pathology
collaborative project. J Mol Diagn. 2010;12(6):835-846.
164. Pratt VM, Everts RE, Aggarwal P, et al. Characterization of 137 genomic DNA reference materials for 28
pharmacogenetic genes: A GeT-RM Collaborative Project. J Mol Diagn. 2016 Jan;18(1):109-123.

165. Infinity UGT1A1 Assay [Directional Package Insert]. In: AutoGenomics I, ed. Vol Doc EM-34028E
(English) Rev. G (CO 2307). Available at: http://www.autogenomics.com/englishpi/
EM-34028E-(English)-G,-Package-Insert-for-UGT1A1.pdf. Accessed December 18, 2015.
166. Langley MR, Booker JK, Evans JP, et al. Validation of clinical testing for warfarin sensitivity: comparison
of CYP2C9-VKORC1 genotyping assays and warfarin-dosing algorithms. J Mol Diagn. 2009;11(3):216-
225.
167. Wetterstrand KA. DNA sequencing costs: Data from the NHGRI Genome Sequencing Program (GSP).
Available at: http://www.genome.gov/sequencingcosts/. Accessed December 18, 2015.
CHAPTER
4
The Pharmacogenetics of Drug
Metabolism
Ali Roberts, PharmD, BCPS; Landry K. Kamdem, PharmD, PhD; and

G. Scott Weston, RPh, PhD

After completing this chapter, the reader ADVERSE DRUG EVENT OR REACTION (ADE
should be able to OR ADR)—a response to a medicine that
is noxious, unintended, and occurs at
Briefly explain the potential health
doses normally used in man.1
impacts of pharmacogenetic differences
influencing drug-metabolizing enzymes CYTOCHROME P450S (CYP450S OR
on the rate of adverse drug reactions or CYPS, MICROSOMAL MIXED FUNCTION
events that occur in patients receiving OXIDASES)—a family of heme-containing
drug therapy. monooxygenase enzymes, many
polymorphic, that are major players in
List and discuss several commonly
drug metabolism.
used therapeutic agents whose safe
and effective usage is impacted by ISOFORM—a protein having a similar function
pharmacogenetic differences affecting and sequence as another protein but
drug-metabolizing enzymes. arising from a different gene (or from a
splice variant of the same gene).
Identify the primary phenotypes, or
metabolizer status, resulting from NARROW THERAPEUTIC INDEX (OR RATIO)
common pharmacogenetic differences (NTI) AGENT—an agent for which there is
that affect drug-metabolizing enzymes. less than a twofold difference in median
lethal dose (LD50) and median effective
Give examples of common medication
dose (ED50) values, or there is less than
dosing adjustments or therapeutic
a twofold difference in the minimum
changes that may be needed due to
toxic concentration and minimum
pharmacogenetic differences impacting
effective concentration in the blood, and
drug-metabolizing enzymes.
for which the safe and effective use of
the drug products require careful titration
and patient monitoring.2
STAR ALLELE—a standardized annotation
nomenclature for denoting genetic
polymorphisms (e.g., CYP2D6*4C).
XENOBIOTIC—a substance foreign to the
human body.
107
INTRODUCTION
A patient’s safe and effective use of therapeutic agents is the goal of every healthcare
professional. The advent of precision, or personalized, medicine allows this goal to be
made even more specific—namely, to determine which patients are most likely to benefit
from a given therapy, which patients may have inappropriate therapies, and whether or not
a therapeutic agent’s typical dosing regimen should be adjusted.3-5 However, patient-to-
patient differences in responses to drug therapy that impact these therapeutic objectives
are common. Studies indicate that the most frequently used pharmaceuticals are effec-
tive in only 25-60% of patients.6 It should also be noted that patients experience adverse
drug events (ADEs) or reactions (ADRs) at rates that are alarmingly high. According to the
National Action Plan for Adverse Drug Event Prevention,7 it has been estimated that in the
United States, ADEs are responsible for one third of the total hospital adverse events,8
impact approximately 2 million hospital stays annually,8,9 and extend the length of hospi-
tal stays by 1.7 to 4.6 days.9-11 The same report also notes that ADRs account for over 3.5
million outpatient office visits,12 as many as 1 million emergency department visits,13 and
approximately 125,000 hospital admissions each year in the United States.13 In addition
to the human costs of ADRs, the economic costs are also significant. National estimates
suggest that ADRs to therapeutic agents add as much as $3.5 billion to U.S. healthcare
costs each year.14
Differences in patient responses to medications, including those differences that lead
to potential ADRs, may arise from many sources, including environmental-, genetic-, and
disease-based factors. Of the many genetic factors that may influence the way patients
respond to therapeutic agents, differences in the enzymes involved in drug metabolism are
known to play a major role. Studies have suggested that the majority of ADEs might be
preventable with the appropriate use of pharmacogenetic profiling of drug-metabolizing
enzymes (DMEs).15-19 For example, antithrombotic agents are involved in most of the fatal
ADRs.20 Warfarin is one of the most commonly used antithrombotic drugs in the world,
but individual patient responses to this agent vary widely. A large part of the interpatient
variability in the response to warfarin therapy is believed to be due to pharmacogenetic
differences in two key polymorphic enzymes: the CYP450 2C9 (CYP2C9, one of the primary
metabolic enzymes responsible for inactivating warfarin) and the vitamin K epoxide reduc-
tase complex subunit 1 (VKORC1, the target through which warfarin exerts its therapeutic
effects).21,22 Increasing evidence suggests the use of pharmacogenetic profiling of these
two polymorphic enzymes, in combination with other clinical data, results in more appropri-
ate warfarin dosing regimens that may reduce potential ADRs.23-25
CLINICAL PEARL
A majority of ADEs or ADRs may be preventable with the appropriate use
of pharmacogenetic profiling, for example, by using CYP2C9 and VKORC1
genotyping in patients prior to the initiation of warfarin therapy to select
optimal dosing regimens.
Several factors complicate our understanding of the impact that pharmacogenetic

differences in DMEs play in a patient’s responses to pharmaceutical therapy. First, metabo-
lism is only one of the many things that occur in vivo on exposure to a therapeutic agent.
Chapter 4 • The Pharmacogenetics of Drug Metabolism 109
Other biochemical processes—including absorption, distribution, and elimination as well

as the sensitivity and number of drug targets available (in addition to other factors such
as disease state, age, drug interactions, etc.)—also have a direct impact on a patient’s
response to drug therapy. For example, one recent study found that genetic variations in
the CYP3A4/5 enzymes accounted for only a minimal (<2%) proportion of the significant
pharmacokinetic variability observed in patients receiving transdermal fentanyl therapy.26

Even when focusing only on the contribution of metabolism to drug reponse variability, it
is important to note that any specific therapeutic agent may be metabolized by a number
of different enzymes in the body. For example, tamoxifen (a common antiestrogen agent
used in the treatment of hormone-dependent breast cancers) is metabolized by the CYP
isoforms 1B1, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1, and 3A4 and also by the flavin monooxy-
genases (FMOs).27,28 Conversely, each of the enzymes involved in human drug metabolism
may be involved in the metabolism of more than one therapeutic agent.27-33 For example,
it is estimated that CYP2D6 is responsible for the metabolism of approximately 15-30% of
the currently used drugs; however, specific literature estimates differ by the date and the
set of drugs evaluated (Figure 4-1).27-29
Another complicating factor is that DMEs such as CYP2D6 are known to be saturable.34
In other words, their capacity to act on substrates can be overwhelmed in cases of high
substrate concentration or low enzymatic activity. Pharmacogenetically based metabolic
differences in drug responses are typically most pronounced in cases where (1) the elimi-
nation of the pharmaceutical agent is primarily due to metabolism; (2) the metabolism of
a drug is primarily due to a single polymorphic enzyme; and (3) major confounding factors
(i.e., co-administration of enzyme inducers/inhibitors, hepatic failure, gastrointestinal block-
age) are absent. These differences are of the most significance to clinicians in situations
3% 10% 1% 1A1
3% 1A2
4%
1B1
2A6
25% 2B6
2C subfamily
15%
2D6
3% 2E1
3A4
FIGURE 4-1 Relative involvement of human CYP450 isoforms in drug metabolism.27

Source: Substrate data from reference 27, Table 6, page 362.
involving certain commonly used therapeutic agents (e.g., codeine for analgesia) or phar-
maceutical agents with a narrow therapeutic index (NTI) (e.g., the anticoagulant warfarin).
CLINICAL PEARL
Pharmacogenetic differences in drug metabolism are of particular

significance to clinicians when they involve certain commonly used
therapeutic agents (e.g., codeine for analgesia) or pharmaceutical agents
with an NTI or ratio (e.g., the anticoagulant warfarin).
CASE STUDY–RESPIRATORY DEPRESSION AND

LETHARGY WITH CODEINE USE
J.M., a 52-year-old male of Arab descent, recently had an outpatient surgical procedure
to repair a partially torn rotator cuff. He received a postoperative prescription for a
combination analgesic containing 300 mg of acetaminophen and 30 mg of codeine
phosphate per tablet with instructions for taking one tablet every 4-6 hours as needed
for pain. Both before and after the procedure, all of J.M.’s vital signs and laboratory values
were within normal range. He reported being on no other over-the-counter or prescription
medications and had no other concomitant diseases. A day and a half after surgery,
J.M.’s spouse reported that, even though he had taken the medication as directed, he was
increasingly lethargic and having some difficulty breathing. An ambulance took J.M. to
the nearest emergency room where he received treatment with naloxone, which quickly
relieved his dyspnea and lethargy.
Questions
1. What is the primary reason that J.M. experienced toxicity from the analgesic medication that
he was prescribed?
2. What is the major human metabolic enzyme involved in the biotransformation of codeine to
the active analgesic morphine?
3. What are the clinical implications for patients who have an ultrarapid phenotype for this
enzyme and are prescribed a codeine-containing product for analgesia?
4. What genotypic or phenotypic tests are currently available for determining the level of activ-
ity of the major codeine biotransformation enzyme present in a patient?
5. What alternative therapeutic options for analgesia would be safer for a patient such as J.M.?
Discussion
Codeine, a narcotic antitussive agent, is also commonly used as an analgesic agent
for the relief of mild-to-moderate pain. However, some patients experience toxicity
when using codeine-based products. The analgesic effects of codeine depend on the
biotransformation (specifically, the O-demethylation) of this compound to morphine. (See
the section on CYP2D6 that follows for additional information.35-37) Patients who have an
ultrarapid 2D6 phenotype produce higher levels of morphine more quickly from codeine,
putting them at risk for opioid toxicity. Common symptoms of opioid toxicity include
respiratory depression and sedation.
ENZYMES INVOLVED IN HUMAN DRUG METABOLISM

Both the enzymes and types of biochemical reactions involved in human drug metabolism
have historically been divided into two major classes or phases: Phase I and Phase II (Table
4-1).28-32 In general, Phase I metabolic reactions involve the oxidative, reductive, or hydro-
lytic exposure or addition of chemical functional groups that frequently act as “handles”
(or attachment sites) for subsequent Phase II conjugation reactions. Phase II reactions, such
as glucuronidation and sulfation, involve the transfer of additional chemical moieties onto
the functional group “handles” present either in the original compound or added in Phase
I. Phase II reactions usually, but not always (e.g., methylation), result in making a compound
more polar or hydrophilic to facilitate its elimination from the body. However, it should be
noted that not all therapeutic agents go through both Phase I and Phase II metabolism.
For example, some pharmaceuticals (e.g., digoxin, isoflurane) do not undergo any significant
metabolism in the human body and are excreted primarily as the unchanged drug, while
others may primarily undergo only Phase I metabolism (e.g., succinylcholine) or only Phase
II metabolism (e.g., lorazepam) but not both (Figure 4-2).38-41
Types of Phase I biotransformations include oxidation, reduction, and hydrolysis reac-
tions (Table 4-1).28-32 Specific enzymes in Phase I include the superfamily, various dehydro-
genases, epoxide hydrolases, esterases, amidases, and the FMO superfamily. The largest
and most notable subgroup of Phase I enzymes is the cytochrome P450 (CYP450) enzymes
(microsomal mixed function oxidases), a superfamily of heme-containing monooxygenases
involved in the metabolism of not only drugs but also other xenobiotics and endogenous
compounds (Figure 4-1).27,49-51 CYPs are widely distributed throughout all areas of the
body, but the highest levels of activity are found in the liver and gastrointestinal tract. At
least 18 different families of these biological catalysts, including more than 50 individual
enzymes, have been identified in humans but only four families (CYP450s 1, 2, 3, and 4) are
currently known to play a major role in drug metabolism (Figure 4-1).27,49-51 Little informa-
tion about the CYP4 family’s role in drug metabolism is available at this time. There are
multiple CYP isoforms, and many individual examples of these enzymes have the ability to
metabolize a wide variety of types of substrates. The nomenclature used for P450s (CYPs
Table 4-1
Enzymes Involved in the Major Phases of Human Drug Metabolism27-33
Phase I Phase II
Alcohol dehydrogenase Glutathione-S-transferases (GSTs)

Aldehyde dehydrogenase Methyltransferases (MTs)
Cytochrome P450s (CYP450s or CYPs) N-acetyltransferases (NATs)
Epoxide hydrolases Sulfotransferases (SULTs)
Esterases UDP-glucuronosyltransferases (UGTs)

Flavin-containing monooxygenases (FMOs)
NADPH-quinone oxidoreductase (NQO)
Oxidoreductases
ABSORPTION METABOLISM ELIMINATION
PHASE I PHASE II
Isoflurane (99.8%
Isoflurane
unchanged) (39)
butyryl–
cholinesterase Inactive Inactive
Succinylcholine
& esterase Metabolites Metabolites

(90%) (40)
UGTs Inactive Inactive

Lorazepam Metabolites
Metabolites
(88%) (41)
~5–10% Reactive
Acetaminophen of dose Metabolite GSTs Inactive Inactive
CYP2E1 (Napqi) Metabolites Metabolites
(2-4%) (42,43)
UGTs Inactive
Inactive
SULTs Metabolites Metabolites
(85%) (42,43)
Ethinyl CYP3A4 Inactive UGTs Inactive
Estradiol Metabolites SULTs Metabolites Inactive
Metabolites
(44–46)
UGTs Active Metabolite Active
Morphine
(~10%) (M6G) Metabolite
(~10%) (M6G)
(47)
Active
CYP2D6 Metabolite Active Metabolite
Fluoxetine (Norfluoxetine)
(Norfluoxetine)
(~7%) (48)
(~7%)
FIGURE 4-2 Potential routes of drug biotransformation using selected examples. The
examples given here are not exhaustive, but they are intended to be illustrative of the many
possible paths of biotransformation taken by different pharmaceutical agents. Percentages,
where noted, refer to the percent of the original dose of the compound administered.
Literature references are noted in parentheses.
GSTs, glutathione-S-transferases; M6G, morphine-6-beta-glucuronide; NAPQI, N-acetyl-p-
benzoquinone imine; SULTs, sulfotransferases; UGTs, UDP-glucuronosyltransferases.
or CYP450s), which is based on gene sequence analysis, uses a number to designate each
family followed by a letter to denote the appropriate subfamily and a number at the end
that identifies the specific gene product (Figure 4-3).50-51 For example, CYP2D6, one of
the best-studied polymorphic members of this metabolic enzyme superfamily, is in CYP
family 2—subfamily D—and was the sixth gene product to be identified within this group. A
standardized annotation nomenclature for denoting genetic polymorphisms, known as the
star (*) allele system, is commonly used to distinguish the different known variants of a
particular gene (Figure 4-3).52
CLINICAL PEARL
Cytochrome P450s (CYPs or CYP450s), the major type of Phase I

biotransformation enzymes, have been studied the most regarding both
potential drug-drug interactions and pharmacogenetic influences on drug
metabolism.
FIGURE 4-3 CYP and star allele nomenclature. Italics are used in referring to the gene.
Phase II metabolic biotransformations involve transfer or conjugation reactions (Figure

4-2).27-33 Most, but not all (e.g., methylation), Phase II reactions result in the formation of
a more polar (or hydrophilic) product that is easier to eliminate in the urine. Metabolic
enzymes included in the Phase II group include the glutathione-S-transferases, methyl-
transferases, the N-acetyltransferases, sulfotransferases, and uridine diphosphate (UDP)-
glucuronosyltransferases (UGTs) (Table 4-1 and Figure 4-2).28-33
SPECIFIC EXAMPLES OF PHARMACOGENETIC DIFFERENCES

IN DRUG METABOLISM
As noted previously, pharmacogenetically based metabolic differences in drug responses
are usually most pronounced in cases where elimination of the pharmaceutical agent is
primarily due to metabolism and metabolism of the drug is primarily due to a single poly-
morphic enzyme. In more practical terms, healthcare providers are mostly concerned with
cases in which these metabolic differences may lead to potentially serious (or even lethal)
outcomes and cases that involve commonly used pharmaceutical agents. In many cases,
pharmacogenetic differences in metabolism lead to a lower rate or absence of enzyme
activity (i.e., a poor metabolizer [PM] phenotype). A lower rate of biotransformation of a
pharmaceutical agent typically causes reduced deactivation and clearance of the agent,
which may lead to higher concentrations of the parent drug being present in the body.
Higher concentrations of the drug in the body may ultimately cause a higher rate of ADEs
or toxicities, especially for agents with a narrow margin of safety, such as warfarin and
the hydantoin anticonvulsants. Alternatively, some genetic differences lead to increased
enzymatic activity (i.e., rapid [RM], ultrarapid [UM], or fast [FM] metabolizer phenotypes).
Examples here include CYP2D6 increased copy number variation and alleles CYP2B6*4,
CYP2C19*17, and NAT2*12.28-32 In FMs, the increased rate of drug clearance may necessitate
frequent or larger doses to achieve optimal therapeutic effects.
CLINICAL PEARL
For drugs converted to less active metabolites, pharmacogenetic variations
in metabolic enzymes that lead to lower rates or absence of drug

metabolism may increase the risk of ADRs for patients and necessitate
reduced dosing regimens. Conversely, in RM phenotypes, the improved
rate of clearance may lead to an increased risk of pharmacotherapy failure
and require higher or more frequent dosing.
At a molecular level, phenotypes that display reduced metabolic enzyme activity are
often a result of amino acid residue changes (including, but not limited to, single nucleotide
polymorphisms [SNPs]), which lead to reduced drug substrate binding, reduced enzymatic
activity, or both. For example, the 2935A>C polymorphism (CYP2D6*7 allele, rs5030867,
1.5% frequency in Caucasians) in CYP2D6 is based on a single nucleotide change that leads
to a complete loss of activity.53 Many other changes including multiple residue mutations,
deletions (of entire genes, domains, or specific residues), splicing defects, crossovers, gene
duplications, and frameshift mutations in the CYP2D6 gene are known to occur. Most,
but not all, of these variations have a similar negative impact on this enzyme’s metabolic
activity. (See reference 50 for a current list.) The distribution of 2D6 alleles that give rise
to reduced or no enzymatic activity varies across ethnic groups. European Caucasians, for
example, have a significant but relatively lower frequency (~27-30%) of reduced or null
activity CYP2D6 alleles compared to patients of Asian and African descent (each ~50%)54,55
(Table 4-2).
At the genetic level, different combinations (e.g., homozygous, heterozygous) of the
various possible alleles of each polymorphic metabolic enzyme are possible. These differ-
Table 4-2
CYP2D6 Phenotypes and Characteristics54-58
Phenotype Frequency Genetic Bases Implications for Pharmaceutical Agents
Activated by 2D6 Deactivated by 2D6
Poor metabolizer ~5-10% No functional Lack of therapeutic Dosage reductions

(PM) alleles present. effects may be may be required.
observed.
Intermediate ~20-40% Either one May show reduced Therapeutics
metabolizer functional and therapeutic effects may be
(IM) one mutant/ effects. achievable at
deficient allele lower-than-
or two partially normal doses.
active/deficient
alleles present.
Extensive ~60-80% Either two active Therapeutic effects Therapeutic effects
or normal alleles or a should be should be
metabolizer combination of achievable with achievable with
(EM) one active and normal dosing. normal dosing.
one partially
active allele
present.
Ultrarapid ~1-5% Three or more Increased risk of Increased risk
metabolizer active alleles toxicity (for any of therapeutic
(UM) present. agents with failure; higher
active, toxic, doses may
or reactive be required
metabolites); to achieve
dosage therapeutic
reductions may effects.
be required.
Listed in order of increasing enzymatic activity. Frequency of occurrence is shown in a sample ethnic population
(Caucasians).
ent genotypic combinations can produce a spectrum of phenotypes, from a lack of enzy-
matic activity (the PM that carries no functional alleles) to enhanced enzymatic activity
(the extensive or normal metabolizer with two functional alleles and the UM with more
than two functional alleles). Those that fall in between (the intermediate metabolizer
[IM]) have a combination of either one defective and one functional allele or one or two
reduced function alleles.54-57 For the CYP2D6 enzyme, at least four different phenotypes
have been identified (Table 4-2), although it should be noted that the specific definitions
for each phenotype may vary according to the scoring system used. Two of the major
scoring systems in current use are those from the Clinical Pharmacogenetics Implementa-
tion Consortium, which recently completed a terminology standardization project, and the
Dutch Pharmacogenetics Working Group.
Although the number of genotypic assays available to profile one or more polymor-
phic DMEs continues to increase, it is important for clinicians to note the current limita-
tions of such genetic tests. First, genotypic testing determines only the patient’s gene
profile (genotype), not the patient’s actual metabolic enzyme activity (phenotype). Also,
genotype assays are aimed at testing for the presence of specific SNPs or alleles, typically
either the most common polymorphisms (e.g., the CYP2D6*4 PM allele found in ~12–20%
of Caucasians) or those that are known to be associated with deficits in enzyme activity.58
Other polymorphisms that the assay does not detect, but may still influence the activity
of the metabolic enzymes being probed, may be present but not detected; the laboratory
may report these SNPs as wild type (or as the *1 allele, in most cases). Finally, commer-
cial clinical tests for many of the polymorphisms that impact drug metabolism are either
not yet available or are still prohibitively expensive, thus limiting access to objective data
that might be used to adjust pharmacotherapeutic regimens. For example, the CYP2B6*4
(A785G) genetic variant is presently one of the only SNPs that is known to increase the
activity of this enzyme.59 However, at this time, currently no commercial genotypic or
phenotypic assays are readily available for this allele. In these cases, however, an indirect
phenotypic approach using either the serum blood levels of a therapeutic agent or a differ-
ent surrogate molecule metabolized by the same polymorphic enzyme (e.g., dextrometho-
rphan as a probe drug for CYP2D6 activity60) may be used as an indirect indicator of the
pharmacogenetic profile of a patient’s DMEs. Alternatively, a representative marker of drug
action (e.g., international normalized ratios for anticoagulants or blood glucose levels for
antidiabetic agents) may be used.
The polymorphic Phase I metabolic enzymes of most clinical relevance include several
members of the CYP450 superfamily (CYP1A2, CYP2B6, CYP2C9, CYP2C19, CYP2D6),
dihydropyrimidine dehydrogenase (DPYD), glucose-6-phosphate dehydrogenase (G6PD),
and butyrylcholinesterase (BChE) (Table 4-3).27-33,49-51,61,62 Phase II polymorphic enzymes of
most clinical importance include N-acetyltransferase 2 (NAT2), the thiopurine S-methyl-
transferase (TPMT), and UGT1A1.27-33,63,64 It should be noted that this is not an exhaustive
list, because many other polymorphic members of both groups of metabolic enzymes are
known and more are being discovered every year.27-33
Phase I Metabolic Enzymes

Cytochrome P450s (CYPs or CYP450s)
The aryl hydrocarbon hydroxylase CYP1A2 (EC 1.14.14.1), although having one of the higher
expression levels of the P450 isoforms in the human liver, is estimated to be involved in
Table 4-3
Examples of Polymorphic Drug-Metabolizing Enzymes of Clinical
Importance27-33
Phase I Phase II
BChE NAT2
CYP1A2 TPMT
CYP2B6 UGT1A1
CYP2C9
CYP2C19
CYP2D6
DPD
BChE, butyrylcholinesterase; DPD, dihydropyrimidine dehydrogenase; NAT2, N-acetyltransferase-2; TPMT, thiopurine-
S-methyltransferase; UGT1A1, UDP-glucuronosyltransferase 1A1.
the metabolism of only about 10% of therapeutic agents.27,65 The 1A2 isoform plays a major
role in the metabolic clearance of the bronchodilator theophylline (N-demethylation and
aromatic hydroxylation), the atypical antipsychotics clozapine (N-demethylation), and olan-
zapine (N-demethylation and aromatic hydroxylation).66-69 With regard to 1A2 substrates,
clozapine is of particular note because it is subject to restricted distribution via a U.S. Food
and Drug Administration (FDA)-mandated Risk Evaluation and Mitigation Strategy due
to the possibility of potentially lethal toxicities. The case of clozapine is complicated by
the fact that this NTI is also metabolized by CYPs 3A4 and 2D6. The FDA has noted the
potential impact of each of these DMEs on clozapine levels in the prescribing information
for clozapine-containing products. CYP450 1A2 preferentially binds substrates that are
polar heterocyclic compounds (e.g., theophylline) and aryl amines (e.g., clozapine, olanzap-
ine).27,70-72
Relatively few variants in the CYP1A2 gene actually linked to phenotypic variabil-
ity have been identified.73 Although more than 20 variant CYP1A2 alleles are currently
known to exist, there are only two major SNPs that have been characterized in humans.33,74
The CYP1A2*1C allele (resulting from the single point mutation −3860G>A) is associated
with decreased metabolic enzyme activity relative to the wild-type *1A allele, while the
CYP1A2*1F allele (resulting from the single point mutation −163C>A) is associated with
increased enzyme induction compared to the wild-type (or non-variant) allele.75 Population
studies done to date indicate that the homozygous and heterozygous *1F allele combina-
tions are more prevalent, suggesting that high CYP1A2 induction is the most common
phenotype here.76
Of particular importance to clinicians is the predominant role that CYP1A2 plays in
the clearance of clozapine, an atypical antipsychotic agent with several potentially fatal
toxicities, including agranulocytosis, seizures, and myocarditis.77,78 Patients who are poor
1A2 metabolizers are potentially at greater risk of clozapine toxicity due to their reduced
clearance of this agent. For example, some studies have highlighted the increased risk
of side effects such as tardive dyskinesia in clozapine-treated patients who have both
reduced CYP1A2 activity (homozygous *1C genotype) and mutations in the D3 dopami-
nergic receptor (although conflicting studies have also been published).79-81 Conversely,
patients with higher-than-normal levels of CYP1A2 activity (due either to the presence of
a high induction phenotype, exposure to a 1A2 inducer such as cigarette smoke, or both)
are at increased risk of treatment failure with clozapine due to their rapid clearance of this
agent.82-84 Another recent study found an association between the homozygous CYP1A2*1F
phenotype and an increased risk of seizures with clozapine use.85
As an alternative to direct genotype testing, several researchers have suggested utiliz-
ing caffeine as a metabolic probe to assess CYP1A2 phenotype because 1A2 is involved in
over 95% of caffeine’s metabolism.66,69,86,87
CLINICAL PEARL
Patients taking clozapine, an atypical antipsychotic agent with several
potentially fatal toxicities, may benefit from CYP1A2 genotypic or
phenotypic testing to modify dosing regimens, if needed. However,
studies have been conflicting, and environmental factors may play a more
important role in CYP1A2 phenotype expression. Exposure to cigarette
smoke, dietary cruciferous vegetables, polyamine hydrocarbons from
grilled meats, and proton pump inhibitors (PPIs) (e.g., omeprazole) have
each been shown to induce CYP1A2. Conversely, oral contraceptives,
fluvoxamine, and fluoroquinolone antibiotics have each been found to
inhibit CYP1A2 activity.88
CYP2B6 (EC 1.14.14.1) is another example of a polymorphic enzyme involved in Phase I

biotransformation reactions. The cytochrome P450 2B6 enzyme is involved in the
metabolism of approximately 4% of current pharmaceuticals, including the antidepres-
sant bupropion, anticancer agents ifosfamide and cyclophosphamide, the opioid analgesic
methadone, and the antiviral non-nucleoside reverse transcriptase inhibitors efavirenz and
nevirapine.27,89-96 The substrate preferences observed for this CYP isoform are for nonpla-
nar structures, usually containing at least one aromatic ring with relatively high lipophilicity
and one or two hydrogen bond-forming groups.70-72 Although less well-studied than some
of the better-known CYP isoforms, the CYP2B6 gene appears to be one of the more poly-
morphic cytochrome genes in man, with well over 100 different SNPs identified to date.97,98
The most frequently occurring variant allele, CYP2B6*6 (which consists of 15631G>T and
18053A>G polymorphisms, rs3745274),99 has a variable distribution across different ethnic
groups ranging from a 14% frequency in Koreans, over 40% in West Africans, and 62% in
Papua New Guineans.97,100,101 Most of the variant alleles identified to date contribute to
lower expression levels and/or lower activity rates of CYP2B6, particularly when present
in a homozygous combination.97 However, at least one SNP, CYP2B6*4, has been linked to
increased CYP2B6 expression in vivo and one SNP, CYP2B6*22, has been shown to increase
2B6 expression in vitro.102
One area of potential clinical importance regarding agents metabolized by CYP2B6 is

the individual variation in the pharmacokinetic parameters of the anticancer agents ifos-
famide and cyclophosphamide.103 Each of these nitrogen mustard derivatives is adminis-
tered as a prodrug dependent on the 2B6 isoform to help create the therapeutically active
species (initially, 4-hydroxycyclophosphamide/aldophosphamide [4-OH-CP] and, subse-
quently, the phosphoramide mustard) in vivo.104,105 Most studies have shown a lower level
of formation of the active 4-OH-CP metabolite and a higher rate of elimination (each of
which may potentially lead to a lower therapeutic benefit) of these agents in patients with
variant CYP2B6 alleles.106 Complicating factors include the number of enzymes involved in
cyclophosphamide metabolism, inducibility of CYP2B6, and relatively small sample sizes
used in most clinical studies to date. However, a recent study identified a clear linkage
between the 2B6 PM phenotype and cyclophosphamide treatment outcomes.107
The polymorphic CYP enzyme 2C9 (EC 1.14.13.80) is the major human isoform of
the CYP2C subfamily, which consists of CYPs 2C8, 2C9, 2C18, and 2C19.50,108,109 Together,
CYP2C9 and CYP2C19 are involved in the metabolism of approximately 25% of currently
used therapeutic agents (Figure 4-1).27 Drug substrates (e.g., phenytoin) for CYP2C9 usually
are weakly acidic and have one or more aromatic rings (Figure 4-4 and Table 4-4).27,70-72
There are two major inherited SNPs of the 2C9 gene, leading to the alleles CYP2C9*2
(430C>T, rs1799853, resulting in the enzyme mutation R144C) and CYP2C9*3 (1075A>C,
rs1057910, resulting in the enzyme mutation I359L), each of which decreases 2C9 func-
tion.108,109 Patients having one of these alleles maintain CYP2C9 activity but at a reduced
rate (i.e., the PM phenotype), and thus may require lower than normal doses of 2C9-metab-
olized agents (especially NTI agents such as the hydantoin anticonvulsants and warfarin) to
optimize therapeutic response and minimize toxicity (Figure 4-5). Population-based stud-
ies indicate that CYP2C9 PM phenotype is more common (~14% frequency) in Caucasian
patients than in Asians or Africans (~4% in each group).108,109
CLINICAL PEARL
Population-based studies indicate that ~14% of Caucasian patients are
deficient in the activity of CYP2C9, a metabolic enzyme that plays a major
role in the biotransformation of several nonsteroidal anti-inflammatory
drugs (NSAIDs), oral sulfonylurea hypoglycemic agents, and NTI agents
such as the hydantoin anticonvulsants and warfarin (Figure 4-4). When
these patients are treated with therapeutic agents metabolized by CYP2C9,
dosage reductions may be required to avoid side effects and toxicity.
CYP2C9 plays a key role in drug metabolism due to its involvement in the biotrans-
formation of more than 100 currently used therapeutic agents, including several oral
sulfonylurea hypoglycemic agents, NSAIDs such as the COX-2 inhibitor celecoxib, angio-
tensin II receptor blockers, and NTI agents such as the hydantoin anticonvulsants and
warfarin (Figure 4-4).21-25,110-117 There are several issues of clinical importance regarding 2C9
polymorphisms, including adverse reactions that are possible with some commonly used
therapeutic agents such as the NSAIDs and sulfonylurea antidiabetic agents. For example,
patients deficient in CYP2C9 activity are up to five times more likely to experience hypo-
glycemia on initiation of antidiabetic pharmacotherapy with sulfonylurea-based drugs at
typical doses due to their decreased clearance of these agents.117 Similarly, patients having
at least one of the two major 2C9 SNPs (resulting in a PM phenotype) have a much higher
risk of gastrointestinal bleeding associated with NSAID therapy.118
Of particular interest to clinicians here are ADEs arising from the use of NTI agents
such as warfarin and hydantoin anticonvulsants in patients with decreased levels of
CYP2C9 activity. Antithrombotic agents, such as warfarin, are involved in a majority of fatal
FIGURE 4-4 CYP2C9-mediated metabolism of phenytoin (top) and celecoxib (bottom) in

man.29,110,114-116
ADRs.20 Warfarin is the most common cause of emergency department visits due to ADEs
for patients who are 65 and older.119 To avoid potential toxicities, dosage reductions for both
warfarin and the hydantoin anticonvulsants may be needed in patients who are deficient
in 2C9 activity. This issue has attracted the attention of the FDA, which in 2007 approved
new pharmacogenetic tests to measure the activity of CYP2C9 (which plays a key role in
the metabolism of warfarin) and VKORC1, the enzymatic target of warfarin in vivo.120 At the
Table 4-4
Selected Examples of CYP2C9 Substrates from Some Examples of
Commonly Used Drugs
Reaction(s) Catalyzed by
Generic Brand Name CYP2C9 (Metabolites Inactive Clinical Implications of
Name (Example) Unless Otherwise Specified) Reduced CYP2C9 Activity
Carvedilol Coreg® O-methylation (of the Probably minimal (2D6 is the

S[-]-isomer)125 major metabolic enzyme for
carvedilol)125
Celecoxib Celebrex® Oxidation of methyl group Greater risk of adverse
to primary alcohol and cardiovascular events122,126
continued oxidation to
carboxylic acid (Figure
4-4)110
Losartan Cozaar® Oxidation of primary alcohol Lower therapeutic effects due
to carboxylic acid (active to lower levels of the active
metabolite [E-3174] that is metabolite111-113
10-40x more potent than
parent compound)111
Montelukast Singulair® Hydroxylation of methyl Greater potential risk of
group127 neuropsychiatric events127
Rosiglitazone Avandia® N-demethylation128 Minimal (2C8 is the primary
isoform involved in
deactivation)128
Valsartan Diovan® 4-hydroxylation129 Minimal (4-hydroxy metabolite
represents only ~9% of
original dose; 80% of
original dose eliminated as
unchanged drug)129
Salt information is not shown. Not all possible reactions for CYP2C9 or for each of the selected compounds are shown.
CYP2C9 may not be the only enzyme involved in the metabolism of the drugs shown here.
same time, the FDA also updated the prescribing information for this drug to encourage
healthcare professionals’ use of these tests to optimize warfarin pharmacotherapy.121 The
FDA also currently recommends that clinicians consider the CYP2C9 phenotype of patients
receiving the COX-2 inhibitor celecoxib, another agent whose metabolic clearance is
dependent on this CYP450 isoform (Figure 4-4).120 Given the known cardiovascular adverse
events of the COX-2 inhibitors that led to the removal of most members of this class of
agents from the market, this advice seems particularly prudent.122
It should be noted that warfarin and celecoxib are not the only therapeutic agents that
include pharmacogenetic information in their labeling. One study of FDA-approved drugs
from 1945–2005 found that 69 different agents from this period contained human genomic
information as a part of the prescribing information, with a majority (62%) of these cases
referring specifically to CYP enzymes.123 The FDA currently lists over 130 drug products on
the market in the United States that contain pharmacogenomic information as part of their
labeling.120 Recent FDA guidance in this area suggests that this trend is likely to increase.124
In spite of the recognized therapeutic importance of genetic variation in this enzyme,
several factors have slowed the adoption of 2C9 genotypic testing by clinicians as a
10
Warfarin dose (mg/day)

0
wt/wt wt/*2 wt/*3 *2/*2 *2/*3 *3/*3
Genotype
FIGURE 4-5 Relationship between CYP2C9 genotype and warfarin dose requirement
for 200 randomly selected Caucasian individuals.57,130 Samples were genotyped only for
CYP2C9*2 and CYP2C9*3. The median dose for each genotype is indicated by the horizontal
bars. A significant difference in dosing requirements between genotypes was found
(p = 0.0002, Kruskal-Wallis test).
Source: Used with permission from Daly AK, King BP. Pharmacogenetics of oral
anticoagulants. Pharmacogenetics. 2003;13:247-252.
standard of care.131 One major factor currently limiting the more widespread adoption
of these tests (and pharmacogenomic tests, in general) is the lack of long-term studies
demonstrating their clinical benefits across a variety of patient groups.132,133 Another factor
is cost. Currently, it may cost up to several hundred dollars to obtain CYP2C9 genotyping
test results from a commercial clinical laboratory for a patient, and insurance coverage
for these tests varies. An additional concern is that 2C9 testing alone is not sufficient for
optimizing the dosing regimens of therapeutic agents, because, in many cases, CYP2C9 is
not the only metabolic enzyme involved in the clearance of these drugs. For example, in
different studies, the CYP2C9 genotype alone accounted for only 10-32% of the observed
variation in the maintenance doses of warfarin.134,135 In this case, it is clear that other infor-
mation, such as a patient’s VKORC1 phenotype or the activity level of other CYP enzymes,
must also be taken into account to more accurately design warfarin dosing regimens.21-25
Additionally, some studies have identified a linkage between CYP2C9 and CYP2C8 genetic
polymorphisms; therefore, CYP2C8 may also contribute to the expression variability
observed for the 2C9 enzyme.136
Another member of the CYP450 2C subfamily that demonstrates polymorphism (with
at least 35 different alleles known) in humans is the 2C19 isoform (ECs 1.14.13.48, 1.14.13.49,
and 1.14.13.80) (Figure 4-1).50,137 CYP2C19 (previously referred to as the S-mephenytoin
hydroxylase) is known to be involved in the biotransformation of a number of commonly
used therapeutic agents, including the antiplatelet agent clopidogrel, hydantoin anti-
convulsants, PPIs, and certain antidepressants (Table 4-5 and Figure 4-6).114-116,138-148 This
CYP isoform displays a preference for aromatic, heteroaromatic, and heteroalkyl drug
substrates, with aromatic hydroxylation and O- and N-dealkylations as prototypical
2C19-catalyzed biotransformation reactions.27,70-72 As with many of the other CYP450
Table 4-5
Selected Examples of CYP2C19 Substrates from Some Examples of
Commonly Used Drugs
Generic Brand Name CYP2C19 (Metabolites Inactive Clinical Implications of
Name (Example) Unless Otherwise Specified) Reduced CYP2C19 Activity
Clopidogrel Plavix® Prodrug is transformed first to Greater risk of cardiovascular

the inactive 2-oxo metabolite events in ACS and PCI
and, subsequently, to the patients due to lower levels
active thiol form139-148 of the active form of the
drug139-148
Escitalopram Lexapro® N-demethylation to Minimal (the 2D6 and 3A4 CYP
S-desmethylcitalopram isoforms also metabolize
(active metabolite with 1/7 escitalopram)149
the activity of the parent
compound)149
Esomeprazole Nexium® O-demethylation150-152 Greater inhibition of gastric acid
secretion and higher H. pylori
eradication rates150-152
Formoterol Foradil® O-demethylation153 Minimal (glucuronidation is the
major metabolic route)153
Salt information is not shown. Not all possible reactions for CYP2C19 or for each of the selected compounds are shown.
CYP2C19 may not be the only enzyme involved in the metabolism of the drugs shown here.
ACS, acute coronary syndrome; PCI, percutaneous coronary intervention.
FIGURE 4-6 CYP2C19-mediated metabolism of omeprazole (or esomeprazole) in man.150-152
enzymes, the major 2C19 polymorphic alleles, CYP2C19*2 (681G>A and 3-4 additional SNPs,
rs4244285) and CYP2C19*3 (636G>A and two additional variant positions, rs4986893),
that have been reported result in a PM phenotype.137 Population-based studies suggest a
higher frequency of the 2C19-deficient phenotype in patients of Asian and Pacific Islander
ancestry (~14% of Chinese and up to 70% of some Pacific Islander populations) than in
other ethnic groups.137 CYP2C19*17 (−806C>T, rs12248560) is currently the only known SNP
associated with increased 2C19 function.154
There are several issues of clinical importance regarding 2C19 polymorphisms. The
most notable of these include potential adverse reactions with the anticonvulsant NTI
agent phenytoin and with the commonly used anticoagulant clopidogrel. For phenytoin
(and also its prodrug fosphenytoin), the clinical implications of being a 2C19 PM are most
pronounced in patients who are also 2C9 deficient or are receiving another medication that
acts as a competitive substrate (or inhibitor) of either CYP 2C9 or 2C19.137 Both CYP2C9
and CYP2C19 are involved in the metabolism and clearance of phenytoin; therefore, dosage
reductions may be required in patients with lower activity (due either to genetic reasons or
competitive substrates or inhibitors) of both of these biotransformation enzymes. Modifi-
cations to dosing regimens may also be needed for patients with PM phenotype for both
CYP2C9 and CYP2D6 who are being treated with tricyclic antidepressants.137
The impact of CYP2C19 polymorphisms on the adverse effects experienced by patients
treated with the widely used antiplatelet agent clopidogrel has received a great deal of
clinical attention.140-148 In part, this is due to the fact that clopidogrel (like warfarin) can
cause serious ADRs that may lead to hospitalization and even death.142,143 Clopidogrel is
administered in the form of a prodrug that requires activation by CYP2C19 to produce the
active form of this agent (Table 4-5).142,143 Thus, patients who are 2C19 PMs may produce
less of the active form of clopidogrel and display resistance to the antiplatelet effects of
this agent.140-148 Given the widespread usage of clopidogrel for a number of different condi-
tions, failure to achieve a full therapeutic effect may have significant deleterious health
consequences for the patient. In particular, the FDA has noted the risks of diminished
effectiveness of clopidogrel in PMs diagnosed with acute coronary syndrome or undergo-
ing percutaneous coronary interventions.
The impact of reduced CYP2C19 activity on the effectiveness of clopidogrel therapy
is magnified with the concurrent use of other agents that compete for this metabolic
enzyme. For example, the PPIs—a family of therapeutic agents frequently used to reduce
the production of stomach acid—act as competitive substrates or inhibitors of the CYP450
2C19 isoform.150,151 Due to the increased risk of gastrointestinal bleeding in patients receiv-
ing anticoagulants such as clopidogrel, PPIs have, in the past, been commonly prescribed
as concurrent therapy for the prevention of gastric ulcers. Increasing clinical evidence
from a number of large studies shows that concurrent use of clopidogrel with omeprazole
or esomeprazole leads to a higher rate of clopidogrel treatment failures.155-157 If an acid-
reducing agent is needed in combination with clopidogrel, pantoprazole appears to be a
safer choice among the PPIs.158 Alternatively, an H2-receptor antagonist, such as famoti-
dine, may also be used.159 The antiplatelet agent prasugrel, although both structurally and
mechanistically related to clopidogrel, does not appear to require CYP2C19 for activation
in vivo.160-162 Thus, prasugrel may represent another viable option for patients who are
deficient in 2C19 activity but require concurrent anticoagulant and gastric acid reduction
therapy. With regard to the concerns that genotype testing increases healthcare costs,
Desta and colleagues have suggested CYP2C19 genotype testing of Asian patients (who
are more likely to be poor 2C19 metabolizers) with gastric or duodenal ulcers could actu-
ally save up to $5,000 for every 1,000 patients tested due to the use of lower doses of the
relatively expensive PPIs to achieve the same therapeutic outcomes.137
Clinicians should be aware that the FDA has also issued guidance suggesting the
usefulness of 2C19 genotyping for patients receiving the azole antifungal voriconazole.120
CLINICAL PEARL
Patients who are deficient in CYP2C19 activity should be warned about
the increased risks of adverse effects with clopidogrel therapy in the
treatment of acute coronary syndrome or following percutaneous coronary
interventions. These risks are magnified with concurrent use of either
of the PPIs omeprazole or esomeprazole. Clinicians should consider
pantoprazole as an alternative PPI or an H2-receptor antagonist (with the
exception of cimetidine) for use with clopidogrel in patients with normal
levels of 2C19 activity. In addition, providers could opt for prasugrel as an
alternative antiplatelet agent in patients who are CYP2C19 PMs or who are
receiving concurrent treatment with an agent that acts as a competitive
substrate or inhibitor of 2C19.
CYP2D6 (EC 1.14.14.1), formerly known as the debrisoquine/sparteine hydroxylase, was

one of the first specific human DMEs identified as being polymorphic.163-166 The P450 2D6
isoform is among the most prolific with regard to the number of different alleles (currently
>100) that have been identified in man.50 Population studies indicate that reduced-function
alleles (such as CYP2D6*9, CYP2D6*10, and CYP2D6*17) are more common in African Amer-
icans and Asians, while completely nonfunctional alleles (such as CYP2D6*3, CYP2D6*4,
CYP2D6*5, and CYP2D6*6) are more common in Caucasians.58 However, with regard to drug
metabolism, the specific phenotype that exists in a patient is dependent on the combina-
tion of alleles present, as previously noted.
CYP2D6 displays a preference for lipophilic amine substrates.27,58 Due to the large
number of drugs that fit this structural pattern, the 2D6 isoform plays a key role in
biotransformation of an estimated 15-50% of commonly used therapeutic agents (Figure
4-1).27-29,58 Specific metabolic reactions catalyzed by CYP2D6 include aromatic hydroxyl-
ation and O- and N-dealkylations.27-29,58 CYP2D6 drug substrates include agents used as
antidepressants, antipsychotics, antihypertensives, opioids, and antiarrhythmics (Table 4-6
and Figure 4-7).167-177 As noted earlier, the multiplicity of CYP2D6 alleles results in the exis-
tence of at least four major phenotypes:54-57,178
Poor metabolizer (PM)—no functional alleles present, resulting in a lack of CYP2D6 activity.
Intermediate metabolizer (IM)—either one functional allele and one mutant/deficient allele
present or two partially active/deficient alleles present, resulting in diminished CYP2D6
activity.
Extensive or normal metabolizer (EM)—either two active alleles or a combination of one active
and one partially active allele present, resulting in normal CYP2D6 activity.
Ultrarapid metabolizer (UM)—three or more active alleles present due to gene duplication,
resulting in higher than normal CYP2D6 activity.
Table 4-6
Selected Examples of CYP2D6 Substrates from Some Examples of
Commonly Used Drugs
Generic Brand Name CYP2D6 (Metabolites Inactive Clinical Implications of
Name (Example) Unless Otherwise Specified) Reduced CYP2D6 Activity
Atomoxetine Strattera® 4-hydroxylation179 Higher rates of adverse

(4-hydroxy-atomoxetine reactions179
reported to be equipotent
to parent compound180),
N-demethylation180
Carvedilol Coreg® 4’-hydroxylation (produces an Greater risk of hypotension due
active metabolite with 13 to decreased clearance
times the β-blocking potency
of the parent compound125)
and 5’-hydroxylation (both
isomers)125
Darifenacin Enablex® Hydroxylation181 (primarily on Minimal (also metabolized by
dihydrobenzofuran ring,182,183 3A4 isoform)
to produce a metabolite with
1/9 the potency of the parent
compound)184
Duloxetine Cymbalta® Hydroxylation of naphthyl Probably minimal (also
ring (at least two active metabolized by 1A2
metabolites reported, with isoform)185
one having 1/5 the potency of
the parent compound)185-189
Olanzapine Zyprexa® Hydroxylation of 2-methyl Minimal (glucuronidation
group190 and 1A2-mediated
N-demethylation are the
major pathways)191
Oxycodone Oxycontin® O-demethylation (to the 40- Potentially lower analgesia195,196
fold192 more active compound (due to less formation of
oxymorphone)193,194 the more active metabolite,
but this is not the major
metabolic pathway)
Venlafaxine Effexor® O-demethylation (to the None (the PM phenotype shifts
equipotent active the ratios of the parent
metabolite desvenla-faxine compound and active
[O-desmethyl-venlafaxine])197 metabolite, but the two
compounds are equipotent
and the total amounts of the
two are similar)197
Salt information is not shown. Not all possible reactions for CYP2D6 or for each of the selected compounds are shown.
CYP2D6 may not be the only enzyme involved in the metabolism of the drugs shown here.
PM, poor metabolizer.
FIGURE 4-7 CYP2D6-mediated metabolism of fluoxetine in man.198
The frequency of CYP2D6 phenotypes varies across different ethnic groups, with the PM
phenotype reported to be more common (~5-14%) in Caucasians, while UMs are found more
frequently among Saudi Arabians (~15-21%) and Ethiopians (~30%).54-57,178,199,200 PMs and UMs
are of the most clinical interest, due to the potential for either treatment failure or adverse
events on exposure to agents metabolized by CYP2D6. Codeine, for example, is a narcotic
antitussive agent that is also commonly used for the relief of mild-to-moderate pain. The
analgesic effects of codeine depend on the 2D6-catalyzed biotransformation (specifically,
the O-demethylation) of this compound to morphine.35-37 Patients who are deficient in 2D6
activity are biochemically unable to adequately convert codeine into the active analgesic
morphine and are, thus, likely to experience treatment failure (a lack of pain relief, in this
case).35-37 When codeine treatment failure occurs, even in the absence of explicit genotype
or phenotype information, clinicians should consider the use of pain-relieving agents that
do not require CYP2D6 activation such as fentanyl (metabolized by CYP3A4), morphine
(which is metabolized primarily by glucuronidation), hydromorphone, oxymorphone, or
diclofenac (which is metabolized by other CYPs and UGT2B7).47,201-203Tramadol should be
avoided, as it is converted to the active metabolite O-desmethyltramadol via CYP2D6.204
Oxycodone and hydrocodone are also converted to active metabolites (oxymorphone and
hydromorphone, respectively) via CYP2D6. Evidence regarding the impact of genetic poly-
morphisms on hydrocodone response is sparse; therefore, hydrocodone may be another
potential alternative in these cases.205
Conversely, as noted in the case study, patients who have the UM phenotype will
convert codeine to morphine more quickly than normal, resulting in a higher level of
exposure to morphine. Even though UMs occur much less frequently than the other
CYP2D6 phenotypes in most ethnic groups (with the apparent exceptions of some Middle
Eastern and North African populations), this is a clinical issue that has resulted in patient
deaths.199,200 For example, nursing mothers ingesting codeine who are 2D6 UMs may inad-
vertently cause the infant to overdose on morphine contained in the breast milk.206,207
Other cases of life-threatening or fatal opioid intoxication in patients due to the ultrarapid
metabolism of codeine have also been reported.208-210
CLINICAL PEARL
Even in the absence of explicit genotype or phenotype information, when

patients experience analgesic treatment failure with codeine, tramadol,
or oxycodone, clinicians should consider the use of pain-relieving agents
that do not require CYP2D6 activation such as fentanyl, morphine,
hydromorphone, oxymorphone, or diclofenac. Hydrocodone may also be
an alternative, but treatment response should be closely monitored.
Another clinically relevant example of a therapeutic agent dependent on the poly-

morphic CYP2D6 enzyme for its bioactivation is the anticancer agent tamoxifen, which is
converted to the major active metabolite (4-hydroxy-N-desmethyl tamoxifen or endoxifen)
in vivo.211 Tamoxifen, an antiestrogen and selective estrogen receptor modulator, remains
the drug of choice for the prevention and treatment of hormone-dependent breast
cancers.212 Unlike other hormonal anticancer agents such as aromatase inhibitors (letrozole,
anastrozole, and exemestane), tamoxifen is used in both pre- and postmenopausal women.
However, tamoxifen is a prodrug and requires bioactivation by the polymorphic CYP2D6
enzyme into a more active metabolite called endoxifen (which is 100-fold more potent
than tamoxifen) (Figure 4-8).213,214 As a consequence, patients with a significantly reduced or
lack of CYP2D6 activity (i.e., individuals homozygous for the CYP2D6 PM allele or patients
taking strong CYP2D6 inhibitors such as antidepressants like fluoxetine or paroxetine)
are at an increased risk for a poor tamoxifen clinical response and treatment failure.215-219
Although there is currently no FDA consensus on CYP2D6 genetic testing as a prerequisite
for tamoxifen treatment, healthcare providers should avoid the concomitant use of strong
CYP2D6 inhibitors with tamoxifen and consider using other agents, particularly aromatase
inhibitors, for the treatment of hormone-dependent breast cancer in postmenopausal
women who are CYP2D6 PMs.220 For premenopausal women who are CYP2D6 PMs, no
proven alternatives to tamoxifen are available; therefore, utilizing a higher dose of this
agent might be considered. Genotypic or phenotypic testing of patients prior to the initia-
tion of tamoxifen therapy may improve therapeutic outcomes by guiding the selection of
appropriate agents and dosing regimens.
CLINICAL PEARL
An increasing number of clinical reports have shown a correlation
between the CYP2D6 PM phenotype and both a shorter time to recurrence
and a worsening of relapse-free survival in breast cancer patients treated
with tamoxifen. Genotypic or phenotypic testing of patients prior to the
initiation of tamoxifen therapy may improve therapeutic outcomes for
patients with reduced 2D6 activity. Concurrent drug therapy should also
be monitored in patients treated with tamoxifen.
For patients being treated with therapeutic agents whose metabolic clearance, rather
than bioactivation (e.g., codeine use for analgesia), is dependent on 2D6 activity (such as
FIGURE 4-8 Biotransformation pathways of tamoxifen to its secondary metabolites,

including endoxifen, which is 100-fold more potent than tamoxifen.
Source: Adapted from reference 221.
the serotonin-selective reuptake inhibitor paroxetine and the tricyclic antidepressants

nortriptyline and desipramine), the potential clinical issues are reversed.222-224 In these
cases, patients who are deficient in CYP2D6 activity will metabolize these agents more
slowly, leading to potential ADEs due to increased drug levels over longer periods of time
in the body. In contrast, UMs are at greater risk of experiencing treatment failures with
these drugs as a result of the rapid clearance of these agents from the body. For drugs
that are converted by the 2D6 isozyme to active metabolites such as bupropion, clopido-
grel, and oxycodone (Table 4-6 and Figure 4-7), the impact of either deficient or overactive
CYP2D6 activity becomes more complicated and is determined by both the relative levels
of the administered drug and active metabolite in the body, as well as the relative activity
ratios of these two compounds.198,225,226
The pharmacology of the active metabolite is also an important consideration. For
example, the antipsychotic risperidone is metabolized to an equally active metabolite,
9-hydroxyrisperidone. Although clinical studies have been conflicting, some reports have
demonstrated an increased risk of adverse effects (e.g., QTc interval prolongation, weight
gain) in poor and intermediate 2D6 metabolizers treated with risperidone. Given multiple
clinical reports of both patient deaths and treatment failures with agents that CYP2D6
converts to active metabolites in vivo, caution in this area is still warranted.227-229 With
regard to specific agents, the FDA currently suggests that 2D6 genotyping may be useful in
patients treated with a wide variety of therapeutic agents, including most of the currently
used antidepressant agents and many cardiovascular agents, among others.120
Other Polymorphic Phase I Enzymes

In addition to the CYP450 enzymes discussed earlier, some other Phase I metabolic
enzymes are known to be polymorphic in man. Among those with current clinical relevance
are DPYD, G6PD, and BChE (Table 4-3).29,61-62,230
DPYD (EC 1.3.1.2) catalyzes the initial and rate-limiting step in the catabolism of pyrimi-
dines (uracil and thymine) of both endogenous and exogenous origin.231 This polymorphic
enzyme is of clinical importance because it plays a key role in the degradation of the
fluoropyrimidine-based anticancer agents 5-fluorouracil (5-FU), floxuridine (fluorodeoxy-
uridine), and capecitabine (each of which is a prodrug that requires bioactivation by other
enzymes in vivo) (Figure 4-9).28,61,232-234 Both floxuridine and capecitabine are derivatives of
5-FU, which are eventually converted to 5-FU in the body.28 DPYD metabolizes more than
80% of the 5-FU ultimately present in the body (regardless of the specific fluoropyrimi-
dine agent administered).235 5-FU is a mainstay in treating numerous types of solid tumors
and is frequently used in combination with other agents (e.g., the FOLFOX [leucovorin/5-
FU/oxaliplatin] and FOLFIRI [leucovorin/5-FU/irinotecan] chemotherapy regimens used in
the treatment of advanced colorectal cancer).236,237 Capecitabine, an orally administered
prodrug of 5-FU, is also seeing increased clinical usage.
Although the number of large-scale pharmacogenomic studies currently available for
the DPYD gene is more limited than the number of CYP gene studies, numerous sequence
variants of the DPYD gene have been identified to date. It appears that the DPYD*2A (the
IVS14+1G>A change at the 5′-splice site of intron 14) mutation is the one most commonly
associated with DPYD deficiency.238,239 Regarding genotypic frequency, it has been esti-
mated that ~3.5% of the population is heterozygous for, and 0.1% is homozygous for, DPYD
alleles with low DPYD activity.240,241 The DPYD phenotype has been found to vary by both
ethnicity and gender.242
Deficiencies in the level of DPYD activity are of clinical importance due to the serious
(and potentially lethal) side effects and toxicities associated with the fluoropyrimidine-
based anticancer agents.61,232-234 Patients with <70% of the normal level of DPYD activity
are considered to be at risk.243 As many as one in three patients receiving 5-FU-based
therapeutics experience dose-limiting ADEs ranging from mucositis, diarrhea, and leuko-
penia to more severe hematological, gastrointestinal, and neurological toxicities.238,244
FIGURE 4-9 The DPYD-mediated metabolism of 5-FU in man.28
Although it is clear that the level of DPYD activity is not the only factor involved in all
5-FU toxicities, multiple reports have demonstrated a strong association between DPYD
deficiency and fluoropyrimidine-related ADRs for a significant number of patients.61,245-253
For example, Raida and colleagues reported that individuals with the DPYD polymorphism
IVS14+1G>A (the DYPD*2A allele) have a sevenfold increased risk of Grade 3 or 4 toxicity
during 5-FU therapy.254 Separately, Morel et al. found that 60% of patients with either
of two SNPs (IVS14+1G>A or 2846A>T) in the DYPD gene experienced early Grade 3 or 4
toxicity on treatment with 5-FU.61 Recognizing this association, the FDA currently suggests
that clinicians evaluate a patient’s DPYD status with regard to fluoropyrimidine therapy.120
However, it is evident that additional, large-scale studies are needed to fully elucidate the
specific factors (including the influence of various genes, gender, epigenetic contributions,
and common concurrent drug therapy regimens) that contribute to and influence fluoro-
pyrimidine toxicity.242,249,255
Conversely, low levels of DPYD activity (along with other enzymes) are associated with
improved responses to 5-FU-based chemotherapy, due to a lower rate of clearance of the
active drug from the body. Salonga and colleagues reported that low expression levels
of DPYD (along with low expression levels of thymidylate synthase and thymidine phos-
phorylase) are positively correlated with the response of colorectal tumors to treatment
with 5-FU.256 Another study showed that DPYD levels were a determinant of capecitabine
treatment response in metastatic colorectal cancer.257 Thus, determination of a patient’s
level of DPYD activity is not only important to assess potential toxicity from 5-FU-based
treatments, but also to evaluate whether or not patients are more likely to respond to
these therapies.
CLINICAL PEARL
For cancer patients eligible for treatment with fluoropyrimidine-based
agents, early determination of DPYD status (through either genotypic or
phenotypic testing) should allow for identification of those individuals at
greatest risk for fluoropyrimidine-associated toxicities as well as patients
who are more likely to positively respond and also assist in subsequent
dose adjustments or in selection of other treatment modalities.
Regarding determination of a patient’s DPYD activity levels, in addition to available

genotypic assays, other more phenotypic and indirect approaches (e.g., use of a C-labeled
uracil breath test and the assay of DPYD activity in peripheral blood mononuclear cells)
have also been proposed as a means of assessing this indicator of potential toxicity and
drug response.18,258,259
The gene for the G6PD enzyme was one of the first to be linked to variability in drug
response.260 It is also remarkably polymorphic, given that more than 300 million people in
the world have a deficiency in G6PD activity.230 This enzyme is expressed in all cells, but
its role is particularly important in red blood cells.261 In fact, it was in red blood cells that
the initial detection of G6PD deficiency occurred several decades ago, due to the hemolytic
anemia caused by exposure of G6PD PMs to the antimalarial drug primaquine.262 It is now
recognized that G6PD deficiency is of concern for not only exposure to primaquine and other
quinine-based agents but also for other commonly used pharmaceuticals (such as aspirin
and other NSAIDs), many sulfa-based agents, and the anti-infective agent nitrofurantoin.255
Interestingly, it appears that the trait of G6PD deficiency has been preserved in populations
where malaria is endemic, as it confers a survival benefit for this tropical disease.263
In terms of clinical application, G6PD deficiency is a more prominent concern with
exposure to quinine-based antimalarial agents. As noted above, the lack of G6PD activity
has been evolutionarily preserved in population groups from geographic locations in which
malaria is prevalent—that is, Africa, Southern Europe, the Mediterranean region, the Middle
East, Southeast Asia, and Oceania. Patients whose ancestry traces back to one of these
people groups or regions should have G6PD activity levels profiled prior to the initiation of
therapy with any quinine-based agent. In addition to the available genotyping assays for
G6PD, many phenotyping assays are available for use; they include the fluorescent spot
test,264 spectrophotometric assay,265 and cytochemical assay.266
BChE (formerly known as pseudocholinesterase) is another example of a non-CYP
Phase I enzyme known to exhibit polymorphism.28-33,62,267 This mutation was among the
earliest variation in DMEs recognized, dating back to the work of Kalow and colleagues in
the 1950s.268-270 Interpatient differences in the activity of BChE are of clinical interest due
to the involvement of this enzyme in metabolism of ester-based therapeutic agents, includ-
ing some neuromuscular blockers (e.g., succinylcholine/suxamethonium), local anesthetics
(e.g., procaine), drugs of abuse (e.g., cocaine), and prodrugs (Figure 4-10).40,271-276 Individuals
who have lower levels of BChE activity are known to experience slower clearance of ester-
based neuromuscular blocking agents (such as succinylcholine/suxamethonium), leading
to prolonged apnea and muscle paralysis when these medications are used.40 Genetic
polymorphism in other key biomolecules, including the ryanodine receptor 1 (RYR1)277 and
calcium channel CACNA1S,278 are also known to impact the toxicity risks of succinylcholine;
therefore, multiple factors should be assessed by clinicians considering the use of this agent.
In addition, evidence suggests that BChE-deficient patients may receive less of a
therapeutic benefit from ester-based prodrugs such as irinotecan.247,279 Finally, individuals
who have a PM phenotype for BChE may be at greater risk for toxicity on exposure to the
ester-containing drug of abuse cocaine; whereas, heroin toxicity may be a concern in indi-
viduals with increased BChE activity.275,280
FIGURE 4-10 The BChE-mediated metabolism of succinylcholine/suxamethonium

in man.40
CLINICAL PEARL
Patients with a PM phenotype for BChE are more likely to experience
slower clearance of ester-based neuromuscular blocking agents (such as
succinylcholine/suxamethonium), leading to prolonged apnea and muscle
paralysis when these medications are used. BChE PMs may also be at
greater risk for treatment failure with the use of ester-based prodrugs and
for toxicity on exposure to cocaine.
As with most of the metabolic enzymes discussed earlier, the majority of the vari-
ants in the BCHE gene that have been identified are associated with either a reduction or
complete absence of BChE activity.281,282 There are two major BCHE alleles: the A (atypi-
cal) form (leading to the D70G residue change in BChE and virtually no amount of active
enzyme present) and the K form (a quantitative variant leading to the A539T residue
change in BChE, resulting in both lower enzyme activity and lower levels of enzyme pres-
ent). The A form, while found less frequently, may carry more clinical relevance. The A form
results in a mutant BChE enzyme displaying only ~30% of the activity that the wild-type
enzyme demonstrates; thus, it has more of a potential impact on pharmacotherapeutic
agents.283-285 Although few, if any, large-scale population studies of BChE genotype or
phenotype distribution have been conducted, it has been estimated that approximately
1 in 2,500 Caucasian patients have the homozygous AA BCHE genotype.283-285 Standard
clinical BChE phenotyping methods involve the direct measurement of BChE activity along
with assessment of the degree of BChE inhibition by sodium fluoride, dibucaine, and the
dimethylcarbamate RO-02-0683.286
Phase II Metabolic Enzymes

Phase II metabolic biotransformations involve transfer or conjugation reactions and most,
but not all (e.g., methylation), of these changes result in the formation of a more polar (or
hydrophilic) product for easier excretion in urine (Figure 4-2).27-33 The phase II polymorphic
enzymes of the most clinical importance include NAT2, TPMT, and UGT1A1.27-33,63,64,287
NAT2 (arylamine N-acetyltransferase, EC 2.3.1.5) is involved in the metabolism of a
variety of therapeutic agents, including several hydrazine-based drugs (e.g., hydralazine,
isoniazid) and arylamine-containing agents (e.g., procainamide, dapsone) (Figure 4-11).27-33
Like CYP2D6, NAT2 is one of the few examples of DMEs identified to date that has multiple
known phenotypes—in this case, FM/RM, IM, and slow metabolizers.27-33 The recogni-
tion of interpatient variations in the activity level of NAT2 dates back to pharmacokinetic
studies with isoniazid (INH) in the 1960s (prior to the identification of the NAT2 enzyme
or the idea of genotyping).288,289 In addition to the wild-type allele (NAT2*4, in this case),
at least 88 genetic variations of the NAT2 gene have been identified to date.290 Patients
carrying fast/rapid NAT2 alleles (NAT2*12, NAT2*13) are fast acetylators who are capable of
clearing NAT2 drug substrates more quickly than individuals who are slow acetylators.291
The NAT2 alleles most commonly associated with the slow acetylator phenotype are
NAT2*5, NAT2*6, NAT2*7, and NAT2*14.291 Phenotype distribution studies indicate that more
than 50% of Europeans and Africans are slow acetylators, while <20% of Asians fit this
profile.29,292 Fast acetylators are found more often in Asians (80-90%) and less frequently
in Caucasians (30-45%).27
FIGURE 4-11 The NAT2-mediated metabolism of hydralazine in man.28
NAT2 phenotypes directly impact the pharmacokinetics of substrates metabolized by

this enzyme. For example, one pharmacokinetic study on the use of INH in children found
that the in vivo half-life of this agent in slow acetylators was more than double that in fast
acetylators.293 Another study reported that NAT2 slow acetylators were 3.8-fold more like-
ly to develop INH-associated hepatotoxicity as compared to rapid acetylators, presumably
due to reduced clearance of this agent and corresponding higher drug levels over longer
periods of time in the body.294 With regard to specific therapeutic agents, the FDA currently
suggests that NAT2 genotyping may be useful in tuberculosis patients treated with INH.120
The Bidil™ (isosorbide and hydralazine fixed-dose product) drug label currently acknowl-
edges that fast acetylators may have reduced exposure to hydralazine and slow acetylators
may have higher bioavailability, but the clinical consequences are not addressed.
CLINICAL PEARL
Genotypic or phenotypic profiling of NAT2 may be helpful in optimizing
the dosing regimens of tuberculosis patients receiving INH therapy.
Patient-to-patient variations in the activity level of TPMT are of clinical importance

due to involvement of this enzyme in deactivation of thiopurine-based therapeutic agents
(6-mercaptopurine [6-MP], azathioprine [a prodrug of 6-MP], and thioguanine [6-TG])
(Figure 4-12).64,290,295,296 Thiopurines, which serve as antimetabolites of endogenous purine
nucleic acid bases, are used in the treatment of solid organ transplant rejection, certain
autoimmune diseases (e.g., rheumatoid arthritis), and various types of cancer.297-299 The
toxicities, including myelosuppression and hepatotoxicity, associated with the use of
thiopurine-based pharmaceutical agents are serious and can be life threatening. Patients
with lower levels of TPMT activity are at greater risk for experiencing ADRs when exposed
to these agents.300,301 The FDA, recognizing these safety issues, currently recommends
that clinicians determine the TPMT activity levels of patients (through either genotypic or
phenotypic assays) before beginning treatment with a thiopurine to adjust dosing regimens
and minimize the number of potential ADRs.120
FIGURE 4-12 The TPMT-mediated metabolism of 6-MP in man.64
CLINICAL PEARL
The toxicities, including myelosuppression and hepatotoxicity, associated
with the use of thiopurine-based pharmaceutical agents are serious and
can be life threatening. Patients with lower levels of TPMT activity are at
greater risk for experiencing ADRs when exposed to these agents.
As with other DMEs, most of the known TPMT allelic variants lead to a lower level of
TPMT activity.302 Variations in the frequencies of the most common TPMT alleles across
different ethnic groups have also been noted. For example, the TPMT*3A allele (G460>A
and A719>G) has been reported to be the most common (4.5%) reduced-function variant
in Caucasians (contributing to a PM phenotype in 0.6% of this ethnic group), while the
TPMT*3C allele (A719>G) seems to be more common (2.3%) in Asians, Africans, and African
Americans.303-305
In terms of drug substrates, the UGTs are second only to the CYPs in number of phar-
maceutical agents that are biotransformed by this group of enzymes.306,307 The human
UGT superfamily is comprised of two families (UGT1 and UGT2) and three subfamilies
(UGT1A, UGT2A, and UGT2B).306-308 Genetic polymorphism has been described for at least
ten of the human UGT enzymes.308-313 Within this metabolic enzyme family, the major
clinical interest is in the polymorphism of UGT1A1 due to its role in inactivation of the
active form (SN-38) of anticancer prodrug irinotecan (Figure 4-13).314-317 A frequent UGT1A1
polymorphism, the UGT1A1*28 allele, involves a specific mutation ([TA]6 →[TA]7) in the
promoter region (the TATA box) of this gene that leads to reduced gene expression and
impaired enzyme activity.318 This variant allele is common in many ethnic groups, ranging
from a frequency of 26-39% in Caucasians to 9-16% in Asians and 42-56% in Africans.319-321
Multiple studies have shown that impaired UGT1A1 activity in patients who are homozygous
for the UGT1A1*28 allele results in severe, dose-limiting toxicity (diarrhea, neutropenia)
during irinotecan therapy.314-317,322-325 These findings ultimately led to an update in irinote-
can labeling information to include dosing recommendations based on the presence of a
UGT1A1*28 allele.120,325 The FDA currently recommends an assessment of a patient’s level
of UGT1A1 activity prior to the initiation of irinotecan therapy.120
FIGURE 4-13 The UGT1A1-mediated metabolism of irinotecan in man.314
CLINICAL PEARL
Patients who are homozygous for the UGT1A1*28 allele have impaired
metabolism of the active form (SN-38) of the anticancer agent irinotecan
that results in severe, dose-limiting toxicity (diarrhea, neutropenia). The
FDA currently recommends an assessment of a patient’s level of UGT1A1

activity prior to the initiation of irinotecan therapy.
Over the past several years, multiple genetic assays have been cleared or approved,
representing a movement toward greater utilization of companion diagnostics to improve
the selection of appropriate agents and dosing regimens. This trend has been driven by the
increasing recognition that personalized medicine can improve patient care and decrease
healthcare costs. Pharmaceutical companies are increasingly developing companion diag-
nostics in the early stages of the drug development process.326,327 Currently, the FDA has
approved a dozen different diagnostic tests targeted at specific DMEs for use by clinicians
to guide and inform pharmacotherapeutic regimens.328 Oncology, in particular, has become
an area of intense development of companion diagnostics due to relatively high treatment
costs, risks of potential treatment failures, and possibilities for serious ADRs. Several FDA
initiatives, including the Critical Path Initiative and the Drug-Diagnostic Co-Development
Concept Paper, have sought to promote and encourage the identification and use of valid
biomarkers to help guide therapeutic decisions.124,329
SUMMARY
As noted at the beginning of this chapter, a patient’s safe and effective use of therapeutic
agents is the goal of every healthcare professional. The advent of precision, or person-
alized, medicine allows this goal to be made even more specific—namely, to determine
which patients are most likely to benefit from a given therapy, for which patients a given
therapy may be inappropriate, and whether or not the typical dosing regimen of a thera-
peutic agent should be adjusted.3-5 In selected cases (e.g., the use of UGT1A1 genotyping
for patients who are candidates for irinotecan therapy), utilization of pharmacogenetic
profiling for DMEs has already contributed to this goal. However, for the potential of this
approach to be fully realized, more work is required. A more thorough understanding of
what factors should be considered in the selection of therapeutic agents and dosing regi-
mens will require rigorous, large-scale studies evaluating multiple nongenetic factors (e.g.,
gender, concomitant disease states, concurrent pharmacotherapy). In addition, such clini-
cal analyses will also need to include the pharmacogenetic profiling of drug transporters,
DMEs, and drug targets (and off targets) for specific pharmaceutical agents. For example,
outcomes associated with warfarin therapy are affected by more variables than simply the
activity of a single DME. In addition, as also illustrated by the example of warfarin, the wide-
spread acceptance of pharmacogenetic testing by both clinicians and insurance providers
will largely rest on the demonstration that such testing is cost effective and improves
clinically meaningful outcomes. Interestingly, patients may become advocates over time
for more pharmacogenomic testing as some genetic testing companies follow the direct-
to-consumer marketing model used by many pharmaceutical companies to advertise their
tests directly to consumers. Furthermore, many pharmacies are beginning to offer genetic
testing directly to patients. Pharmacists, as drug information experts, are well positioned
to take the lead in implementing and using pharmacogenetic tools and information to
improve patient care.330
REFERENCES
1. World Health Organization. Safety of Medicines: A Guide to Detecting and Reporting Adverse Drug
Reactions. Geneva, Switzerland: World Health Organization; 2002. Available at: http://whqlibdoc.who.
int/hq/2002/WHO_EDM_QSM_2002.2.pdf.
2. U.S. Office of the Federal Register. 21 CFR 320.33(c).
3. Burke W, Psaty BM. Personalized medicine in the era of genomics. JAMA. 2007;298:1682-1684.
4. Evans WE, Relling MV. Moving towards individualized medicine with pharmacogenomics. Nature.
2004;429:464-468.
5. Sadée W, Dai Z. Pharmacogenetics/genomics and personalized medicine. Human Molecular Genetics.
2005;14(Review Issue 2):R207-R214.
6. Spear BB, Heath-Chiozzi M, Huff J. Clinical application of pharmacogenetics. Trends Mol Med.
2002;7:201-204.
7. U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion.
(2014). National Action Plan for Adverse Drug Event Prevention. Washington, DC.
8. U.S. Department of Health and Human Services Office of Inspector General (OIG). Adverse Events in
Hospitals: National Incidence Among Medicare Beneficiaries. Washington, DC: 2010 November. Report
No.: OEI-06-09-00090.
9. Lucado J (Social & Scientific Systems, Inc.), Paez K (Social & Scientific Systems, Inc.), Elixhauser A
(AHRQ). Medication-Related Adverse Outcomes in U.S. Hospitals and Emergency Departments, 2008.
HCUP Statistical Brief #109. April 2011. Agency for Healthcare Research and Quality, Rockville, MD.
Available at: http://www.hcup-us.ahrq.gov/reports/statbriefs/sb109.pdf.
10. Bates DW, Spell N, Cullen DJ, et al. The costs of adverse drug events in hospitalized patients. Adverse
Drug Events Prevention Study Group. JAMA. 1997;277(4):307-311.
11. Classen DC, Pestotnik SL, Evans RS, et al. Adverse drug events in hospitalized patients. Excess length
of stay, extra costs, and attributable mortality. JAMA. 1997;277(4):301-306.
12. Bourgeois FT, Shannon MW, Valim C, et al. Adverse drug events in the outpatient setting: an 11-year
national analysis. Pharmacoepidemiol Drug Saf. 2010;19(9):901-910.
13. CDC, unpublished data. Updated numbers for Budnitz DS, Pollock DA, Weidenbach KN, et al.
National surveillance of emergency department visits for outpatient adverse drug events. JAMA.
2006;296(15):1858-1866.
14. Institute of Medicine Committee on Identifying and Preventing Medication Errors. Preventing Medica-
tion Errors: Quality Chasm Series. Washington, DC: The National Academies Press; 2006.
15. Phillips KA, Veenstra DL, Oren E, et al. Potential role of pharmacogenomics in reducing adverse drug
reactions. JAMA. 2001;286:2270-2279.
16. Severino G, Del Zompo M. Adverse drug reactions: role of pharmacogenomics. Pharm Res.
2004;49:363-373.
17. Pirmohamed M, Park BK. Cytochrome P450 enzyme polymorphisms and adverse drug reactions. Toxi-
col. 2003;192:23-32.
18. Ingelman-Sundberg M. Pharmacogenetic biomarkers for prediction of severe adverse drug reactions. N
Engl J Med. 2008;358:638.
19. Wilke RA, Lin DW, Roden DM, et al. Identifying genetic risk factors for serious adverse drug reactions:
current progress and challenges. Nat Rev Drug Discov. 2007;6:904-916.
20. Wester K, Jönsson AK, Spigset O, et al. Incidence of fatal adverse drug reactions: A population based
study. Br J Clin Pharmacol. 2007;65:573-579.
21. Hirsh J, Fuster V, Ansell J, et al. American Heart Association/American College of Cardiology Founda-
tion guide to warfarin therapy. J Am Coll Cardiol. 2003;41:1633-1652.

23. Klein TE, Altman RB, Eriksson N, et al. Estimation of the warfarin dose with clinical and pharmacoge-
netic data. N Engl J Med. 2009;360:753-764.
24. Gage BF, Eby C, Milligan PE, et al. Use of pharmacogenetics and clinical factors to predict the mainte-
nance dose of warfarin. Thromb Haemost. 2004;91:87-94.
25. Gage BF, Eby C, Johnson JA, et al. Use of pharmacogenetics and clinical factors to predict the thera-
peutic dose of warfarin. Clin Pharm Ther. 2008;84:326-331.
26. Barratt DT, Bandak B, Klepstad P, et al. Genetic, pathological and physiological determinants of
transdermal fentanyl pharmacokinetics in 620 cancer patients of the EPOS study. Pharmacogenet
Genomics. 2014;24:185-194.
27. Rendic S. Summary of information on human CYP enzymes: human metabolism data. Drug Metab Rev.
2002;34:83-448.
28. Williams DA. Drug metabolism. In: Williams DA, Lemke TL, eds. Foye’s Principles of Medicinal Chemis-
try. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2008:253-326.
29. Kramer SD, Testa B. The biochemistry of drug metabolism—an introduction. Part 6. Inter-individual
factors affecting drug metabolism. Chem Biodiv. 2008;5:2465-2578.
30. Gonzalez FJ, Tukey RH. Drug metabolism. In: Brunton LL, Lazo JS, Parker KL, eds. Goodman and
Gilman’s The Pharmacological Basis of Therapeutics. 11th ed. New York, NY: McGraw-Hill Professional;
2006:71-92.
31. Taniguchi C, Guengerich FP. Drug metabolism. In: Golan DE, Tashjian AH, Armstrong EJ, et al., eds.
Principles of Pharmacology: The Pathophysiological Basis of Drug Therapy. 2nd ed. Philadelphia, PA:
Lippincott Williams & Wilkins; 2008:49-61.
32. Correia MA. Drug biotransformation. In: Katzung BG, ed. Basic and Clinical Pharmacology. 11th ed. New
York, NY: McGraw-Hill Medical; 2009:53-66.
33. Smith DA, van de Waterbeemd H, Walker DK. Pharmacokinetics and Metabolism in Drug Design.
Weinheim, Germany: Wiley-VCH; 2006.
34. Benitez J, Dahl M-L, Spina E, et al. Genetic and environmental factors causing variability in psycho-
tropic drug response. In: Pacifici GM, Pelkonen O, eds. Interindividual Variability in Human Drug Metab-
olism. London, UK: Taylor and Francis; 2001:85-128.
35. Williams DG, Patel A, Howard RF. Pharmacogenetics of codeine metabolism in an urban population of
children and its implications for analgesic reliability. Br J Anaesth. 2002;89:839-845.
36. Drendel A. Pharmacogenomics of analgesic agents. Clin Ped Emerg Med. 2007;8:262-267.
37. Flores CM, Mogil JS. The pharmacogenomics of analgesia: toward a genetically-based approach to
pain management. Pharmacogenomics. 2001;2:177-194.
38. Marcus FI, Kapadia GJ, Kapadia GG. The metabolism of digoxin in normal subjects. J Pharmacol Exp
Ther. 1964;145:203-209.
39. Forane® (prescribing information/package insert). Deerfield, IL: Baxter Healthcare Corporation; 2006.
40. Anectine® (prescribing information/package insert). Research Triangle Park, NC: GlaxoSmith- Kline;
1999.
41. Elliott HW. Metabolism of lorazepam. Br J Anesth. 1976;48:1017-1023.
42. Brodie BB, Axelrod J. The fate of acetanilide in man. J Pharmacol Exp Ther. 1948;94:29-38.
43. Gelotte CK, Auiler JF, Lynch JM, et al. Disposition of acetaminophen at 4, 6, and 8 g/day for 3 days in
healthy young adults. Clin Pharm Ther. 2007;81:840-848.
44. Helton ED, Williams MC, Goldzieher JW. Human urinary and liver conjugates of 17-alphaethinylestra-
diol. Steroids. 1976;27:851-867.
45. Ebner T, Remmel RP, Burchell B. Human bilirubin UDP-glucuronosyltransferase catalyzes the gluc-
uronidation of ethinylestradiol. Mol Pharm. 1993;43:649-654.
46. Orme MLE, Back DJ, Ball S. Interindividual variation in the metabolism of ethinylestradiol. Pharmacol
Ther. 1989;43:251-260.
47. Skarke C, Lotsch J. Morphine metabolites: clinical implications. Seminars in anesthesia. Perioperative
Medicine and Pain. 2002;21:258-264.
48. DeVane CL. Metabolism and pharmacokinetics of selective serotonin reuptake inhibitors. Cell Mol
Neurobiol. 1999;19:443-466.
49. Nelson DR. Human cytochrome P450s. Available at: http://drnelson.uthsc.edu/human.P450.table.html.
Accessed November 2, 2015.

50. Human Cytochrome P450 (CYP) Allele Nomenclature Committee. Allele nomenclature for Cyto-
chrome P450 enzymes. Available at: http://www.cypalleles.ki.se. Accessed November 2, 2015.
51. Nelson DR, Koymans L, Kamataki T, et al. P450 superfamily: update on new sequences, gene mapping,
accession numbers and nomenclature. Pharmacogenetics. 1996;6:1-42.
52. Sim SC, Ingelman-Sundberg M. Update on allele nomenclature for human cytochromes P450
and the Human Cytochrome P450 Allele (CYP-allele) Nomenclature Database. Methods Mol Biol.
2013;987:251-259.
53. Evert B, Griese E-U, Eichelbaum M. A missense mutation in exon 6 of the CYP2D6 gene leading to a
histidine 324 to proline exchange is associated with the poor metabolizer phenotype of sparteine.
NaunynSchmiedeberg’s Arch Pharmacol. 1994;350:434-439.
54. Bradford LD. CYP2D6 allele frequency in European Caucasians, Asians, Africans, and their descen-
dants. Pharmacogenomics. 2002;3:229-243.
55. Bradford LD, Kerlin WG. Polymorphism of CYP2D6 in black populations: implications for psychophar-
macology. Int J Neuropsychopharmacol. 1988;1:173-185.
56. Yengi LG. Pharmacogenetics and pharmacogenomics. In: Nassar AF, ed. Drug Metabolism Handbook:
Concepts and Applications. New York, NY: John F. Wiley & Sons; 2009:65-88.
57. Daly AK. Pharmacogenetics. In: Pearson PG, Wienkers LC, eds. Handbook of Drug Metabolism. 2nd ed.
(Drugs and the Pharmaceutical Sciences). New York, NY: Informa Healthcare; 2008:179-202.
58. Ingelman-Sundberg M. Genetic polymorphisms of cytochrome P450 2D6 (CYP2D6): clinical conse-
quences, evolutionary aspects and functional diversity. Pharmacogenomics J. 2005;5:6-13.
59. Wang H, Tompkins LM. CYP2B6: New insights into a historically overlooked cytochrome P450 isozyme.
Curr Drug Metab. 2008;9:598-610.
60. Gogtay NJ, Mali NB, Iyer K, et al. Evaluation of cytochrome P450 2D6 phenotyping in healthy adult
Western Indians. Indian J Pharmacol. 2014;46:266-269.
61. Morel A, Boisdron-Celle M, Fey L, et al. Clinical relevance of different dihydropyrimidine dehy-
drogenase gene single nucleotide polymorphisms on 5-fluorouracil tolerance. Mol Cancer Ther.
2006;5:2895-2904.
62. Jensen FS, Schwartz M, Viby-Mogensen J. Identification of human plasma cholinesterase variants
using molecular biological techniques. Acta Anaesth Scand. 1995;39:142-149.
63. Agundez JAG. Polymorphisms of human N-acetyltransferases and cancer risk. Curr Drug Metab.
2008;9:520-531.
64. Reuther LO, Vainer B, Sonne J, et al. Thiopurine methyltransferase (TPMT) genotype distribution in
azathioprine-tolerant and -intolerant patients with various disorders. The impact of TPMT genotyping
in predicting toxicity. Eur J Clin Pharmacol. 2004;59:797-801.
65. Shimada T, Yamazaki H, Mimura M, et al. Interindividual variations in human liver cytochrome P-450
enzymes involved in the oxidation of drugs, carcinogens and toxic chemicals: studies with liver micro-
somes of 30 Japanese and 30 Caucasians. J Pharmacol Exp Ther. 1994;270:414-423.
66. Ozdemir V, Kalow W, Posner P, et al. CYP1A2 activity as measured by a caffeine test predicts clozap-
ine and active metabolite norclozapine steady-state concentration in patients with schizophrenia. J
Clin Psychopharmcol. 2001;21:398-407.
67. Ha HR, Chen J, Freiburghaus AU, et al. Metabolism of theophylline by cDNA-expressed human cyto-
chromes P-450. Br J Clin Pharmacol. 1995;39:321-326.
68. Doude van Troostwijk LJAE, Koopmans RP, Vermeulen HDB, et al. CYP1A2 activity is an important
determinant of clozapine dosage in schizophrenic patients. Eur J Pharm Sci. 2003:20:451-457.
69. Shirley KL, Hon YY, Penzak SR, et al. Correlation of cytochrome P450 (CYP) 1A2 activity using
caffeine phenotyping and olanzapine disposition in healthy volunteers. Neuropsychopharmacology.
2003;28:961-966.
70. Lewis DFV. Guide to Cytochromes P450: Structure and Function. London, UK: Taylor and Francis; 2001.
71. Brown CM, Reisfeld B, Mayeno AN. Cytochromes P450: a structure-based summary of biotransforma-
tions using representative substrates. Drug Metabolism Reviews. 2008;40:1-100.
72. Guengerich FP. Human cytochrome P450 enzymes. In: Ortiz de Montellano PR, ed. Cytochrome P450:
Structure, Mechanism, and Biochemistry. 3rd ed. New York, NY: Kluwer Academic; 2005:377-382.
73. Ghotbi R, Christensen M, Roh HK, et al. Comparisons of CYP1A2 genetic polymorphisms, enzyme
activity and the genotype-phenotype relationship in Swedes and Koreans. Eur J Clin Pharmacol.
2007;63:537-46.
74. Skarke C, Kirchhof A, Geisslinger G, et al. Rapid genotyping for relevant CYP1A2 alleles by pyro-
sequencing. Eur J Clin Pharmacol. 2005;61:887-892.
75. Omari AA, Murry DJ. Pharmacogenetics of the cytochrome P450 enzyme system: review of current
knowledge and clinical significance. J Pharm Pract. 2007;20:206-218.
76. Solus JF, Arietta BJ, Harris JR, et al. Genetic variation in eleven phase I drug metabolism genes in an
ethnically diverse population. Pharmacogenomics. 2004;5:895-931.
77. Doude van Troostwijk LJAE, Koopmans RP, Vermeulen HDB, et al. CYP1A2 activity is an important
determinant of clozapine dosage in schizophrenic patients. Eur J Pharm Sci. 2003;20:451-457.
78. Clozaril® (prescribing information/package insert). East Hanover, NJ: Novartis Pharmaceuticals Corpora-
tion; 2009.
79. Basile VS, Masellis M, Potkin SG, et al. Pharmacogenomics in schizophrenia: the quest for individual-
ized therapy. Human Mol Genetics. 2002;11:2517-2530.
80. Sachse C, Brockmoller J, Baue S, et al. Cytochrome P450 2D6 variants in a Caucasian population:
allele frequencies and phenotypic consequences. Am J Hum Genet. 1997;60:284-295.
81. Schulze TG, Schumacher J, Muller DJ, et al. Lack of association between a functional polymorphism
of the cytochrome P450 1A2 (CYP1A2) gene and tardive dyskinesia in schizophrenia. Am J Med Genet.
2001;105:498-501.
82. Bender S, Eap CB. Very high cytochrome P4501A2 activity and nonresponse to clozapine. Arch Gen
Psychiatry. 1998;55:1048-1050.
83. Bozikas VP, Papakosta M, Niopas I, et al. Smoking impact on CYP1A2 activity in a group of patients
with schizophrenia. Eur Neuropsychopharmacol. 2004;14:39-44.
84. Papetti F, Morel-Pingault V, Buisse V, et al. Clozapine-resistant schizophrenia related to an increased
metabolism and benefit of fluvoxamine: four case reports. Encephale. 2007;33:811-818.
85. Kohlrausch FB, Severino-Gama C, Lobato MI, et al. The CYP1A2 -163C>A polymorphism is associated
with clozapine-induced generalized tonic-clonic seizures in Brazilian schizophrenia patients. Psychiatry
Res. 2013;209:242-245.
86. Webster E, McIntyre J, Choonara I, et al. The caffeine breath test and CYP1A2 activity in children.
Paediatric and Perinatal Drug Ther. 2004;5:28-33.
87. Bertilsson L, Carrillo JA, Dahl ML, et al. Clozapine disposition covaries with CYP1A2 activity determined
by a caffeine test. Br J Clin Pharmacol. 1994;38:471-473.
88. Gunes A, Dahl ML. Variation in CYP1A2 activity and its clinical implications: influence of environmental
factors and genetic polymorphisms. Pharmacogenomics. 2008;9:625-637.
89. Faucette SR, Hawke RL, Lecluyse EL, et al. Validation of bupropion hydroxylation as a selective marker
of human cytochrome P450 2B6 catalytic activity. Drug Metab Dispos. 2000;28:1222-1230.
90. Chang TK, Weber GF, Crespi CL, et al. Differential activation of cyclophosphamide and ifosphamide by
cytochromes P-450 2B and 3A in human liver microsomes. Cancer Res. 1993;53:5629-5637.
91. Crettol S, Deglon JJ, Besson J, et al. Methadone enantiomer plasma levels, CYP2B6, CYP2C19, and
CYP2C9 genotypes, and response to treatment. Clin Pharmacol Ther. 2005;78:593-604.
92. Totah RA, Sheffels P, Roberts T, et al. Role of CYP2B6 in stereoselective human methadone metabo-
lism. Anesthesiology. 2008;108:363-374.
93. Haas DW, Smeaton LM, Shafer RW, et al. Pharmacogenetics of long-term responses to antiretroviral
regimens containing efavirenz and/or nelfinavir: an Adult AIDS Clinical Trials Group study. J Infect Dis.
2005;192:1931-1942.
94. Haas DW, Ribaudo HJ, Kim RB, et al. Pharmacogenetics of efavirenz and central nervous system side
effects: an Adult AIDS Clinical Trials Group study. AIDS. 2004;18:2391-2400.
95. Rotger M, Colombo S, Furrer H, et al. Influence of CYP2B6 polymorphism on plasma and intracel-
lular concentrations and toxicity of efavirenz and nevirapine in HIV-infected patients. Pharmacogenet
Genomics. 2005;15:1-5.
96. Zanger UM, Klein K, Saussele T, et al. Polymorphic CYP2B6: molecular mechanisms and emerging
clinical significance. Pharmacogenomics. 2007;8:743-759.
97. Mo SL, Liu YH, Duan W, et al. Substrate specificity, regulation, and polymorphism of human cyto-
chrome P450 2B6. Curr Drug Metab. 2009;10:730-753.
98. Zanger UM, Klein K. Pharmacogenetics of cytochrome P450 2B6 (CYP2B6): advances on polymor-
phisms, mechanisms, and clinical relevance. Front Genet. 2013;4:1-12.

99. Klein K, Lang T, Saussele T, et al. Genetic variability of CYP2B6 in populations of African and Asian
origin: allele frequencies, novel function variants, and possible implications for anti-HIV therapy with
efavirenz. Pharmacogenet Genomics. 2005;15:861-873.
100. Cho JY, Lim HS, Chung JY, et al. Haplotype structure and allele frequencies of CYP2B6 in a Korean
population. Drug Metab Disp. 2004;32:1341-1344.
101. Mehlotra RK, Ziats MN, Bockarie MJ, et al. Prevalence of CYP2B6 alleles in malaria-endemic popula-
tions of West Africa and Papua New Guinea. Eur J Clin Pharmacol 2006;62:267-275.
102. Zukunft J, Lang T, Richter T, et al. A natural CYP2B6 TATA box polymorphism (-82T--> C) lead-
ing to enhanced transcription and relocation of the transcriptional start site. Mol Pharmacol.
2005;67(5):1772-1782.
103. Yule SM, Boddy AV, Cole M, et al. Cyclophosphamide pharmacokinetics in children. Br J Clin Pharma-
col. 1996;41:13-19.
104. Zhang J, Tian Q, Zhou S-F. Clinical pharmacology of cyclophosphamide and ifosfamide. Curr Drug
Ther. 2005;1:55-84.
105. de Jonge ME, Huitema ADR, van Dam, SM, et al. Population pharmacokinetics of cyclophosphamide
and its metabolites 4-hydroxycyclophosphamide, 2-dechloroethylcyclophosphamide, and phos-
phoramide mustard in a high-dose combination with thiotepa and carboplatin. Ther Drug Monitoring.
2005;27:756-765.
106. Afsharian P, Terelius Y, Hidestrand M, et al. The role of human CYP2B6 polymorphism in the
bioactivation of cyclophosphamide using cDNA expressed enzymes. Biol Blood Marrow Trans.
2007;13(Suppl.1):70-71.
107. Bachanova V, Shanley R, Malik F, et al. Cytochrome P450 2B6*5 increases relapse after cyclophos-
phamide-containing conditioning and autologous transplantation for lymphoma. Biol Blood Marrow
Transplant. 2015;21:944-948.
108. Kirchheiner J, Tsahuridu M, Jabrane W, et al. The CYP2C9 polymorphism: from enzyme kinetics to
clinical dose recommendations. Personalized Med. 2004;1:63-84.
109. Kirchheiner J, Brockmuller J. Clinical consequences of cytochrome P450 2C9 polymorphisms. Clin
Pharmacol Ther. 2005;77:1-16.
110. Sandberg M, Yasar U, Strömberg P, et al. Oxidation of celecoxib by polymorphic cytochrome P450 2C9
and alcohol dehydrogenase. Br J Clin Pharmacol. 2002;54:423-429.
111. Yasar U, Tybring G, Hidestrand M, et al. Role of CYP2C9 polymorphism in losartan oxidation. Drug
Metab Disp. 2001;29:1051-1056.
112. Sica DA, Gehr TWB, Ghosh S. Clinical pharmacokinetics of losartan. Clin Pharmacokinetics.
2005;44:797-814.
113. Joy MS, Dornbrook-Lavender K, Blaisdell J, et al. CYP2C9 genotype and pharmacodynamic
responses to losartan in patients with primary and secondary kidney diseases. Eur J Clin Pharmacol.
2009;65:947-953.
114. Giancarlo GM, Venkatakrishnan K, Granda BW, et al. Relative contributions of CYP2C9 and 2C19 to
phenytoin 4-hydroxylation in vitro: inhibition by sulfaphenazole, omeprazole, and ticlopidine. Eur J Clin
Pharmacol. 2001;57:31-36.
115. Karlen B, Garle M, Rane A, et al. Assay of the major (4-hydroxylated) metabolites of diphenylhydan-
toin in human urine. Eur J Clin Pharmacol. 1975;8:359-363.
116. Eadie MJ, Tyrer JH, Bochner F, et al. The elimination of phenytoin in man. Clin Exp Pharmacol Physiol.
2007;3:217-224.
117. Holstein A, Plaschke A, Ptak M, et al. Association between CYP2C9 slow metabolizer genotypes and
severe hypoglycemia with sulfonylurea hypoglycemic agents. Br J Clin Pharmacol. 2005;60:103-106.
118. Pilotto A, Seripa D, Franceschi M, et al. Genetic susceptibility to nonsteroidal anti-inflammatory drug–
related gastroduodenal bleeding: role of cytochrome P450 2C9 polymorphisms. Gastroenterology.
2007;133:465-471.
119. Budnitz DS, Lovegrove MC, Shehab N, et al. Emergency hospitalizations for adverse drug events in
older Americans. N Engl J Med. 2011;365:2002-2012.
120. U.S. Food and Drug Administration. Table of Pharmacogenomic Biomarkers in Drug Labeling. Available
at: http://www.fda.gov/Drugs/ScienceResearch/ResearchAreas/Pharmacogenetics/ucm083378.htm.
121. Coumadin® (prescribing information/package insert). Princeton, NJ: Bristol-Myers Squibb; 2007.
122. Psaty BM, Furberg CD. COX-2 inhibitors—lessons in drug safety. N Engl J Med. 2005;352:1133-1135.
123. Frueh FW, Amur S, Mummaneni P, et al. Pharmacogenomic biomarker information in drug labels
approved by the United States Food and Drug Administration: Prevalence of related drug use.
Pharmacotherapy. 2008;28:992-998.
124. U.S. Department of Health and Human Services. Guidance for industry: Definitions for genomic
biomarkers, pharmacogenomics, pharmacogenetics, genomic data, and sample coding categories.
Washington, DC: US Department of Health and Human Services; 2008. Available at: http://www.fda.
gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm073162.pdf.
125. Coreg® CR (prescribing information/package insert). Research Triangle Park, NC: GlaxoSmithKline;
2009.
126. Celebrex® (prescribing information/package insert). New York, NY: Pfizer; 2009.
127. Chiba M, Xu X, Nishime JA, et al. Hepatic microsomal metabolism of montelukast, a potent leukotriene
D4 receptor antagonist, in humans. Drug Metab Disp. 1997;25:1022-1031.
128. Baldwin SJ, Clarke SE, Chenery RJ. Characterization of the cytochrome P450 enzymes involved in the
in vitro metabolism of rosiglitazone. Br J Clin Pharmacol. 1999;48:424-432.
129. Nakashima A, Kawashita H, Masuda N, et al. Identification of cytochrome P450 forms involved in the
4-hydroxylation of valsartan, a potent and specific angiotensin II receptor antagonist, in human liver
microsomes. Xenobiotica. 2005;35:589-602.
130. Daly AK, King BP. Pharmacogenetics of oral anticoagulants. Pharmacogenetics. 2003;13:247-252.
131. Bylander J. Personalized medicine not paying the bills yet; more science needed. The Pink Sheet. July
7, 2008;18-19.
132. Allingham-Hawkins D. Successful genetic tests are predicated on clinical utility. Gen Eng News. 2008;
August 6-9.
133. Bylander J. Clinicians split on warfarin testing as a new standard of care. The Pink Sheet Daily. August
6, 2008.
134. Hillman MA, Wilke RA, Caldwell MD, et al. Relative impact of covariates in prescribing warfarin accord-
ing to CYP2C9 genotype. Pharmacogenetics. 2004;14:539-547.
135. Wadelius M, Sorlin K, Wallerman O, et al. Warfarin sensitivity related to CYP2C9, CYP3A5, ABCB1
(MDR1) and other factors. Pharmacogenomics J. 2004;4:40-48.
136. Yasar U, Lundgren S, Eliasson E, et al. Linkage between the CYP2C8 and CYP2C9 genetic polymor-
phisms. Biochem Biophys Res Comm. 2002;299:25-28.
137. Desta Z, Zhao X, Shin J-G, et al. Clinical significance of the cytochrome P450 2C19 genetic polymor-
phism. Clin Pharmacokinet. 2002;41:913-958.
138. Goldstein JA, Faletto MB, Romkes-Sparks M, et al. Evidence that CYP2C19 is the major (S)-mephenyt-
oin 4’-hydroxylase in humans. Biochemistry. 1994;33:1743-1752.
139. Xie HG, Kim RB, Wood AJ, et al. Molecular basis of ethnic differences in drug disposition and response.
Annu Rev Pharmacol Toxicol. 2001;41:815-850.
140. Mega JL, Close SL, Wiviott SD, et al. Cytochrome P-450 polymorphisms and response to clopidogrel.
N Engl J Med. 2009;360:354-362.
141. Collet JP, Hulot JS, Pena A, et al. Cytochrome P450 2C19 polymorphism in young patients treated
with clopidogrel after myocardial infarction: a cohort study. The Lancet. 2009;373:309-317.
142. Simon T, Verstuyft C, Mary-Krause M, et al. Genetic determinants of response to clopidogrel and
cardiovascular events. N Engl J Med. 2009;360:363-375.
143. Plavix® (prescribing information/package insert). Bridgewater, NJ: Bristol-Myers Squibb/Sanofi Pharma-
ceuticals; 2011.
144. Shuldiner AR, O’Connell JR, Bliden KP, et al. Association of cytochrome P450 2C19 genotype with the
antiplatelet effect and clinical efficacy of clopidogrel therapy. JAMA. 2009;302:849-857.
145. Kim KA, Park PW, Hong SJ, et al. The effect of CYP2C19 polymorphism on the pharmacokinetics and
pharmacodynamics of clopidogrel: a possible mechanism for clopidogrel resistance. Clin Pharmacol
Ther. 2008;84:236-242.
146. Scott SA, Sangkuhl K, Gardner EE, et al. Clinical Pharmacogenetics Implementation Consortium guide-
lines for cytochrome P450–2C19 (CYP2C19) genotype and clopidogrel therapy. Clin Pharmacol Ther.
2011;90:328-332.
antiplatelet effect and clinical efficacy of clopidogrel therapy. JAMA. 2009;302:849-857.
148. Mega JL, Simon T, Collet J-P, et al. Reduced-function CYP2C19 genotype and risk of adverse clinical
outcomes among patients treated with clopidogrel predominantly for PCI. A meta-analysis. JAMA.
2010;304:1821-1830.
149. von Moltke LL, Greenblatt DJ, Giancarlo GM, et al. Escitalopram (S-citalopram) and its metabolites in
vitro: cytochromes mediating biotransformation, inhibitory effects, and comparison to R-citalopram.
Drug Metab Disp. 2001;29:1102-1109.
150. Chong E, Ensom MHH. Pharmacogenetics of the proton pump inhibitors: A systematic review. Pharma-
cotherapy. 2003;23:460-471.
151. Furuta T, Shirai N, Sugimoto M, et al. Pharmacogenomics of proton pump inhibitors. Pharmacogenomics.
2004;5:181-202.
152. Andersson T, Hassan-Alin M, Hasselgren G, et al. Pharmacokinetic studies with esomeprazole, the
(S)-isomer of omeprazole. Clin Pharmacokinetics. 2001;40:411-426.
153. Foradil® Aerolizer (prescribing information/package insert). Basel, Switzerland: Novartis Pharma AG;
2002.
154. Li-Wan-Po A, Girard T, Farndon P, et al. Pharmacogenetics of CYP2C19: functional and clinical implica-
tions of a new variant CYP2C19*17. Br J Clin Pharmacol. 2010;69:222-230.
155. Aubert RE, Epstein RS, Teagarden JR, et al. Proton pump inhibitors effect on clopidogrel effectiveness:
The clopidogrel Medco outcomes study. Circulation. 2008;118:S815.
156. Khalique SC, Cheng-Lai A. Drug interaction between clopidogrel and proton pump inhibitors. Cardiol
Rev. 2009;17:198-200.
157. Stanek EJ. Possible “class effect” for proton-pump inhibitors on top of clopidogrel therapy. Oral
presentation at: Society for Cardiovascular Angiography and Interventions (SCAI) 2009 Meeting; May
6, 2009; Las Vegas, NV.
158. Fontes-Carvalho R, Albuquerque A, Araujo C, et al. Omeprazole, but not pantoprazole, reduces the
antiplatelet effect of clopidogrel: a randomized clinical crossover trial in patients after myocardial
infarction evaluating the clopidogrel-PPIs drug interaction. Eur J Gastroenterol Hepatol. 2011;23:396-
404.
159. Aschenbrenner DS. Drug interaction decreases clopidogrel’s effectiveness. Am J Nursing. 2010;110:55-
56.
160. Mega JL, Close SL, Wiviott SD, et al. Cytochrome P450 genetic polymorphisms and the response
to prasugrel: relationship to pharmacokinetic, pharmacodynamic, and clinical outcomes. Circulation.
2009;119:2553-2560.
161. Petersen KU. Relevance of metabolic activation pathways: the example of clopidogrel and prasugrel.
Arzneimittelforschung. 2009;59:213-227.
162. Jakubowski JA, Winters KJ, Naganuma H, et al. Prasugrel: a novel thienopyridine antiplatelet agent. A
review of preclinical and clinical studies and the mechanistic basis for its distinct anti-platelet profile.
Cardiovasc Drug Rev. 2007;25:357-374.
163. Mahgoub A, Idle JR, Dring DG, et al. Polymorphic hydroxylation of debrisoquine in man. Lancet.
1977;2:584-586.
164. Tucker GT, Silas JH, Iyun AO, et al. Polymorphic hydroxylation of debrisoquine in man. Lancet.
1977;2:718.
165. Eichelbaum M, Spannbrucker N, Steinke B, et al. Defective N-oxidation of sparteine in man: a new
pharmacogenetic defect. Eur J Clin Pharmacol. 1979;16:183-187.
166. Eichelbaum M, Bertilsson L, Säwe J, et al. Polymorphic oxidation of sparteine and debrisoquine related
pharmacogenetic entities. Clin Pharmacol Ther. 1982;31:184-186.
167. Kirchheiner J, Nickchen K, Bauer M, et al. Pharmacogenetics of antidepressants and antipsychotics: the
contribution of allelic variations to the phenotype of drug response. Mol Psychiatry. 2004;9:442-473.
168. Olesen OV, Linnet K. Metabolism of the tricyclic antidepressant amitriptyline by cDNA-expressed
human cytochrome P450 enzymes. Pharmacology. 1997;55:235-243.
169. Stahl SM. Essential Psychopharmacology of Depression and Bipolar Disorder. Cambridge, UK:
Cambridge University Press; 2000.
170. Scordo MG, Spina E, Facciolà G, et al. Cytochrome P450 2D6 genotype and steady state plasma levels
of risperidone and 9-hydroxyrisperidone. Psychopharmacology (Berl). 1999;147:300-305.

171. Laika B, Leucht S, Heres S, et al. Intermediate metabolizer: increased side effects in psychoactive drug
therapy. The key to cost-effectiveness of pretreatment CYP2D6 screening? Pharmacogenomics J.
2009;9:1-10.
172. Bijl MJ, Visser LE, van Schaik RH, et al. Genetic variation in the CYP2D6 gene is associated with a
lower heart rate and blood pressure in beta-blocker users. Clin Pharmacol Ther. 2009;85:45-50.
173. Holtzman NA. Clinical utility of pharmacogenetics and pharmacogenomics. In: Rothstein MA, ed. Phar-
macogenomics: Social, Ethical, and Clinical Dimensions. Hoboken, NJ: John Wiley & Sons; 2003:163-
186.
174. Shin J, Johnson JA. Pharmacogenetics of beta-blockers. Pharmacotherapy. 2007;27:874-887.
175. Kirchheiner J, Heesch C, Bauer S, et al. Impact of the ultrarapid metabolizer genotype of cyto-
chrome P450 2D6 on metoprolol pharmacokinetics and pharmacodynamics. Clin Pharmacol Ther.
2004;76:302-312.
176. Buchert E, Woosley RL. Clinical implications of variable antiarrhythmic drug metabolism. Pharmaco-
genetics. 1992;2:2-11.
177. Siddoway LA, Thomspon KA, McAllister CB, et al. Polymorphism of propafenone metabolism and
disposition in man: clinical pharmacokinetic consequences. Circulation. 1987;75:785-791.
178. Zhou S-F, Liu J-P, Chowbay B. Polymorphism of human cytochrome P450 enzymes and its clinical
impact. Drug Metab Rev. 2009;41:89-295.
179. Strattera® (prescribing information/package insert). Indianapolis, IN: Eli Lilly & Company; 2009.
180. Witcher JW, Long A, Smith B, et al. Atomoxetine pharmacokinetics in children and adolescents with
attention deficit hyperactivity disorder. J Child Adolesc Psychopharmacol. 2003;13:53-63.
181. Enablex® (prescribing information/package insert). Stein, Switzerland: Novartis Pharma AG; 2006.
182. Beaumont KC, Cussans NJ, Nichols DJ, et al. Pharmacokinetics and metabolism of darifenacin in the
mouse, rat, dog and man. Xenobiotica. 1998;28:63-75.
183. Skerjanec A. The clinical pharmacokinetics of darifenacin. Clin Pharmacokinet. 2006;45:325-350.
184. Kerbusch T, Milligan PA, Karlsson MO. Assessment of the relative in vivo potency of the hydroxylated
metabolite of darifenacin in its ability to decrease salivary flow using pooled population pharmaco-
kinetic–pharmacodynamics data. Br J Clin Pharmacol. 2004;57:170-180.
185. Cymbalta® (prescribing information/package insert). Indianapolis, IN: Eli Lilly & Company; 2009.
186. Karpa KA, Cavanaugh JE, Lakoski JM. Duloxetine pharmacology: profile of a dual monoamine modula-
tor. CNS Drug Reviews. 2002;8:361-376.
187. Bymaster FP, Thomas CL, Knadler MP, et al. The dual transporter inhibitor duloxetine: A review of
its preclinical pharmacology, pharmacokinetic profile, and clinical results in depression. Curr Pharm
Design. 2005;11:1475-1493.
188. Caccia S. Metabolism of the newest antidepressants: comparisons with related predecessors. Drugs.
2004;7:143-150.
189. Kuo F, Gillespie TA, Kulanthaivel P, et al. Synthesis and biological activity of some known and putative
duloxetine metabolites. Bioorg Med Chem Lett. 2004;14:3481-3486.
190. Ring BJ, Catlow J, Lindsay TJ, et al. Identification of the human cytochromes P450 responsible for the
in vitro formation of the major oxidative metabolites of the antipsychotic agent olanzapine. J Phar-
macol Exp Ther. 1996;27:658-666.
191. Zyprexa® (prescribing information/package insert). Indianapolis, IN: Eli Lilly & Company; 2009.
192. Chen ZR, Irvine RJ, Somogyi AA, et al. Mu receptor binding of some commonly used opioids and their
metabolites. Life Sci. 1991;48:2165-2171.
193. Lalovic B, Phillips B, Risler LL, et al. Quantitative contribution of CYP2D6 and CYP3A to oxycodone
metabolism in human liver and intestinal microsomes. Drug Metab Disp. 2004;32:447-454.
194. Lalovic B, Kharasch E, Hoffer C, et al. Pharmacokinetics and pharmacodynamics of oral oxycodone in
healthy human subjects: role of circulating active metabolites. Clin Pharmacol Ther. 2006;79:461-479.
195. Susce MT, Murray-Carmichael E, de Leon J. Response to hydrocodone, codeine and oxycodone in a
CYP2D6 poor metabolizer. Prog Neuro-Psychopharmacol Biol Psych. 2006;30:1356-1358.
196. Foster A, Mobley E, Wang Z. Complicated pain management in a CYP450 2D6 poor metabolizer. Pain
Practice. 2007;7:352-356.
197. Effexor XR® (prescribing information/package insert). Philadelphia, PA: Wyeth Pharmaceuticals; 2009.
198. Lemberger L, Bergstrom RF, Wolen RL, et al. Fluoxetine: clinical pharmacology and physiologic dispo-
sition. J Clin Psychiatry. 1985;46:14-19.

199. Aklillu E, Persson I, Bertilsson L, et al. Frequent distribution of ultrarapid metabolizers of debrisoquine
in an Ethiopian population carrying duplicated and multiduplicated functional CYP2D6 alleles.
200. McLellan RA, Oscarson M, Seidegard J, et al. Frequent occurrence of CYP2D6 gene duplication in
Saudi Arabians. Pharmacogenetics. 1997;7:187-191.
201. Ultram® ER (prescribing information/package insert). Raritan, NJ: Ortho-McNeil Inc.; 2007.
202. Christup LL. Morphine metabolites. Acta Anaesthesiol Scand. 1997;41:116-122.

203. Tang W. The metabolism of diclofenac—enzymology and toxicology perspectives. Curr Drug Metab.
2003;4:319-329.
204. Subrahmanyam V, Renwick AB, Walters DG, et al. Identification of cytochrome P-450 isoforms
responsible for cis-tramadol metabolism in human liver microsomes. Drug Metab Disp. 2001;29:1146-
1155.
205. Lotsch J, Skarke C, Liefhold J, et al. Genetic predictors of the clinical response to opioid analgesics:
Clinical utility and future perspectives. Clin Pharmacokinetics. 2004;43:983-1013.
206. Madadi P, Koren G, Cairns J, et al. Safety of codeine during breastfeeding: Fatal morphine poisoning in
the breastfed neonate of a mother prescribed codeine. Can Fam Physician. 2007;53:33-35.
207. Madadi P, Ross CJD, Hayden MR, et al. Pharmacogenetics of neonatal opioid toxicity following mater-
nal use of codeine during breastfeeding: a case–control study. Clin Pharmacol Ther. 2008;85:31-35.
208. Gasche Y, Daali Y, Fathi M, et al. Codeine intoxication associated with ultrarapid CYP2D6 metabolism.
N Engl J Med. 2004;351:2827-2831.
209. Ferreiros N, Dresen S, Hermanns-Clausen M, et al. Fatal and severe codeine intoxication in 3-year-old
twins—interpretation of drug and metabolite concentrations. Int J Legal Med. Published online: 07
April 2009.
210. Hermanns-Clausen M, Weinmann W, Auwarter V, et al. Drug dosing error with drops—severe clinical
course of codeine intoxication in twins. Eur J Pediatr. 2009;168:819-824.
211. Lim YC, Desta Z, Flockhart DA, et al. Endoxifen (4-hydroxy-N-desmethyl-tamoxifen) has anti-estro-
genic effects in breast cancer cells with potency similar to 4-hydroxy-tamoxifen. Cancer Chemother
Pharmacol. 2005;55:471-478.
212. Burstein HJ, Temin S, Anderson H, et al. Adjuvant endocrine therapy for women with hormone recep-
tor-positive breast cancer: American Society of Clinical Oncology Clinical Practice Guideline Focused
Update. J Clin Oncol. 2014;32:2255-2269.
213. Teft WA, Mansell SE, Kim RB. Endoxifen, the active metabolite of tamoxifen, is a substrate of the
efflux transporter P-glycoprotein (multidrug resistance 1). Drug Metab Dispos. 2011;39:558-562.
214. Hoskins JM, Carey LA, McLeod HL. CYP2D6 and tamoxifen: DNA matters in breast cancer. Nat Rev
Cancer. 2009;9:576-586.
215. Schroth W, Goetz MP, Hamann U, et al. Association between CYP2D6 polymorphisms and outcomes
among women with early stage breast cancer treated with tamoxifen. JAMA. 2009;302:1429-1436.
216. Goetz MP, Rae JM, Suman VJ, et al. Pharmacogenetics of tamoxifen biotransformation is associated
with clinical outcomes of efficacy and hot flashes. J Clin Oncol. 2005;23:9312-9318.
217. Jin Y, Desta Z, Stearns V, et al. CYP2D6 genotype, antidepressant use, and tamoxifen metabolism
during adjuvant breast cancer treatment. J Natl Cancer Inst. 2005;97:30-39.
218. Goetz MP, Knox SK, Suman VJ, et al. The impact of cytochrome P450 2D6 metabolism in women
receiving adjuvant tamoxifen. Breast Cancer Res Treat. 2007;101:113-121.
219. Higgins MJ, Rae JM, Flockhart DA, et al. Pharmacogenetics of tamoxifen: who should undergo CYP2D6
genetic testing? J Natl Compr Canc Netw. 2009;7:203-213.
220. Swen JJ, Nijenhuis M, de Boer A, et al. Pharmacogenetics: from bench to byte—an update of guidelines.
Clin Pharmacol Ther. 2011;89:662-673.
221. Desta Z, Ward BA, Soukhova, NV, et al. Comprehensive evaluation of tamoxifen sequential biotrans-
formation by the human cytochrome P450 system in vitro: prominent roles for CYP3A and CYP2D6.
222. Greenblatt DJ, von Moltke LL, Harmatz JS, et al. Human cytochromes and some newer antidepres-
sants: kinetics, metabolism, and drug interactions. J Clin Psychopharmacol. 1999;19(suppl 1);23S-35S.
223. Caccia S. Metabolism of newer antidepressants: an overview of the pharmacological and pharmaco-
kinetic implications. Clin Pharmacokinet. 1998;34:281-302.
224. Paxil® (prescribing information/package insert). Research Triangle Park, NC: GlaxoSmithKline; 2009.
225. Effexor XR® (prescribing information/package insert). Philadelphia, PA: Wyeth Pharmaceuticals; 2009.
226. Mannens G, Huang ML, Meuldermans W, et al. Absorption, metabolism, and excretion of risperidone in
humans. Drug Metab Dispos. 1993;21:1134-1141.
227. Sallee FR, DeVane CL, Ferrell RE. Fluoxetine-related death in a child with cytochrome P-450 2D6
genetic deficiency. J Child Adolesc Psychopharmacol. 2000;10:27-34.
228. de Leon J, Susce MT, Pan RM, et al. The CYP2D6 poor metabolizer phenotype may be associated with
risperidone adverse drug reactions and discontinuation. Eur J Pediatr. 2009;168:819-824.
229. Lessard E, Yessine MA, Hamelin BA, et al. Influence of CYP2D6 activity on the disposition and cardio-
vascular toxicity of the antidepressant agent venlafaxine in humans. Pharmacogenetics. 1999;9:435-443.
230. Nkhoma ET, Poole C, Vannappagari V, et al. The global prevalence of glucose-6-phosphate dehydro-
genase deficiency: a systematic review and meta-analysis. Blood Cells Mol Dis. 2009;42:267-278.
231. Gonzalez FJ, Fernandez-Salguero P. Diagnostic analysis, clinical importance and molecular basis of
dihydropyrimidine dehydrogenase deficiency. Trends Pharmacol Sci. 1995;16:325-327.
232. FUDR® (prescribing information/package insert). Paramus, NJ: Faulding Pharmaceutical Company;
2002.
233. Xeloda® (prescribing information/package insert). Nutley, NJ: Roche Laboratories Inc.; 2006.
234. Soong R, Diasio RB. Advances and challenges in fluoropyrimidine pharmacogenomics and pharmaco-
genetics. Pharmacogenomics. 2005;6:835-847.
235. Ho DH, Townsend L, Luna MA, et al. Distribution of dihydrouracil dehydrogenase activities using
5-fluorouracil as a substrate. Anticancer Res. 1986;6:781-784.
236. Goldberg RM. Therapy for metastatic colorectal cancer. The Oncologist. 2006;11:981-987.
237. Pasetto LM, Jirillo A, Iadicicco G, et al. FOLFOX versus FOLFIRI: a comparison of regimens in the treat-
ment of colorectal cancer metastases. Anticancer Res. 2005;25(1B):563-576.
238. Podoltsev NA, Saif MW. Dihydropyrimidine dehydrogenase gene (DPD) polymorphism among Cauca-
sian patients (pts) with 5-FU and capecitabine (CAP)-related toxicity. J Clin Oncol. 2009;27(May 20
Supplement):e14588.
239. Saif MW, Ezzeldin H, Vance K, et al. DPD*2A mutation: the most common mutation associated with
DPD deficiency. Cancer Chemother Pharmacol. 2007;60:503-507.
240. Milano G, Etienne MC. Potential importance of dihydropyrimidine dehydrogenase (DPD) in cancer
chemotherapy. Pharmacogenetics. 1994;4:301-306.
241. Milano G, Etienne MC. Dihydropyrimidine dehydrogenase (DPD) and clinical pharmacology of 5-fluoro-
uracil. Anticancer Res. 1994;14:2295-2297.
242. Amstutz U, Farese S, Aebi S, et al. Dihydropyrimidine dehydrogenase gene variation and severe 5-fluo-
rouracil toxicity: a haplotype assessment. Pharmacogenomics. 2009;10:931-944.
243. Van Kuilenburg AB, Meinsma R, Zoetekouw L, et al. Increased risk of grade IV neutropenia after admin-
istration of 5-fluorouracil due to a dihydropyrimidine dehydrogenase deficiency: high prevalence of
the IVS14+1g>a mutation. Int J Cancer. 2002;101:253-258.
244. Harris BE, Carpenter JT, Diasio RB. Severe 5-fluorouracil toxicity secondary to dihydropyrimidine dehy-
drogenase deficiency. Cancer. 1991;68:499-501.
245. Lecomte T, Ferraz J-M, Zinzindohoué F, et al. Thymidylate synthase gene polymorphism predicts
toxicity in colorectal cancer patients receiving 5-fluorouracil-based chemotherapy. Clin Cancer Res.
2004;10:5880-5888.
246. Pullarkat ST, Stoehlmacher J, Ghaderi V, et al. Thymidylate synthase gene polymorphism determines
response and toxicity of 5-FU chemotherapy. Pharmacogenomics J. 2001;1:65-70.
247. Ichikawa W, Takahashi T, Suto K, et al. Orotate phosphoribosyltransferase gene polymorphism
predicts toxicity in patients treated with bolus 5-fluorouracil regimen. Clin Cancer Res. 2006;12:3928-
3934.
248. Schwab M, Zanger UM, Marx C, et al. Role of genetic and non-genetic factors for fluorouracil treat-
ment related severe toxicity: a prospective clinical trial by the German 5-FU toxicity study group. J Clin
Oncology. 2008;26:2131-2138.
249. Ezzeldin HH, Diasio RB. Predicting fluorouracil toxicity: can we finally do it? J Clin Oncology.
2008;26:2080-2082.
250. Gross E, Busse B, Riemenschneider M, et al. Strong association of a common dihydropyrimidine dehy-
drogenase gene polymorphism with fluoropyrimidine-related toxicity in cancer patients. PLoS One.
2008;3:e4003.
251. Omura K. Clinical implications of dihydropyrimidine dehydrogenase (DPD) activity in 5-FU-based
chemotherapy: mutations in the DPD gene, and DPD inhibitory fluoropyrimidines. Int J Clin Oncol.
2003;8:132-138.
252. van Kuilenburg AB, Vreken P, Beex LV, et al. Severe 5-fluorouracil toxicity caused by reduced dihydro-
pyrimidine dehydrogenase activity due to heterozygosity for a G-->A point mutation. J Inherit Metab
Dis. 1998;21:280-284.
253. Wei X, McLeod HL, McMurrough J, et al. Molecular basis of the human dihydropyrimidine dehydroge-
nase deficiency and 5-fluorouracil toxicity. J Clin Invest. 1996;98:610-615.
254. Raida M, Schwabe W, Hausler P, et al. Prevalence of a common point mutation in the dihydropyrimi-
dine dehydrogenase (DPD) gene within the 5′-splice donor site of intron 14 in patients with severe
5-fluorouracil (5-FU)-related toxicity compared with controls. Clinical Cancer Research. 2001;7:2832.
255. Mattison LK, Fourie J, Desmond RA, et al. Increased prevalence of dihydropyrimidine dehydrogenase
deficiency in African-Americans compared with Caucasians. Clin Cancer Res. 2006;12:5491-5495.
256. Salonga D, Danenberg KD, Johnson M, et al. Colorectal tumors responding to 5-fluorouracil have low
gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine

phosphorylase. Clin Cancer Res. 2000;6:1322-1327.
257. Vallbohmer D, Yang DY, Kuramochi H, et al. DPD is a molecular determinant of capecitabine efficacy in
colorectal cancer. Int J Oncol. 2007;31:413-418.
258. Mattison LK, Fourie J, Hirao Y, et al. The uracil breath test in the assessment of dihydropyrimidine
dehydrogenase activity: pharmacokinetic relationship between expired 13CO2 and plasma [2-13C]dihy-
drouracil. Clin Cancer Res. 2006;12:549-555.
259. Furuhata T, Kawakami M, Okita K, et al. Plasma level of a 5-fluorouracil metabolite, fluoro-beta-
alanine correlates with dihydropyrimidine dehydrogenase activity of peripheral blood mononuclear
cells in 5-fluorouracil treated patients. J Exp Clin Cancer Res. 2006;25:79-82.
260. McDonagh EM, Thorn CF, Bautista JM, et al. PharmGKB summary: very important pharmacogene infor-
mation for G6PD. Pharmacogenet Genomics. 2012;22:219-228.
261. Mason PJ, Bautista JM, Gilsanz F. G6PD deficiency: the genotype-phenotype association. Blood Rev.
2007;21:267-283.
262. Alving AS, Carson PE, Flanagan CL, et al. Enzymatic deficiency in primaquine-sensitive erythrocytes.
Science. 1956;124:484-485.
263. Ruwende C, Khoo SC, Snow RW, et al. Natural selection of hemi- and heterozygotes for G6PD defi-
ciency in Africa by resistance to severe malaria. Nature. 1995;376:246-249.
264. Tan IK, Whitehead TP. Automated fluorometric determination of glucose-6-phosphate dehydroge-
nase (G6PD) and 6-phosphogluconate dehydrogenase (6PGD) activities in red blood cells. Clin Chem.
1969;15:467-478.
265. Beutler E. A series of new screening procedures for pyruvate kinase deficiency, glucose-6-phosphate
dehydrogenase deficiency, and glutathione reductase deficiency. Blood. 1966;28:553-562.
266. Van Noorden CJ, Vogels IM. A sensitive cytochemical staining method for glucose-6-phosphate dehy-
drogenase activity in individual erythrocytes. II. Further improvements of the staining procedure and
some observations with glucose-6-phosphate dehydrogenase deficiency. Br J Haematol. 1985;60:57-
63.
267. iHOP (information linked over proteins) web site. Butyrylcholinesterase. Available at: http://www.ihop-
net.org/UniPub/iHOP/gs/86703.html. Accessed November 3, 2015.
268. Kalow W, Staron N. On distribution and inheritance of atypical forms of human serum cholinesterase,
as indicated by dibucaine number. Can J Biochem. 1957;35:1306-1317.
269. Kalow W, Gunn DR. Some statistical data on atypical cholinesterase of human serum. Ann Hum
Genet. 1959;23:239-250.
270. Kalow W, Gunn DR. The relation between dose of succinylcholine and duration of apnea in man. J
Pharmacol Exp Ther. 1957;120:203-214.
271. Ostergaard D, Viby-Mogensen J, Rasmussen SN, et al. Pharmacokinetics and pharmacodynamics of
mivacurium in patients phenotypically homozygous for the atypical plasma cholinesterase variant—
effect of injection of human cholinesterase. Anesthesiology. 2005;102:1124-1132.
272. Barta C, Sasvari-Szekely M, Devai A, et al. Analysis of mutations in the plasma cholinesterase gene
of patients with a history of prolonged neuromuscular block during anesthesia. Mol Genet Metab.
2001;74:484-488.
273. Jensen FS, Viby-Mogensen J. Plasma cholinesterase and abnormal reaction to succinylcho-
line—20 years experience with the Danish Cholinesterase Research Unit. Acta Anaesthesiol Scand.
1995;39:150-156.
274. Wood M. Pharmacogenetics and anesthetic toxicity. In: Rice SA, Fish KJ, eds. Anesthetic Toxicity. New
York, NY: Raven Press Ltd; 1994:199-218.
275. Hoffman RS, Henry GC, Wax PM, et al. Decreased plasma cholinesterase activity enhances cocaine
toxicity in mice. Pharmacol Exp Ther. 1992;263:698-702.
276. Morton CL, Wadkins RM, Danks MK, et al. The anticancer prodrug CPT-11 is a potent inhibitor of
acetylcholinesterase but is rapidly catalyzed to SN-38 by butyrylcholinesterase. Cancer Research.
1999;59:1458-1463.
277. Hopkins PM. Malignant hyperthermia: pharmacology of triggering. Br J Anaesth. 2011;107:48-56.
278. Fiszer D, Shaw M-A, Fisher NA, et al. Next-generation sequencing of RYR1 and CACNA1S in malignant
hyperthermia and exertional heat illness. Anesthesiology. 2015;5:1033-1046.
279. Leiderer BM, Borchardt RT. Enzymes involved in the bioconversion of ester-based prodrugs. J Pharm
Sci. 2006;95:1177-1195.
280. Hoffman RS, Henry GC, Howland MA, et al. Association between life-threatening cocaine toxicity and
plasma cholinesterase activity. Ann Emerg Med. 1992;21:247-253.
281. Souza RLR, Mikami LR, Maegawa EA, et al. Four new mutations in the BCHE gene of human butyryl-
cholinesterase in a Brazilian blood donor sample. Mol Genet Metab. 2005;84:349-353.
282. Parmo-Follani F, Nunes K, Lepienski LM, et al. Two new mutations of the human BCHE gene (IVS3-
14T>C and L574fsX576). ChemicoBiological Interactions. 2008;175:135-137.
283. Lando G, Mosca A, Bonora R, et al. Frequency of butyrylcholinesterase gene mutations in individuals
with abnormal inhibition numbers: an Italian population study. Pharmacogenetics. 2003;13:265-270.
284. Pestel G, Sprenger H, Rothhammer A. Frequency distribution of dibucaine numbers in 24,830 patients.
Anaesthesist. 2003;52:495-499.
285. Bartels CF, Jensen FS, Lockridge O, et al. DNA mutation associated with the human butyrylcholines-
terase K-variant and its linkage to the atypical variant mutation and other polymorphic sites. Am J
Hum Genet. 1992;50:1086-1103.
286. Evans RT. Cholinesterase phenotyping: clinical aspects and laboratory applications. CRC Crit Rev Clin
Lab Sci. 1986;23:35-64.
287. Argikar UA, Upendra A, Iwuchukwu OF, et al. Update on tools for evaluation of uridine diphosphogluc-
uronosyltransferase polymorphisms. Exp Opin Drug Metab Toxicol. 2008;4:879-894.
288. Price Evans DA, Manley KA, McKusick VA. Genetic control of isoniazid metabolism in man. Br Med J.
1960;2:485-491.
289. Price Evans DA, Storey PB, Wittstadt FB. Determination of the isoniazid inactivator phenotype. Am
Respir Dis. 1960;82:853-861.
290. Database of Arylamine N-Acetyltransferases. Available at: http://nat.mbg.duth.gr/Human%20
NAT2%20alleles_2013.htm. Accessed November 3, 2015.
291. Ma MK, Woo MH, McLeod HL. Genetic basis of drug metabolism. Am J Health-Syst Pharm.
2002;59:2061-2069.
292. Lin HJ, Han CY, Lin BK, et al. Ethnic distribution of slow acetylator mutations in the polymorphic
N-acetyltransferase (NAT2) gene. Pharmacogenetics. 1994;4:125-134.
293. Rey E, Gendrel D, Treluyer JM, et al. Isoniazid pharmacokinetics in children according to acetylator
phenotype. Fund Clin Pharmacol. 2001;15:355-359.
294. Cho H-J, Koh W-J, Ryu Y-J, et al. Genetic polymorphisms of NAT2 and CYP2E1 associated with anti-
tuberculosis drug-induced hepatotoxicity in Korean patients with pulmonary tuberculosis. Tuberculosis.
2007;87:551-556.
295. Coulthard SA, Hogarth LA, Little M, et al. The effect of thiopurine methyltransferase expression on
sensitivity to thiopurine drugs. Mol Pharmacol. 2002;62:102-109.
296. McLeod HL, Siva C. The thiopurine S-methyltransferase gene locus—implications for clinical pharma-
cogenomics. Pharmacogenomics. 2002;3:89-98.
297. Purinethol® (prescribing information/package insert). Sellersville, PA: Gate Pharmaceuticals; 2007.
298. Imuran® (prescribing information/package insert). San Diego, CA: Prometheus Laboratories Inc; 2008.
299. Tabloid® (prescribing information/package insert). Research Triangle Park, NC: GlaxoSmithKline; 2009.
300. Weinshilboum R. Thiopurine pharmacogenetics: clinical and molecular studies of thiopurine methyl-
transferase. Drug Metab Dispos. 2001;29:601-605.
301. Teml A, Schaeffeler E, Schwab M. Pretreatment determination of TPMT—state of the art in clinical
practice. Eur J Clin Pharmacol. 2009;65:219-221.
302. Relling MV, Gardner EE, Sandborn WJ, et al. Clinical Pharmacogenetics Implementation Consortium
guidelines for thiopurine methyltransferase genotype and thiopurine dosing. Clin Pharmacol Ther.
2011;89:387-391.
303. Schaeffeler E, Fischer C, Brockmeier D, et al. Comprehensive analysis of thiopurine S-methyl-transfer-

ase phenotype-genotype correlation in a large population of German-Caucasians and identification
of novel TPMT variants. Pharmacogenetics. 2004;14:407-417.
304. Xin HW, Xiong H, Wu XC, et al. Relationships between thiopurine S-methyltransferase polymorphism
and azathioprine-related adverse drug reactions in Chinese renal transplant recipients. Eur J Clin
Pharmacol. 2009;65:249-255.
305. TPMT. Genetics Home Reference, National Institutes of Health. Available at: http://ghr.nlm.nih.gov/
gene/TPMT. Accessed November 3, 2015.
306. King CD, Rios GR, Green MD, et al. UDP-glucuronosyltransferases. Curr Drug Metab. 2000;1:143-161.
307. Kiang TK, Ensom MH, Chang TK. UDP-glucuronosyltransferases and clinical drug-drug interactions.
Pharmacol Ther. 2005;106:97-132.
308. De Wildt SN, Kearns GL, Leeder JS, et al. Glucuronidation in humans. Pharmacogenetic and develop-
mental aspects. Clin Pharmacokinet. 1999;36:439-452.
309. Mackenzie PI, Miners JO, McKinnon RA. Polymorphisms in UDP glucuronosyltransferase genes: func-
tional consequences and clinical relevance. Clin Chem Lab Med. 2000;38:889-892.
310. Miners JO, McKinnon RA, Mackenzie PI. Genetic polymorphisms of UDP-glucuronosyltransferases and
their functional significance. Toxicology. 2002;181:453-456.
311. Burchell B. Genetic variation of human UDP-glucuronosyltransferase: implications in disease and drug
glucuronidation. Am J Pharmacogenomics. 2003;3:37-52.
312. Gullemette C. Pharmacogenomics of human UDP-glucuronosyltransferase enzymes. Pharmaco-
genomics J. 2003;3:136-158.
313. Stingl JC, Bartels H, Viviani R, et al. Relevance of UDP-glucuronosyltransferase polymorphisms for
drug dosing: A quantitative systematic review. Pharmacol Ther. 2014;141:92-116.
314. Mani S. UGT1A1 polymorphism predicts irinotecan toxicity: evolving proof. AAPS PharmSci. 2001;3:1.
315. Innocenti F, Iyer L, Ratain MJ. Pharmacogenetics of anti-cancer agents: lessons from amonafide and
irinotecan. Drug Metab Dispos. 2001;29:596-600.
316. Ando Y, Saka H, Ando M, et al. Polymorphisms of UDP-glucuronosyltransferase gene and irinotecan
toxicity: a pharmacogenetic analysis. Cancer Res. 2000;60:6921-6926.
317. Iyer L, King CD, Whitington PF, et al. Genetic predisposition to the metabolism of irinotecan (CPT-11).
J Clin Invest. 1998;101:847-854.
318. Bosma PJ, Chowdhury JR, Bakker C, et al. The genetic basis of the reduced expression of bilirubin
UDP-glucuronosyltransferase 1 in Gilbert’s syndrome. N Engl J Med. 1995;333:1171-1175.
319. Hall D, Ybazeta G, Destro-Bisol G, et al. Variability at the uridine diphosphate glucuronosyltransferase
1A1 promoter in human populations and primate. Pharmacogenetics. 1999;9:591-599.
320. Beutler E, Gelbart T, Demina A. Racial variability in the UDP-glucuronosyltransferase 1 (UGT1A1)
promoter: a balanced polymorphism for regulation of bilirubin metabolism? Proc Natl Acad Sci USA.
1998;95:8170-8174.
321. PharmGKB: The Pharmacogenomics Knowledge Base. Important Variant Information for UGT1A1. Avail-
able at: https://www.pharmgkb.org/gene/PA420?tabType=tabVip#ImportantVariantInformationforUGT
1A1-28. Accessed November 3, 2015.
322. Rouits E, Boisdron-Celle M, Dumont A, et al. Relevance of different UGT1A1 polymorphisms in irinote-
can-induced toxicity: a molecular and clinical study of 75 patients. Clin Cancer Res. 2004;10:5151-5159.
323. Iyer L, Das S, Janisch L, et al. UGT1A1*28 polymorphism as a determinant of irinotecan disposition and
toxicity. Pharmacogenomics J. 2002;2:43-47.
324. Kim TW, Innocenti F. Insights, challenges, and future directions in irinogenetics. Ther Drug Monit.
2007;29:265-270.
325. Camptosar® (prescribing information/package insert). New York, NY: Pharmacia & Upjohn Co; 2008.
326. Kuhlmann J. The applications of biomarkers in early clinical drug development to improve decision-
making processes. Ernst Schering Res Found Workshop. 2007;59:29-45.
327. Bakhtiar R. Biomarkers in drug discovery and development. J Pharm Tox Methods. 2008;57:85-91.
328. U.S. Food and Drug Administration. “Drug Metabolizing Enzymes” section of page on Nucleic Acid
Based Tests. Available at: http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/InVitro-
Diagnostics/ucm330711.htm. Accessed November 3, 2015.
329. U.S. Food and Drug Administration. Critical Path Initiative. Available at: http://www.fda.gov/ScienceRe-
search/SpecialTopics/CriticalPathInitiative/. Accessed November 3, 2015.
330. Collins S. Pharmacists called to take lead in pharmacogenomics—the bridge to personalized

medicine. Pharmacy Today-Health System Edition. November 1, 2015.
CHAPTER
5
Pharmacogenomics of
Transport Proteins
Arthur G. Cox, PhD

After completing this chapter, the reader ENTEROCYTE—cells lining the intestine that
should be able to contain transport proteins with protective
roles.
Classify drug transport proteins by their
location in tissues and cells and their LINKAGE DISEQUILIBRIUM—a nonrandom
physiological function. association of alleles, causing a certain
combination to occur more or less
Discuss the general clinical relevance of
frequently than otherwise expected.
the transporters and their inhibitors.
MISSENSE MUTATION—a polymorphism that
Describe classes of drugs whose
results in a different amino acid being
pharmacokinetics are affected by
expressed in the protein.
individual transporters.
NONSENSE MUTATION—a polymorphism that
Relate genetic variants of transport
results in a premature stop codon.
proteins to variations in drug action.
NONSYNONYMOUS MUTATION—nucleotide
substitution in a gene that results in a
change in the amino acid sequence of a
protein.
SINGLE NUCLEOTIDE POLYMORPHISM
(SNP)—a point mutation occurring in >1%
of the population.
SYNCYTIOTROPHOBLAST—multinucleated
cells in the placenta containing transport
proteins that serve a protective function
for the fetus.
SYNONYMOUS MUTATION—the substitution
of a nucleotide within a gene that does
not result in a change in an amino acid in
the expressed protein.
TRANSPORTER—a protein embedded in a
cell membrane responsible for either

removing substances from a cell or
bringing them into the cell or membrane-
bound vesicle within the cell.
151
INTRODUCTION
Genetic variability of drug-metabolizing enzymes has long been recognized as a factor
in both altered therapeutic response and adverse effects in individuals and in patient
populations. The cytochrome family of enzymes is particularly important in this regard.
Another area where genotype can strongly affect drug response is that of transport
proteins. This chapter will discuss the importance of transport proteins in drug absorp-
tion and response and also review recent information on the effect of genetic variability
on these transporters.
Transporters are those proteins that carry either endogenous compounds or xeno-
biotics across biological membranes. They can be classified into either efflux or uptake
proteins, depending on the direction of transport. The extent of expression of genes coding
for transport proteins can have a profound effect on the bioavailability and pharmaco-
kinetics of various drugs. Additionally, genetic variation such as single nucleotide polymor-
phisms (SNPs) of the transport proteins can cause differences in the uptake or efflux of
drugs. In terms of cancer chemotherapy, tumor cells expressing these proteins can have
either enhanced sensitivity or resistance to various anticancer drugs.1
Polymorphisms in the CYP450 drug-metabolizing isoforms are known to cause large
variations in pharmacokinetic parameters for systemic drug exposure. Genetic variation
of membrane transporters typically results in less than a threefold alteration of systemic
drug levels. However, intracellular drug levels may be altered significantly more.2 Trans-
porters that serve as efflux pumps on a cell membrane can remove drugs from the cell
before they can act. Transport proteins that are responsible for the vital influx of ions
and nutrients, such as glucose, can promote growth of tumor cells if overexpressed or
lead to increased susceptibility to a drug if the transporter carries that drug into the cell.
Additionally, genetic variants of transport proteins can cause or contribute to a number of
diseases such as cystic fibrosis, retinal degeneration, hypercholesterolemia, bile transport
defects, and anemia.3
Two superfamilies of transport proteins have important effects on the absorption, distri-
bution, and excretion of drugs: the ATP-binding cassette (ABC) and the solute carrier (SLC)
superfamilies. With the advent of high-throughput screening methods, the screening of
large volumes of samples of SNPs has become viable. Public databases of the genetic vari-
ants that have been discovered are available and include those maintained by the Human
Genome Organization Gene Nomenclature Committee (HGNC), National Center for Biotech-
nology Information (NCBI) SNP database (dbSNP), the National Human Genome Research
Institute haploid map (HapMap), the Japanese SNP database (JSNP), the Online Mendelian
Inheritance in Man (OMIM) database from Johns Hopkins Hospital, and the pharmaco-
genetics and pharmacogenomics knowledge base at Stanford University (PharmGKB).
Additionally, information related specifically to the pharmacogenomics of human trans-
porters has recently been gathered into a curated database known as the Human Trans-
porter Database.4
CASE STUDY—IRINOTECAN
C.B. is a 44-year-old Caucasian female who was diagnosed with colon cancer 6 months
ago. Due to her limited financial resources, she decided to seek treatment in another
country where medical care is much less expensive but not necessarily cutting edge. Her
Chapter 5 • Pharmacogenomics of Transport Proteins 153
new oncologist initiated therapy with irinotecan at 150 mg/m2 dosage level. After her
first infusion, C.B. soon developed mouth ulcers that were very slow to heal. On traveling
back to the United States, she consulted her general practitioner. He ordered blood tests,
which revealed her blood contained an unusually low number of neutrophils; in fact,
she was suffering from neutropenia. The oral ulcers were diagnosed as resulting from
an infection due to her inadequate white blood cells. Further genetic testing showed that
she was homozygous for a polymorphism in the promoter region of the UDP-glucuronyl
transferase gene (UGT1A1*28/*28). UGT1A1 is the primary enzyme responsible for
inactivation of irinotecan’s active metabolite (SN-38). It has been suggested that patients
carrying this polymorphism may benefit from a reduction in the starting dose of
irinotecan.
A complicating factor is that C.B. was also found to have a polymorphism of a specific
transport protein, which may compound the liability for toxicity associated with
irinotecan treatment (ABCB1 [1236C>T]). In fact, when treatment was continued at a
lower dosage level, the cancer was successfully treated with greatly reduced neutropenia
compared to her initial treatment.
Irinotecan is widely used in cancer chemotherapy, but it has been associated with
unpredictable severe toxic reactions such as myelosuppression and delayed-type
diarrhea. Polymorphisms of the drug-metabolizing enzyme family UGT1A is a known
contributor to varied response and toxicity of irinotecan in different individuals.
Polymorphisms in genes coding for drug transport proteins such as ABCB1, ABCC1,
ABCC2, ABBC5, ABCG1, ABCG2, and SLC01B1 have been suggested to possess additive or
synergistic effects with UGT1A1.5-9
Questions
1. What patient parameters should normally be considered when determining the correct dose
for irinotecan?
2. How can genetic polymorphisms affect patient response to irinotecan?
3. How can knowledge of pharmacogenomics improve the therapeutic use and safety profile of
irinotecan?
INDIVIDUAL TRANSPORTERS OF PHARMACOGENOMIC

INTEREST
ABC Transporters
ABC transporters are present in cellular and intracellular membranes and can be respon-
sible for either importing (influx) or removing (efflux) substances from cells and tissues.
They often use the hydrolysis of ATP to drive the transport substances against a concen-
tration gradient. There are at least 48 ABC transporter genes, which are divided into seven
different families (A–G) based on sequence similarity. Members of three of these seven
gene families are particularly important for drug transport and multiple drug resistance in
tumor cells10:
1. ABCB1 gene, encoding multidrug resistance protein-1 (MDR1), which is also known as
P-glycoprotein (P-gp)
2. ABCC family (ABCC1 through ABCC6) or multidrug resistance proteins (MRPs)
3. ABCG2 (breast cancer resistance protein)
ABC transporters are characterized by the homology of their ATP-binding regions. All
families except one (ABCG2) contain two ATP-binding regions and two transmembrane
domains. The transmembrane domains contain multiple alpha helices, which span the lipid
bilayer. The number of alpha helices in a transmembrane domain differs depending on
the family. The ATP-binding regions are located on the cytoplasmic side of the membrane
(Figure 5-1). In addition to being important mediators of resistance in human chemotherapy,

ABC transporters are found in bacteria and can contribute to the development of resistance
to multiple antibiotics. The localization of the proteins depends on the cell type such as
hepatocyte, enterocyte, and renal proximal tubule (Figure 5-2). The majority of ABC trans-
porters move compounds from the cytoplasm to the outside of a cell, although some
move compounds into an intracellular compartment such as the endoplasmic reticulum,
mitochondria, or peroxisome.
The exact mechanism used by ABC transporters to transport substrates has not been
fully elucidated. It has been proposed that there is an ATP-dependent conformational
change in the protein that causes the substrate to be pumped across the membrane. This
hypothesis has been supported by recent X-ray crystallographic studies, which have shown
that both import and export proteins oscillate between two conformations: one in which the
substrate binding site is open to the cytoplasm, and one in which the binding site faces the
opposite side of the membrane.11 ATP binding and hydrolysis are proposed to play separate
roles in the cycle. ATP binding favors the outward facing orientation, while ATP hydrolysis
returns the transporter back to the inward facing conformation (Figure 5-3).11,12 In this way,
ATP can be used to drive the transport of a substance against its concentration gradient.
ABCB1 Transporters: P-glycoprotein
The ABCB1 gene codes for a glycosylated membrane protein originally detected in cells
that had developed resistance to cancer chemotherapy agents. The protein is commonly
NH2 out
Membrane
NBD
in
NBD
CO2H
FIGURE 5-1 General structure of ATP-binding cassette (ABC) transporters showing

transmembrane and nucleotide binding domains (NBD). Individual members of the

superfamily contain differing numbers of transmembrane helices within the transmembrane
domains. The example shown here illustrates MRP1 (ABCC1). ABCG2 transporters differ from
the rest of the members of the superfamily in that they have only 1 ATP-binding domain. The
alpha helices making up the transmembrane segments and the nucleotide binding regions
are critical to the function of ABCs (see Figure 5-3).15 Figure adapted from reference 15.
A
BLOOD
MRP3 MRP1
Enterocyte
ATP ATP
ATP ATP ATP
BCRP MRP2 MDR1 PEPT1 OCTN2

GUT LUMEN
B
BLOOD
OATP-B OATP-C OATP8
Hepatocyte ATP ATP Hepatocyte
MDR1 MRP2
ATP ATP
Bile MDR3 BCRP Bile
ATP ATP ATP ATP
MRP1 MRP3 MRP4 MRP5 OCT1 OAT2

BLOOD
C
BLOOD
MRP1 MRP3 MRP5 OAT1 OAT2 OAT3 OCT2
ATP ATP ATP
Proximal Tubule
ATP ATP ATP
MRP2 MRP4 MDR1 PEPT2 OAT4 OCTN1 OCTN2

URINE
Prepared by Holly R. Fischer, MFA
FIGURE 5-2 Localization of transporters in differing cell types: in A, small intestine

enterocyte; in B, hepatocyte with canaliculi; and in C, renal proximal tubule. In addition

to those transporters discussed in the text, other transport proteins with protective and
possible pharmacogenomic relevance are shown. OCTN1 and OCTN2: novel organic
cation transporters-1 and 2 (SLC22A4, SLC22A5), OATP-B: organic anion transporting
polypeptide-B (SLCO2B1), OATP-C (SLCO1B1), OATP8 (SLCO1B3), OCT1 (SLC22A1), OAT2
(SLC22A7).19 Figure adapted from reference 19.
exterior
transmembrane
1. domain
cytoplasm nucleotide binding domain

ATP
ATP
substrate ATP
exterior
2. (a)
cytoplasm ATP ATP
(b) ATP
substrate
exterior
3.
cytoplasm ATP ATP
ATP
substrate
FIGURE 5-3 Schematic illustration of the function of ATP efflux transporters.12 In 1, the
two transmembrane domains that make up the functional protein are attached to nucleotide
binding domains (NBDs) that are widely separated. In 2, ATP and the substrate bind to
their domains. Highly lipophilic substrates may diffuse through the plasma membrane (a).
Otherwise, they can diffuse from the cytoplasm to the binding pocket (b). In 3, the nucleotide
binding regions containing ATP undergo a conformational shift, bringing them close
together. In 4, the conformational shift of the NBDs has caused a change in the conformation
of the substrate binding pocket, which opens a pocket to the outside of a cell and allows efflux
of the substrate. In 5, ATP is hydrolyzed to ADP and pyrophosphate (PPi). The protein can
then return to its resting state, with the substrate binding site directed inward.
exterior
4.
cytoplasm
ATP ATP
substrate
exterior
5.
cytoplasm
ATP
ATP
ATP + PPi
substrate
FIGURE 5-3 Continued
referred to P-gp, PGY1, or MDR1. It is designated as an MRP due to the fact that its expres-
sion in a cell may confer resistance to multiple classes of drugs with differing chemical
structures and mechanisms of action. Various cancers have been shown to display low
initial levels of P-gp with levels of expression increasing after chemotherapy and relapse.
A wide variety of transport proteins have been discovered and studied, but P-gp is the best
characterized in terms of distribution and function. Some drugs (e.g., cyclosporine) act as
both substrates and inhibitors of P-gp. Other drugs act only as substrates or only as inhibi-
tors. The substrates for P-gp are often hydrophobic drugs with a polyaromatic skeleton
and a neutral or positive charge.13 P-gp functions as a dimer of 1280 residue polypeptides,
forming a pore across the cell membrane.

In addition to cytotoxic chemotherapeutic agents, many other drugs are transported
across membranes by P-gp. These include protease inhibitors, immunosuppressants,
calcium channel blockers, b-blockers, statins, steroids, antihistamines, anticonvulsants, and
antidepressants. The importance of P-gp for pharmacotherapy has led to great interest
in its pharmacogenomics.14-16 The mechanism of MDR1 upregulation during chemotherapy

has been researched extensively. One mechanism by which its expression is increased is
through activation of the pregnane-X receptor.17 This receptor is also known to upregulate
CYP3A4; as such, it serves as a protective mechanism for the body in the case of toxic
insult (e.g., during cancer chemotherapy).
CLINICAL PEARL
P-gp translocates multiple structurally unrelated drugs out of
cells including anticancer drugs, immunosuppressants, human
immunodeficiency virus protease inhibitors, cardiac drugs, and
β-adrenoreceptor antagonists. Expression of P-gp in a cell may result in
resistance to the effects of a wide variety of drugs, and genetic variation
of the protein may result in differing susceptibility to pharmacotherapy.
Ethnic background can also increase or decrease the likelihood of
interaction between P-gp and a drug.
In addition to being expressed in cancer cells, P-gp is expressed in multiple normal

tissues with excretory or protective function including intestine, kidney, liver, blood-
brain barrier, spinal cord, testes, and placenta. P-gp plays an important role in forming a
protective barrier against absorption of xenobiotics in these tissues. The broad substrate
specificity of P-gp is shared with cytochrome P450 3A4 (CYP3A4), which is well known to
metabolize a diverse set of drugs. This broad specificity, coupled with the tissue localiza-
tion and function of both proteins, has led to the hypothesis that they work in concert by
protecting the body from absorption of harmful compounds and acting synergistically in
the small intestine.
Significant interindividual variability of the amount of P-gp expressed (2−8 fold) has
been demonstrated in healthy volunteers during intestinal biopsy, suggesting the possibil-
ity of variable bioavailability of its substrates. Numerous SNPs of the human MDR1 gene
have been discovered and studied during systematic screening. The frequencies of these
SNPs in a population can vary according to racial/ethnic background.18 At least 29 SNPs
have been found, 19 of them located in exonic regions and 11 of them coding for nonsyn-
onymous mutations.19 Interest in the clinical and functional relevance of polymorphisms of
MDR1 has led to a number of recent reviews.19-21
Two SNPs of particular interest are a mutation in exon 26 at position 3435 (3435C>T)
and a mutation in exon 21 (2677G>T/A). The 3435C>T SNP has been extensively studied
since it is associated with differences in expression or function of P-gp. Change in nucleo-
tide sequence from C to T at position 3435 does not result in a change of amino acids but is
a silent mutation located in the wobble position of the codon. Although there is no change
in the expressed protein, both the level of its expression and function can be variable. For
instance, a two-fold reduction of intestinal P-gp was observed in patients who were homo-
zygous for 3435T.22 In a number of studies, reduction in P-gp level has been correlated with
differences in pharmacokinetic parameters for substrates such as digoxin. A possible expla-

nation for the reduction in P-gp expression with homozygous 3435T genotype is a reduction
in translation of the protein.23
SNPs AND ANTICANCER AGENTS. Other studies have probed the relationship of SNPs in
MDRI with various anticancer agents that are substrates for P-gp transport. For example,
one recent study investigated the correlation of MDR1 polymorphisms with clinical response
to docetaxel-cisplatin in non-small cell lung cancer (NSCLC) in Han Chinese patients. This
study found the 2677 GG genotype was associated with significantly better response to
chemotherapy compared with the combined 2677 GT and TT genotypes.24 The haplotype
of 2677G−3435C was also found to be a significant predictor of treatment response in this
same study. A demonstrated linkage disequilibrium between the synonymous SNP C3435T
and the nonsynonymous SNP 2677G>T/A may explain observed functional differences in
P-gp that have previously been attributed to the 3435C>T.25
The variation in frequency of SNPs for MDR1 has been studied in different racial/ethnic
populations. It has been found that the allelic frequency can differ among these groups.
The incidence of C/T and C/C genotypes at position 3435 has been found to be much
higher in African than Caucasian or Asian populations. In one study 83% of Ghanaians and
61% of African Americans were homozygous for the C allele, while only 26% of Caucasians
and 34% of Japanese shared this trait.26 Individuals who are homozygous for the T allele
have substantially lower intestinal P-gp than those who are homozygous for the C allele.22
Lower intestinal P-gp may increase the bioavailability of P-gp substrates. This seems to
be supported by studies showing that the maximum plasma concentrations of the P-gp
substrate cyclosporine is substantially lower in African Americans than Caucasians.27 The
higher frequency of the C/C genotype in African populations compared to Japanese or
Caucasians could result from a selective advantage of this genotype against gastrointestinal-
tract infections endemic to tropical regions.26 On the other hand, the high frequency of
the C3435 allele in African populations may explain a high prevalence of more aggressive
tumors in breast cancer and the high incidence of resistance to cancer chemotherapy seen
in African populations.18,28,29
EFFECT ON DIGOXIN ABSORPTION. The effect of MDR1 polymorphism on digoxin absorp-

tion has been probed in a number of studies.30,31 Because digoxin is not subject to metabolic
transformation, it has been used as a model substrate for the study of phenotype-genotype
relationships of MDR1 polymorphs. A sizable Dutch study (195 elderly patients) involving
chronic dosing of digoxin rather than single-dose kinetics examined the effect of MDR1
genotype on digoxin levels.30 The 3435C>T, 1236C>T, and 2677G>T/A SNPs were identified
in peripheral blood deoxyribonucleic acid (DNA). All three variants were associated with an
increase in serum digoxin concentration of 0.18−0.21 mg/L per additional T allele. The asso-
ciation was even stronger for the 1236-2677-3435 TTT haplotype and absent from other
haplotypes examined. These study results agree with a study in healthy Japanese subjects32
in which a single oral dose of digoxin was administered, and the serum concentration of
digoxin was monitored. Individuals harboring a T allele at 3435 had significantly lower AUC4
(Area Under the Curve for 0–4 hours) than those homozygous for C at this position.
Not all studies support the association of the 3435C>T SNP with reduced P-gp func-
tion or clinical outcome of patients treated with known P-gp substrates. For instance, in a
study conducted in Korea of 200 patients with acute myeloid leukemia (AML) undergoing a
standard induction chemotherapeutic regimen, no correlation was found between 3435C>T
polymorphism and P-gp function in leukemic blasts or in clinical outcomes.33 This inconsis-
tency in correlating clinical outcomes with the 3435C>T polymorphism in AML and other
diseases suggests that other genetic or nongenetic factors also play an important role.
In addition to race and ethnicity, gender can also significantly affect the expression
of P-gp. Hepatic P-gp levels are 2−2.4 fold lower in females than males.34 In the case of
antineoplastics such as vinca alkaloids, etoposide, doxorubicin, and docetaxel, this means
increased risk for myelosuppression and gastrointestinal toxicity in females as well as
prolonged drug exposure.35 Therefore, females may have an increased response to a drug
in addition to increased toxicity compared to males.
CLINICAL PEARL
A patient’s gender can influence the rate of clearance and efficacy for
drugs that are transported by P-gp.
IRINOTECAN. Another anticancer drug that has been extensively studied with respect
to pharmacogenomics is irinotecan.5-9 Irinotecan is a prodrug, transformed to the active
metabolite 7-ethyl-10-hydroxycamptothecan (SN-38) by carboxylesterase enzymes. SN-38
is thought to be responsible for most of the activity of irinotecan. SN-38 is transformed in
phase II metabolism to the glucuronide conjugate by UDP-glucuronosyltransferase (UGT)
enzymes. The resultant conjugate is more hydrophilic than the parent and is subsequently
eliminated in the bile or urine by transport proteins. These proteins include ABCB1, ABCC1,
ABCC2, ABCC5, ABCG1, ABCG2, and SLC01B1 (OATP1B1).
Standard dosing regimens of irinotecan rely on calculation of patient body surface
area, which correlates with blood volume. However, there is tremendous interindividual
variability in response to irinotecan, with some patients developing severe life-threatening
diarrhea and neutropenia. Correct dosing is critical since reduced plasma levels may not
provide effective treatment, while elevated levels may produce toxicity. Modifications of
dosing regimens are recommended based on the observed individual toxicity. Polymor-
phisms of UGT1A1 that reduce glucuronidation and thus increase plasma levels have been
definitively identified. Because of this, the package labeling was revised in 2005 to recom-
mend reduced dosing in patients known to be homozygous for a polymorphism in the
promoter region of the UGT1A1 gene (UGT1A1*28). This includes approximately 10% of the
North American population. In 2005, the U.S. Food and Drug Administration also approved
a genetic test to aid the detection and identification of UGT1A1*28 (Invader UGT1A1 by Third
Wave Technologies Inc.). Polymorphisms of transport proteins with reduced activity would
naturally be expected to further modify pharmacokinetics and possibly increase toxic-
ity. This supposition has been supported for ABCB1 (1236C>T), ABCC2 (3972T>C), ABCG2
(delCTCA -19572-19576 and 421C>A), and SLC01B1*1b in various ethnic groups.36-42 The data
suggest that testing for transporter polymorphisms may further improve quality of treat-
ment for irinotecan.
ANTIDEPRESSANTS AND OPIOID ANALGESICS. Besides anticancer agents, polymorphisms

of ABCB1 have also been associated with pharmacogenomic relevance for both antide-
pressants and opioid analgesics. In users of antidepressants that are substrates for ABCB1,
the A allele of the rs2032588 SNP was associated with a lower number of side effects
after adjusting for gender, age, dosage, and duration of use.43 Other early results suggest
that antidepressant treatment may be optimized by clinical application of ABCB1 genotyp-
ing.44 A Dutch study (n = 424) provides agreement to the assertion that ABCB1 genomics
may be associated with side effects for selective serotonin reuptake inhibitors (SSRIs). In
this study, six ABCB1 variants (1236T>C, 2677G>T/A, 3435T>C, rs2032583, rs2235040, and
rs2235015) were examined and correlated with the serotonergic adverse events sleepless-
ness, gastrointestinal complaints, and sexual effects. Both rs2032583 and rs2235040 were
significantly correlated with these effects.45
A particular polymorphism has also been associated with opioid-induced respiratory
depression (rs9282564).46 Children with GG and GA genotypes had higher risks of respira-
tory depression after receiving intravenous morphine, which resulted in prolonged hospi-
tal stays. Each additional copy of the minor allele (G) increased the odds of a prolonged
hospital stay due to respiratory depression by 4.7 fold. Other transporters have also been
found to have roles in the pharmacokinetics of morphine in children, notably ABCC3 and
OCT1. In these two transporters, significance has been discovered for several genotypes.47
The action of another opioid, methadone, has also been found to depend at least partially
on variants of the ABCB1 gene. When SNPs of ABCB1 were considered in conjunction with
variants of the mu opioid receptor OPRM1, it was observed that some variants of ABCB1
were associated with reduced methadone requirements.48
ABCC Transporter Family
The protein products of ABCC genes are commonly known as MRPs. In contrast to the
neutral and cationic hydrophobic compounds that P-gp transports, MRPs often transport
anionic compounds. Ten members of the MRP family are known, and at least seven may
be involved in conferring resistance to cancer chemotherapeutics (MRP1 to MRP7).15 MRP1
likely has the greatest significance in clinical anticancer drug resistance. MRPs are located
in various tissues such as the brain, liver, kidney, and intestines and serve protective and
excretory functions. They transport a structurally diverse set of endogenous substances,
xenobiotics, and metabolites. Genetic polymorphisms of ABCC1-5 have recently been
subject to intensive study.50
ABCC1 Transporters
The ABCC1 (MRP1) transport protein has broad substrate specificity and is expressed in
many tissues of the body. It was originally discovered in small-cell lung cancer cells that
showed multidrug resistance without overexpressing ABCB1 (MDR1). Similarly to MDR1, it is
able to confer resistance to anthracyclines and vinca alkaloids. MRP1 transports primarily
neutral and anionic hydrophobic compounds and their glutathione, sulfate, and glucuro-
nide conjugates. A few cationic substances can also be transported. Many unconjugated
substances are co-transported with reduced glutathione (G-SH). The oxidized form of
glutathione (G-SS-G) is also transported by MRP1. In most polarized cells, localization of
the protein is on basolateral membranes for efflux of substrates into the blood. ABCC1 is
also found in many epithelial tissues (e.g., testes, skeletal muscle, heart, kidney, lung) and
may also have a protective role for the central nervous system. Physiologically relevant
endogenous compounds that are transported by MRP1 include leukotriene C4, which is
important for inflammatory reactions.
A number of nonsynonymous genetic variants of the transporter have been studied
for functional significance by in vitro methods. For instance, Arg433Ser decreased the
transport of leukotriene C4 and estrone sulfate but not estradiol 17-β glucuronide.51 This
same SNP conferred a 2.1 fold resistance to doxorubicin compared to cells expressing the
wild-type MRP1. Another SNP, Cys43Ser, has been associated with a decrease in vincristine
resistance. In this case, the polymorphism led to loss of localization to the correct cell
membrane.52 Polymorphisms in the promoter region of ABCC1 have also been found, raising
the possibility of differences in promoter activity and gene expression.53 The usefulness
of identifying specific polymorphisms of ABCC1 and SLC22A11 has been highlighted in at
least one study (of 233 rheumatoid arthritis patients) where identification of these SNPs
was used to predict which patients would not respond to methotrexate.54 However, further
validation of these results is indicated.
ABCC2 Transporters
The ABCC2 transporter is also known as MRP2 or canalicular multispecific organic anion
transporter (cMOAT). It is the most studied member of the ABCC family. This protein is
expressed in the liver, kidneys, and intestines. It plays an important role in chemoprotec-
tion by transporting the products of phase II metabolism out of cells. Thus glucuronide,
glutathione, and sulfate conjugates of drugs are the predominant substrates of MRP2.
These conjugates are transported from hepatic cells into the canaliculi and then to the bile
for excretion. Unconjugated drugs are also transported, as are the conjugates of bilirubin.
Unlike other members of the ABCC family, ABCC2 is expressed in apical membranes of
absorptive and excretory cells such as hepatocytes, enterocytes, renal proximal tubules,
and syncytiotrophoblasts of the placenta.
Mutations in the ABCC2 gene are associated with the rare autosomal recessive disor-
der Dubin-Johnson syndrome (DJS). These mutations may cause DJS through a variety
of mechanisms. The most obvious is the formation of nonfunctional forms of the protein,
which results in the inability of hepatocytes to secrete conjugated bilirubin into the bile.
Many of the mutations associated with DJS occur on the ATP-binding region, which is criti-
cal for protein function. Other mutations result in impaired transcription and localization
of the protein or reduced substrate binding. The results of the dysfunction are conjugated
hyperbilirubinemia and consequent deposition of pigment into hepatocytes.
Occurrence of DJS is most common in males, but pregnancy or oral contraceptive use
in women with DJS may result in jaundice. The prevalence of DJS varies among racial/
ethnic populations, and it is most commonly seen in Iranian Jewish patients. Besides modi-
fication of hepatic function, DJS patients have been thought to have reduced expression
and function of intestinal MRP2, although there is little evidence of it.55 Wide-ranging
studies concerning the effect of DJS polymorphisms on drug pharmacokinetics are not yet
available, but some small-scale studies have been completed.
In a case study of a DJS patient being treated for large B-cell lymphoma with metho-
trexate, a 3-fold reduction in methotrexate elimination rate was observed that resulted
in elevated systemic methotrexate concentration and reversible nephrotoxicity. Genetic
analysis of the ABCC2 gene revealed a heterozygous SNP Arg412Gly, which occurs in a
region of the protein associated with substrate binding. Functional analysis revealed that
this mutation conferred loss of transport activity.56 This case illustrates where effective
pharmacogenomic screening might have been successfully applied to improve patient care.
Another polymorphism of ABCC2, −24C>T (rs717620), has been reported to contribute to
altered methotrexate pharmacokinetics in individuals who are not necessarily positive for
DJS. This allele was significantly associated with higher risks of both hematologic (leuko-
penia, anemia) and non-hematologic (gastrointestinal, mucosal) methotrexate toxicities.57

Other studies have attempted to correlate the expression of MRP2 with both intrin-
sic and acquired resistance to other cancer chemotherapeutics; an example is cisplatin
in the treatment of pancreatic cancer.58 In resected pancreatic cancer tissues, only MRP2
messenger ribonucleic acid (mRNA)—and not MRP1 or MRP3—was expressed and was
overexpressed compared to normal pancreatic tissue. In this same study, when pancreatic
cancer cells were cultured in the presence of cisplatin, they began to overexpress MRP2
but not MRP1 or MRP3 proteins.
Besides the association of ABCC2 with altered chemotherapy response and DJS, a few
other drug classes have been examined. Drug resistance associated with ABCC2 genotype
has been associated with antiepileptic drugs in at least one study.59 In a population of
537 Chinese patients receiving anticonvulsants, multiple polymorphs of both ABCB1 and
ABCC2 were examined for reduced drug response. The ABCC2-24C>T and 3972C>T SNPs,
and one ABCC2 haplotype, were found to be associated with resistance. ABCC2 1249G>A
and ABCB1 3435C>T polymorphisms were not associated with antiepileptic drug resis-
tance in this particular study. A second study that examined gene variants of ABCC2 in 381
Caucasian patients with epilepsy was also unable to find an association with drug response
in contrast to older studies.60
ABCC3 Transporters
The ABCC3 gene, which codes for MRP3, has not been studied as extensively as either
MDR1 or MDR2. In contrast to MDR1, MRP3 does not transport glutathione and is a poor
transporter for glutathione conjugates.61 Glucuronide conjugates are transported, such
as estradiol-17-β-glucuronide. MRP3 is localized in the liver, kidneys, and intestines.
Location in polarized cells is in basolateral membranes, similar to MRP1. A number of
different polymorphs have been investigated for their effect on MRP3 expression levels.
One of the SNPs frequently found in the promoter region, 211C>T, has possible relevance
for pharmacotherapy and disease progression.62 Individuals who are homozygous or
heterozygous for this SNP showed significantly lower MRP3 mRNA levels than individu-
als with a wild-type allele. This SNP has been studied for its association with AML as a
predictor for disease predisposition or prognosis.63 It was found that individuals with the
211C>T SNP had a poorer prognosis as measured by overall survival estimates. Conflict-
ing results have been obtained for the correlation of 211C>T with treatment outcome in
childhood AML.64
ABCC4 and ABCC5 Transporters
These proteins, also known as MRP4 and MRP5, respectively, are much less studied than
MRP1, MRP2, and MRP3. Tissue localization is shown in Table 5-1. Substrates for both
transporters are anticancer/antiviral nucleoside and nucleotide analogs as well as vari-
ous organic anions. A number of SNPs have been identified in the genes encoding these
transporters, and some of them may have relevance for pharmacotherapy. For instance,
the SNP in MRP4, rs3765534, was found to dramatically reduce MRP4 function through
impairment of membrane localization.65 This SNP is relatively common in Japanese
patients (>18%) and may play a role in the high sensitivity that some patients have for
thiopurines. Polymorphism of ABCC5 has been shown to have a possible contribution to
anthracycline-induced cardiotoxicity in survivors of childhood acute lymphoblastic leuke-
mia (CALL). In a study of 251 CALL patients, individuals with the ABCC5 TT-1629 genotype
had an 8−12% decrease in two markers for cardiotoxicity.66 ABCC4 has been found to act
as an independent regulator of the levels of intracellular cyclic nucleosides, such as cAMP.
As such, it mediates cAMP-dependent signal transduction to the nucleus.67 It is also over-
expressed in several cancers, and its inhibition results in striking effects against cancer
progression and drug resistance.68
Table 5-1
Transporter Localizations and Polymorphisms
Transporter Gene Name Tissue Localization Example
(Common (Systematic and Position in Representative Polymorphisms and

Name) Protein Name) Polarized Cells Substrates Phenotype Effect
MDR1, P-gp ABCB1 Apical: kidney, liver, Anthracyclines, 3435C>T (↓ intes-
brain, intestine, cyclosporine, tinal expres-
placenta taxanes, vinca sion, ↓ substrate
alkaloids, doxo- bioavailability)22
rubicin139 2677G>T/A (↑
response to
docetaxel/
cisplatin)24
MRP1 ABCC1 Lung, ubiquitous Anthracyclines, Arg433Ser (↑
on basolat- vinca alkaloids, doxorubicin
eral membrane methotrexate, resistance)51
epithelial: e.g., glutathione conju- Cys433Ser (↓
choroid plexus gates, leukotriene vincristine
(blood-cerebro- C4, bilirubin, resistance)52
spinal fluid barri- glutathione,
er), testes saquinavir, ritona-
vir, difloxacin
MRP2, ABCC2 Apical: liver, proxi- Bilirubin conjugates, ↑ Cisplatin
cMOAT mal tubule, small glucuronide, resistance58
intestine, placenta sulfate and gluta- 2302C>T, 2439T>C
thione conjugates (Dubin-Johnson
of various drugs, syndrome)140
unconjugated c.3972C>T (↑
anionic drugs hepatocellular
(e.g., metho- carcinoma)141
trexate): broad
substrate speci-
ficity
MRP3 ABCC3 Basolateral: liver, Glucuronidated −211C>T (↓ expres-
kidneys, intes- substrates (e.g., sion),62 (worsen
tines acetaminophen,142 prognosis: lung
morphine,143 cancer)144
estradiol, biliru- Arg1381Ser,
bin) Ser346Phe, &
Ser607Asn (↓
transport
activity)145
MRP4 & ABCC4 & Prostate146 (basolat- Azidothymidine, MRP4: rs3765534 (↑
MRP5 ABCC5 eral), kidney,147 mercaptopurine, thiopurine
lung,148 brain,149 thioguanine, sensitivity),
pancreas,150 cladribine, abaca- Gly187Trp,
lymphocytes,151 vir50 Gly487Glu (↓
platelets,152 heart azidothymidine
(MRP5)153 transport),154
A3463G
(↓ tenofovir
efflux)155
MRP6 ABCC6 Basolateral: liver, Glutathione conju- Many: e.g.,
kidney gates, leukotriene c.3421C>T
C4 (pseudoxanthoma
elasticum)156
Table 5-1 (continued)

Example
Transporter Gene Name Tissue Localization Polymorphisms

(Common (Systematic and Position in Representative and Phenotype
Name) Protein Name) Polarized Cells Substrates Effect
MRP8 ABCC11 Testes, breast, skin Bile acids, conju- 538G >A (GA and GG
apocrine glands gated steroids, give wet vs. dry
(apical) cyclic nucleotides, earwax type)70
5-fluorouracil69
BCRP, MXR, ABCG2 Placenta syncy- Doxorubicin, dauno- ↑ Anthracyclines,

ABCP tiotrophoblasts, rubicin, mitoxan- mitoxantrone,
hepatocyte trone, topotecan, SN-38 resistance
canalicular, prazosin, uric 421C>A (worsen
apical intestinal acid82 prognosis:
epithelia, vascular lung cancer &
endothelia cisplatin),144 (↑
gefitinib-induced
diarrhea)80
Gln141Lys (↑
chemotherapy-
induced
diarrhea)157
Serotonin SLC6A4 Neurons, heart Serotonin “l” allele (↑ psycho-
transporter valve, intestine pathology)159
(apical)158 “s” allele (↓ antide-
pressant efficacy,
citalopram-
induced
diarrhea)96
Reduced SLC19A1 Apical: kidney, Methotrexate, 80AA (↑ methotrex-
folate carri- leukemic cells, leucovorin, ate polyglutama-
er (RFC-1) wide distribution pemetrexed tion)160
OATP1A2 SLCO1A2 Basolateral: liver, Methotrexate, E184K, D185N,
brain, lung, imatinib, bile T259P, and
kidney, testes acids, steroidal D288N
compounds (↓ methotrexate
transport)161
OATP1B1 SLCO1B1 Basolateral: liver, Simvastatin, pravas- 521T>C
brain tatin, atorvastatin, (↓ pravastatin
lovastatin, ceriv- AUC)162
astatin, bilirubin,
digoxin, estradiol,
thyroid hormones,
mycophenolate
OCT2 SLC22A2 Apical: kidney Tryptophan, cispla- 808G>T (↓ cisplatin
tin toxicity)
OATP1B3 SLCO1B3 Basolateral: liver Methotrexate, 334T>G (GG ↑

glucuroni- mycophenolate
dated estradiol, AUC)163
mycophenolate

Example
Transporter Gene Name Tissue Localization Polymorphisms

(Common (Systematic and Position in Representative and Phenotype
Name) Protein Name) Polarized Cells Substrates Effect
PEPT1 and SLC15A1, PEPT1: small Cephalexin, Arg57His (transport

PEPT2 SLC15A2 intestine, other β-lactam function loss)165
duodenum antibiotics,
(apical) ACE inhibitors,
PEPT2: broad valacyclovir,
distribution164 peptides
RFC-1 SLC19A1 Broad distribution Methotrexate 80A>G (AA ↑
plasma folate)119
(↑ remission
of rheumatoid
arthritis with
methotrexate)120
CNT1, CNT2, SLC28A1, Intestinal/renal Didanosine, Unclear
CNT3 SLC28A2, epithelia, liver, idoxuridine, relevance for
SLC28A3 macrophages, zidovudine, polymorphisms
leukemic cells2 cladribine,
fludarabine,
gemcitabine,
capecitabine2
ENT1, ENT2, SLC29A1, Intestine, liver, Pyrimidine and/ Unclear
ENT3, SLC29A2, kidney, or purine relevance for
ENT4 SLC29A3, placenta166 nucleosides, polymorphisms
SLC28A4 adenosine,
gemcitabine,
cladribine,
fludarabine
ACE, angiotensin-converting enzyme; AUC, area under the curve.
ABCC6 Transporters
ABCC6 encodes MRP6 protein, also known as MRP-like protein 1 (MLP-1), anthracycline
resistance−associated protein (ARAP), and multispecific organic anion transporter-E
(MOAT-E). It is expressed primarily in the liver and kidneys. Mutations in the ABCC6 gene
are associated with pseudoxanthoma elasticum, a disease that causes mineralization of
elastic fibers in some tissues.
ABCC11 Transporters
ABCC11, also known as MRP 8 (MRP8), is an efflux transporter for anionic lipophilic
compounds. It has been found to confer resistance to antiviral and anticancer agents. One
missense variant has been found to be associated with a markedly reduced ability to transport
5-fluorouracil.69 An observable physiological trait has been linked to a variant to ABCC11,
which is the association of the nonsynonymous SNP rs17822931 in exon 4 with either wet
or dry earwax.70 The difference in type of earwax is due to alterations in apocrine secretory
cell function. This same polymorphism has been associated with lack of body odor and
has been used to predict lower frequency of deodorant use.71 It is also a potential marker
for risk of mastopathy.72
ABCG2 Transporters
ABCG2 is alternatively known as breast cancer resistance protein (BCRP), placenta-specific
ABC transporter (ABCP), and mitoxantrone resistance protein (MTX). It was originally iden-
tified in a resistant breast cancer cell line. It is very important in limiting bioavailability of
certain drugs, concentrating drugs in breast milk, and protecting the fetus from drugs in
maternal circulation.10 It is highly expressed in the gastrointestinal tract, liver, and placenta,
and it influences the absorption and distribution of a wide variety of drugs and organic
anions.73-76 The substrate specificity for ABCG2 is broad and overlaps that of P-gp but is
distinct from it. In contrast to the rest of the ABC transporter family, ABCG2 contains only
one binding site for ATP and one transmembrane domain rather than two of each. It is
assumed to function as a dimer and is, therefore, referred to as a half-transporter. ABCG2
confers resistance to a broad range of hydrophobic anticancer drugs, similar to P-gp and
MDR1, and is considered one of the most important ABC transporters mediating multidrug
resistance in cancer cells. Resistance can be brought about by either reduced absorption or
increased biliary excretion of the drug.
Various polymorphisms of ABCG2 are known to exist, some of which are associated
with increased resistance to anticancer drugs such as mitoxantrone, the anthracyclines, and
camptothecin derivatives. Some SNPs that have been associated with altered transport
activity are Arg428Gly and Arg428Thr, Cys421Ala,77 Val12Met, Gln141Lys, and Gln126X.78
Other drugs that act as inhibitors of ABCG2 are antiviral nucleoside analogs such as
zidovudine (AZT), lopinavir, and nelfinavir.79 One SNP of ABCG2 has been associated with
adverse reactions in patients treated with gefitinib, an inhibitor of the epidermal growth
factor receptor tyrosine kinase that is used in NSCLC.80 Therefore, 44% of patients who
were heterozygous for the Cys421Ala polymorphism developed diarrhea after treatment
with gefitinib versus 12% of patients homozygous for the wild-type allele. In situations
where a patient possesses a nonfavorable haplotype, it may be possible to increase the
dose of the relevant drug in order to increase response rates.81
Besides being associated with adverse drug reactions and variations in therapeutic effi-
cacy, SNPs of ABCG2 have been found to be highly predictive of plasma uric acid levels. In
one large study,82 a genome-wide scan was made for SNPs associated with serum uric acid
concentration and gout. The study used phenotype and genotype results from a cohort
of the Framingham Heart Study as well as a Rotterdam cohort. SNPs identified as being
associated with uric acid concentration and gout were identified in ABCG2, SLC17A3, and
SLC2A9. The results of this study were used to calculate a risk score for an individual based
on whether he or she has the polymorphisms associated with hyperuricemia. The risk score
was generated based on the number of alleles associated with high uric acid concentration.
Mean uric acid concentration rose linearly with the number of risk alleles. For individuals
with no risk alleles, the prevalence of gout was 1–2% across the cohorts examined. The
prevalence increased to 8–12% with six risk alleles. Individual common genetic variants
were found to confer only a modest risk of gout, but their combination resulted in a large
association with uric acid and gout. Ultimately the risk score may be used to help identify
patients with asymptomatic hyperuricemia and guide therapeutic intervention. Further
validation that ABCG2 is a key determinant of response to drugs used for the treatment
of gout, such as allopurinol, continues to accumulate.83 Another transporter with pharma-

cogenomic relevance for uricosuric is the urate transporter SLC22A12 (see below).
One of the primary tissues in which ABCG2 is expressed in humans is in the placenta.
The precise physiological function of the protein in this location is not clear, but it likely
plays an important role in protecting the fetus from xenobiotics, toxins, and metabolites
by expelling them across the placental barrier.84 Wide variation in the expression level
of BCRP has been found in human placenta, which may lead to considerable variation in
fetal exposure to drugs and xenobiotics. Such variation may be caused by polymorphisms
in BCRP.85
Solute Carrier Proteins

SLC proteins are important in the transport of ions and organic substances across biological
membranes for the maintenance of homeostasis. Members of the SLC superfamily consist
of membrane channels, facilitative transporters, and secondary active transporters. Exam-
ples of some of the endogenous solutes that are transported include steroid hormones,
thyroid hormones, leukotrienes, and prostaglandins. Additionally, SLCs are important in the
transport of a large number of drugs. The SLC class includes the organic anion transporters
(OATs), organic anion transporting polypeptides (OATPs, which are structurally different from
OATs), organic cation transporters (OCTs), and peptide transport proteins (PepTs). In all, more
than 40 families of transporters make up the SLC superfamily. Members within an individual
SLC family have >20–25% sequence homology. However, the homology between families is
low or nonexistent. Thus, inclusion of a family into the SLC group is not dependent on evolu-
tionary or structural relationship but rather is a functional classification. Individual members
of the SLCs are expressed in a variety of tissues such as liver, kidney, brain, and intestine and
transport substances either into or out of cells.
SLCO1A2
SLCO1A2 (also known as OATPA or OATP1A2) is an organic anion transporter that carries
ions across the epithelium. It has been identified in the human liver, brain, lung, kidney, and
testes. High-throughput pharmacogenomics has been used to identify polymorphisms of
SLCO1A2 that may be used to identify the risk for febrile neutropenia (FN) in breast cancer
patients treated with combinations of doxorubicin with docetaxel or cyclophosphamide.86
Two SNPs, rs4762699 and rs2857468, located on the SLCO1A2 gene were found to be
associated with neutropenia. Haplotype T-T was associated with a high risk of FN; 83% of
patients with at least one copy of T-T suffered this side effect versus 33% of patients with
other haplotypes.
SLCO1B1
Genetic variants of SLC proteins, such as SLCO1B1, have been associated with pharmaco-
genomic relevance. A genome-wide scan of 300,000 genetic markers in a study of statin-
induced myopathy found a strong correlation with the rs4363657 SNP located within the
SLCO1B1 gene.87 SLCO1B1 encodes the sodium-independent OATP OATP1B1. An increased
risk of myopathy was associated with simvastatin use in patients expressing this particu-
lar variation. Polymorphisms of solute transporter genes have also been associated with
pharmacokinetic variance for other statin drugs. For example, altered uptake of pravastatin
into the liver has been associated with polymorphisms of SLC21A6 (OATP-C) and SLC22A8
(OAT3).88 A variant of SLCO1B1 has also been associated with functionally relevant SNPs
important for the pharmacokinetics of other drugs such as the irinotecan metabolite
SN-38, estrone 3-sulfate, and estradiol 17-beta glucuronide. Methods have been developed
to rapidly identify the relevant SNPs.89
Other variants of the SLCO1B1 gene have been found to affect simvastatin transport. In
a study with 646 participants, SLCO1B1 521T/C and 521C/C individuals were found to have
71% and 248% higher 12-hour plasma levels of the metabolite simvastatin acid, respec-
tively, compared to SLCO1B1 521T/T participants.90 Further exploration of these differences

may give information about different risk-benefit profiles for individuals with these vari-
ants. One SNP (rs4149056T>C) has been associated with the risk of muscle toxicity from
simvastatin in numerous studies. Evidence for this, along with therapeutic recommenda-
tions based on SLCO1B1 genotype, has been recently reviewed.91 It has been suggested
that SLCO1B1, working in conjunction with ABCC3 and SLCO1B3, forms a liver-blood shut-
tling loop for bilirubin glucuronide in which ABCC3 secretes conjugated bilirubin back into
the blood from the liver; the SLC proteins reabsorb it in downstream hepatocytes.92 This
provides an efficient detoxification mechanism.
SLC1A1
SLC1A1 is a member of the SLC family that resides in neuronal/epithelial tissues and is
a high affinity glutamate transporter. It is also the primary route for neuronal uptake of
cysteine, which is the rate-limiting step in glutathione synthesis.93 Variants of this protein
appear to interact with environmental factors (life stress) in providing resistance to phar-
macological treatment of obsessive-compulsive disorder.94
SLC6 Family
Members of the SLC6 family are sodium-dependent transporters for neurotransmitters
such as dopamine, serotonin, norepinephrine, glycine, and GABA. The SLC6A4 gene codes
for the serotonin transporter (SERT). The best evidence for pharmacogenomic relevance
within the SLC6 family has been found for SERT. SERT is a cotransporter for serotonin and
sodium ions. Its physiological function at the synapse is serotonin reuptake and the termi-
nation of serotonin signaling. Because this protein is the site of action of the serotonin
reuptake inhibitors, there has been much interest in the effect of SERT polymorphisms on
drug action and pathology.
The 5HTTLPR (SERT-linked promoter region) of the SLC6A4 gene has been exten-
sively studied for association with neuropsychiatric disorders. This polymorphism occurs in
the promoter region of the gene, rather than the protein coding region. It is associated with
short (“s”) and long (“l”) repeats in this region. The short variation contains fourteen repeats
of a particular sequence, while the long version contains sixteen repeats. The short version
leads to reduced promoter activity and less transcription of SLC6A4, while the l allele has
the opposite effect. A number of studies and meta-analyses have found that the ss geno-
type or s allele is predictive of reduced antidepressant efficacy, while the ll genotype is
associated with better response to therapy.
Other studies have found that the presence of the s allele is associated with greater
numbers of side effects during treatment of depression with SSRIs.95 In one large study, the
adverse effects of citalopram were strongly associated with the 5HTTLPR s allele and ss
genotype. Interestingly, this study found no difference in therapeutic responses in patients
with the different alleles.96 In summary, evidence suggests that genetic screening may soon
help to predict if a given antidepressant will be effective or produce adverse effects in a
patient. This would be a major advancement for individualizing the pharmacotherapy of
depression. Details that may complicate interpretation include observations of gene-gene

interactions between 5-HTTLPR and serotonin receptor subtype variants, such as HTR2A.97
The SLC6A3 gene encodes for the dopamine transporter DAT1. Polymorphisms in this
gene have been observed, and an attempt has been made to correlate genotype with
neuropsychiatric disorders such as attention-deficit hyperactivity disorder (ADHD).98 One
study suggested an association of a particular haplotype with adult ADHD.99 The haplo-
type implicated was the 9-6 SLC6A3-haplotype, formed by the 9-repeat allele of the vari-
able number of tandem repeat (VNTR) polymorphism in the 3’ untranslated region of the
gene and the 6-repeat allele of the VNTR in intron 8 of the gene. Polymorphisms of DAT1
have also been implicated in variability of response to methylphenidate in ADHD, although
there have been conflicting results presented by a number of studies. In one meta-analysis,
a significant relationship was seen between low rates of methylphenidate response and
a homozygotic 10R VNTR polymorphism.100 Greater efficacy of both methylphenidate and
L-DOPA in Parkinson’s disease patients has also been associated with SLC6A3 variants.101
The SLC6A2 gene encodes for the norepinephrine transporter (NET). Because of the
wide implications of this neurotransmitter in neuropsychiatric disorders and drug action,
many studies have focused on finding polymorphisms for this gene and correlating them to
therapeutic response. One study that examined predictive antidepressant response to the
mixed serotonin /norepinephrine reuptake inhibitor milnacipran found that a polymorphism
of NET was associated with superior response.102 Variants of SLC6A2 were also noted to be
associated with response to atomoxetine in children with ADHD, along with variants of the
adrenergic receptor ADRA2A.103 Substantially more research is needed in this area to make
firm predictions regarding psychiatric drug response.
SLC7 and SLC8 Family
SLC7 and SLC8 family members are amino acid transporters. The amino acid transporters
LAT1 and LAT2 are encoded by the SLC7A5 and SLC7A8 genes. Both of these transport-
ers are principal mediators of cellular uptake of the nitrogen mustard chemotherapeutic
melphalan. One SNP in the first intron of SLC7A5, rs4240803, has been found to be signifi-
cantly associated with gastrointestinal toxicity after high-dose melphalan. Presence of this
SNP predicts the resultant need for total parenteral nutrition and suggests that variability
in melphalan transport affects mucosal injury.104
SLC15 Family
The PEPT1 and PEPT2 transporters (SLC15A1 and SLC15A2) are proton-coupled oligo-
peptide cotransporters. They carry small peptides of two to three residues as well as
peptide-like drugs that would otherwise not cross lipid membranes. The physiological
role of SLC15A1, which is found in the intestine, is the transport of small peptides arising
from digestion of dietary proteins. Intestinal PEPT1 is involved in uptake of cephalexin and
other β-lactams. The nucleoside prodrugs valacyclovir and valganciclovir have enhanced
bioavailability due to transport by PEPT1.105 Significant interindividual variation in intes-
tinal absorption of valacyclovir suggests the presence of genetic factors.106 Angiotensin-
converting enzyme inhibitors are also often considered to be substrates for PEPT1 and
PEPT2; however, data supporting this claim are inconsistent.107,108 Besides being localized
in the intestine, PEPT transporters are found in the kidney and liver; in the case of PEPT2,
they are found in the central and peripheral nervous system, lung, heart, kidney, and
mammary glands.109
Clinical relevance for genetic variation of PEPT1 or PEPT2 remains murky, but several
researchers have studied polymorphisms of these loci. In one study of 44 ethnically diverse
individuals, nine nonsynonymous and four synonymous polymorphisms were identified in
PEPT1.110 When transfected into an immortal cell line and analyzed for transport capacity,
only one rare SNP (Pro586Leu) was found to be associated with reduced activity, which
resulted from post-translational reduction of protein expression in the plasma membrane.

The results of this study have been confirmed and extended to 247 individuals of various
ethnic origins.111 This study found additional genetic variants of PEPT1, but concluded genetic
factors played only a small role in determining interindividual variation in PEPT1 transport
activity in the intestine. Because of the vital role that PEPT1 plays in normal homeostasis,
mutations that result in loss of activity likely have a high negative evolutionary selection
pressure. This does not, however, preclude future discovery of polymorphs with variation
in expression or activity. In the case of PEPT2, polymorphs have been identified that lack
transport function and have differing affinity and pH sensitivity. Variable mRNA expression
has also been observed, likely due to cis-acting polymorphisms.112 Thus, variability in the
PEPT2 gene is considerable, with a possible influence on the pharmacokinetics of drugs
transported by PEPT2. The roles of the various members of the SLC15 family have recently
been reviewed.113
SLC17A8
SLC17A8 is a sodium-dependent phosphate cotransporter. Some studies suggest the trans-
porter gene SLC17A8, which encodes the VGLUT3 protein (vesicular glutamate transporter),
may play a role in genetic predisposition to cocaine abuse. In mice, the disruption of the
VGLUT3 gene markedly increased cocaine self-administration.114 At the same time, the
amount of dopamine release in the nucleus accumbens was strongly augmented.
SLC17A8 has also been implicated in a type of hearing loss (autosomal-dominant
sensorineural hearing loss). It has been found that mice lacking VGLUT3 are profoundly
deaf due to absence of glutamate being released at the first synapse of the auditory
pathway.115
SLC18A2
The SLC18A2 gene encodes another vesicular amine transporter, known as vesicular amine
transporter 2 (VAT2) or vesicular monoamine transporter 2 (VMAT2). This transporter
moves cytosolic monoamines into synaptic vesicles using a proton gradient. It is important
to the action of several drugs, such as reserpine and tetrabenazine.116 Several polymor-
phisms of SLC18A2 have been studied in attempts to correlate the side effect of tardive
dyskinesia with antipsychotic use. A number of these SNPs have been found to be at least
nominally correlated.117
SLC19A1
SLC19A1, also known as reduced folate carrier-1 (RFC-1), is involved in the transport of folate
and antifolate drugs into human cells. Resistance to folate anticancer drugs may be medi-
ated by point mutations of this transporter. Because lack of nutritional folate is strongly
associated with birth defects such as cleft palate, it would be expected that variants of
the folate carrier might also be associated with these defects. Although one study failed
to show a strong correlation between genetic variants of RFC-1 and cleft palate, this same
study did show modest evidence for an interaction between infant RFC-1 genotype and risk
of certain congenital heart defects.118 The specific variant examined was the SNP 80A>G,
which results in the replacement of a histidine residue with an arginine in the protein.119
The functional result of this replacement on the transport protein is unknown, but higher
plasma folate levels were observed in individuals with the 80AA genotype compared to
individuals with the 80GG genotype.
Methotrexate is an example of a drug that is transported by the RFC-1. The 80G>A
polymorphism in RFC-1 has been associated with altered treatment efficacy in patients
with rheumatoid arthritis treated with methotrexate. In one study, the probability of remis-
sion was 3.3 fold higher in patients with the 80AA genotype compared to those with the
80GG genotype. The frequency of the A allele was also found to be 14% higher in patients
who responded to methotrexate compared to nonresponders. Additionally, aminotransfer-
ase activity was noted more frequently in patients with the 80AA genotype.120 All of this
information suggests that evaluation of RFC-1 polymorphism could be useful for optimiza-
tion of methotrexate therapy.121 Alternatively, it may be useful for prediction of gastrointes-
tinal toxicity associated with methotrexate treatment.122
Another study examined the effect of the Gly80Ala polymorphism in RFC-1 in relation
to risk for thrombosis.123 Since folate lowers homocysteine, which is thrombogenic, reduc-
tion in the transport of folate might be expected to have an effect on the prevalence of
thrombosis. This study did find a significant protective effect of the A allele against throm-
bosis. No effect on homocysteine plasma level was observed, but an increased extracel-
lular to intracellular ratio of folate was seen. This is consistent with the biological role of
RFC-1 and may explain the protective effect of the polymorph against thrombosis.
SLC22A1
SLC22A1 encodes an organic cation transporter known as OCT1. It has substrate selectiv-
ity for a number of endogenous ligands such as dopamine, serotonin, and choline. It is
also known to transport the cationic forms of the drugs metformin, cimetidine, imatinib,
oxaliplatin, tramadol, and agmatine. Its general pharmacogenomics relevance has been
recently reviewed.124,125 Polymorphisms in this gene have been hypothesized to influence
the interindividual variation of response to the glucose-lowering activity of metformin. This
has been found in various studies with conflicting results in different patient populations.
In one study of type 2 diabetes mellitus patients of South Indian descent, patients with
two copies of the A allele in rs622342 were 5.6 times more likely to respond to metformin
treatment.126
Genetic variability in SLC22A1 may play a role in the large interindividual variability to
morphine response and adverse events. It has been observed that Caucasian children have
more adverse effects and slower clearance of morphine than African-American children. A
possible explanation for this has been their genotype for OCT1, because Caucasian children
are seen to have higher allelic frequencies of variants with loss of function.127
SLC22A2
The SLC22A2 transporter is also known as the organic cation transporter 2 (OCT2). It is
found primarily in the kidney and is involved in elimination of many small endogenous
organic cations and exogenous xenobiotics. The amino acid tryptophan has recently been
shown to be a substrate.128 The polymorphism rs316019 (c.808G>T; Ser270Ala) has been
found to be significantly associated with protection from cisplatin-induced ototoxicity in
both pediatric and adult patients.129 This suggests a critical role of SLC22A2 for cisplatin
transport in the body.
SLC22A4 and SLC22A5

The SLC22A4 and SLC22A5 genes code for the organic cation transporters OCTN1 and
OCTN2, respectively. Specific genotypes of SLC22A4 have been associated with suscep-
tibility to rheumatoid arthritis, and specific genotypes of SLC22A5 have been associated
with systemic primary carnitine deficiency.130,131 Polymorphisms in both of these genes have
been found to be potentially important predictors of time to progression in gastrointestinal

stromal tumor patients receiving imatinib therapy.132 Variants of OCTN1, particularly 917C>T,
along with the OCT2 808 G>T SNP have also been identified as potentially useful in titrat-
ing optimum metformin dosage.133
SLC22A9
The human organic anion transporter 7 (OAT7, encoded by SLC22A9) is a hepatic protein
that has been poorly characterized thus far. Pravastatin has been identified as the first
substrate.134 Substantial, interindividual variation in hepatic OAT7 may contribute to pravas-
tatin pharmacokinetics and might affect drug response.
SLC22A12
SLC22A12 encodes the urate transporter URAT1, or organic anion transporter 4-like protein.
Polymorphisms of this transporter have been studied with respect to their effect on
hypertensive patients with hyperuricemia.135 In 101 individuals, the frequency of rs3825016
(C/T) was significantly higher in hypertensive patients with hyperuricemia than in healthy
controls. This gene variant was also found to influence the uricosuric action of losartan.
Another study found a possible relationship of the rs11602903 SNP of SLC22A12 with
metabolic syndrome in hypertensive Caucasians, where it accounted for 7% of the variance
in body-mass index.136 This same relationship was not found for African Americans.
SLC28
There are three members of the SLC28 gene family in humans: SLC28A1, SLC28A2, and
SLC28A3. All of them encode nucleoside transporters coupled to ion gradients (i.e.,
concentrative nucleoside transporters [CNT1, CNT2, and CNT3]). CNT1 translocates pyrimi-
dines, while CNT2 translocates purines. Both of them are coupled to sodium ion transport.
The CNT3 transporter has broad selectivity and can transport nucleosides coupled to
either sodium ions or protons. This coupling allows the transporters to move nucleosides
against their concentration gradient. Besides transporting naturally occurring nucleosides
such as adenosine, these transporters are vitally important in transporting anticancer/anti-
viral nucleosides into cells. Because nucleoside drugs are mostly hydrophilic molecules,
cellular uptake is dependent on transport proteins. Along with the SLC29 family, SLC28
members are important for salvage processing of nucleosides. SNPs of the CNTs have been
identified, but pharmacogenomic relevance is currently poorly defined.
SLC29
The SLC29 genes code for equilibrative nucleoside transporter (ENT) proteins. There are
four members of this gene family in humans, SLC29A1 to SLC29A4 (ENT1 to ENT4). ENT1
is independent of sodium ion concentration, in contrast to CNTs. This transporter plays a

role in cellular uptake of anticancer nucleoside analogs.137 The clinical relevance of this
finding is currently unclear, although a number of polymorphisms at this gene have been
uncovered. In one study on a population of 256 Japanese cancer patients, 39 variations of
the gene were found; the highest frequency was 0.051.138
SUMMARY
Evidence is growing regarding the many ways in which the genomics of individual transport
proteins affects the response to a variety of medications. Additionally, the profound effect
of racial/ethnic heritage on distribution of genotypic variation can no longer be ignored.
This chapter has reviewed some of the accumulated evidence for pharmacogenomic
considerations of transporters. Many other transporters, which have been identified, may
also prove to be important for determining individual therapeutic response. Additionally,
many more SNPs for transport proteins have been identified than have been studied in vivo.
The advent of inexpensive broad genetic screening for transport protein polymor-
phisms will no doubt be instrumental in a new era of truly personalized therapy. For
instance, DNA chips are now available that screen 100,000 SNPs in a matter of hours.
In order to solidify treatment guidelines for genetically diverse populations, significant
amounts of research continue to be needed in this area. Because few drugs are transported
by just one carrier protein, other carriers may compensate for a deleterious SNP. Thus, a
single SNP is often not capable of altering the pharmacokinetics of a drug. For this reason,
future studies that are more comprehensive in scope will offer a greater insight into the
genetics of drug response.
REFERENCES
1. Huang Y, Sadée W. Membrane transporters and channels in chemoresistance and sensitivity of tumor
cells. Cancer Lett. 2006;239(2):168-182.
2. Lai Y, Varma M, Feng B, et al. Impact of drug transporter pharmacogenomics on pharmacokinetic and
pharmacodynamic variability—considerations for drug development. Expert Opin Drug Metab Toxicol.
2012; 8(6):723-743.
3. Klein I, Sarkadi B, Varadi A. An inventory of the human abc proteins. Biochim Biophys Acta.
1999;1461(2):237-262.
4. Ye AY, Liu QR, Li CY, et al. Human transporter database: Comprehensive knowledge and discovery
tools in the human transporter genes. PloS One. 2014;9(2) e88883.
5. de Jong FA, de Jonge MJA, Verweij J, et al. Role of pharmacogenetics in irinotecan therapy. Cancer
Lett. 2006;234(1):90-106.
6. Smith NF, Figg WD, Sparreboom A. Pharmacogenetics of irinotecan metabolism and transport: An
update. Toxicol In Vitro. 2006;20(2):163-175.
7. Takane H, Kawamoto K, Sasaki T, et al. Life-threatening toxicities in a patient with UGT1a1*6/*28
and SLCO1b1*15/*15 genotypes after irinotecan-based chemotherapy. Cancer Chemother Pharmacol.
2009;63(6):1165-1169.
8. Takane H, Miyata M, Burioka N, et al. Severe toxicities after irinotecan-based chemotherapy in a patient
with lung cancer: A homozygote for the SLCO1b1*15 allele. Ther Drug Monit. 2007;29(5):666-668.
9. Chen S, Villeneuve L, Jonker D, et al. ABCC5 and ABCG1 polymorphisms predict irinotecan-induced
severe toxicity in metastatic colorectal cancer patients. Pharmacogenet Genomics. 2015;25(12):573-583.
10. Williams JA, Andersson T, Andersson TB, et al. PhRMA white paper on ADME pharmacogenomics.
J Clin Pharmacol. 2008;48(7):849-889.
11. Oldham ML, Davidson AL, Chen J. Structural insights into ABC transporter mechanism. Curr Opin
Struct Biol. 2008; 18(6):726-733.
12. Davidson AL, Maloney PC. ABC transporters: How small machines do a big job. Trends Microbiol.
2007;15(10):448-455.
13. Rabow AA, Shoemaker RH, Sausville EA, et al. Mining the National Cancer Institute’s tumor-
screening database: Identification of compounds with similar cellular activities. J. Med. Chem. 2002;
45(4):818-840.
14. Martin F, Fromm ME. Pharmacogenomics of human P-glycoprotein. In: Wong M-L, Licinio J, ed. Phar-
macogenomics: The Search for Individualized Therapies. New York, NY; Wiley-Blackwell; 2003:159-178.
15. Ambudkar SV, Kimchi-Sarfaty C, Sauna ZE, et al. P-glycoprotein: From genomics to mechanism. Onco-
gene. 2003;22(47):7468-7485.
16. Leschziner GD, Andrew T, Pirmohamed M, et al. ABCB1 genotype and Pgp expression, function and
therapeutic drug response: A critical review and recommendations for future research. [review]. Phar-
macogenomics J. 2007;7(3):154-179.
17. Pondugula SR, Flannery PC, Abbott KL, et al. Diindolylmethane, a naturally occurring compound, induc-
es CYP3a4 and MDR1 gene expression by activating human PXR. Toxicol Lett. 2015;232(3):580-589.
18. Ameyaw MM, Regateiro F, Li T, et al. MDR1 pharmacogenetics: Frequency of the c3435t mutation in
exon 26 is significantly influenced by ethnicity. Pharmacogenetics. 2001; 11(3): 217-221.
19. Marzolini C, Paus E, Buclin T, et al. Polymorphisms in human MDR1 (p-glycoprotein): Recent advances
and clinical relevance. Clin Pharmacol Ther. 2004;75(1):13-33.
20. Pauli-Magnus C, Kroetz DL. Functional implications of genetic polymorphisms in the multidrug resis-
tance gene MDR1 (ABCB1). Pharm Res. 2004;21(6):904-913.
21. Ieiri I, Takane H, Otsubo K. The MDR1 (ABCB1) gene polymorphism and its clinical implications. Clin
Pharmacokinet. 2004;43(9):553-576.
22. Hoffmeyer S, Burk O, von Richter O, et al. Functional polymorphisms of the human multidrug-resis-
tance gene: Multiple sequence variations and correlation of one allele with p-glycoprotein expression
and activity in vivo. Proc Natl Acad Sci U S A. 2000;97(7):3473-3478.
23. Eichelbaum M, Fromm MF, Schwab M. Clinical aspects of the MDR1 (ABCB1) gene polymorphism. Ther
Drug Monit. 2004;26(2):180-185.
24. Pan JH, Han JX, Wu JM, et al. MDR1 single nucleotide polymorphism g2677t/a and haplotype are
correlated with response to docetaxel-cisplatin chemotherapy in patients with non-small-cell lung
cancer. Respiration. 2009;78(1):49-55.
25. Kim RB, Leake BF, Choo EF, et al. Identification of functionally variant MDR1 alleles among European
Americans and African Americans. Clin Pharmacol Ther. 2001;70(2):189-199.
26. Schaeffeler E, Eichelbaum M, Brinkmann U, et al. Frequency of c3435t polymorphism of MDR1 gene in
African people. Lancet. 2001;358(9279):383-384.
27. Min DI, Lee M, Ku YM, et al. Gender-dependent racial difference in disposition of cyclosporine among
healthy African American and white volunteers. Clin Pharmacol Ther. 2000;68(5) 478-486.
28. Elmore JG, Moceri VM, Carter D, et al. Breast carcinoma tumor characteristics in black and white
women. Cancer. 1998; 83(12):2509-2515.
29. Cross CK, Harris J, Recht A. Race, socioeconomic status, and breast carcinoma in the u.S: What have
we learned from clinical studies. Cancer. 2002;95(9):1988-1999.
30. Aarnoudse A-J, Dieleman JP, Visser LE, et al. Common ATP-binding cassette B1 variants are associated
with increased digoxin serum concentration. Pharmacogenet Genomics. 2008;18(4):299-305.
31. Verstuyft C, Schwab M, Schaeffeler E, et al. Digoxin pharmacokinetics and MDR1 genetic polymor-
phisms. Eur J Clin Pharmacol. 2003;58(12):809-812.
32. Sakaeda T, Nakamura T, Horinouchi M, et al. MDR1 genotype-related pharmacokinetics of digoxin after
single oral administration in healthy Japanese subjects. Pharm Res. 2001; 18(10):1400-1404.
33. Hur E-H, Lee J-H, Lee MJ, et al. C3435t polymorphism of the MDR1 gene is not associated with
p-glycoprotein function of leukemic blasts and clinical outcome in patients with acute myeloid leuke-
mia. Leuk Res. 2008;32(10):1601-1604.
34. Schuetz EG, Furuya KN, Schuetz JD. Interindividual variation in expression of p-glycoprotein in normal
human liver and secondary hepatic neoplasms. J Pharmacol Exp Ther. 1995;275(2):1011-1018.
35. Davis M. Gender differences in p-glycoprotein: Drug toxicity and response. J Clin Oncol.
2005;23(26):6439-6440.
36. Sai K, Kaniwa N, Itoda M, et al. Haplotype analysis of ABCB1/MDR1 blocks in a Japanese population
reveals genotype-dependent renal clearance of irinotecan. Pharmacogenetics. 2003;13(12):741-757.
37. de Jong FA, Scott-Horton TJ, Kroetz DL, et al. Irinotecan-induced diarrhea: Functional significance of
the polymorphic ABCC2 transporter protein. Clin Pharmacol Ther. 2007;81(1):42-49.

38. Mathijssen RHJ, Marsh S, Karlsson MO, et al. Irinotecan pathway genotype analysis to predict pharma-
cokinetics. Clin Cancer Res. 2003;9(9):3246-3253.
39. Mathijssen RHJ, de Jong FA, van Schaik RHN, et al. Prediction of irinotecan pharmacokinetics by use of
cytochrome p450 3a4 phenotyping probes. J Natl Cancer Inst. 2004;96(21):1585-1592.
40. Innocenti F, Kroetz DL, Schuetz E, et al. Comprehensive pharmacogenetic analysis of irinotecan
neutropenia and pharmacokinetics. J Clin Oncol. 2009; 27(16):2604-2614.
41. Han JY, Lim HS, Yoo YK, et al. Associations of ABCB1, ABCC2, and ABCG2 polymorphisms with irino-
tecan-pharmacokinetics and clinical outcome in patients with advanced non-small cell lung cancer.
Cancer. 2007;110(1):138-147.
42. Balram C, Li J, Zhou QY, et al. Molecular mechanisms of interethnic differences in irinotecan disposi-
tion: Impact of variants in ABCG2. J Clin Oncol (Meeting Abstracts). 2005; 23(16suppl): 018.
43. Bet PM, Verbeek EC, Milaneschi Y, et al. A common polymorphism in the ABCB1 gene is associated
with side effects of Pgp-dependent antidepressants in a large naturalistic Dutch cohort. Pharmacoge-
nomics J. (19 May 2015) | doi:10.1038/tpj.2015.38.
44. Breitenstein B, Scheuer S, Pfister H, et al. The clinical application of ABCB1 genotyping in antidepres-
sant treatment: A pilot study. CNS Spectrums. 2014;19(2):165-175.
45. de Klerk OL, Nolte IM, Bet PM, et al. ABCB1 gene variants influence tolerance to selective serotonin
reuptake inhibitors in a large sample of Dutch cases with major depressive disorder. Pharmacogenom-
ics J. 2013;13(4):349-353.
46. Sadhasivam S, Chidambaran V, Zhang X, et al. Opioid-induced respiratory depression: ABCB1 trans-
porter pharmacogenetics. Pharmacogenomics J. 2015;15(2):119-126.
47. Venkatasubramanian R, Fukuda T, Niu J, et al. ABCC3 and OCT1 genotypes influence pharmacokinetics
of morphine in children. Pharmacogenomics. 2014;15(10):1297-1309.
48. Barratt DT, Coller JK, Hallinan R, et al. ABCB1 haplotype and OPRM1 118a > g genotype interaction in
methadone maintenance treatment pharmacogenetics. Pharmacogenomics Pers Med. 2012;5:53-62.
49. Zhu P, Zhu Q, Zhang Y, et al. ABCB1 variation and treatment response in AIDS patients: Initial results
of the Henan cohort. PloS one. 2013; 8(1): e55197.
50. Gradhand U, Kim RB. Pharmacogenomics of MRP transporters (ABCC1-5) and BCRP (ABCG2). Drug
Metab Rev. 2008;40(2):317-354.
51. Conrad S, Kauffmann HM, Ito KI, et al. A naturally occurring mutation in MRP1 results in a selec-
tive decrease in organic anion transport and in increased doxorubicin resistance. Pharmacogenetics.
2002;12(4):321-330.
52. Leslie EM, Letourneau IJ, Deeley RG, et al. Functional and structural consequences of cysteine
substitutions in the nh2 proximal region of the human multidrug resistance protein 1 (MRP1/ABCC1).
Biochemistry. 2003;42(18):5214-5224.
53. Wang Z, Wang B, Tang K, et al. A functional polymorphism within the MRP1 gene locus identified
through its genomic signature of positive selection. Hum. Mol. Genet. 2005; 14(14):2075-2087.
54. Lima A, Bernardes M, Azevedo R, et al. Pharmacogenomics of methotrexate membrane transport
pathway: Can clinical response to methotrexate in rheumatoid arthritis be predicted? Int J Mol Sci.
2015;16(6):13760-13780.
55. Nakamura T, Yamamori M, Sakaeda T. Pharmacogenetics of intestinal absorption. Curr Drug Deliv.
2008;5(3):153-169.
56. Hulot J-S, Villard E, Maguy A, et al. A mutation in the drug transporter gene ABCC2 associated with
impaired methotrexate elimination. Pharmacogenet Genomics. 2005; 15(5):277-285.
57. Liu Y, Yin Y, Sheng Q, et al. Association of ABCC2 -24c>t polymorphism with high-dose methotrex-
ate plasma concentrations and toxicities in childhood acute lymphoblastic leukemia. PloS one.
2014;9(1):e82681.
58. Noma B, Sasaki T, Fujimoto Y, et al. Expression of multidrug resistance-associated protein 2 is
involved in chemotherapy resistance in human pancreatic cancer. Int J Oncol. 2008; 33(6):1187-1194.
59. Qu J, Zhou BT, Yin JY, et al. ABCC2 polymorphisms and haplotype are associated with drug resistance
in Chinese epileptic patients. CNS Neurosci Ther. 2012;18(8):647-651.
60. Hilger E, Reinthaler EM, Stogmann E, et al. Lack of association between ABCC2 gene variants and
treatment response in epilepsy. Pharmacogenomics. 2012;13(2):185-190.
61. Grant CE, Gao M, DeGorter MK, et al. Structural determinants of substrate specificity differences
between human multidrug resistance protein (MRP) 1 (ABCC1) and MRP3 (ABCC3). Drug Metab Dispos.
2008;36(12):2571-2581.
62. Lang T, Hitzl M, Burk O, et al. Genetic polymorphisms in the multidrug resistance-associated protein 3
(ABCC3, MRP3) gene and relationship to its mRNA and protein expression in human liver. Pharmaco-
genetics. 2004;14(3):155-164.
63. Müller P, Asher N, Heled M, et al. Polymorphisms in transporter and phase II metabolism genes as
potential modifiers of the predisposition to and treatment outcome of de novo acute myeloid leuke-
mia in Israeli ethnic groups. Leuk Res. 2008;32(6):919-929.
64. Doerfel CA, Rump AB, Sauerbrey AC, et al. In acute leukemia, the polymorphism -211c>t in the promot-
er region of the multidrug resistance-associated protein 3 (MRP3) does not determine the expression
level of the gene. Pharmacogenet Genomics. 2006;16(2):149-150.
65. Krishnamurthy P, Schwab M, Takenaka K, et al. Transporter-mediated protection against thiopurine-
induced hematopoietic toxicity. Cancer Res. 2008;68(13):4983-4989.
66. Krajinovic M, Elbared J, Drouin S, et al. Polymorphisms of ABCC5 and NOS3 genes influence doxorubi-
cin cardiotoxicity in survivors of childhood acute lymphoblastic leukemia. Pharmacogenomics J. 2015

Sep 8. doi: 10.1038/tpj.2015.63 [Epub ahead of print].
67. Sassi Y, Lipskaia L, Vandecasteele G, et al. Multidrug resistance-associated protein 4 regulates
camp-dependent signaling pathways and controls human and rat smc proliferation. J Clin Invest.
2008;118(8):2747-2757.
68. Wen J, Luo J, Huang W, et al. The pharmacological and physiological role of multidrug-resistant
protein 4. J Pharmacol Exp Ther. 2015;354(3):358-375.
69. Arlanov R, Lang T, Jedlitschky G, et al. Functional characterization of common protein variants in the
efflux transporter ABCC11 and identification of t546m as functionally damaging variant. Pharmacoge-
nomics J. 2015;Apr 21[epub ahead of print]
70. Yoshiura K, Kinoshita A, Ishida T, et al. A SNP in the ABCC11 gene is the determinant of human earwax
type. Nature Genetics. 2006;38(3):324-330.
71. Brown S. The pharmacogenetics of body odor: As easy as ABCC? J Invest Dermatol. 2013; 133(7):1709-
1711.
72. Ishikawa T, Toyoda Y, Yoshiura K, et al. Pharmacogenetics of human ABC transporter ABCC11: New
insights into apocrine gland growth and metabolite secretion. Front Genet. 2012;3:306 doi: 10.3389/
fgene.2012.00306.
73. Cusatis G, Sparreboom A. Pharmacogenomic importance of ABCG2. Pharmacogenomics.
2008;9(8):1005-1009.
74. Mao Q, Unadkat JD. Role of the breast cancer resistance protein (ABCG2) in drug transport. AAPS J.
2005;7(1):118-133.
75. Krishnamurthy P, Schuetz JD. Role of ABCG2/BCRP in biology and medicine. Annu Rev Pharmacol
Toxicol. 2006;46:381-410.
76. Lee HK, Hu M, Lui S, et al. Effects of polymorphisms in ABCG2, SLCO1b1, SLC10a1 and CYP2c9/19 on
plasma concentrations of rosuvastatin and lipid response in Chinese patients. Pharmacogenomics.
2013;14(11):1283-1294.
77. Kimberly K. Adkison SSV, Lee DY, et al. The ABCG2 c421a polymorphism does not affect oral nitrofu-
rantoin pharmacokinetics in healthy Chinese male subjects. Br J Clin Pharmacology. 2008;66(2):233-
239.
78. Kim HS, Sunwoo YE, Ryu JY, et al. The effect of ABCG2 v12m, q141k and q126x, known functional vari-
ants in vitro, on the disposition of lamivudine. British J Clin Pharmacol. 2007;64(5):645-654.
79. Weiss J, Rose J, Storch CH, et al. Modulation of human BCRP (ABCG2) activity by anti-HIV drugs. Jour-
nal of Antimicrobial Chemotherapy February. 2007;59(2):238-245.
80. Cusatis G, Gregorc V, Li J, et al. Pharmacogenetics of ABCG2 and adverse reactions to gefitinib. J Natl
Cancer Inst. 2006;98(23):1739-1742.
81. Delord M, Rousselot P, Cayuela JM, et al. High imatinib dose overcomes insufficient response associ-
ated with ABCG2 haplotype in chronic myelogenous leukemia patients. Oncotarget. 2013;4(10):1582-
1591.
82. Dehghan A, Köttgen A, Yang Q, et al. Association of three genetic loci with uric acid concentration and
risk of gout: A genome-wide association study. Lancet. 2008;372(9654);1953-1961.
83. Wen CC, Yee SW, Liang X, et al. Genome-wide association study identifies ABCG2 (BCRP) as an allo-
purinol transporter and a determinant of drug response. Clin Pharmacol Ther. 2015;97(5):518-525.
84. Mao Q. BCRP/ABCG2 in the placenta: Expression, function and regulation. Pharmaceutical Research .
2008;25(6):1244-1255.
85. Kondo C, Suzuki H, Itoda M, et al. Functional analysis of SNPs variants of BCRP/ABCG2. Pharm Res.
2004;21(10):1895-1903.
86. Callens C, Debled M, Delord M, et al. High-throughput pharmacogenetics identifies SLCO1a2 polymor-
phisms as candidates to elucidate the risk of febrile neutropenia in the breast cancer rapp-01 trial.
Breast Cancer Res Treat. 2015;153(2):383-389.
87. SEARCH Collaborative Group. SLCO1B1 variants and statin-induced myopathy—a genomewide study. N
Engl J Med. 2008;359(8):78-99.
88. Nishizato Y, Ieiri I, Suzuki H, et al. Polymorphisms of OATP-c (SLC21a6) and OAT3 (SLC22a8) genes:
Consequences for pravastatin pharmacokinetics. Clin Pharmacol Ther. 2003;73(6):554-565.
89. Rohrbacher M, Kirchhof A, Skarke C, et al. Rapid identification of three functionally relevant polymor-
phisms in the OATP1B1 transporter gene using pyrosequencing. Pharmacogenomics. 2006;7(2):167-176.
90. Luzum JA, Theusch E, Taylor KD, et al. Individual and combined associations of genetic variants in
CYP3a4, CYP3a5, and SLCO1b1 with simvastatin and simvastatin acid plasma concentrations. J Cardio-
vasc Pharmacol. 2015;66(1):80-85.
91. Ramsey LB, Johnson SG, Caudle KE, et al. The clinical pharmacogenetics implementation consor-
tium guideline for slco1b1 and simvastatin-induced myopathy: 2014 update. Clin Pharmacol Ther.
2014;96(4):423-428.
92. van de Steeg E, Stranecky V, Hartmannova H, et al. Complete OATp1b1 and OATp1b3 deficiency causes
human rotor syndrome by interrupting conjugated bilirubin reuptake into the liver. J Clin Invest.
2012;122(2):519-528.
93. Berman AE, Chan WY, Brennan AM, et al. N-acetylcysteine prevents loss of dopaminergic neurons in
the eaac1-/- mouse. Ann Neurol. 2011;69(3):509-520.
94. Real E, Gratacos M, Labad J, et al. Interaction of SLC1a1 gene variants and life stress on pharmacologi-
cal resistance in obsessive-compulsive disorder. Pharmacogenomics J. 2013;13(5):470-475.
95. Murphy DL, Fox MA, Timpano KR, et al. How the serotonin story is being rewritten by new gene-based
discoveries principally related to SLC6a4, the serotonin transporter gene, which functions to influence
all cellular serotonin systems. Neuropharmacology. 2008;55(6):932-960.
96. Hu X-Z, Rush AJ, Charney D, et al. Association between a functional serotonin transporter promoter
polymorphism and citalopram treatment in adult outpatients with major depression. Arch Gen Psychi-
atry. 2007;64(7):783-792.
97. Lohoff FW, Narasimhan S, Rickels K. Interaction between polymorphisms in serotonin transporter
(SLC6a4) and serotonin receptor 2a (HTR2a) genes predict treatment response to venlafaxine xr in
generalized anxiety disorder. Pharmacogenomics J. 2013;13(5):464-469.
98. Pasini A, Sinibaldi L, Paloscia C, et al. Neurocognitive effects of methylphenidate on adhd children
with different dat genotypes: A longitudinal open label trial. Eur J Paediatr Neurol. 2013;17(4):407-414.
99. Franke B, Hoogman M, Arias Vasquez A, et al. Association of the dopamine transporter (SLC6a3/DAT1)
gene 9-6 haplotype with adult ADHD. Am J Med Genet B Neuropsychiatr Genet. 2008;147B(8):1576-9.
100. Purper-Ouakil D, Wohl M, Orejarena S, et al. Pharmacogenetics of methylphenidate response in
attention deficit/hyperactivity disorder: Association with the dopamine transporter gene (SLC6a3). Am
J Med Genet B Neuropsychiatr Genet. 2008;147B(8):1425-1430.
101. Moreau C, Meguig S, Corvol JC, et al. Polymorphism of the dopamine transporter type 1 gene modifies
the treatment response in Parkinson’s disease. Brain. 2015;138(Pt 5):1271-1283.
102. Yoshida K, Takahashi H, Higuchi H, et al. Prediction of antidepressant response to milnacipran by
norepinephrine transporter gene polymorphisms. Am J Psychiatry. 2004;161(9):1575-1580.
103. Yang L, Qian Q, Liu L, et al. Adrenergic neurotransmitter system transporter and receptor genes associ-
ated with atomoxetine response in attention-deficit hyperactivity disorder children. J Neural Transm.
2013;120(7):1127-1133.
104. Giglia JL, White MJ, Hart AJ, et al. A single nucleotide polymorphism in SLC7a5 is associated with
gastrointestinal toxicity after high-dose melphalan and autologous stem cell transplantation for
multiple myeloma. Biol Blood Marrow Transplant. 2014;20(7):1014-1020.
105. Fujun Li HMTA. Prodrugs of nucleoside analogues for improved oral absorption and tissue targeting. J
Pharm Sci. 2008;97(3):1109-1134.
106. Phan DD, Chin-Hong P, Lin ET, et al. Intra- and interindividual variabilities of valacyclovir oral
bioavailability and effect of coadministration of an hpept1 inhibitor. Antimicrob Agents Chemother.
2003;47(7):2351-2353.
107. Knutter I, Wollesky C, Kottra G, et al. Transport of angiotensin-converting enzyme inhibitors by h+/
peptide transporters revisited. J Pharmacol Exp Ther. 2008;327(2):432-441.
108. Brandsch M, Knutter I, Bosse-Doenecke E. Pharmaceutical and pharmacological importance of peptide
transporters. J Pharm Pharmacol. 2008;60(5):543-585.
109. Liu W, Liang R, Ramamoorthy S, et al. Molecular cloning of PEPT 2, a new member of the h+/peptide
cotransporter family, from human kidney. Biochim Biophys Acta. 1995;1235(2):461-466.
110. Zhang EY, Fu D-J, Pak YA, et al. Genetic polymorphisms in human proton-dependent dipeptide trans-
porter PEPT1: Implications for the functional role of pro586. J Pharmacol Exp Ther. 2004;310(2):437-
445.
111. Anderle P, Nielsen CU, Pinsonneault J, et al. Genetic variants of the human dipeptide transporter
pept1. J Pharmacol Exp Ther. 2006;316(2):636-646.
112. Pinsonneault J, Nielsen CU, Sadee W. Genetic variants of the human h+/dipeptide transporter PEPT2:
Analysis of haplotype functions. J Pharmacol Exp Ther. 2004;311(3):1088-1096.

113. Smith DE, Clemencon B, Hediger MA. Proton-coupled oligopeptide transporter family slc15: Physi-
ological, pharmacological and pathological implications. Mol Aspects Med. 2013;34(2-3):323-336.
114. Sakae DY, Marti F, Lecca S, et al. The absence of VGLUT3 predisposes to cocaine abuse by increasing
dopamine and glutamate signaling in the nucleus accumbens. J Mol Psychiatry. 2015;20(11):1448-1459.
115. Ruel J, Emery S, Nouvian R, et al. Impairment of SLC17a8 encoding vesicular glutamate transporter-3,
VGLUT3, underlies nonsyndromic deafness DFNA25 and inner hair cell dysfunction in null mice. Am J
Hum Genet. 2008;83(2):278-292.
116. Peter D, Finn JP, Klisak I, et al. Chromosomal localization of the human vesicular amine transporter
genes. Genomics. 1993;18(3):720-723.
117. Zai CC, Tiwari AK, Mazzoco M, et al. Association study of the vesicular monoamine transporter gene
SLC18a2 with tardive dyskinesia. J Psych Res. 2013;47(11):1760-1765.
118. Shaw GM, Zhu H, Lammer EJ, et al. Genetic variation of infant reduced folate carrier (a80g) and risk of
orofacial and conotruncal heart defects. Am J Epidemiol. 2003;158(8):747-752.
119. Chango A, Emery-Fillon N, de Courcy GP, et al. A polymorphism (80g->a) in the reduced folate carrier
gene and its associations with folate status and homocysteinemia. Mol Genet Metab. 2000;70(4):310-
315.
120. Drozdzik M, Rudas T, Pawlik A, et al. Reduced folate carrier-1 80g>a polymorphism affects methotrex-
ate treatment outcome in rheumatoid arthritis. Pharmacogenomics J. 2007;7(6):404-407.
121. Bohanec Grabar P, Leandro-Garcia LJ, Inglada-Perez L, et al. Genetic variation in the SLC19a1 gene and
methotrexate toxicity in rheumatoid arthritis patients. Pharmacogenomics. 2012;13(14):1583-1594.
122. Lima A, Bernardes M, Sousa H, et al. SLC19a1 80g allele as a biomarker of methotrexate-relat-
ed gastrointestinal toxicity in Portuguese rheumatoid arthritis patients. Pharmacogenomics.
2014;15(6):807-820.
123. Yates Z, Lucock M. G80a reduced folate carrier snp modulates cellular uptake of folate and affords
protection against thrombosis via a non homocysteine related mechanism. Life Sciences. 2005;
7(22):2735-2742.
124. Goswami S, Gong L, Giacomini K, et al. PharmGKB summary: Very important pharmacogene informa-
tion for SLC22a1. Pharmacogenet Genomics. 2014;24(6):324-328.
125. Motohashi H, Inui K. Organic cation transporter OCTs (SLC22) and MATEs (SLC47) in the human kidney.
AAPS J. 2013;15(2):581-588.
126. Umamaheswaran G, Praveen RG, Damodaran SE, et al. Influence of SLC22a1 rs622342 genetic poly-
morphism on metformin response in South Indian Type 2 diabetes mellitus patients. Clin Exp Med.
2014;15(4):511-517.
127. Fukuda T, Chidambaran V, Mizuno T, et al. OCT1 genetic variants influence the pharmacokinetics of
morphine in children. Pharmacogenomics. 2013;14(10):1141-1151.
128. Song IS, Lee do Y, Shin MH, et al. Pharmacogenetics meets metabolomics: Discovery of tryptophan as
a new endogenous OCT2 substrate related to metformin disposition. PloS One. 2012;7(5):e36637.
129. Lanvers-Kaminsky C, Sprowl JA, Malath I, et al. Human OCT2 variant c.808g>t confers protection
effect against cisplatin-induced ototoxicity. Pharmacogenomics. 2015;16(4):323-332.
130. Tokuhiro S, Yamada R, Chang X, et al. An intronic SNP in a RUNX1 binding site of SLC22a4, encoding
an organic cation transporter, is associated with rheumatoid arthritis. Nature Genetics. 2003;35(4):341-
348.
131. Wang Y, Korman SH, Ye J, et al. Phenotype and genotype variation in primary carnitine deficiency.
Genet Med. 2001;3(6):387-392.
132. Angelini S, Pantaleo MA, Ravegnini G, et al. Polymorphisms in OCTN1 and OCTN2 transporters genes
are associated with prolonged time to progression in unresectable gastrointestinal stromal tumours
treated with imatinib therapy. Pharmacol Res. 2013;68(1):1-6.
133. Yoon H, Cho HY, Yoo HD, et al. Influences of organic cation transporter polymorphisms on the popula-
tion pharmacokinetics of metformin in healthy subjects. AAPS J. 2013;15(2):571-580.
134. Emami Riedmaier A, Burk O, van Eijck BA, et al. Variability in hepatic expression of organic anion trans-
porter 7/SLC22a9, a novel pravastatin uptake transporter: Impact of genetic and regulatory factors.
Pharmacogenomics J. 2015 (55) [epub ahead of print].
135. Sun H, Qu Q, Qu J, et al. URAT1 gene polymorphisms influence uricosuric action of losartan in hyper-
tensive patients with hyperuricemia. Pharmacogenomics. 2015;16(8):855-863.
136. Shafiu M, Johnson RJ, Turner ST, et al. Urate transporter gene SLC22a12 polymorphisms associ-
ated with obesity and metabolic syndrome in Caucasians with hypertension. Kidney Blood Press Res.
2012;35(6):477-482.
137. Huang Y, Anderle P, Bussey KJ, et al. Membrane transporters and channels: Role of the transportome
in cancer chemosensitivity and chemoresistance. Cancer Res. 2004;64(12):4294-4301.
138. Kim S-R, Saito Y, Maekawa K, et al. Thirty novel genetic variations in the SLC29a1 gene encoding
human equilibrative nucleoside transporter 1 (HENT1). Drug Metab Pharmacokinet. 2006;21(3):248-256.
139. Lal S, Wong ZW, Sandanaraj E, et al. Influence of ABCB1 and ABCG2 polymorphisms on doxorubicin
disposition in Asian breast cancer patients. Cancer Sci. 2008;99(4):816-823.
140. Toh S, Wada M, Uchiumi T, et al. Genomic structure of the canalicular multispecific organic anion-
transporter gene (MRP2/CMOAT) and mutations in the ATP-binding-cassette region in Dubin-Johnson
syndrome. Am J Hum Genet. 1999;64(3):739-746.
141. Hoblinger A, Grunhage F, Sauerbruch T, et al. Association of the c.3972c>t variant of the multidrug
resistance-associated protein 2 gene (MRP2/ABCC2) with susceptibility to bile duct cancer. Digestion.
2009;80(1):36-39.
142. Zamek-Gliszczynski MJ, Nezasa K-I, Tian X, et al. Evaluation of the role of multidrug resistance-
associated protein (MRP) 3 and MRP4 in hepatic basolateral excretion of sulfate and glucuronide
metabolites of acetaminophen, 4-methylumbelliferone, and harmol in ABCC3-/- and ABCC4-/- mice.
J Pharmacol Exp Ther. 2006;319(3):1485-1491.
143. Zelcer N, van de Wetering K, Hillebrand M, et al. Mice lacking multidrug resistance protein 3 show
altered morphine pharmacokinetics and morphine-6-glucuronide antinociception. Proc Natl Acad Sci
USA. 2005;102(20):7274-7279.
144. Muller PJ, Dally H, Klappenecker CN, et al. Polymorphisms in ABCG2, ABCC3 and CNT1 genes and their
possible impact on chemotherapy outcome of lung cancer patients. Int J Cancer. 2009;124(7):1669-
1674.
145. Kobayashi K, Ito K, Takada T, et al. Functional analysis of nonsynonymous single nucleotide polymor-
phism type ATP-binding cassette transmembrane transporter subfamily C member 3. Pharmacogenet
Genomics. 2008;18(9):823-833.
146. Ho LL, Kench JG, Handelsman DJ, et al. Androgen regulation of multidrug resistance-associated
protein 4 (MRP4/ABCC4) in prostate cancer. Prostate. 2008;68(13):1421-1429.
147. El-Sheikh AAK, van den Heuvel JJMW, Koenderink JB, et al. Effect of hypouricaemic and hyperuri-
caemic drugs on the renal urate efflux transporter, multidrug resistance protein 4. Br J Pharmacol.
2008;155(7):1066-1075.
148. Torky A-RW, Stehfest E, Viehweger K, et al. Immuno-histochemical detection of MRPS in human lung
cells in culture. Toxicology. 2005; 207(3): 437-450.
149. Nies AT, Jedlitschky G, Konig J, et al. Expression and immunolocalization of the multidrug resistance
proteins, MRP1-MRP6 (ABCC1-ABCC6), in human brain. Neuroscience. 2004;129(2):349-360.
150. Konig J, Hartel M, Nies AT, et al. Expression and localization of human multidrug resistance protein
(ABCC) family members in pancreatic carcinoma. Int J Cancer. 2005; 115(3):359-367.
151. Schuetz JD, Connelly MC, Sun D, et al. MRP4: A previously unidentified factor in resistance to nucleo-
side-based antiviral drugs. Nat Med. 1999;5(9):1048-1051.
152. Jedlitschky G, Tirschmann K, Lubenow LE, et al. The nucleotide transporter MRP4 (ABCC4) is highly
expressed in human platelets and present in dense granules, indicating a role in mediator storage.
Blood. 2004;104(12):3603-3610.
153. Dazert P, Meissner K, Vogelgesang S, et al. Expression and localization of the multidrug resistance
protein 5 (MRP5/ABCC5), a cellular export pump for cyclic nucleotides, in human heart. Am J Pathol.
2003;163(4):1567-1577.
154. Abla N, Chinn LW, Nakamura T, et al. The human multidrug resistance protein 4 (MRP4, ABCC4): Func-
tional analysis of a highly polymorphic gene. J Pharmacol Exp Ther. 2008;325(3): 859-868.
155. Kiser JJ, Aquilante CL, Anderson PL, et al. Clinical and genetic determinants of intracellular tenofo-
vir diphosphate concentrations in hiv-infected patients. J Acquir Immune Defic Syndr. 2008; 47(3):
298-303.
156. Shi Y, Terry SF, Terry PF, et al. Development of a rapid, reliable genetic test for pseudoxanthoma elas-
ticum. J Mol Diagn. 2007;9(1):105-112.
157. Kim I-S, Kim H-G, Kim DC, et al. ABCG2 q141k polymorphism is associated with chemotherapy-induced
diarrhea in patients with diffuse large B-cell lymphoma who received frontline rituximab plus cyclo-
phosphamide/doxorubicin/vincristine/prednisone chemotherapy. Cancer Sci. 2008; 99(12):2496-2501.
158. Gill RK, Pant N, Saksena S, et al. Function, expression, and characterization of the serotonin transport-
er in the native human intestine. Am J Physiol Gastrointest Liver Physiol. 2008;294(1):254-262.
159. Goldberg TE, Kotov R, Lee AT, et al. The serotonin transporter gene and disease modification in
psychosis: Evidence for systematic differences in allelic directionality at the 5-HTTLPR locus.
Schizophr Res. 2009;111(1-3):103-108.
160. Dervieux T, Kremer J, Lein DO, et al. Contribution of common polymorphisms in reduced folate carrier
and gamma-glutamylhydrolase to methotrexate polyglutamate levels in patients with rheumatoid
arthritis. Pharmacogenetics. 2004;14(11):733-739.
161. Zhou F, Zheng J, Zhu L, et al. Functional analysis of novel polymorphisms in the human SLCO1a2 gene
that encodes the transporter OATP1a2. AAPS J. 2013; 15(4):1099-1108.
162. Ho RH, Choi L, Lee W, et al. Effect of drug transporter genotypes on pravastatin disposition in Euro-
pean- and African-American participants. Pharmacogenet Genomics. 2007;17(8):647-656.
163. Miura M, Satoh S, Inoue K, et al. Influence of SLCO1B1, 1B3, 2B1 and ABCC2 genetic polymorphisms on
mycophenolic acid pharmacokinetics in Japanese renal transplant recipients. Eur J Clin Pharmacol.
2007;63(12):1161-1169.
164. Rubio-Aliaga I. Daniel H. Peptide transporters and their roles in physiological processes and drug
disposition. Xenobiotica. 2008;38(7):1022-1042.
165. Terada T, Irie M, Okuda M, et al. Genetic variant ARG57his in human H+/peptide cotransporter 2
causes a complete loss of transport function. Biochem Biophys Res Commun. 2004; 316(2):416-420.
166. Govindarajan R, Bakken AH, Hudkins KL, et al. In situ hybridization and immunolocalization of concen-
trative and equilibrative nucleoside transporters in the human intestine, liver, kidneys, and placenta.
Am J Physiol Regul Integr Comp Physiol. 2007; 293(5): R1809-R1822.
CHAPTER
6
Pharmacodynamics and
Pharmacogenomics
Kathy D. Webster, PharmD, PhD and Samit Shah, PhD, RPh, MBA

After completing this chapter, the reader BIOMARKER—an indicator of a particular
should be able to disease state or a particular state of an
organism.
List the common types of drug
targets that are affected by genetic DEOXYRIBONUCLEIC ACID (DNA)—a self-
polymorphism. replicating material present in nearly all
Describe the consequences of genetic
of chromosomes.
variability on drug action.
DRUG TARGET—an enzyme or receptor
Describe the types of drug targets that
protein that is a direct target of a drug,
have demonstrated or are associated with
a signal transduction, or downstream
genetic variability and give examples of protein that mediates a drug response, or
each type. a protein associated with a disease that
Explain how genetic variability can is modified by a drug.
indirectly modulate the overall drug EPIGENETICS—heritable changes in gene
response. function that occur without change in the
Discuss the role of genetic variability on sequence of nuclear DNA.
the incidence of adverse drug reactions. GENOTYPE—set of two alleles carried by an
List common examples of genetic individual at a given polymorphic site.
polymorphism and adverse drug GERMLINE MUTATIONS—genetic variation
reactions associated with them. that is detectable and occurs in germ
Discuss the role of disease state cells; a heritable mutation.
biomarkers and drug response. HAPLOTYPE—group of alleles or
polymorphisms on a single chromosome
that are closely linked and inherited as
a unit.
LINKAGE DISEQUILIBRIUM—multiple single
nucleotide polymorphisms that always
appear together; the alleles are inherited
as a unit, also known as a haplotype
block.
PHENOTYPE—physical manifestation of a
genetic trait or a general constitutional
manifestation of an individual’s health or
disease.
183
POLYMORPHISM—genetic variation in the DNA sequence with a measurable frequency of

detection >1%.
SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—genetic variation at a single DNA base with a
measurable frequency of detection >1%.
SOMATIC CELL MUTATIONS—genetic variation that occurs only in an affected organ or

disease locus (tumor); acquired mutation.
INTRODUCTION
The use of pharmacogenetic markers or biomarkers to identify patients who will most
likely benefit from a given treatment is an important new tool for optimizing drug therapy.
Pharmacogenetic studies associate a characteristic drug response (i.e., a phenotype) with a
genetic polymorphism (i.e., a genotype). These polymorphisms may include single nucleo-
tide polymorphisms (SNPs), nucleotide sequence repeats, insertions/deletions (I/Ds), or
grouped mutations (haplotypes), which are characteristic of the genotype. A patient is
considered homozygous if both copies of the gene, each allele, is the same and hetero-
zygous if each of the alleles is different. Mutations can occur in both germline (inherited)
deoxyribonucleic acid (DNA) or somatic (tumor or tissue-specific) DNA.
Pharmacodynamic-based pharmacogenetic studies are those that focus on drug
targets. Altered drug response has been associated with allelic variants of genes encoding
for molecular targets and their associated modifiers, or key proteins in the pathophysiology
of systems affected by the drug. Over- or under-expression of normal proteins or produc-
tion of variant proteins for receptors, enzymes, ion channels, transcription factors, and
intracellular or extracellular signaling proteins can all modify drug response.
Our understanding of pharmacokinetic-based pharmacogenetics is more advanced
than our understanding of pharmacodynamic-based pharmacogenetics. Unlike pharmaco-
kinetic studies where a genetic variation in an enzyme or transporter might affect many
drugs or disease states, pharmacodynamic studies generally focus on one drug, one drug
class, or one disease-specific target at a time. This more narrow focus slows the progress
of pharmacodynamic-based pharmacogenetics. The correlation of pharmacogenetic vari-
ability in drug targets with clinical outcomes has shown inconsistent results that may be
due, in part, to attempts to relate single SNPs to pharmacodynamic variability where a
more complex relationship may be present.
Multiple SNPs or haplotypes may be better predictors of drug target response. In
addition, the complexity of a drug response typically involves multiple components. For
example, a cell surface receptor generally utilizes a protein receptor that interfaces with
a multicomponent second messenger cascade. Genetic variation in one or more of these
components may affect the ultimate outcome of the drug response. Many physiological
responses are likewise mediated by a balance between multiple receptor types; thus, the
complexity increases when trying to pinpoint the impact of a single genetic mutation. It is
likely that a genetic profile of relevant genes would be more helpful than the genotyping
of an SNP or haplotype in predicting therapeutic outcome.1
Variability in the drug response phenotype can also be due to more than one polymor-
phism that affects both the pharmacokinetic and pharmacodynamic properties of the drug
Chapter 6 • Pharmacodynamics and Pharmacogenomics 185
(Figure 6-1).2 Pharmacokinetic characteristics will determine how much drug gets to the
site of action, while pharmacodynamic characteristics will determine the responsiveness
of the drug target. Two recently characterized examples, warfarin and β-blockers, illustrate
this concept.
Warfarin, an anticoagulant drug with a narrow therapeutic index, shows significant
variability in its dose response. Inappropriate dosing for an individual patient can lead to
serious adverse outcomes; too much warfarin will result in excess bleeding, while insuf-
ficient warfarin can result in blood clot formation. Both of these conditions may be life-
threatening.
At least one pharmacokinetic and one pharmacodynamic polymorphism have been
shown to affect the therapeutic response of warfarin (Figure 6-2).3 S-Warfarin, the most
potent enantiomer of the racemic mixture, is predominantly metabolized by CYP2C9. The
cytochrome P450 enzyme, CYP2C9, has a well-characterized genetic polymorphism (see
Chapter 4). The CYP2C9*2 and *3 alleles are significantly less active than the wild-type allele,
CYP2C9*1. An individual who carries one or more of the variant alleles will have reduced
metabolism of warfarin with a concomitant increase in drug levels and risk of hemorrhaging.
The molecular target for warfarin is vitamin K epoxide reductase complex 1 (VKORC1), which
Drug target
Drug-
Drug metabolizing Drug binds to target
enzyme and triggers response
Example of genetic variant

Genetic influencing drug
polymorphism in drug- pharmacokinetics
metabolizing enzyme Change in amount of drug
available at site of action
resulting in altered response
Example of genetic variant

influencing drug
pharmacodynamics
Genetic polymorphism
in drug target (or downstream
signaling protein) alters
response to drug
Example of genetic variants

influencing both drug
pharmacokinetics and
pharmacodynamics
FIGURE 6-1 Variability in drug response can be due to polymorphisms in proteins that
affect the pharmacokinetic or pharmacodynamic properties of the drug. Warfarin and
β- blockers are examples of drugs, where both pharmacokinetic- and pharmacodynamics-
related genetic polymorphisms influence drug response.
Coagulation
pathway
Pharmacogenetic Target
Non-functional
Coagulation Dicarboxylated
Factors Coagulation
Factors
Gamma Glutamyl Carboxylase
Calumenin
OH O
CH3 CH3
O
R R
OH O
Reduced Vitamin K Vitamin K Epoxide
Vitamin K Epoxide
Warfarin-7- Reductase Complex
Hydroxyl
CYP2C9 Inhibits
Warfarin
FIGURE 6-2 Warfarin inhibits the vitamin K epoxide reductase complex (VKORC1),
decreasing the availability of reduced vitamin K. Reduced vitamin K is required for
synthesis of the coagulation factors, which must be activated to produce coagulation. Both
pharmacokinetics and pharmacodynamics affect patient variability. The pharmacokinetic
polymorphism is in cytochrome P450 2C9 (CYP2C9), and the pharmacodynamic
polymorphisms are found in VKORC1.
converts vitamin K epoxide to the active reduced form of vitamin K required for synthesis
of coagulation factors. Warfarin inhibits this enzymatic conversion.
A series of characterized haplotypes have been associated with high- or low-dose
warfarin requirements (see the section on enzymes as drug targets). The warfarin dose
requirement is indicative of VKORC1 enzymatic activity. A haplotype that requires low-

dose warfarin has a much lower intrinsic enzymatic activity compared to a haplotype that
requires high-dose warfarin. Therefore, a patient with both decreased drug metabolism
(CYP2C9 activity) and enhanced inhibition of the target (VKORC1) would require a much
lower dose of warfarin compared to the average patient in order to achieve the desired
therapeutic response. The combination of CYP2C9 and VKORC1 genetic polymorphisms,
along with other risk factors, accounts for more than 50% of dosing variability seen with
warfarin.4
The clinical efficacy of β-blockers or β-adrenergic receptor antagonists is also affected
by both pharmacokinetic and pharmacodynamic genetic variations. Many β-blockers are

metabolized by another cytochrome P450 isoform, CYP2D6. Phenotypes for CYP2D6
metabolism can be classified as poor metabolizers (PMs), intermediate metabolizers (IMs),
extensive or normal metabolizers (EMs), and ultra-rapid metabolizers (UMs).5 With chronic
administration of an equivalent dose of the β-blocker metoprolol, PMs had approximately
fivefold higher blood levels compared to non-PMs.6 Metoprolol produced greater reduc-
tions in heart rate, diastolic blood pressure, and mean arterial pressure in the PMs. Several
polymorphisms of the β1-adrenergic receptor have also been characterized (see the section
on cell surface receptors as drug targets). Some studies have shown increased response to
β-blockers in patients homozygous for the Arg389 allele.7 It appears likely that the varia-
tion in response to β-blockers is determined by both alterations in metabolism as well as
receptor response. However, no study has yet documented the clinical interaction between
these genes.
Pharmacodynamic variability associated with germline or somatic DNA mutations
may result in increased or decreased susceptibility to disease. A significant change in
a drug-target protein that causes loss or excessive functional activity may result in a
disease phenotype. For example, β-adrenergic receptor polymorphism has been associated
with asthma,8 genetic variation of serotonin receptors may predict the risk of developing
depression,9 and mutated K+ channels can result in cardiac arrhythmias.10 The gene for
APOE4 has been associated with increased progression of both coronary heart disease
and Alzheimer’s disease.11 Somatic mutations of growth factor receptors are associated
with highly aggressive forms of cancer.12 Natural selection favors emergence of genetic
mutations that provide protection from specific diseases; therefore, glucose-6-phosphate
dehydrogenase (G6PD) polymorphisms were thought to have developed in order to provide
protection against malaria.13
The goal of pharmacogenetics is to optimize drug efficacy and minimize toxicity based
on an individual’s genetic profile. Ideally, the clinician will be able to predict the patient’s
drug response based on his or her DNA and use this information to pick the right drug
and the right dose. Genetic polymorphisms in proteins that are drug targets will obviously
mediate variability in drug response. More subtle changes in proteins that affect the physi-
ological context, provide secondary targets, or mediate the development or severity of a
disease will also indirectly cause variations in drug response.
CLINICAL PEARL
Pharmacodynamic-based genetic variability is more complicated and less
well understood compared to pharmacokinetic-based genetic variability.
CASE STUDY
M.K. is a 70-year-old Asian female who weighs 60 kg. She presents to the clinic with
complaints of a sensation of a “racing heart.” On an electrocardiogram, M.K. is found to be in
atrial fibrillation. She states that she started having these symptoms approximately 3 days
ago. M.K. is being considered for warfarin therapy. The warfarin product label contains
the following recommendation for initial dose, if genotype information is available.
VKORC1
Genotypes
(−1639 G>A) CYP2C9 Genotypes
*1/*1 *1/*2 *1/*3 *2/*2 *2/*3 *3/*3

G/G 5−7 mg/day 5−7 mg/day 3−4 mg/day 3−4 mg/day 3−4 mg/day 0.5−2 mg/day
A/G 5−7 mg/day 3−4 mg/day 3−4 mg/day 3−4 mg/day 0.5−2 mg/day 0.5−2 mg/day
A/A 3−4 mg/day 3−4 mg/day 0.5−2 mg/day 0.5−2 mg/day 0.5−2 mg/day 0.5−2 mg/day
Questions
1. What dosage range will you consider for M.K. if she has an A/G VKORC1 genotype and a *2/*2
CYP2C9 genotype?
2. Is the variability in response to warfarin due to polymorphisms that affect the pharmacoki-
netics or to the pharmacodynamic property of the drug; or is it both?
3. How can genetic polymorphisms in VKORC1 lead to variability in response to warfarin therapy?
4. Are there any known genetic polymorphisms besides CYP2C9 and VKORC1 that may affect
the response to warfarin therapy?
CONSEQUENCES OF GENETIC VARIABILITY AND DRUG

PHARMACODYNAMICS
Pharmacogenetics can profoundly affect the efficacy and safety of a drug. Functional
variations in the coding sequence or expression level of a drug target protein can affect
responses to a drug by altering drug-target interactions and/or subsequent signaling
events, thus producing insufficient or excessive pharmacological activity. In some situa-
tions, the genetic variant can result in a total lack of response to the drug entity. Somatic
mutations in tumors can provide a unique target that is not found in normal tissues, and so
provide a drug effect that is specific for the tumor tissue.
Changes in genes that affect the physiological environment (i.e., variations that
produce an underlying disease state or more subtle changes in homeostasis) can result in
variability in the therapeutic response or an increased or decreased incidence of a particu-
lar adverse drug reaction (ADR). Genetic variation in nontarget receptors can also result in
variable risk for ADRs.
A full understanding of the role that genetic variation can play in pharmacodynamics
will optimize therapeutic decision-making. Drugs that will not work or that will precipitate
an ADR can be avoided. Patients who need to start at a higher or lower dose can be identi-
fied and treated accordingly.
GENETIC VARIABILITY OF DRUG TARGETS

A polymorphism in a gene coding for a protein that functions as a drug target can result in
enhanced or reduced drug response. Typical drug targets include cell surface or cytosolic
receptor proteins or enzymes. A list of drug targets associated with genetic variants that
result in a pharmacodynamic change are summarized in Table 6-1.
Cell Surface Receptors and Signal Transduction Systems

ß-Adrenergic Receptors
Variability in drug response can be related to changes in a cell surface receptor and/or its
second messenger proteins. A classic example can be found with the G-protein-coupled
β-adrenergic receptors. The effect of genetic polymorphism on the activity of these
β-adrenergic receptors has been extensively studied, and modified drug response has been
documented for both β1 and β2 receptor subtypes.14
β2-adrenergic receptors (ADRB2) are important mediators of bronchodilation, veno-
dilation, and lipid metabolism. Although studies on genetic polymorphism of ADRB2
have many inconsistencies, variants in the gene coding for ADRB2 have been associated
with significant variability in the response of asthma patients to short-acting β2-agonists
(SABA). No consistent results have been seen with the long-acting β2-agonists (LABA). The
ADRB2 has also shown variable response to the vascular effects of β-agonists.15
At least 13 SNPs have been documented in the ADRB2 gene, and two important modified
receptor proteins have been identified, Arg16Gly and the Gln27Glu.6,16 There is some link-
age disequilibrium with these two SNPs in that the Arg16 SNP is usually linked to the Glu27
SNP, but the Gly16 SNP can be linked with either the Glu27 or the Gln27 SNP.17 Individuals
with Gly16/Gly16 gene were found to have higher rates of drug response and increased
receptor down regulation with albuterol compared to patients with the Arg16/Arg16
gene.18,19 Studies have documented a decline in peak expiratory flow rates and exacerbation
Table 6-1
Drug Targets Associated with Genetic Variability to Drug Response
Target Type Affected Drug/Class
β2-adrenergic receptor Cell surface receptor β2 receptor agonists8,14-25

β1-adrenergic receptor second Cell surface receptor β1 receptor agonists and
messenger antagonists7,26-33
Chemokine (C-C motif) Cell surface receptor CCR5 inhibitors34-40
receptor 5 (CCR5)
Mu opioid receptor Cell surface receptor Opioids41-44
Serotonin transporter and receptor Cell surface receptor SSRIs45-55
Estrogen receptors Cytosolic receptor Estrogen/tamoxifen56-58
VKORC1 Enzyme Warfarin59-62
5-Lipoxygenase Enzyme Zileuton63-65
Angiotensin-converting enzyme Enzyme ACE inhibitors66–70

Cyclo-oxygenase 2 Enzyme NSAIDs71
ACE, angiotensin-converting enzyme; NSAIDs, nonsteroidal anti-inflammatory drugs; SSRIs, selective serotonin
reuptake inhibitors; VKORC1, vitamin K epoxide reductase complex 1.
of asthma with regular use of SABAs in patients with the ArgArg16 receptor.18,20,21 Limited
studies on the long-term use of LABA in conjunction with inhaled corticosteroid use report-
ed no difference with the Arg16 allele.22 Another low frequency variant allele, Thr164Ile,
has been characterized both functionally and clinically.23,24 The Ile164 receptor protein has
a threefold reduction in affinity for β-agonists and a decrease in basal and agonist-stimu-
lated adenylyl cyclase activity.23 In the dorsal hand vein model, the Ile164 variant protein
was associated with a fivefold reduction in sensitivity to ADRB2 agonist (isoproterenol)-
mediated vasodilation while vasoconstrictor (epinephrine) sensitivity was increased.24
Although some studies have supported the genetic associations described above,
other studies have shown no difference or have had opposing results. The effect of poly-
morphisms on β2-agonist response is variable in different ethnic groups and, as a result, it
is likely that an examination of haplotypes containing these SNPs might be more useful in
predicting the clinical response to β-agonists.16,25
ADRB1s are important in regulating heart rate and contractility as well as renin release
by the kidney. ADRB1 is coupled to the stimulatory G-protein, Gs, which activates adenylyl
cyclase and other non-cAMP pathways including various ion channels (Figure 6-3). The two
most common ADRB1 genetic polymorphisms, Ser49Gly and Arg389Gly, modify resting
heart rate and blood pressure and enhance the response to β-agonists and antagonists.26
The Ser49Gly polymorphism showed variation in agonist-mediated down regulation as
evidenced by increased receptor desensitization after exposure to isoproterenol compared
to the wild type.27 The Arg389Gly polymorphism is located in the intracellular cytoplas-
mic tail. The Arg389 receptor showed increased coupling to the G protein with enhanced
adenylyl cyclase activity. As a result, second messenger activity increased under both basal
and isoproterenol-stimulated conditions.28 Patients homozygous for the Arg389 allele
had greater reduction in diastolic blood pressure in response to metoprolol, a β-blocker,
compared to patients who were carriers of the Gly389 allele.29 Hypertensive patients with
the Ser49Arg389/Ser49Arg389 haplotype protein were associated with the best systolic
blood pressure response to metoprolol in both Caucasian and Chinese populations.7,29
The pharmacodynamic variability of several cardiac parameters in response to β-blocker
treatment has also been examined. Heart rate response to metoprolol was not affected by
these polymorphisms, but heart failure patients with the Arg389 variant or Ser49Arg389
haplotype proteins showed the greatest improvement in left ventricular ejection fraction
after initiation of β-blocker therapy.30,31
Several studies have looked at polymorphisms in the Gs protein associated with the
ADRB1 receptor.32,33 One study examined the response to atenolol in hypertensive patients
with the 825C/T variant in the β subunit of the gene for the Gs protein (GNB3) of ADRB1.
The T allele has been associated with increased signal transduction. Females with the CC
alleles showed a significant fall in blood pressure compared to the T carriers. An earlier
study showed an enhanced response to β-blockers in Caucasian hypertensive patients with
an SNP, 131T/C, in the gene for β subunit of the Gs protein (GNAS) of ADRB1. No specific
β-blocker was used. The 131C allele was associated with good responders; however, the
functional basis of the variation in drug response is unclear. The ADRB1 receptor (Figure
6-3) is a model for altered drug response due to genetic variation in the cell surface recep-
tor as well as the signal transduction system.
Pharmacogenetic target
Ser49Gly
Arg389Gly
Receptor
Adenylyl
cyclase
α β
γ
ATP cAMP
G-protein
131T/C
825C/T
FIGURE 6-3 The Ser49 and the Arg389 alleles in the receptor protein show enhanced
receptor activity and increased response to metoprolol and other β-blockers. The 131C allele
in the a subunit protein and the 825C allele in the β subunit protein of the coupled Gs protein
also show an increased response to β-blockers.
HIV Therapeutics
Drug response in human immunodeficiency virus type 1 (HIV-1) infections is dependent on
both virus and host genetics.34 To infect host cells HIV-1 must first bind to a CD4 antigen,
a receptor on the cell surface of human helper T-cells, monocytes/macrophages, dendritic
cells, and glial cells. A co-receptor is also necessary for HIV-1 entry into the host cell.
The human chemokine receptors CCR5 and CXCR4 are the main co-receptors used by the
Macrophage tropic (M- or CCR5-tropic) and the T-cell-tropic (T- or CXCR4-tropic) HIV-1
strains, respectively, for entering their CD4+ target cells. A protein on the surface of the HIV
particle, glycoprotein 120, binds specifically to CD4 and the co-receptor forming a complex
that allows the viral envelope to join with the host cell membrane and the virus to enter
the host cell.
An allelic mutation of the CCR5 receptor has been identified that provides resistance
against HIV-1 by blocking attachment to the receptor, thus denying entry of the virus
into the macrophage.35 The mutation consists of a 32 nucleotide deletion that prevents
expression of the receptor on the cell surface. The homozygous variant provides almost
complete resistance to HIV-1 infection, while heterozygous alleles show partial resistance
with slower disease progression. The CCR5-Delta32 deletion variant has been found with
a high frequency in Caucasian populations (10%) but has not been found in African, Asian,
Middle Eastern, and American Indian populations.36
The observation that a chemokine receptor mutation can slow or delay the onset of
acquired immunodeficiency syndrome (AIDS) in patient populations suggested that inhibi-
tors of CCR5 might prohibit the entrance of the virus into the human cell and provide a
novel form of antiretroviral therapy. Several CCR5 inhibitors have been developed, but only
maraviroc (Selzentry®) has been approved for treatment of AIDS.34,37 In patients infected
with the CCR5-tropic HIV-1 strain, maraviroc in combination with optimized antiretroviral
therapy was shown to be effective and generally well tolerated for at least 48 weeks.38
Appropriate use of maraviroc requires that the virus is the CCR5-tropic HIV-1 strain.37 The
impact of host CCR5 polymorphisms on the efficacy of maraviroc is being evaluated.39
Therefore, genetic variants of both the pharmacodynamic receptor and the infecting agent
may be important for determining drug efficacy against HIV.
Central Nervous System Targets
The OPRM1 gene encodes the protein for the mu opioid receptor, which is responsible for
mediating the analgesic effects of most opioid drugs. A SNP in exon 1, A118G, results in an
amino acid change from asparagine to aspartate at position 40. The 40Asp allele occurs
with a frequency of 10–19% in the Caucasian population.40 The 40Asp allele has been asso-
ciated with decreased messenger ribonucleic acid (mRNA) expression and receptor protein
levels, as well as a decreased analgesic response to morphine.41,42 The mu opioid receptor
has also been implicated in the reward properties of several addictive substances including
cocaine, alcohol, and opioids. Mu receptor antagonists like naltrexone have been used to
treat alcohol and opioid dependence, and it has been suggested that variability in response
to naltrexone for treatment of alcoholism has a genetic component.43 Alcohol-dependent
patients with the 40Asp variant were shown to be more responsive to naltrexone and less
likely to relapse into alcoholism.44 Thus OPRMI genotyping might be useful in selecting
treatment for alcoholism as well as pain management.
A second type of central nervous system (CNS) cell surface drug target is the sero-
tonin transporter (SERT, 5-HTT, SLC6A4) protein, which is a prime target for antidepressant
drugs. Inhibition of this transporter results in increased synaptic levels of serotonin that
modulates neuronal activity by allowing serotonin to bind other targets. Kirchheiner et al.
recently reviewed the impact of pharmacogenetics on antidepressants and antipsychot-
ics.45 Polymorphisms in the SLC6A4 gene were correlated with an increased response
to selective serotonin reuptake inhibitors (SSRIs) in a European population.46 SLC6A4
polymorphisms resulted in variable expression of the transporter protein.47 The 44 base
pair I/D (5HTTLPR) in the promoter region of the gene results in a long (l) or short (s)
allele. The (l) allele shows twofold higher expression and is associated with an increased
response to the SSRIs compared to the (s) allele. Some studies in Asian populations have
shown better response to fluoxetine with carriers of the (s) allele.48 The frequency of
the (s) allele is much higher in Asians (79%) than Caucasians (42%); therefore, ethnicity
may explain the discrepancy in results.45,49 A second SLC6A4 polymorphism found in the
deleted portion of the gene, the rs25531 A/G allele, shows a decreased expression of the
protein similar to the deletion or (s) allele.50
Another variant associated with antidepressant efficacy, the variable number tandem
repeat polymorphism for SLC6A4, has three alleles with 9, 10, and 12 copies of the tandem
repeat in the second intron. The 12-copy tandem repeat allele had higher expression and
was associated with better response to the SSRIs.48 Reynolds et al. recently published
a summary of SLC6A4 polymorphisms and their impact on antidepressant activity.51
Although the serotonin receptor is not the direct pharmacological target for SSRIs, inhibi-
tion of SLC6A4 by the SSRIs will increase the levels of serotonin in the synapse and thus
increase stimulation of serotonin receptors. Antidepressant activity is thought to result

from the increase in serotonin activity at the receptors. Therefore, polymorphisms of the
serotonin receptors could also affect the antidepressant activity of SSRIs.
The 102T/C (rs6313) polymorphism in the 5-HT2A receptor gene showed better anti-
depressant response with one or two 102C alleles compared to the T/T homozygote.52
The 5-HT1A receptor (HTR1A) is an auto receptor that inhibits neurotransmitter activity.
Downregulation of this receptor results in increased neurotransmitter activity, which may
produce both antidepressant and antipsychotic effects. A well-studied polymorphism,
1019C/G (rs6295), has been associated with decreased antidepressant and antipsychotic
response and increased depression and suicidality.53-55 The G allele blocks the action of
inhibitory transcription factors resulting in an increase in HTR1A expression with a concomi-
tant decrease in serotonergic neurotransmission. This is an example of how polymorphisms
in genes that are either direct (SLC6A4, HTR1A) or indirect (5-HT2A) drug targets can
modify pharmacodynamic response.
Cytosolic Receptors
Estrogen Receptors
Cytosolic estrogen receptors (ER) are members of the nuclear steroid receptor superfamily.
Two receptor types have been identified, ER-α and ER-β, these proteins are products of the
ER1 and ER2 genes. Estrogen’s positive effects on heart disease have been associated with
its ability to raise plasma levels of high-density lipoprotein (HDL) cholesterol.56 Herrington
et al. evaluated 10 different variants of ER-α in postmenopausal women with coronary
disease.57 Women with the IVS1 (intervening sequence on intron 1) 401 C/C genotype
and several other closely linked intron 1 polymorphisms had an increase in HDL choles-
terol levels with hormone replacement therapy, which was more than twice the increase
observed in other women.
More recently, ER polymorphisms and the lipid effects of tamoxifen were investigated
in women treated for breast cancer.58 Tamoxifen is considered a selective estrogen receptor
modifier because it has both agonist and antagonist activity at various ERs. Genetic variants
for both ER-α and ER-β were associated with differences in tamoxifen-mediated changes in
HDL cholesterol, low-density lipoprotein cholesterol, and triglycerides. Tamoxifen effects
differed in premenopausal and postmenopausal women. The mechanism for and the clini-
cal significance of the ER-mediated changes in lipids is unclear.
Enzyme Targets
Warfarin Response
There are numerous examples of genetic variation in enzymes used as drug targets. As
previously discussed, warfarin’s primary drug target is the enzyme VKORC1 (see Figure
6-2). Inhibition of VKORC1 interferes with vitamin K reduction and vitamin K-dependent
carboxylation of clotting factors II, VII, IX, and X as well as proteins C and S. Depletion
of vitamin K results in the production of nonfunctional coagulation factors and loss of
coagulation. Genetic polymorphism in the VKORC1 gene illustrates a clinically important

example of variation in drug response associated with haplotypes rather than a single
SNP.59 Promoter and intronic VKORC1 variants −1639G/A (3673), 1173C/T (6484), 1542G/C
(6853), and 2255C/T (7566) show strong linkage disequilibrium; they are inherited together.
Haplotypes with the variant SNPs require low-dose warfarin, are associated with lower
gene expression, and are called haplotype group A; the other haplotypes are considered
group B. Haplotypes that contain variations in the VKORC1 gene generally result in patients
who require high doses, BB (6.2 mg/day); moderate doses, AB (4.9 mg/day); or low doses,
AA (2.7 mg/day) to achieve anticoagulation, depending on the level of enzyme expression.
Recently, additional targets within the vitamin K complex have been associated with
warfarin dosing variations. Polymorphisms in gamma glutamyl carboxylase (GGCX), the
enzyme responsible for vitamin K-dependent γ-carboxylation, and calumenin (CALU), a
calcium-binding chaperone protein that inhibits GGCX, have been shown to influence
warfarin dose requirements. The GGCX rs11676382 C>G variant, located in intron 14, corre-
lated with lower warfarin doses in several studies involving Caucasian cohorts.60,61 The
CALU variant rs339097A>G is associated with a higher warfarin dose requirement in African
Americans.62
Asthma Therapy
Variability in the gene ALOX5 provides an example of a mutation in the promoter region
that affects the expression of the normal enzyme protein, namely 5-lipoxygenase.
5-Lipoxygenase is the enzyme responsible for conversion of arachidonic acid to leukotri-
enes. Leukotrienes mediate inflammation, vasoconstriction, and bronchoconstriction, and
they have been implicated in the pathophysiology of asthma.63 Zileuton, an inhibitor of
5-lipoxygenase, can be used to decrease airway inflammation in patients with asthma.
However, not all patients respond to this drug. The promoter region of the ALOX5 gene
shows variation in the number of tandem Sp1 binding motifs (5′GGGCGG3′).64 Transcription
factors SP1 and Egr-1 bind to this sequence and up-regulate ALOX5 transcription. Genetic
variants in the promoter region may change the binding of these transcription factors and,
therefore, the rate of 5-lipoxygenase transcription and the activation of leukotrienes under
inflammatory conditions.
The most common or wild-type allele (frequency of 77%) has five repeats of the
tandem Sp1 binding motif. Several mutant alleles contain three, four, or six tandem
Sp1-binding motifs. These variants were associated with reduced transcription of the
5-lipoxygenase gene, as compared with the common allele.65 Patients with at least one
copy of the wild-type allele responded to therapy, but asthma patients who were homo-
zygous for the variant alleles had a decreased response to treatment with antileukotriene
drugs. This indicates a pharmacogenetic effect of the promoter sequence on response to
treatment.
ACE Inhibitors
The association between polymorphisms in the angiotensin-converting enzyme (ACE) and
the antihypertensive effects of ACE inhibitors has been extensively studied.66,67 The I/D
variant in intron 16 is characterized by the absence or presence of 287 nucleotides.68 The
DD genotype has been associated with elevated levels of serum ACE.68 Variations in clini-
cal effects associated with ACE inhibition polymorphisms include blood pressure reduc-
tion, left ventricular hypertrophy, expression of angiotensin II type 1 receptor (AT1R) mRNA,
arteriole stiffness, heart rate, and renoprotection.67 Carriers of the I allele treated with ACE
inhibitors have been associated with a reduced regression of left ventricular hypertrophy,
greater reduction of the glomerular filtration, and decreased expression of AT1R mRNA as
well as a greater reduction of diastolic blood pressure with AT1R antagonists.
Conflicting results have been seen regarding the influence of ACE inhibitors and the
I/D polymorphism on reduction in blood pressure with some results showing no relation-
ship, and others showing an increased reduction in blood pressure with either the I or the
D allele. Ethnic diversity might explain the variable results that were observed. If an allele
is more prevalent, it may be easier to find a relationship between drug response and poly-
morphism. The I allele has a higher frequency in the Asian (62%) and the African-American
(60%) populations compared to the Caucasian population (50%).67,69 A reduced response
to ACE inhibitors and AT1R antagonists among hypertensive African-American patients is
well documented.70 However, the relationship between the ACE I/D polymorphism, ethnic-
ity, and response to ACE inhibitors remains unclear.
Inflammation
Cyclooxygenase (COX) 1 and 2 are important enzymes in the conversion of arachidonic acid
to prostaglandins. Prostaglandins have an important role in mediation of inflammation and
pain. Lee et al. examined the role of haplotypes of the COX1 (PTGS1) and COX2 (PTGS2)
genes on enzyme expression and response to postsurgical pain relief of ibuprofen, a
nonselective COX inhibitor, and rofecoxib, a selective COX-2 inhibitor.71 A PTGS2 haplotype
block with four promoter and intron SNPs (−1290A/G,-765G/C, 3629G/A, 4068T/C) caused
a differential response to the COX inhibitors. The homozygous major haplotype showed
increased COX-2 expression in response to surgery and increased pain relief with rofecoxib,
while the homozygous minor and the heterozygous haplotype showed decreased COX-2
expression and a better response to the nonselective inhibitor ibuprofen. This suggests that
postsurgical pain was mediated by COX-2 in the patients homozygous for the major PTGS2
haplotype and that postsurgical pain in other patient populations was predominantly
mediated by COX-1. This provides an example of haplotype variants resulting in different
enzyme activity and, thus, inhibiting drug response.
POLYMORPHISMS THAT INDIRECTLY MODULATE DRUG

RESPONSE
Pharmacodynamic-based genetic variability can produce undesired or unexpected
changes in drug response that are not directly due to changes in the drug target. Often
these genetic variants result in ADRs, but they can also result in protection from ADRs
or modification of drug efficacy. The polymorphism may result in a patient who is more
susceptible to an adverse effect, because this variant decreases his or her capacity to
handle the stress of the drug therapy. This is illustrated by the decreased capacity to
handle oxidative stress in individuals with reduced G6PD activity. On the other hand, a
genetic variant might provide added protection in some individuals who have an increased
capacity to handle the drug-induced stress. Thus, pharmacogenetics may offer an explana-
tion for some idiosyncratic reactions.
Another type of indirect change in drug response can occur when genetic variation
results in the increased sensitivity of a nontarget drug receptor, enzyme, or signaling
pathway. Drug interactions with a nontarget protein can result in an increased side effect
in genetically sensitive individuals as seen with the increased nausea associated with the
SSRI, paroxetine. Examples of pharmacodynamic polymorphisms that indirectly affect drug
response are summarized in Table 6-2 and discussed in further detail below.
CLINICAL PEARL
Idiosyncratic reactions may result from pharmacodynamic-based genetic

variability.
Antineoplastic Response and Epigenetic Polymorphisms

Genetic differences may have indirect effects on drug response that are related to modi-
fication of the drug-target interactions. Epigenetics refers to changes in the phenotype or
protein expression caused by mechanisms other than changes in the DNA sequence. Post-
translational methylation variability has been seen in cancer cells. Increased or decreased
methylation can modulate expression of enzymes and cell-cycle regulators. This, in turn,
can impact prognosis or drug response. Improved antineoplastic response can be seen
with mutations in enzymes involved in tumor DNA repair. Methylguanine methyltransfer-
ase (MGMT) is a DNA repair enzyme. Increased activity of this enzyme can repair alkylated
tumor DNA, resulting in poor response to chemotherapeutic alkylating agents. Methyla-
Table 6-2
Genetic Variability That Modifies Drug Response
Mutation Physiological Response Affected Drug/Class
G6PD Redox state of red blood cells Drugs that cause oxidative
stress13
Factor V and prothrombin Coagulation Oral contraceptives75-79
HLA Hypersensitivity reaction Abacavir, carbamazepine,
ximelagatran80-84
Cardiac potassium and sodium QT-prolonging Drugs that block the potassium
channels channels10,86-92
Serotonin receptor Nausea SSRIs95-96
Dopamine receptor Tardive dyskinesia Antipsychotics98
Dopamine and serotonin Weight gain Antipsychotics
receptors
Bradykinin receptor Cough ACE inhibitors105
Apolipoproteins, cholesterol Dyslipidemia, lipodystrophy Anti-retroviral therapy109-112
transport proteins,
mitochondrial proteins,
resistin
Ryanodine receptor Malignant hyperthermia Volatile anesthetics and
succinylcholine113,114
ACE, angiotensin-converting enzyme; G6PD, glucose-6-phosphate dehydrogenase; HLA, human leukocytic antigen;
SSRIs, selective serotonin reuptake inhibitors.
tion of the promoter region of the MGMT gene has been shown to decrease expression,
decrease DNA repair, and improve response of gliomas to treatment with alkylating agents
carmustine or temozolomide.72,73
A more recent study has examined the epigenetics of the response of gastric cancer
to 5-fluorouracil-based drugs.74 Methylation of the promoter region of p16, a cell cycle
regulator gene, results in decreased expression of p16 and an increased response to the
5-fluorouracil analogs. The increased response to the 5-fluorouracil analogs is thought
to be due to disruption of the cell cycle resulting in an increased numbers of cells in the
S-phase, the phase most susceptible to antimetabolites. Profiling of gene methylation
patterns should, therefore, allow better decision-making in selecting antineoplastic drugs.
Increased Incidence of Adverse Drug Reactions

G6PD and Oxidative Stress
One of the earliest and most common examples of genetic polymorphisms resulting in
increased adverse drug reactions (ADRs) are the functional variants of G6PD.13 More than
400 million people carry polymorphic genes for this enzyme. Greater than 400 biochemi-
cal variants of this X-linked gene have been described, most of which result in defective
enzyme activity. G6PD is an enzyme that catalyzes the first step in the pentose phosphate
pathway. The product of this reaction is a reduced form of nicotinamide adenine dinucleo-
tide phosphate hydrogen (NADPH). NADPH is essential for the activity of enzymes that
counterbalance oxidative stress. Defective G6PD results in decreased tolerance to oxida-
tive stress. As a result, drugs that increase oxidative stress may precipitate hemolytic
anemia in the presence of this defective enzyme. Individuals with the variant G6PD have
experienced increased incidence of hemolytic anemia when treated with drugs like anti-
malarials, sulfonamides, sulfones, aspirin, ciprofloxacin, vitamin K analogs, chlorampheni-
col, and nitrofurantoin. The highest frequency G6PD variants are found in Africa, Asia, the
Mediterranean, and the Middle East.
Thromboembolic Disorders
It has long been known that some women taking oral contraceptives are susceptible to
thromboembolic disorders such as deep vein thrombosis (DVT) or pulmonary embolism.
Enhanced coagulation is thought to be related to elevation in prothrombin (factor II)
levels.75 Genetic variations in coagulation factors such as factor V and prothrombin may
contribute to an increase in this risk.76,77 Factor V is converted to the active form, factor Va,
by thrombin. Activated factor V serves as an essential protein in the coagulation pathway
and acts as a cofactor for the conversion of prothrombin to thrombin by factor Xa. Factor
Va is inactivated by activated protein C (APC). APC resistance was associated with a vari-
ant allele identified as 1691G/A, a SNP found in exon 10 of the factor V (F5) gene, which
results in a coding change Arg506Gln.78 The 1691A allele is often referred to as factor V
Leiden because it was characterized in a family in Belgium. Individuals with one or two
alleles are more susceptible to clot formation. Another important variant in the coagula-
tion pathway is the SNP found in the 3′ untranslated region of prothrombin, 20210G/A.
Carriers of the 20210A allele have higher plasma prothrombin levels than controls with
the normal 20210GG genotype and have a 2.8-fold increased risk of venous thrombosis.79
The use of oral contraceptives in patients with either the 1691A allele or the 20210A allele
results in a markedly higher incidence of DVTs because the combinations are synergistic
rather than additive.76,77
Hypersensitivity Reactions
Hypersensitivity reactions are a common ADR; in some cases, the allergic reaction can
be related to a gene variant of the immune system. The human leukocyte antigen (HLA)
system is the major histocompatibility complex in humans. Hypersensitivity reactions have
been associated with variant alleles of several HLA genes. About 5% of the patients taking
the nucleoside reverse transcriptase inhibitor (NRTI) abacavir experience a systemic hyper-
sensitivity reaction that includes gastrointestinal (GI) distress, rash, fever, and fatigue.80 The
HLA-B*57:01 and the HLA-DR7 genes were associated with the abacavir-related hypersen-
sitivity reaction.81,82 A more serious hypersensitivity reaction, Stevens-Johnson syndrome,
has occurred with the antiepileptic drug carbamazepine. This particular ADR has been
associated with HLA-B*15:02 in Han Chinese.83 Recently a thrombin inhibitor, ximelagatran,
was removed from the market because of hepatotoxicity; studies have revealed a strong
association between this hepatotoxicity and the HLA-DRB1*07:01 gene.84
QT-Interval Prolongation
Genetic variation in ion channel genes that are not drug targets can also have a role
in predisposing patients to toxic effects of drugs. Individuals with variant alleles in ion
channels may have substantial morbidity or mortality resulting from drugs that cause
QT-interval prolongation and may be inherently more susceptible to cardiac arrhythmias.
QT-prolongation is a common life-threatening adverse effect that has resulted in removal
of numerous drugs from the market.85 Polymorphisms in potassium or sodium channels
can allow accumulation of excess intracellular positive ions in cardiac cells, leading to
delayed repolarization or decrease in “repolarization reserve.”86 Drug-induced or acquired
QT-prolongation has been identified in individuals with the KCNQ1 and KCNE1 genes (IKs
potassium channel), HERG and KCNE2 genes (IKr potassium channel), and the SCN5A
gene (sodium channel) variants.87 Some anti-arrhythmic drugs are expected to affect the
QT-interval, but many noncardiovascular drugs also affect the QT-interval. The voltage-
gated potassium channel (IKr) involved in cardiac repolarization current appears to be the
main culprit in QT-prolongation–based arrhythmias known as torsades de pointes.10 Drugs
that block this potassium channel can precipitate cardiac arrhythmias; several examples are
listed in Table 6-3. Understanding the genetic basis for QT-prolongation–based arrhythmias
could help improve drug development as well as patient safety.
Serotonin Receptors and GI Side Effects
Drug side effects may be caused by a drug binding to ancillary receptors that are not the
primary therapeutic target. A significant number of patients taking SSRIs like paroxetine
(Paxil®) experience nausea as an undesired side effect. Even though paroxetine is believed
to exert its antidepressant effects by inhibiting SLC6A4, it is thought that activation of the
5-hydroxytryptamine 3B or 2A receptors (5-HT3B or 5-HT2A) mediates the SSRI-induced
GI side effects. The 5-HT3B receptors located in the small intestinal mucosa and the
chemotrigger zone in the CNS mediate the vomiting reflex, and the 5-HT2A receptor affects
gut motility.93 Studies using 5-HT3 antagonists, cisapride and ondansetron, showed a
reduction in SSRI-induced GI side effects, a finding that suggests that 5-HT3 receptors are
involved in these GI effects.94 The Tyr129Ser polymorphism of the HTR3B gene has been
shown to significantly affect the incidence of nausea in a population of Japanese psychi-
atric patients treated with paroxetine. Patients with the Tyr/Tyr genotype had a fourfold
higher risk of developing nausea compared to patients with the Ser allele. There was no
correlation with the incidence of nausea and HTR3A gene or the CYP2D6 gene polymor-
Table 6-3
Noncardiac Drugs That Bind to Cardiac Potassium Channels
Drug Class Drug Target Channel Mutation Reference
Clarithromycin Antibiotic IKr blocker MiRP1(KCNE2) a

Q9E 88
subunit of the IKr –
potassium channel
Sulfamethoxazole Antibiotic IKr blocker MiRP1(KCNE2)a T8A 89
subunit of the IKr –
potassium channel
Cisapride Prokinetic agent IKr blocker SCN5A sodium L1825P 90
channel
Terfenadine Antihistamine IKr blocker KCNH2 91
Sibutramine SSRI IKr blocker KCNQ1 T265I 92
aMinK-related peptide 1-MiRP1 is also called KCNE2, subunit of I , inward rectifying potassium channel.
Kr
SSRI, selective serotonin reuptake inhibitor.
phisms.95 Murphy et al. showed a strong correlation between polymorphisms in the 5-HT2A
receptor and side effect severity in patients treated with paroxetine.96
The HTR2A 102T/C SNP showed a frequency of 0.6:0.4, T:C, in a U.S. Caucasian
population. Patients with the CC genotype showed a significantly higher incidence of GI
complaints and discontinuation rate. The increased incidence of GI side effects associ-
ated with genetic variations in serotonin receptors provides an example of a shift in drug
response due to a polymorphism in a secondary target.
Tardive Dyskinesia
Another example of the pharmacogenetics of a secondary target modifying drug response
can be seen in the association of dopamine three (D3) receptor polymorphisms with an
increased incidence of tardive dyskinesia (TD). TD is a serious, irreversible movement disor-
der that affects at least 20% of the patients taking traditional antipsychotic agents.97 Many
typical antipsychotics are thought to work through their antagonism of dopamine two (D2)
receptors; however, these drugs can also antagonize other dopamine receptors including
the D3 receptor. Numerous studies have identified a variant of the D3 receptor gene (DRD3
Ser9Gly, rs6280) in patients who develop TD. Patients homozygous for the DRD3Gly allele
show an increased incidence of TD, and in vitro studies show higher affinity of dopamine for
cells expressing the variant gene product.98 Recent meta-analysis has produced conflicting
results possibly due to increased use of second-generation antipsychotics, which have less
activity at the D3 receptors.51,99,100
Weight Gain
The metabolic adverse effect of weight gain seen with numerous antipsychotics is asso-
ciated with a number of SNPs. Dopamine and serotonin receptors associated with anti-
psychotic effects are the best candidates for mutations resulting in an increased weight
gain.101,102 The Taq 1A (rs1800497) polymorphism decreases the expression of the D2 recep-
tor (DRD2). Decreased DRD2 receptors found in carriers of the 1A allele have been associ-
ated with a decreased response to the antipsychotic and an increased incidence of weight
gain.101 Antagonism of the 5-HTC2 receptor has been implicated in weight gain associated
with some antipsychotics. An increased incidence of weight gain is associated with a poly-
morphism in the promoter region of the HTRC2 gene (−759 C/T, rs3813929). This SNP is
believed to be in linkage disequilibrium with several other variants.102
The −759C/T polymorphism appears to disrupt DNA-protein interactions that decreas-
es HTRC2 expression, resulting in increased antipsychotic-induced weight gain.103

Bradykinin-Mediated Cough
Although the primary therapeutic effect of ACE inhibitors results from blocking the
conversion of angiotensin I to angiotensin II, ACE inhibitors also block the breakdown
of bradykinin. Bradykinin is thought to mediate some of the adverse effects of the ACE
inhibitors, including the dry persistent cough seen in about 10% of the patients taking ACE
inhibitors.104 Variants in the bradykinin B2 receptor have been implicated in a predisposi-
tion for this ADR. Mukae et al. looked at the incidence of cough and genetic polymorphism
of genes for ACE, angiotensin II type I and II receptor, and the bradykinin B2 receptor in a
hypertensive Japanese population.105 Patients with one or two copies of the −58T allele of
the bradykinin B2 receptor have a higher incidence of cough compared to patients with the
−58CC genotype. This promoter region variant resulted in higher expression of the brady-
kinin B2 receptor protein.
Anti-Retroviral Therapy and Dyslipidemia
Anti-retroviral therapy (ART) can cause elevations in serum triglycerides, decreases in HDL,
insulin resistance, and lipodystrophy, which is the loss of peripheral adipose tissue with a
concomitant increase in abdominal adiposity.106 These effects on lipids are independent
of the drug target used to suppress HIV proliferation. Protease inhibitors and NRTIs have
been shown to alter the expression of adipogenic transcription factors, genes involved in
lipid metabolism, and cell cycle control.107 NRTIs deplete adipocyte mitochondrial DNA
(mtDNA) with an associated decrease in mtDNA copy number, and they stimulate mito-
chondrial proliferation, fat wasting, and adipocyte loss.108 Not all patients experience the
same degree of dyslipidemia and lipodystrophy with ART, suggesting a genetic component.
Susceptibility to ART lipid and mitochondrial disturbances shows a complex relation-
ship with a number of genetic polymorphisms; patient populations with compromised lipid
metabolism and adipocyte mitochondrial function may be more sensitive to these adverse
effects.109 Variants of apolipoproteins APOA5 and APOC3, interacting with APOE geno-
types, have been associated with an increase in severity of ART-induced dyslipidemia and
lipodystrophy. Genetic polymorphisms of the nuclear transcription-factor sterol response
element binding proteins (SREBP1c) and of tumor necrosis factor-alpha (TNFα) genes,
which can also affect lipid metabolism, have yielded contrasting results in patients treated
with ART. In addition to supporting the role of APOA5, APOC3, and APOE polymorphisms,
Arnedo et al. demonstrated that variant alleles of the genes for ATP-binding cassette
(ABC) protein cholesterol transporter (ABCA1) and cholesteryl ester transfer protein (CETP)
also contributed to elevation of plasma triglyceride and decreased HDL-cholesterol levels
related to ART exposure.110

Recently, other allele variants have been associated with ART-induced lipidemias. The
hemochromatosis gene (HFE) is known to impact mitochondrial function, Hulgen et al.
found that a specific variant in the HFE gene, HFE 187C/G, was protective against fat loss in
patients treated with ART.111 An extensive evaluation of genes in patients treated with ART
showed a strong association of ART-induced metabolic complications with a genetic varia-

tion in resistin, a gene previously implicated in obesity and insulin resistance.112 Although
the genetic variability associated with ART-induced adverse lipid metabolism remains
unclear, identification of all predisposing factors remains an important goal in optimizing
therapy.
Ryanodine Receptor and Malignant Hyperthermia

Certain patients given volatile anesthetics like halothane or succinylcholine may develop
malignant hyperthermia (MH). MH is a syndrome associated with rigidity, muscle break-
down, respiratory acidosis, hyperthermia, tachycardia, and arrhythmias.113 This hypermeta-
bolic state is associated with elevated cytosolic Ca++ and often related to a defective
ryanodine receptor (RyR1) located on calcium channels in the sarcoplasmic reticulum
(SR). Volatile anesthetics can cause uncontrolled release of intracellular Ca++ from the SR;
evidence indicates that this can explain the signs and symptoms of MH. Seventy percent of
the families susceptible to MH possess one of 34 RYR1 alleles.114 Genes for other proteins
associated with the RyR1 protein are also being examined.113
Disease States and Biomarkers

Certain disease states have been associated with a genetic variant, and often this same
variant has been associated with altered efficacy to a particular drug therapy. A genetic
variant that is associated with a disease or a particular subpopulation of that disease is
termed a biomarker. The relationship of the biomarker to therapeutic outcome may be
directly related to a genetic variant in the drug target, indirectly through changes in the
ability of the drug to reach the drug target, or due to the linkage of this variant to other
pathological changes. Below are several examples that can be used to illustrate the direct
association between a disease biomarker and the therapeutic outcome based on a specific
drug target.
Alzheimer’s Disease
Alzheimer’s disease has been associated with a genetic polymorphism of APOE, a protein
involved in the expression of choline acetyltransferase.115 Choline acetyltransferase is
required for synthesis of acetylcholine. The APOE4 allele (Cys112Arg) is associated with
a poor prognosis and a decreased response to tacrine (Cognex), an acetylcholinesterase
inhibitor that is used to increase endogenous acetylcholine levels.11 A beneficial response to
tacrine and a better prognosis is seen in individuals with the APOE2 and APOE3 alleles. This
differential response seems to be more important in women than men.116 Two other acetyl-
cholinesterase inhibitors, rivastigmine (Aricept) and galantamine (Razadyne), also show
increased response rates with the APOE2/3 allele compared to patients with the APOE4
allele.117,118 It is thought that the number of functional presynaptic cholinergic fibers in the
brains of patients with Alzheimer’s disease carrying the APOE4 allele may be too low to
benefit from the neurochemical and therapeutic action of acetylcholinesterase inhibitors.119
Cystic Fibrosis
Cystic fibrosis (CF) is a disease characterized by defective mucus production. The disease
is typically associated with a defective protein called CF transmembrane conductance
regulator (CFTR).120 CFTR is a critical regulator of fluid and ion transport that is responsible
for transport of chloride and bicarbonate ions. In CF, various autosomal recessive vari-
ants in CFTR result in decreased biosynthesis, processing, and breakdown of the protein
or nonfunctional/dysfunctional chloride conduction. Ivacaftor was recently developed to

enhance CFTR-mediated chloride transport and, thus, it seems to alter the disease’s course
and increase length of survival.121 Although the exact mechanism of ivacaftor is unclear, it is
thought to increase the length of time the chloride channel is open and increase chloride
secretion. Ivacaftor only works in a patient with one allele that produces a semi-functional
CFTR because it potentiates the channel’s activity.

Ivacaftor was initially approved for use in patients with one or two G551D-CFTR vari-
ant proteins, rs75527207 genotype AA or AG.122 This mutation is associated with a gating
or reduced chloride secretion defect. The U.S. Food and Drug Administration (FDA) recently
approved the use of ivacaftor in patients with one allele of additional CFTR variants with
gating mutations: G1244E (rs267606723 genotype AA or AG), G1349D (rs193922525 geno-
type AA or AG), G178R (rs80282562 genotype AA or AG), G551S (rs121909013 genotype AA
or AG), S1251N (rs74503330 genotype AA or AG), S1255P (rs121909041 genotype CC or CT),
S549N (rs121908755 genotype AA or AG), and S549R (rs121909005 genotype GG or GT,
rs121908757 genotype CC or CA).123
Cancer
Somatic or acquired mutations in tumor cell DNA can produce individual variabilities
in cancer growth rates and response to cancer treatment. Growth factor biomark-
ers have been used to predict therapeutic outcomes and design-specific drug entities.
Activation of the EGFR (epidermal growth factor receptor), also called human epidermal
growth factor receptor-1 (HER-1), or other ErbB family members promote cell growth and
survival. In addition to EGFR, mutated ErbB receptors found in tumor cells include HER-2
and HER-3, vascular endothelial growth factor receptor (VEGFR), fibroblast growth factor
receptor (FGFR), and platelet-derived growth factor receptor (PDGFR). Receptor function
is often dysregulated in solid tumors due to specific mutations. Excessive EGFR activity
may result from receptor overexpression, overproduction of receptor ligands (transforming
growth factor alpha-TGF-α or epidermal growth factor-EGF), mutations within the receptor,
or mutations in the signal transduction cascade or second-messenger pathway.
Pharmacogenetics of cancer treatment based on targeting growth factor receptors has
been reviewed, and molecularly targeted therapy presents a very complicated and promis-
ing picture.12,124-127 A diagram of the ErbB-type receptor is presented in Figure 6-4. Dimer-
ization of ErbB receptors or EGFR-containing tyrosine kinase results in phosphorylation of
intracellular tyrosine residues on the dimerized receptor. This phosphorylation initiates the
enrollment of other cytosolic proteins that ultimately affect gene expression and various
biological outcomes. Numerous cancers including breast, ovarian, glioma, colon, bladder,
renal cell carcinoma, and non-small cell lung cancer and squamous-cell carcinomas of the
head and neck have been associated with excess activity due to overexpression or mutation
of these receptors.128 Not only is the development of cancer and its prognosis associated
with these genetic mutations, but these mutations provide a prime target for drugs directed
specifically at the tumor cells.
Small molecule tyrosine kinase inhibitors (TKIs) and monoclonal antibodies (MABs)
have provided the main therapeutic options for targeting the ErbB receptors and new
approaches, including the use of second- and third-generation TKIs and MABs, as well as
antibody-based immunoconjugates, antisense oligonucleotides, and other novel agents
expanding the treatment options.127 Strategic combinations of these medications have
Dimerization
Phosphorylation
Tyrosine Kinase
PI3K RAS JAK
AKT RAF STAT

Pharmacogenetic target
MEK (altered protein can
mTOR influence drug response)
ERK Cancer drug targets
Proliferation, apoptosis & other biological outcomes
FIGURE 6-4 Somatic mutations in tumor cells can result in excess activity of the ErbB
receptors. These receptor mutations have provided unique targets for cancer drugs, including
monoclonal antibodies to the receptor and tyrosine kinase inhibitors. Conversely, mutations
in other genes (i.e., genes for downstream second-messengers such as the RAF gene)
have resulted in resistance to these cancer drugs. Thus, new drug targets include second-
messenger protein kinases resulting from these mutant genes.
resulted in better outcomes for certain populations. A summary of these types of drugs
can be found in Table 6-4.
Inhibition of growth factor-mediated gene expression by these drugs results in
decreased tumor cell growth and reduced angiogenesis, which may result in tumor cell
death and improved prognosis. Cancer cells that do not contain these somatic mutations
will not respond to these drugs. Normal cells that do not contain these receptors will
likewise not be affected, thus avoiding many of the typical cytotoxic side effects associ-
ated with cancer chemotherapy. Growth factor receptor inhibitors may not work if addi-
tional mutations are present downstream from the receptor in the signaling cascade. For
example, mutations that result in overexpression of K-ras, a small G-protein downstream
of EGFR, can overcome the positive effects of cetuximab (an EGFR antibody) in treatment
of advanced colorectal cancer.129
Genetic testing for the mutations associated with activated growth factor receptors is
crucial for identifying the patient population that may benefit from a specific treatment.
Trastuzumab is an example of a drug that was approved by the FDA with a required test,
companion diagnostic, to identify overexpression of HER2 protein, a member of the ErbB
Table 6-4
Selected Examples of Growth Factor Inhibitors in Cancer
Drug Type Target Cancer Reference
trastuzumab Monoclonal HER2 Breast 130, 132

Herceptin antibody
ado-trastuzumab conjugated
emtansine with cytotoxic
Kadcyla emtansine
pertuzumab Monoclonal HER2-dimerization Breast 133
Perjeta antibody inhibitor
cetuximab Monoclonal EGFR Colorectal, SCCHN 129, 134
Erbitux antibody
necitumumab
Portrazza
bevacizumab Monoclonal VEGFR Colorectal, NSCLC, 135
Avastin antibody cervical,
glioblastoma
ramucirumab Monoclonal VEGFR2 Gastric adenocarci- 136
Cyramza antibody noma, NSCLC
panitumumab Monoclonal EGFR Colorectal 12
Vectibix antibody
imatinib Tyrosine kinase BCR-ABL, multiple Chronic and acute 137
Gleevec inhibitor kinases myelogenous
leukemia, GIST
(kit+)
gefitinib Tyrosine kinase ErbB1 or EGFR NSCLC 138, 139
Iressa inhibitor
erlotinib
Tarceva
afatinib Second-generation EGFR with exon Metastatic NSCLC, 140-142
Gilotrif tyrosine kinase 19 deletions or breast
lapatinib inhibitor exon 21 (L858R)
Tykerb/Tyverb HER2
osimertinib Third-generation EGFR with T790M NSCLC 143
Tagrisso tyrosine kinase mutation
inhibitor
sorafenib Tyrosine kinase VEGFR, multiple RCC, hepatocellular 135
Nexavar inhibitor kinases carcinoma,
differentiated
thyroid cancer
sunitinib Tyrosine kinase VEGFR, multiple RCC, GIST (c-kit) 135, 139
Sutent inhibitor kinases
cabozantinib Tyrosine kinase VEGFR2, Medullary thyroid 144
Cometriq inhibitor c-MET, cancer
RET-tyrosine
kinase
vandetanib Multi-kinase VEGFR, EGFR, Medullary thyroid 145
Caprelsa inhibitor RET-tyrosine cancer
kinase
lenvatinib Multi-kinase VEGFR2 Differentiated 145
Lenvima inhibitor VEGFR3 Thyroid cancer

Selected Examples of Growth Factor Inhibitors in Cancer
Drug Type Target Cancer Reference
nintedanib ofev, Tyrosine kinase VEGFR, FGFR, Idiopathic pulmo- 146

Vargatef inhibitor PDGFR nary fibrosis,
NSCLC
pazopanib Tyrosine kinase VEGFR, FGFR, RCC, soft tissue 147
Votrient inhibitor PDGFR, c-kit sarcoma
temsirolimus M-TOR inhibitor M-TOR RCC 148
Toresel
everolimus M-TOR inhibitor M-TOR RCC, breast cancer, 149
Zortress, Afinitor neuroendocrine
tumors, renal
angiomyolipoma
with tuberous
sclerosis
complex
vemurafenib Serine/threonine- B-raf Metastatic melano- 150, 151
Zelboraf protein kinase ma with BRAF
dabrafenib inhibitor V600 mutation
Tafinlar
trametinib Tyrosine kinase MEK Metastatic melano- 151, 152
Mekinist inhibitor ma with BRAF
cobimetinib V600 mutation
Cotellic
crizotinib ALK inhibitor EML4-ALK NSCLC 153-155
Xalkori
ceritinib
Zykadia
alectinib
Alecensa
BCR-ABL, breakpoint cluster region-Abelson; B-raf, serine/threonine-specific protein kinase related to retroviral
oncogene RAF (rapidly accelerated fibrosarcoma); EGFR, epidermal growth factor receptor; EML4-ALK, echinoderm
microtubule–associated protein-like 4–anaplastic lymphoma kinase; ErbB1,erythroblastic leukemia viral oncogene
homolog 1; ErbB2, erythroblastic leukemia viral oncogene homolog 2; FGFR, fibroblast growth factor receptor; GIST,
gastrointestinal stromal tumor; HER2, human epidermal growth factor receptor 2; MEK, mitogen-activated protein
kinase; M-TOR, mammalian target of rapamycin; NSCLC, non-small cell lung cancer; PDGFR, platelet-derived
growth factor receptor; RCC, renal cell carcinoma; SCCHN, squamous-cell carcinoma of the head and neck; and
VEGFR, vascular endothelial growth factor receptor.
family.130 More than a dozen companion diagnostic devices used to detect growth factor
mutations are currently approved by the FDA.131 Additional genes and specific mutations
have been identified, which will allow better management of certain cancers either for
initiation of therapy or for follow-up therapy as drug resistance develops in the tumor
cell population. Better understanding of the various mutations found in tumors, enhanced
genetic testing, and development of treatment algorithms will significantly improve cancer
treatment.
SUMMARY
Our understanding of pharmacodynamic-based genetic variation is still in its early
stages. Haplotype and genome- wide studies have provided better indicators of the
importance of specific polymorphisms in drug response. The complexity of drug targets
and their interactions with other proteins have made it difficult to translate the infor-
mation regarding individual genetic variants into useful therapeutic strategies. On the
other hand, pharmacodynamic genetic polymorphisms can provide a better explanation for
idiosyncratic ADRs. Better designed studies and more extensive genetic mapping should
allow improved prediction of optimal drug response and avoidance of ADRs.
REFERENCES
1. Evans WE, Johnson JA. Pharmacogenomics: the inherited basis for interindividual differences in drug
response. Annu Rev Genomics Hum Genet. 2001;2:9-39.
2. Evans WE, McLeod HL. Pharmacogenomics: drug disposition, drug targets, and side effects. N Engl J
Med. 2003;348:538-549.
4. Michaud V, Vanier MC, Brouillette D, et al. Combination of phenotypic assessments and CYP2C9
VKORC1 polymorphisms in the determination of warfarin dose requirements in heavily medicated
patients. Clin Pharmacol Ther. 2007;83:740-748.
5. Sachse C, Brockmöller J, Bauer S, et al. Cytochrome P450 2D6 variants in a Caucasian population:
allele frequencies and phenotypic consequences. Am J Hum Genet. 1997;60:284-295.
6. Rau T, Wuttke H, Michels LM, et al. Impact of the CYP2D6 genotype on the clinical effects of meto-
prolol: a prospective longitudinal study. Clin Pharmacol Ther. 2009;85:269-272.
7. Liu J, Liu ZQ, Tan ZR, et al. Beta1-adrenergic receptor polymorphisms influence the response to meto-
prolol monotherapy in patients with essential hypertension. Clin. Pharmacol Ther. 2006;80:23-32.
8. Reihsaus E, Innis M, MacIntyre N, et al. Mutations in the gene encoding for the beta(2)-adrenergic
receptor in normal and asthmatic subjects. Am J Resp Cell Molec Biol. 1993;8:334-339.
9. Wong M-L, Licinio J. From monoamines to genomic targets: a paradigm shift for drug discovery in
depression. Nat Rev Drug Discov. 2004;3:136-151.
10. Roepke TK, Abbott GW. Pharmacogenetics and cardiac ion channels. Vasc Pharmacol. 2006;44:90-106.
11. Poirier J, Delisle MC, Quirion R, et al. Apolipoprotein ε4 allele as a predictor of cholinergic deficits and
treatment outcome in Alzheimer disease. Proc Natl Acad Sci USA. 1995;92:12260-12264.
12. Krejsa C, Rogge M, Sadee W. Protein therapeutics: new applications for pharmacogenetics. Nat Rev
Drug Discov. 2006;5:507-521.
13. Cappellini MD, Fiorelli G. Glucose-6-phosphate dehydrogenase deficiency. Lancet. 2008;371:64-74.
14. Contopoulos-Ioannidis DG, Kouri I, Ioannidis JP. Pharmacogenetics of the response to beta2 agonist
drugs: a systematic overview of the field. Pharmacogenomics. 2007;8:933-958.
15. Dishy V, Gbenga MD, Sofowora, G. The effect of common polymorphisms of the ß–adrenergic receptor
on agonist-mediated vascular desensitization. N Engl J Med. 2001;345:1030-1035.
16. Drysdale C, McGraw DW, Stack C, et al. Complex promoter and coding region ß2-adrenergic recep-
tor haplotypes alter receptor expression and predict in vivo responsiveness. Proc Natl Acad Sci USA.
2000;97:10483-10488.
17. Dewar JC, Wheatley AP, Venn A, et al. ß2-adrenoceptor polymorphisms are in linkage disequilibrium
but are not associated with asthma in an adult population. Clin Exp Allergy. 1998;28:442-448.
18. Israel E, Chinchilli VM, Ford JG, et al. Use of regularly scheduled albuterol treatment in asthma: geno-
type-stratified, randomised, placebo-controlled cross-over trial. Lancet. 2004;364:1505-1512.
19. Green SA, Turki J, Innis M, et al. Amino-terminal polymorphisms of the human ß2-adrenergic receptor
imparts distinct agonist-promoted regulatory properties. Biochemistry. 1994;33:9414-9419.
20. Israel E, Drazen JM, Liggett SB, et al. The effect of polymorphisms of the ß2-adrenergic receptor on
the response to regular use of albuterol in asthma. Am J Respir Crit Care Med. 2000;162:75-80.
21. Taylor DR, Drazen JM, Herbison GP, et al. Asthma exacerbations during long term ß agonist use: influ-
ence of ß(2)-adrenoceptor polymorphism. Thorax. 2000;55:762-767.
22. Bleecker ER, Postma DS, Lawrance RM, et al. Effect of ADRB2 polymorphisms on response to long
acting ß2-agonist therapy: a pharmacogenetic analysis of two randomised studies. Lancet. 2007;
370:2118-2125.
23. Green SA, Cole G, Jacinto M, et al. A polymorphism of the human ß2-adrenergic receptor within the
fourth transmembrane domain alters ligand binding and functional properties of the receptor. J Biol
Chem. 1993;268:23116-23121.
24. Dishy V, Landau R, Sofowora GG, et al. Beta-2-adrenoceptor Thr164Ile polymorphism is associated
with markedly decreased vasodilator and increased vasoconstrictor sensitivity in vivo. Pharmaco-
genetics. 2004;14:517-522.
25. Liggett SB. ß-Adrenergic receptor pharmacogenetics. Am J Respir Crit Care Med. 2000;161:S197-S201.
26. Johnson JA, Shin J. Pharmacogenetics of ß-blockers. Pharmacotherapy. 2007;27:874-887.
27. Rathz DA, Brown KM, Kramer LA, et al. Amino acid 49 polymorphisms of the human beta1-adrenergic
receptor affect agonist promoted trafficking. J Cardiovasc Pharmacol. 2002;39:155-160.
28. Mason DA, Moore JD, Green SA, et al. A gain-of-function polymorphism in a G-protein coupling
domain of the human beta1-adrenergic receptor. J Biol Chem. 1999;274:12670-12674.
29. Johnson JA, Zineh I, Puckett BJ, et al. ß1-Adrenergic receptor polymorphisms and antihypertensive
response to metoprolol. Clin Pharmacol Ther. 2003;74:44-52.
30. Beitelshees AL, Zineh I, Yarandi HN, et al. Influence of phenotype and pharmacokinetics on beta
blocker drug target pharmacogenetics. Pharmacogenomics J. 2006;6:174-178.
31. Mialet-Perez, J. Rathz DA, Petrashevskaya NN, et al. Beta 1-adrenergic receptor polymorphisms confer
differential function and predisposition to heart failure. Nat Med. 2003;9:1300-1305.
32. Filigheddu F, Reid JE, Troffa C, et al. Genetic polymorphisms of the ß-adrenergic system: association
with essential hypertension and response to ß-blockade. Pharmacogenomics J. 2004;4:154-160.
33. Jia H, Hingorani AD, Sharma P, et al. Association of the G(s) alpha gene with essential hypertension
and response to beta-blockade. Hypertension. 1999;34:8-14.
34. Reiche EMV, Bonametti AM, Voltarelli JC, et al. Genetic polymorphisms in the chemokine and chemo-
kine receptors: impact on clinical course and therapy of the human immunodeficiency virus type 1
infection (HIV-1). Curr Med Chem. 2007;14:1325-1334.
35. Samson M, Libert F, Doranz BJ, et al. Resistance to HIV-1 infection in Caucasian individuals bearing
mutant alleles of the CCR-5 chemokine receptor gene. Nature. 1996;382:722-725.
36. Martinson JJ, Chapman NH, Rees DC, et al. Global distribution of the CCR5 gene 32-basepair deletion.
Nat Genet. 1997;16:100-103.
37. Raphael D. A new class of anti-HIV therapy and new challenges. N Engl J Med. 2008;359:1510-1511.
38. Gulick RM, Lalezari J, Goodrich J, et al. Maraviroc for previously treated patients with R5 HIV-1 infec-
tion. N Engl J Med. 2008;359:1429-1441.
39. Ketas TJ, Kuhmann SE, Palmer A, et al. Cell surface expression of CCR5 and other host factors influ-
ence the inhibition of HIV-1 infection of human lymphocytes by CCR5 ligands. Virology. 2007;364:281-
290.
40. Lotsch J, Geisslinger G. Are mu-opioid receptor polymorphisms important for clinical opioid therapy?
Trends Mol Med. 2005;11:82-89.
41. Zhang Y, Wang D, Johnson AD, et al. Allelic expression imbalance of human mu receptor (OPRM1)
caused by variant A118G. J Biol Chem. 2006;280:32618-32624.
42. Campa D, Gioia A, Tomei A, et al. Association of ABCB1/MDR1 and OPRM1 gene polymorphisms with
morphine pain relief. Clin Pharmacol Ther. 2008;83:559-566.
43. Anton RF, Drobes DJ, Voronin K, et al. Naltrexone effects on alcohol consumption in a clinical labora-
tory paradigm: temporal effects of drinking. Psychopharmacology. 2004;173:32-40.
44. Oslin DW, Berrettini W, Kranzler HR, et al. A functional polymorphism on the mu-opioid receptor gene
is associated with naltrexone response in alcohol-independent patients. Neuropsychopharmacology.
2003;28:1546-1552.
45. Kirchheiner J, Nickchen K, Bauer M, et al. Pharmacogenetics of antidepressants and antipsychotics: the
contribution of allelic variations to the phenotype of drug response. Mol Psychiatry. 2004;9:442-473.
46. Smeraldi E, Zanardi R, Benedetti F, et al. Polymorphism within the promoter of the serotonin trans-
porter gene and antidepressant efficacy of fluvoxamine. Mol Psychiatry. 1998;3:508-511.
47. Lesch KP, Bengel D, Heils A, et al. Association of anxiety-related traits with a polymorphism in the
serotonin transporter gene regulatory region. Science. 1996;274:1527-1531.
48. Kim DK, Lim SW, Lee S, et al. Serotonin transporter gene polymorphism and antidepressant response.
Neuroreport. 2000;11:215-219.
49. Kunugi H, Hattori M, Kato T, et al. Serotonin transporter gene polymorphisms: ethnic difference and
possible association with bipolar affective disorder. Mol Psychiatry. 1997;2:457-462.
50. Hu XZ, Lipsky RH, Zhu G, et al. Serotonin transporter promoter gain-of-function genotypes are linked
to obsessive-compulsive disorder. Am J Hum Genet. 2006;78:815-826.
51. Reynolds GP, McGowan OO, Dalton CF. Pharmacogenomics in psychiatry: the relevance of receptor
and transporter polymorphisms. British J Clin Pharmacol. 2014;77:654-672.
52. Minov C, Baghai TC, Schule C, et al. Serotonin-2A-receptor and –transporter polymorphisms: lack of
association in patients with major depression. Neurosci Lett. 2001;303:119-122.
53. Lemonde S, Turecki G, Bakish D, et al. Impaired repression at a 5-hydroxytryptamine 1A receptor gene
polymorphism associated with major depression and suicide. J Neurosci. 2003;23:8788-8799.
54. Lemonde S, Du L, Bakish D, et al. Association of the C(-1019)G 5-HT1A functional promoter polymor-
phism with antidepressant response. Int J Neuropsychopharmacol. 2004;7:501-506.
55. Reynolds GP, Arranz B, Templeman LA, et al. Effect of 5-HT1A receptor gene polymorphism on nega-
tive and depressive symptom response to antipsychotic treatment of drug-naive psychotic patients.
Am J Psychiatry. 2006;163:1826-1829.
56. Gerhard M, Ganz P. How do we explain the clinical benefits of estrogen? From bedside to bench.
Circulation. 1995;92:5-8.
57. Herrington DM, Howard TD, Hawkins GA, et al. Estrogen-receptor polymorphisms and effects of estro-
gen replacement on high-density lipoprotein cholesterol in women with coronary disease. N Engl J
Med. 2002;346:967-974.
58. Ntukidem NI, Nguyen AT, Stearns V, et al. Estrogen receptor genotypes, menopausal status, and the
lipid effects of tamoxifen. Clin Pharmacol Ther. 2008;83:702-710.
warfarin dose. N Engl J Med. 2005;52:2285-2293.
60. Rieder MJ, Reiner AP, Rettie AE. Gamma-glutamyl carboxylase (GGCX) tagSNPs have limited utility for
predicting warfarin maintenance dose. J Thromb Haemost; 2007;5:2227-2234.
61. King CR, Deych E, Milligan P, et al. Gamma-glutamyl carboxylase and its influence on warfarin dose.
Thromb Haemost. 2010;104:750-754.
62. Voora D, Koboldt DC, King CR, et al. A polymorphism in the VKORC1-regulator calumenin predicts
higher warfarin doses in African-Americans. Clin Pharmacol Ther. 2010;87:445-451.
63. Drazen JM, Israel E, O’Byrne PM. Treatment of asthma with drugs modifying the leukotriene pathway.
N Engl J Med. 1999;340:197-206.
64. In KH, Asano K, Beier D, et al. Naturally occurring mutations in the human 5-lipoxygenase gene
promoter that modify transcription factor binding and reporter gene transcription. J Clin Invest.
1997;99:1130-1137.
65. Drazen JM, Yandava CN, Dube L, et al. Pharmacogenetic association between ALOX5 promoter geno-
type and the response to anti-asthma treatment. Nat Genet. 1999;22:168-170.
66. Schelleman H, Klungel OH, van Duijn CM, et al. Drug-gene interaction between the insertion/deletion
polymorphism of the angiotensin-converting enzyme gene and antihypertensive therapy. Ann Phar-
macother. 2006;40:212-218.
67. Schelleman H, Stickler BHC, de Boer A, et al. Drug-gene interaction between genetic polymorphisms
antihypertensive therapy. Drugs. 2004;64:1801-1816.
68. Rigat B, Hubert C, Alhenc-Gelas F, et al. An insertion/deletion polymorphism in the angiotensin
I-converting enzyme gene accounting for half the variance of serum enzyme levels. J Clin Invest.
1990;86:1343-1346.
69. Wu J, Kraja AT, Oberman A, et al. A summary of the effects of antihypertensive medications on
measured blood pressure. Am J Hypertens. 2005;18:935-942.
70. Frazier L, Turner ST, Schwartz GL, et al. Multilocus effects of the renin-angiotensin aldosterone system
genes on blood pressure response to a thiazide diuretic. Pharmacogenomics J. 2004;4:1723.
71. Lee Y-S, Kim H, Wu T-X, et al. Genetically mediated interindividual variation in analgesic responses to
cyclooxygenase inhibitory drugs. Clin Pharmacol Ther. 2006;79:407-418.
72. Esteller M, Garcia-Foncillas J, Andion E, et al. Inactivation of the DNA-repair gene MGMT and the clini-
cal response of gliomas to alkylating agents. N Engl J Med. 2000;343:1350-1354.
73. Hegi ME, Diserens A-C, Godard S, et al. Clinical trial substantiates the predicative value of MGMT-
methylation in glioblastoma patients treated with temozolomide. Clin Cancer Res. 2004;10:1871-1874.
74. Mitsuno M, Kitajima Y, Ide T, et al. Aberrant methylation of p16 predicts candidates for 5-fluorouracil-
based adjuvant therapy in gastric cancer patients. J Gastroenterol. 2007;42:866-873.
75. Kluft C, Lansik M. Effect of oral contraceptives on haemostasis variables. Thromb Haemost.
1997;78:315-326.
76. Van Hylckama Vlieg A, Rosendaal FR. Interaction between oral contraceptive use and coagulation
factor levels in deep venous thrombosis. J Thromb Haemostasis. 2003;1:2186-2190.
77. Martinelli I, Taioli E, Cetin I, et al. Interaction between the G20210A mutation of the prothrombin gene
and oral contraceptive use in deep vein thrombosis. Arterioscler Thromb Vasc Biol. 1999;19:700-703.
78. Bertina RM, Koeleman BPC, Koster T, et al. Mutation in blood coagulation factor V associated with
resistance to activated protein C. Nature. 1994;369:64-67.
79. Poort SR, Rosendaal FR, Reitsma PH, et al. A common genetic variation in the 3’-untranslated region
of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in
venous thrombosis. Blood. 1996;88:3698-3703.
80. Hetherington S, McGuirk S, Powell G, et al. Hypersensitivity reactions during therapy with the nucleo-
side reverse transcriptase inhibitor abacavir. Clin Ther. 2001;23:1603-1614.
81. Mallal S, Nolan D, Witt C, et al. Association between presence of HLA-B*5701, HLA-DR7, and HLA-DQ3
and hypersensitivity to HIV-1 reverse-transcriptase inhibitor abacavir. Lancet. 2002;359:727-732.
82. Hetherington S, Hughes AR, Mosteller M, et al. Genetic variations in HLA-B region and hypersensitivity
reactions to abacavir. Lancet. 2002;359:1121-1122.
83. Chung WH, Hung SI, Hong HS, et al. Medical genetics: a marker for Stevens-Johnson syndrome.
Nature. 2004 Apr 1;428(6982):486.
84. Kindmark A, Jawaid A, Harbron CG, et al. Genome-wide pharmacogenetic investigation of a hepatic
adverse event without clinical signs of immunopathology suggests an underlying immune patho-
genesis. Pharmacogenomics J. 2008;8:186-195.
85. Wilke RA, Lin DW, Roden D, et al. Identifying genetic risk factors for serious adverse drug reactions:
current progress and challenges. Nat Rev Drug Discov. 2007;6:904-916.
86. Roden DM. Taking the idio out of idiosyncratic: predicting torsades de pointes. Pacing Clin Electro-
physiol. 1998;21:1029-1034.
87. Mank-Seymour AR, Richmond JL, Wood LS, et al. Association of torsades de pointes with novel and
known single nucleotide polymorphisms in long QT syndrome genes. Am Heart J. 2006;152:1116-1222.
88. Abbott GW, Sesti F, Spawski I, et al. MiRP1 forms IKr potassium channels with HERG and is associated
with cardiac arrhythmia. Cell. 1999;97:175-187.
89. Sesti F, Abbott GW, Wei J, et al. A common polymorphism associated with antibiotic-induced cardiac
arrhythmia. Proc Natl Acad Sci USA. 2000;97:10613-10618.
90. Makita N, Horie M, Nakamura T, et al. Drug-induced long-QT syndrome associated with a subclinical
SCN5A mutation. Circulation. 2002;106:1269-1274.
91. Fitzgerald PT, Ackerman MJ. Drug-induced torsades de pointes: the evolving role of pharmacogenetics.
Heart Rhythm. 2005;2:S30-S37.
92. Harrison-Woolrych M, Clark DW, Hill GR, et al. QT interval prolongation associated with sibutramine
treatment. Br J Clin Pharmacol. 2006;6:464-469.
93. Gershon MD. Review article: serotonin receptors and transporters: roles in normal and abnormal
gastrointestinal motility. Aliment Pharmacol Ther. 2004;20:S3-S14.
94. Bergeron R, Blier P. Cisapride for the treatment of nausea produced by selective serotonin reuptake
inhibitors. Am J Psychiatry. 1994;151:1084-1086.
95. Sugai T, Suzuki Y, Sawamura K, et al. The effect of 5-hydroxytryptamine 3A and 3B receptor genes on
nausea induced by paroxetine. Pharmacogenomics J. 2006;6:351-356.
96. Murphy GM, Kremer C, Rodrigues HE, et al. Pharmacogenetics of antidepressant medication intoler-
ance. Am J Psychiatry. 2003;160:1830-1835.

97. Janno S, Holi M, Tuisku K, et al. Prevalence of neuroleptic-induced movement disorders in chronic
schizophrenia inpatients. Am J Psychiatry. 2004;161:160-163.
98. Lerer B, Segman RH, Fangerau H, et al. Pharmacogenetics of tardive dyskinesia: combined analysis of
780 patients supports association with dopamine D3 receptor gene Ser9Gly polymorphism. Neuro-
psychopharmacology. 2002;27:105-119.
99. Bakker PR, van Harten PN, van Os J. Antipsychotic-induced tardive dyskinesia and the Ser9Gly poly-
morphism in the DRD3 gene: meta-analysis. Schizophr Res. 2006;83:185-192.
100. Tsai HT, North KE, West SL, et al. The DRD3 rs6280 polymorphism and prevalence of tardive dys-
kinesia: meta-analysis. Am J Med Genet B Neuropsychiatr Genet. 2010;153B:57-66.
101. Müller DJ, Zai CC, Sicard M, et al. Systematic analysis of dopamine receptor genes (DRD1-DRD5) in
antipsychotic-induced weight gain. Pharmacogenomics J. 2012;12:156-164.
102. Reynolds GP. Pharmacogenomics aspects of antipsychotic drug-induced weight gain—a critical review.
Clin Psychopharmacol Neurosci. 2012;10:71-77.
103. Hill MJ, Reynolds GP. Functional consequences of two HTR2C polymorphisms associated with anti-
psychotic-induced weight gain. Pharmacogenomics. 2011;12:727-734.
104. Karpman I. Cough from ACE inhibitors. Am Heart J. 1988;116:1658.
105. Mukae S, Itoh S, Aoki S, et al. Association of polymorphisms of the renin-angiotensin system and
bradykinin B2 receptor with ACE-inhibitor-related cough. J Human Hypertens. 2002;16:857-863.
106. Behrens G, Dejam A, Schmidt H, et al. Impaired glucose tolerance, beta cell function and lipid metab-
olism in HIV patients under treatment with protease inhibitors. AIDS. 1999;13:63-70.
107. Pacenti M, Barzon L, Favaretto F, et al. Microarray analysis during adipogenesis identifies new genes
altered by antiretroviral drugs. AIDS. 2006;20:1691-1705.
108. Nolan D, Hammond E, Martin A, et al. Mitochondrial DNA depletion and morphologic changes in
adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS. 2003;17:1329-1338.
109. Bonnet E, Genoux A, Bernard J, et al. Impact of genetic polymorphisms on the risk of lipid disorders in
patients on anti-HIV therapy. Clin Chem Lab Med. 2007;45:815-821.
110. Arnedo M, Taffe P, Sahli R, et al. Contribution of 20 single nucleotide polymorphisms of 13 genes to
dyslipidemia associated with antiretroviral therapy. Pharmacogenet Genomics. 2007;17:755-764.
111. Hulgan T, Tebas P, Canter JA, et al. Hemochromatosis gene polymorphisms, mitochondrial
haplogroups, and peripheral lipoatrophy during antiretroviral therapy. J Infect Dis. 2008;197:858-866.
112. Ranade K, Geese WJ, Noor M, et al. Genetic analysis implicates resistin in HIV lipodystrophy. AIDS.
2008;22:1561-1568.
113. Rosenberg H, Pollock N, Schiemann A, et al. Malignant hyperthermia: a review. Orphanet Journal of
Rare Diseases. 2015;10:93. doi:10.1186/s13023-015-0310-1.
114. Sambuughin N, Holley H, Muldoon S, et al. Screening of the entire ryanodine receptor type 1 coding
region for sequence variants associated with malignant hyperthermia susceptibility in the North
American population. Anesthesiology. 2005;102:515–521.
115. Poirier J. Apolipoprotein E in animal models of CNS injury and in Alzheimer’s disease. Trends Neurosci.
1994;17:525-530.
116. Farlow MR, Lahiri DK, Poirier J, et al. Treatment outcome of tacrine therapy depends on apolipoprotein
genotype and gender of the subjects with Alzheimer’s disease. Neurology. 1998;50:669-677.
117. MacGowan SH, Wilcock GK, Scott M. Effect of gender and apolipoprotein E genotype on response to
anticholinesterase therapy in Alzheimer’s disease. Int J Geriatr Psychiatry. 1998;13:625-630.
118. Farlow M, Lane R, Kudaravalli S, et al. Differential qualitative responses to rivastigmine in APOE epsi-
lon 4 carriers and noncarriers. Pharmacogenomics J. 2004;4:332-335.
119. Soininen H, Kosunen O, Helisalmi S, et al. A severe loss of choline acetyltransferase in the frontal
cortex of Alzheimer patients carrying apolipoprotein E4 allele. Neurosci Lett. 1995;187:79-82.
120. Riordan JR. CFTR function and prospects for therapy. Annu Rev Biochem. 2008;77:701–726.
121. Ramsey BW, Davies J, McElvaney NG, et al. VX08-770-102 Study Group A CFTR potentiator in
patients with cystic fibrosis and the G551D mutation. N Engl J Med. 2011;365:1663-1672.
122. Clancy JP, Johnson SG, Yee SW, et al. Clinical Pharmacogenetics Implementation Consortium (CPIC)
guidelines for ivacaftor therapy in the context of CFTR genotype. Clin Pharmacol Ther. 2014;95:592-
597.
123. Baker DE. Approvals, submission, and important labeling changes for US marketed pharmaceuticals.
Hosp Pharm. 2014;49(5):482-484.

124. Roengvoraphoja M, Tsongalis GJ, Dragnev KH, et al. Epidermal growth factor receptor tyrosine kinase
inhibitors as initial therapy for non-small cell lung cancer: Focus on epidermal growth factor receptor
mutation testing and mutation-positive patients. Cancer Treat Rev. 2013;39:839-885.
125. Tan C-S, Cho B-C, Soo RA. Next-generation epidermal growth factor receptor tyrosine kinase inhibitors
in epidermal growth factor receptor mutant non-small cell lung cancer. Lung Cancer. 2016;93:59-68.
126. Fisher R, Larkin J. Individualising treatment choices in a crowded treatment algorithm. Eur J Cancer
Supplements. 2013;11:160-168.
127. Yewale C, Baradia D, Vhora I, et al. Epidermal growth factor receptor targeting in cancer: A review of
trends and strategies. Cancer Treatment Reviews. 2013;39:839-850.
128. Holbro T, Hynes NE. ErbB receptors: directing key signaling networks throughout life. Annu Rev Phar-
macol Toxicol. 2004;44:195-217.
129. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in
advanced colorectal cancer. N Engl J Med. 2008;350:1757-1765.
130. Pegram MD, Konecny G, Slamon DJ. The molecular and cellular biology of HER2/neu gene ampli-
fication/overexpression and the clinical development of herceptin (trastuzumab) therapy for breast
cancer. Cancer Treat Res. 2000;103:57-75.
131. National Cancer Policy Forum; Board on Health Care Services; Institute of Medicine. Policy issues in
the development and adoption of biomarkers for molecularly targeted cancer therapies: workshop
summary. Washington, DC: National Academies Press (US); 2015 Jun 1. Available at: http://www.ncbi.
nlm.nih.gov/books/NBK284967/.
132. Corrigan PA, Cicci TA, Auten JJ, et al. Ado-trastuzumab emtansine: a HER2-positive targeted antibody-
drug conjugate. Ann Pharmacother. 2014;48:1484-1493.
133. Harbeck N, Beckmann M, Rody A, et al. HER2 dimerization inhibitor pertuzumab—mode of action and
clinical data in breast cancer. Breast Care. 2013;8:49-55.
134. Thatcher N, Hirsch FR, Luft AV, et al. Necitumumab plus gemcitabine and cisplatin versus gemcitabine
and cisplatin alone as first-line therapy in patients with stage IV squamous non-small-cell lung
cancer (SQUIRE): an open-label, randomised, controlled phase 3 trial. The Lancet Oncology.
2015;16:763-774.
135. Ellis LM, Hicklin DJ. VEGF-targeted therapy: mechanisms of anti-tumour activity. Nat Rev Cancer.
2008;8:579-591.
136. Camidge DR, Berge EM, Doebele RC, et al. A phase II, open-label study of ramucirumab in combination
with paclitaxel and carboplatin as first-line therapy in patients with stage IIIB/IV non–small-cell lung
cancer. J Thoracic Oncology. 2014;9:1532-1539.
137. Capdeville R, Buchdunger E, Zimmermann J, et al. Glivec (STI571, imatinib), a rationally developed,
targeted anticancer drug. Nat Rev Drug Discov. 2002;1:493-502.
138. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor
underlying responsiveness of non–small-cell lung cancer to gefitinib. N Engl J Med. 2004;350:2129-
2139.
139. Ikediobi ON. Somatic pharmacogenomics in cancer. Pharmacogenomics J. 2008;8:305-314.
140. Miller VA, Hirsh V, Cadranel J, et al. Afatinib versus placebo for patients with advanced, metastatic
non-small-cell lung cancer after failure of erlotinib, gefitinib, or both, and one or two lines of chemo-
therapy (LUX-Lung 1): a phase 2b/3 randomised trial. Lancet Oncol. 2012;13:528-538.
141. Schuler M, Awada A, Harter P, et al. A phase II trial to assess efficacy and safety of afatinib in exten-
sively pretreated patients with HER2-negative metastatic breast cancer. Breast Cancer Res Treat.
2012;134:1149-1159.
142. Tevaarwerk AJ, Kolesar JM. Lapatinib: a small-molecule inhibitor of epidermal growth factor recep-
tor and human epidermal growth factor receptor-2 tyrosine kinases used in the treatment of breast
cancer. Clin Ther. 2009;31(Pt 2):2332-2348.
143. Cross DAE, Ashton SE, Ghiorghiu S, et al. AZD9291, an irreversible EGFR TKI, overcomes T790M-
mediated resistance to EGFR inhibitors in lung cancer. Cancer Discov. 2014;4:1046-1061.
144. Choueiri TK, Pal SK, McDermott DF, et al. A phase I study of cabozantinib (XL184) in patients with
renal cell cancer. Ann Oncol. 2014;25:1603-1608.
145. Cabanillas ME, Habra MA. Lenvatinib: role in thyroid cancer and other solid tumors. Cancer Treat Rev.
2016;42:47-55.
146. Hilberg F, Roth GJ, Krssak M, et al. IBF 1120: triple angiokinase inhibitor with sustained receptor block-
ade and good antitumor efficacy. Cancer Res. 2008;68:4774-4782.
147. Motzer RJ, Hutson TE, Cella D, et al. Pazopanib versus sunitinib in metastatic renal-cell carcinoma. N
Engl J Med. 2013;369:722-731.
148. Hudes G, Carducci M, Tomczak P, et al. Temsirolimus, interferon alfa, or both for advanced renal-cell
carcinoma. N Engl J Med. 2007;356:2271-2281.
149. Motzer RJ, Escudier B, Oudard S, et al. Phase 3 trial of everolimus for metastatic renal cell carcinoma:
final results and analysis of prognostic factors. Cancer. 2010;116:4256-4265.
150. Bollag G, Hirth P, Tsai J, et al. Clinical efficacy of a RAF inhibitor needs broad target blockade in
BRAF-mutant melanoma. Nature. 2010;467:596-599.
151. Flaherty KT, Infante JR, Daud A, et al. Combined BRAF and MEK inhibition in melanoma with BRAF
V600 mutations. N Engl J Med. 2012;367:1694-1703.
152. Templeton IE, Musib L. MEK inhibitors beyond monotherapy: current and future development. Curr
Opin Pharmacol. 2015;23:61-67.
153. Forde PM, Rudin CM. Crizotinib in the treatment of non-small-cell lung cancer. Expert Opin Pharma-
cother. 2012;3:1195-1201.
154. Li S, Qi X, Huang Y, et al. Ceritinib (LDK378): A potent alternative to crizotinib for ALK-rearranged non-
small-cell lung cancer. Clin Lung Cancer. 2015;16:86-91.
155. Song Z, Wang M, Zhang A. Alectinib: a novel second generation anaplastic lymphoma kinase (ALK)
inhibitor for overcoming clinically-acquired resistance. Acta Pharmaceutica Sinica B. 2015;5:34-37.
Chapter 7 • Cardiovascular Disease 213
PART II
Applications of
Pharmacogenomics in
Therapeutics
CHAPTER 7: Cardiovascular Disease.............................................................................. 215
CHAPTER 8: Hematology/Oncology Pharmacogenomics.......................257
CHAPTER 9: Central Nervous System.........................................................................283
CHAPTER 10: Infectious Diseases................................................................................... 309
CHAPTER 11: Respiratory Diseases..................................................................................327
CHAPTER 12: Toxicogenomics.............................................................................................357
CHAPTER 13: The Pharmacogenetics of Addiction......................................... 375
CHAPTER 14: Pharmacogenomics and Diabetes Mellitus.........................405

213
CHAPTER
7
Cardiovascular Disease
Kathryn M. Momary, PharmD, BCPS

After completing this chapter, the reader ARRHYTHMIA—abnormal electrical activity
should be able to within the heart.
Identify genetic variants that have been DOSING ALGORITHM—an equation that
associated with drug therapy response in considers clinical, laboratory, and/or
cardiovascular disease. genetic characteristics to predict dosing
requirements.
Translate data from genetic studies in
cardiovascular disease to clinical practice. DUAL ANTIPLATELET THERAPY—usually
refers to aspirin in combination with
Distinguish between polymorphisms that
clopidogrel or another antiplatelet agent.
have demonstrated clinical utility from
those that are still under investigation. DYSLIPIDEMIA—altered blood lipid
concentrations.
Design a therapeutic management plan
based on genetic information for warfarin HEART FAILURE—a condition where the heart
and clopidogrel therapy. is incapable of providing sufficient blood
Theorize which polymorphisms are flow to the body to meet metabolic
likely to be used in future cardiovascular needs.
practice. HYPERTENSION—elevated blood pressure.
ISCHEMIC HEART DISEASE—reduced blood
supply through the coronary arteries that
supply the heart muscle.
LOSS-OF-FUNCTION ALLELE—a
polymorphism that has been associated
with impaired metabolic function of the
enzyme that it encodes.
PRIMARY PREVENTION—measures taken to
prevent disease.
SECONDARY PREVENTION—measures taken
to prevent recurrence and/or additional
manifestations of a known disease.
215
STENT THROMBOSIS—an acute or delayed thrombus within an intracoronary stent, which

may be associated with a catastrophic outcome (e.g., myocardial infarction).
SURROGATE MARKER—a measurable change in physiology believed to lead to an eventual
outcome.
INTRODUCTION
Despite many advances in the treatment of cardiovascular disease (CVD), it remains the
leading killer of both men and women in the United States. On average, one death occurs
every 37 seconds because of CVD. In addition, it is estimated that the direct and indirect
cost of care for CVD exceeded $316 billion annually in 2011-2012.1
Given the large burden CVD has on society, efforts have been made to improve treat-
ment and outcomes associated with CVD. Many large-scale clinical trials have assessed
treatment modalities for CVD, resulting in the publication of numerous consensus guide-
lines. These guidelines drive the management of CVDs including hypertension, dyslipid-
emia, ischemic heart disease, and heart failure.2-4 Drug therapy recommendations make up
the majority of these guidelines.
A shortcoming of guideline-driven therapy is that it may lead to empiric therapeutic
choices that do not consider patient-specific factors. Given the significant interpatient
variability in response to many drugs used to treat CVD, it is important to understand the
causes of the variability and use that information to make patient-specific choices. The
variability in response is due to many factors such as age, race, sex, concomitant medica-
tions, and concomitant disease states. However, even after these patient-specific factors
are considered, a significant amount of interpatient variability remains in response. This has
led many to assess the role of genetic variability in the response to cardiovascular drugs.
The prevailing opinion is that cardiovascular, guideline-based therapy can also take into
account the individual patient’s genetic makeup.
CASE STUDY
A 58-year-old Asian-American woman with a past medical history of gout, diabetes,
hypertension, and dyslipidemia presents to the emergency department with crushing
sub-sternal chest pain. She is diagnosed with ST-elevation myocardial infarction (STEMI)
and is immediately sent to the cardiac catheterization lab for percutaneous coronary
intervention. At this time, she receives a single dose of aspirin (325 mg) and clopidogrel
(600 mg); two drug-eluting stents are implanted. She is then sent to the coronary care
unit, and her new medication regimen is started. This regimen includes atorvastatin
20 mg po daily, aspirin 325 mg po daily, clopidogrel 75 mg po daily, fosinopril 10 mg
po daily, metoprolol XL 50 mg po daily, esomeprazole 20 mg po daily, and metformin
1 g po daily. While she was in the emergency department, her CYP2C19 genotype was
determined using a new rapid genotyping assay. She was found to be CYP2C19*2/*3.
Questions
1. What effect will her CYP2C19 genotype have on formation of the clopidogrel active metabo-
lite and risk of future adverse cardiovascular events?
2. What drug-drug interactions must be considered, along with genetic testing, in patients
receiving clopidogrel therapy?

3. Given her CYP2C19 genotype, do you feel her therapy should be modified? If so, what therapy
changes do you recommend?
ANTICOAGULANT AND ANTIPLATELET AGENTS

Perhaps the most active research area in cardiovascular pharmacogenomics is the genetic
variation related to agents that affect coagulation or platelet activity. Several different
studies have evaluated the pharmacogenomics of warfarin, direct oral anticoagulants, aspi-
rin, clopidogrel, and the P2Y12 receptor inhibitors.
Warfarin
Warfarin has a narrow therapeutic range and multiple drug-drug and drug-food interactions
but is still widely prescribed despite the approval of newer oral anticoagulants. Numer-
ous clinical and demographic variables influence warfarin dosing including age, nutritional
status, and hepatic function. Even when one considers the known clinical variables that
alter warfarin dosing, the precision of dosing remains low. Polymorphisms in the genes
encoding two enzymes, cytochrome P450 (CYP) 2C9 and vitamin K epoxide reductase
complex subunit 1 (VKORC1), contribute significantly to warfarin dose requirements (see
Figure 7-1).
Warfarin is highly metabolized, and hence its effects can be altered by genetic varia-
tion that modifies drug metabolism. Warfarin is a racemic mixture (R- and S-isomers) with
the S-isomer being about five times more potent. Warfarin is cleared by multiple CYP
isoenzymes, but the S-isomer undergoes metabolism primarily via the CYP2C9 isoenzyme.
The initial research regarding genetic variation with warfarin focused on altered
metabolism via CYP2C9. CYP2C9*1 encodes for the wild-type enzyme that is consistent
with normal extensive metabolism of warfarin. Two common single nucleotide polymor-
phisms (SNPs) have been discovered: CYP2C9*2 and CYP2C9*3. The CYP2C9*2 variant has
an arginine replaced with a cysteine at position 144 in exon 3, which occurs in about 10-20%
of Caucasians and rarely in African Americans, Hispanics, and Asians (see Table 7-1).5 In the
CYP2C9*3 variant, isoleucine is replaced with leucine at exon 7. This occurs in about 5-9%
of the population. Overall, CYP2C9*2 variants have about a 40% reduction in enzymatic
activity corresponding to a 30% reduction in dose if one variant is present.6-9 CYP2C9*3
confers a 75% reduction in activity equivalent to a 47% reduction in dose if at least one
variant is present.10
Other CYP2C9 alleles (CYP2C9*5, *6, *8, and *11) have also been reported with varying
frequencies by race and ethnicity.5 The CYP2C9*5, 6, and 11 alleles occur almost exclusively
in African Americans but at much lower frequencies than CYP2C9*8. The CYP2C9*5, 8, and
11 alleles result from nonsynonymous SNPs in the coding region, while CYP2C9*6 results
from a nucleotide deletion. Decreased enzymatic activity has been seen for CYP2C9*5,
*6, *8, and *11; however, this appears to be substrate specific.11-15 In spite of this, decreased
warfarin dose requirements have been seen for CYP2C9*5, *6, *8, and *11.16-18 Another key
Activated Hypofunctional
Clotting Factors Clotting Factors
Carboxylase
Oxidized Reduced
Vitamin K Vitamin K
VKORC1
S-warfarin R-warfarin
CYP2C9 CYP1A/CYP3A4
FIGURE 7-1 Vitamin K recycling.
point of warfarin research is that altered metabolism also changes time to steady-state,
requiring longer periods before dose adjustment.9,18 Thus, genetic variability in the phar-
macokinetics of warfarin leads not only to decreased dose requirements, but alterations
in other pharmacokinetic parameters that may affect time to the therapeutic international
normalized ratio (INR) and timing of dose adjustments.
CLINICAL PEARL
Genetic variation in CYP2C9 has been associated with warfarin dose
requirements and bleeding risk with warfarin therapy. It is important to
note that CYP2C9 genetic variants also affect time to steady state; thus,
these patients require longer periods before dose adjustment.
This change in pharmacokinetic properties may be why patients possessing a

CYP2C9*2 or *3 allele are at an increased risk of both time above goal INR range and
serious or life-threatening bleeding.9,19,20 Although the risk for bleeding with a CYP2C9
variant is highest during initiation of therapy, evidence suggests that the risk is increased
throughout the treatment’s duration. Thus, patients with variant CYP2C9 alleles need to
be closely monitored.
Table 7-1
Racial Differences in Estimated Allele Frequencies Relevant to
Warfarin Pharmacokinetics and Pharmacodynamics
Allele Frequencies %
European African
Americans American Asian Hispanic
CYP2C9
CYP2C9*1 — 78.8 86.7 92.2 82.2
CYP2C9*2 rs1799853 15.1 2.8 2.9 6.9
CYP2C9*3 rs1057910 5.7 2 3.9 6.4
CYP2C9*5 rs28371686 — 1.5 — 1.5
CYP2C9*6 rs933213 — 1 — 0.5
CYP2C9*8 rs7900194 — 4.7 1 1.5
CYP2C9*11 rs28371685 0.5 1.3 — 1
VKORC1
−1693G — 59.4 89.2 33.3 56.4
−1693A rs992321 40.6 10.8 66.7 43.6
The gene encoding the active site for warfarin, VKORC1, mediates the conversion from
oxidized to reduced vitamin K, which is necessary for the activation of clotting factors II, VII,
IX, X and proteins C and S (Figure 7-1). Binding of warfarin to this enzyme blocks vitamin K
recycling and prevents the activation of clotting factors. Thus, warfarin leads to hypofunc-
tional clotting factors and an anticoagulated state. Altered response to warfarin at VKORC1
changes how patients respond to warfarin therapy. Various VKORC1 polymorphisms have
been identified, but the two most widely studied include C1173T and G-1639A.18 These two
polymorphisms, along with several others, often occur together in Caucasians and are thus
said to be in linkage disequilibrium. Therefore, looking at only one of these polymorphisms
(often the G-1639A SNP) is generally sufficient to characterize the genetic variability in the
Caucasian population. However, less is known about the linkage disequilibrium patterns
in non-Caucasian populations, such as African Americans. The frequencies of the variant
VKORC1 1173T or −1639A alleles are well characterized in several racial groups, and these
variant alleles are more common in the Asian population compared to those of European-
American descent and are uncommon in African Americans (Table 7-1).
In Caucasians, these variant alleles have been associated with decreased VKOR
messenger ribonucleic acid levels.9 Therefore, possession of either variant allele is asso-
ciated with increased warfarin sensitivity that is likely due to decreased available VKOR
to block. VKORC1 −1639AA genotype corresponds with patients who are more sensi-
tive to warfarin, whereas GG corresponds to patients who are less sensitive to therapy.
Mean warfarin dose requirements for patients with AA, GA, and GG genotypes are 2.7,
4.9, and 6.2 mg daily, respectively.21 The effect of the VKORC1 −1639 and 1173 SNPs have
been well studied, and the results have consistently demonstrated lower warfarin dose
requirements with −1639A and 1173T alleles.17,18,22-26 VKORC1 has been sequenced in
African Americans, and a novel SNP was discovered G−8191A.16 Each −8191G allele was
associated with a 5.2-mg/week dose increase and a linear regression model, including
this SNP, and known CYP2C9 SNPs explained 40% of the variability in warfarin dose
requirements in African Americans. This model performed significantly better than
traditional models that explain approximately 25% of the variability in warfarin dose
requirements in this population. Prior to the genetic characterization of VKORC1, warfa-

rin dosing nomograms used race as a predictor of warfarin dose requirements. We now
know that African Americans and Asians generally require higher and lower warfarin
doses, respectively, compared to Caucasians because of the difference in VKORC1 allele
frequencies in these populations. We also know that it is important to assess genetic
variability in racially and ethnically diverse cohorts.
CLINICAL PEARL
Prior to the genetic characterization of VKORC1, warfarin dosing
nomograms used race as a predictor of warfarin dose requirements. We
now know that African Americans and Asians generally require higher and
lower warfarin doses, respectively, compared to Caucasians because of the
difference in VKORC1 allele frequencies in these populations.
The CYP4F2 enzyme is responsible for metabolizing vitamin K1 to hydroxyvitamin K1;

this leads to less vitamin K1 available for reduction to vitamin KH2.27 If there is less vitamin
KH2, there is less vitamin K epoxide available for clotting factor activation. Thus, decreased
CYP4F2 activity leads to increased clotting factor activation. A nonsynonymous SNP in
CYP4F2 V433M leads to lower CYP4F2 protein concentration and thus greater vitamin
K availability. Studies have demonstrated the CYP4F2 433M/M genotype is associated
with approximately 1 mg/day higher warfarin dose requirements compared to the V/V
genotype.28 In Caucasians and Asians, this SNP explains approximately 1–3% of the overall
variability in warfarin dose requirements.29-32 Interestingly, when the CYP4F2 V433M SNP
was studied in African Americans, Indonesians, Egyptians, and children, no association with
warfarin dose requirement was found.17,33-35 The lack of association in African Americans is
likely due to the low frequency of the 433M allele. In addition, because body size plays a
greater role in warfarin dose requirements in children as compared to adults, this may also
affect genotype association. The role of CYP4F2 genotype in warfarin dose requirements
is less well defined than CYP2C9 and VKORC1. Genetic variability in the genes encoding
Calumen and gamma-glutamyl carboxylase have also been associated with warfarin dose
requirements in some patients. However, these associations are also less well defined than
VKORC1 and CYP2C9.36-38
Various dosing algorithms have been developed based on the available pharmaco-
genomic data and other clinical variables (e.g., age, weight, height, gender, race).25,26,39-42
The International Warfarin Pharmacogenetics Consortium (IWPC), a group of international
investigators, was initially founded to create a warfarin dosing equation.26,43 Data from 21
groups representing 11 countries and 4 continents pooled genotype and phenotype data
for over 5,700 chronic warfarin-treated patients. The pharmacogenetic algorithm, which
took into account both genotype and clinical information (drug interactions, body size, race,
and smoking status), accurately identified individuals who required markedly lower weekly
warfarin dosing (<21 mg/week) and those requiring higher weekly dosing (>49 mg/week),
which was necessary in 46.2% of the population. Overall, the pharmacogenetic algorithm,
as compared to the clinical algorithm, better predicted the stable dose of warfarin required
to achieve the target INR. This algorithm explained 50% of the variability in warfarin dose
among Caucasians and approximately 25% among Asians and African Americans. This
dosing algorithm is freely available at www.warfarindosing.org.
Based on the previously described results, warfarin became the first cardiovascular
drug to have a change in its package insert adding pharmacogenetic information, specifi-
cally stating “…lower initiation doses should be considered for patients with certain genetic
variations in CYP2C9 and VKORC1 enzymes” (August 2007).44 The potential benefits of
pharmacogenetic-guided dosing are to achieve the correct INR sooner, maintain the INR
within range better, and prevent complications.
The IWPC algorithm appears to be the most predictive of all of the algorithms
proposed to date. This algorithm is more accurate than the table provided in the prescrib-
ing information for warfarin and those algorithms established in smaller populations.45-47
However, all of these algorithms have some limitations. They do not include all the factors
known to affect variability in warfarin dose requirements, such as vitamin K intake and
many drugs known to interact with warfarin. In addition, most algorithms do not include
genetic variants specific to African Americans, likely leading to poor accuracy in individuals
of African ancestry. Lastly, these algorithms may overestimate doses in elderly patients.
Due to the inability of dosing algorithms to accurately predict dose requirements in all
patients, they should not replace routine INR monitoring and clinical judgment.
CLINICAL PEARL
Although genotype-based warfarin dosing algorithms can be effective,
they are not perfect predictors in all racial groups. Therefore, these
algorithms should not replace routine INR monitoring and clinical
judgment.
This area has been complicated by the simultaneous publication of three prospec-
tive, randomized, controlled trials comparing pharmacogenetic-guided dosing with a clini-
cal algorithm or standard of care.48-50 Only one of these studies was based in the United
States, the Clarification of Optimal Oral Anticoagulation through Genetics (COAG).48 This
study included 1,015 patients who were randomized to receive warfarin dosing according to
an algorithm that contained genotype and clinical variables, including early INR data, or one
with only clinical variables. The clinical variables–only algorithm included all of the same
clinical variables as the genetic and clinical algorithm (age, black race, smoker, body surface
area, amiodarone use, target INR, and deep vein thrombosis/pulmonary embolism [DVT/
PE] as an indication for therapy). In this study, both dose initiation and dose adjustment
for the first 4 weeks were handled using algorithms. The primary endpoint was percent of
time that INR was in the therapeutic range from day 4 or 5 through day 28 of therapy. The
investigators found no significant difference in time in therapeutic range between the two
algorithms, 45.2 ± 26.6% in the genotype plus clinically guided group and 45.4 ± 25.8% in
the clinically guided group; p value 0.91.
There was a significant interaction between race and dosing strategy, with black
patients in the arm that were dosed based on genotype and clinical variables having a
significantly lower time in therapeutic range compared to blacks in the clinically guided
group (35.2 ± 26% versus 43.5 ± 26.5%; p value 0.01). In contrast, the trend was toward
benefit in the primary endpoint with the genotype-guided strategy in the non-black cohort
(48.8 ± 25.9% versus 46.1 ± 25.5%; p = 0.15); however, this was not statistically significant.
Although this study was not powered to evaluate bleeding events, the INR ≥4, major bleed-
ing, clinically relevant nonmajor bleeding, and thromboembolism were assessed. There
were no statistically significant differences in these outcomes.

The European study from the European Pharmacogenetics of AntiCoagulant Therapy
(EU-PACT) included 455 patients.49 These patients were also randomized to genotype-
guided or standard therapy. In contrast to the COAG study where a clinical algorithm was
used, patients in the standard therapy arm in this study were given either 10 or 5 mg of
warfarin for 3 days based on age, and then the warfarin dose was adjusted based on INR.
The primary outcome for this study was percent of time in therapeutic range during the
12 weeks after warfarin initiation. The genotype-guided group had significantly greater
percentage of time in therapeutic range compared to standard of care (67.4 ± 18.1% versus
60.3 ± 21.7%; p <0.001). Patients in the genotype-guided group were also statistically, signifi-
cantly less likely to have an INR ≥4 and had a significantly shorter time to reach therapeutic
INR. Although other safety outcomes were assessed, no major bleeding events occurred
during the study.
The second EU-PACT study evaluated acenocoumarol and phenprocoumon, agents
that are not available in the United States.50 In this study, patients were randomized to
genotype-guided or clinically guided dosing (as opposed to the fixed dose utilized in the
other EU-PACT study). There were 548 patients in this study, and the primary outcome was
percent of time in the therapeutic range during the 12 weeks of coumarin initiation. There
was no difference in the primary outcome between the two groups, 61.6 ± 23.3% in the
genotype-guided group and 60.2 ± 23.2% in the clinically guided group; p = 0.52. The time
in therapeutic range during the first 4 weeks of therapy was higher in the genotype-guided
group compared to the clinically guided group, 52.8% versus 47.5%; p = 0.02.
Several important distinctions between the studies must be considered. The EU-PACT
studies were composed of almost exclusively white patients, but nearly one-third of the
patients enrolled in the COAG study were black. All three studies looked at CYP2C9*2,
CYP2C9*3, and VKORC1 G3673A, which occur far more frequently in Caucasians as
compared to those of African descent. Thus, these alleles explain less of the variation in
warfarin dose requirements in African Americans and are likely less predictive of warfa-
rin dose requirements (as described previously). This is potentially one reason for the
negative results from the COAG study. Another potential reason for the negative COAG
study results is the short duration of followup. Genotype guidance likely has the greatest
benefits during the first 1–2 months of therapy; it is possible that some of the benefit was
missed in the COAG study, as followup was only 28 days. Also, the first EU-PACT study
utilized a fixed, loading dose regimen while COAG used a clinically guided model. It is not
currently standard practice to initiate warfarin with a clinical algorithm, so it is likely that
EU-PACT better represented the “real world” warfarin experience.
In addition, all of these studies have some shared weaknesses. First, the study protocol
for all of the studies required intensive INR monitoring, which may have blurred some of the
difference between genotype-guided and clinically guided therapy. INR was likely checked
more frequently than is currently done in routine clinical practice. Also, none of these stud-
ies was powered to look at the most clinically important outcomes such as bleeding and
thrombotic events.
Due to the unclear benefit seen in these studies, current clinical guidelines for the
treatment of venous thromboembolism do not recommend the use of routine pharma-
cogenetic testing for warfarin.51 The Clinical Pharmacogenetics Implementation Consor-
tium (CPIC) has provided guidelines on how to interpret and apply genetic test results to
warfarin dosing when such results are available.52 The CPIC guideline does not, however,
address when or whom to genotype, leaving this to the discretion of the clinician. The CPIC
guidelines were written in recognition that the available data strongly support a genetic
influence on dose requirements and that the dose should be adjusted when genotype is
known. These guidelines provide a rating of A (strong) for warfarin dosing based on geno-
type when genotyping results are available. However, these guidelines were written prior
to the publication of the COAG and EU-PACT trials.
Given the current state of evidence, routine pharmacogenetic testing for warfarin
cannot be recommended. However, the EU-PACT study suggests that Caucasian patients
are likely to derive some benefit from pharmacogenetic testing using current genetic
algorithms. The greatest benefit for pharmacogenetic testing with warfarin will likely be
in patients who are Caucasian and at a high risk of bleeding or having out-of-range INRs.
However, insurance companies do not routinely reimburse for pharmacogenetic testing
for warfarin, which complicates clinical utility. But one ongoing study entitled the Genetics
Informatics of Warfarin to Prevent DVT may provide some additional clarification on this
topic. This study has a primary outcome of a composite endpoint of nonfatal venous
thromboembolism, nonfatal major hemorrhage, INR ≥4, and death.53 Because this study will
look at hard clinically relevant outcomes, it will add to the currently available literature.
Direct Oral Anticoagulants

Several direct oral anticoagulants have been recently approved for the treatment of
thromboembolic disorders. They include dabigatran, a direct thrombin inhibitor, and rivar-
oxaban and apixaban, which are factor Xa inhibitors. Unlike warfarin, these agents do not
require INR monitoring or dosage adjustment. Only dabigatran has been studied for phar-
macogenetic variation. A sample of patients from the phase three study for approval of
dabigatran, Randomized Evaluation of Long-term Anticoagulation Therapy, was analyzed in
a genome-wide association study (GWAS).54 Specifically, they looked for genetic variants
associated with dabigatran peak and trough concentrations. Although SNPs in CES1 and
ABCB1 were associated with dabigatran concentrations, only one SNP in CES1 was associ-
ated with a clinical outcome. The CES1 SNP rs2244613 was statistically, significantly asso-
ciated with trough concentrations, any bleeding, and minor bleeding with a trend toward
significance for major bleeding with dabigatran therapy. No association existed between
this SNP and bleeding with warfarin. Although these findings are intriguing, they have yet
to be replicated in a separate cohort. Until that time, routine genotyping for dabigatran
cannot be recommended.
Aspirin
Aspirin is a standard therapy in patients with or at high risk of ischemic heart disease or
stroke. It has been shown to reduce the risk of death, MI, and nonfatal stroke in this patient
population.55 Moreover, aspirin in addition to a P2Y12 inhibitor plays an important role in
duel antiplatelet therapy after percutaneous coronary intervention (PCI).
Aspirin exhibits its antiplatelet effect by inhibiting cyclooxygenase-1 (COX-1), which is
responsible for production of thromboxane A2 (TXA2). TXA2 is one of many factors that
promote platelet activation, which subsequently leads to platelet aggregation. Aspirin is

not effective in all patients, and numerous variables may be responsible for this phenom-
enon including platelet agonists beyond TXA2 (e.g., adenosine diphosphate), gender differ-
ences, and patient adherence. The term aspirin resistance has been coined to represent
this phenomenon. This term can be defined further as clinical resistance, where there is the
occurrence of an atherothrombotic event despite aspirin therapy, and biologic/pharmaco-

dynamic resistance, where aspirin fails to inhibit platelet function as determined by labo-
ratory assay. The cause of this resistance is unlikely to be mediated by variability in COX-1
inhibition. Direct measures of COX-1 inhibition in patients on aspirin demonstrate little
variability or heritability.56,57 However, platelet aggregation through non-COX-1 mediated
pathways demonstrates significant heritability and variability. This is further supported by
the conflicting data when looking at polymorphisms of the gene encoding COX-1 (PTGS1)
as a potential cause of impaired response to aspirin. Genetic variation in the promoter
region, specifically the A-842G polymorphism, has been the focus of research associated
with aspirin resistance. Studies in coronary artery disease, using laboratory assays, suggest
the −842G allele is associated with a reduced response to aspirin,58,59 although contradic-
tory data have been published in patients with recurrent ischemic stroke.60,61
Recent studies have found that genetic variation in PEAR1 may be associated with
platelet aggregation and aspirin response. This association has been seen in two separate
GWAS.62,63 However, the implications of PEAR1 genotype on aspirin response and clinical
outcomes is still unknown. Of note, all of these studies used different methods to deter-
mine aspirin resistance, clinically and laboratory based. This highlights the importance
of phenotype choice in pharmacogenomics studies, especially when the phenotype is as
complicated as platelet aggregation and aspirin responsiveness. Due to the inconsistency
of the data currently available, clinical genotyping for aspirin response cannot be recom-
mended.
CLINICAL PEARL
The pharmacogenetics of antiplatelet therapy is complicated by the use of
surrogate markers, such as platelet function testing, to assess response.
This highlights the importance of understanding the phenotype being
studied.
Clopidogrel
Clopidogrel is another important antiplatelet agent commonly used in clinical practice. In
addition to being a viable option in patients who cannot take aspirin, clopidogrel is a stan-
dard of care in patients after PCI with stent placement. Dual antiplatelet therapy (P2Y12
receptor blockade in addition to aspirin) is necessary after stent placement to prevent
in-stent thrombosis. Although other P2Y12 receptor blockers have been recently approved,
the many years of clinical experience with clopidogrel and its price make it still the
mainstay of therapy. However, not all patients respond similarly to clopidogrel therapy.
Studies have shown that anywhere between 4-30% of patients have high on-treatment
platelet reactivity (HOTPR) and increased rates of cardiovascular events.64,65 Patients with
increased platelet aggregation seen on in vitro platelet function tests are determined to
have HOTPR.
Variability in the pharmacokinetics of clopidogrel is significant, beginning with its

absorption in the gastrointestinal tract, which is affected by P-glycoprotein (P-gp) (ABCB1);
see Figure 7-2. Also, 85% of clopidogrel is degraded into inactive carboxylic acid deriva-
tives by hepatic esterases. The most notable of these enzymes is carboxylesterase 1 (CES1).
Additionally, the 15% of remaining clopidogrel is a prodrug and requires conversion to
its active form via numerous CYP450 isoenzymes. This active metabolite then exerts its
effects by inhibiting adenosine diphosphate activity via the P2Y12 receptor. There are vari-
ous ways in which polymorphisms can lead to altered drug effect at the receptor site. In
Clopidogrel
ABCB1 Liver cell

Clopidogrel
apical
CYP2C19
membrane
CYP1A2 CYP2B6
Clopidogrel
2-oxo-clopidogrel
CYP2C9 CYP3A4
CES1
CYP2B6 CYP3A5
PON1 CYP2C19
© Pharr mGKB
B
Intestinal Active metabolite

cell
Inactive metabolite
irreversible
Elimination
(urine; feces) P2RY12
Platelet
FIGURE 7-2 Clopidogrel pharmacokinetic and pharmacodynamic pathway.

Source: Sangkuhl K, Klein TE, Altman RB. Clopidogrel pathway. Pharmacogenet Genomics.
2010 Jul; 20(7):463-465. https://www.pharmgkb.org/pathway/PA154424674; © PharmGKB.
Permission for use has been given by PharmGKB and Stanford University.
fact, up to 30% of natural variation in platelet reactivity is related to genetic inheritance,

as described with aspirin.66 Specifically, genetic variability in the CYP450 system and in
P2Y12 have been studied for their role in clopidogrel response. Much like aspirin, clopidogrel
response can be characterized through clinical outcomes or via platelet aggregation tests.
Genetic variability in the gene encoding the P2Y12 receptor (P2RY12) has been well
studied. Several SNPs in P2RY12 have been assessed for their association with clopidogrel
response. The majority of the SNPs studied were in linkage disequilibrium and occurred
together. Although some studies have found positive results, the largest study to date
assessed the role of the T744C (rs2046934) polymorphism in clopidogrel efficacy as
assessed by both biologic and clinical response.67,68 They found no association between
this SNP and response to clopidogrel. Other studies have found similar results.69 Therefore,
genotyping of P2RY12 polymorphisms is not used to predict clopidogrel response.
Genetic variation in pathways involved in drug absorption and metabolism can also
lead to reduced response to clopidogrel. P-gp is involved in clopidogrel absorption and is
encoded by the gene ABCB1 (ATP-binding cassette subfamily B member 1). As clopidogrel is
absorbed from the intestinal lumen, P-gp pumps a portion of the drug back into the lumen
(decreasing bioavailability). The role of three ABCB1 SNPs (C3435T, G2677T, and C136T) has
been evaluated for their role in clopidogrel responsiveness. The C3425T SNP (rs1045642)
has been the most well-studied polymorphism. This variant is relatively common and has
been shown to result in increased P-gp expression.70 Early studies revealed a lower clopi-
dogrel concentration (both Cmax [maximum drug concentration after a dose at steady-state]
and AUC [area under the curve]) after a single dose of 300 and 600 mg in patients who
were homozygous for the variant 3435T allele.71 Of interest, a larger loading dose of 900
mg overcame this difference. Subsequent studies demonstrated an association between
the ABCB1 3435T allele and increased cardiovascular events with clopidogrel therapy.69,72
However, this change in pharmacokinetic parameters has not been consistently associated
with clinical response to clopidogrel.73 Two recent meta-analyses have also found conflict-
ing results for an association with the 3435T allele with cardiovascular events with clopi-
dogrel therapy.74,75 Although some evidence suggests that the ABCB1 C3435T variant plays
a role in clopidogrel responsiveness, the conflicting evidence makes routine genotyping of
ABCB1 in patients receiving clopidogrel not currently recommended.
Other studies have evaluated the role of altered metabolism with clopidogrel. Vari-
ous isoenzymes of the CYP450 system have been evaluated including CYP3A4, CYP3A5,
CYP2B6, CYP1A2, and CYP2C9. Some associations have been seen with genetic variation
in these CYP enzymes and clopidogrel responsiveness, but the results have been inconsis-
tent.76 These inconsistencies are theoretically due to the redundant mechanisms for clopi-
dogrel activation, making individual enzyme effects quite small. Thus, the current evidence
is not strong enough to warrant genotyping.
However, several well-characterized polymorphisms in the gene encoding CYP2C19
have been studied for their role in clopidogrel response. The CYP2C19*2 and *3 vari-
ant alleles are associated with decreased CYP2C19 function compared to CYP2C19*1.
Approximately 30% of Caucasians, 40% of African Americans, and more than 55% of
East Asians carry one of these variant alleles.77 There are other loss-of-function alleles
such as CYPC19*4-*8 that are generally rare in most populations and, thus, have not been
well studied for their role in clopidogrel responsiveness. There is also a gain-of-function
allele CYP2C19*17, which is in the promoter region of the gene and theoretically results in
increased transcription of the gene.78
The role of CYP2C19*2 in clopidogrel responsiveness has been well character-

ized. Early investigations found decreased active clopidogrel metabolite production
and HOTPR.79-82 Two studies, a genetic analysis of the Trial to Assess Improvement in
Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel-Thrombolysis
in Myocardial Infarction (TRITON-TIMI) 38 and a registry study of young patients status
post-MI, found that possession of at least one loss-of-function CYP2C19 allele was
associated with increased risk of cardiovascular events.83,84 In contrast, data from two
independent trials, Clopidogrel in Unstable Angina to Prevent Recurrent Event (CURE)
and Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events-
Aspirin (ACTIVE-A), showed no evidence of association between CYP2C19*2 or *3 and
cardiovascular events.85
Several recent meta-analyses have provided some clarification on the conflicting
data.86-89 Clopidogrel is used in many different disease states; however, patients under-
going PCI appear to gain the greatest benefit from clopidogrel therapy. Thus, it makes
sense that these patients would see the greatest impact of CYP2C19 genotype on clopi-
dogrel efficacy. Nearly all of the meta-analyses to date have demonstrated a strong asso-
ciation between CYP2C19*2 and *3 genotype and stent thrombosis risk with clopidogrel
therapy. Also, when focusing specifically on patients undergoing PCI, a strong association
has been demonstrated with CYP2C19*2 and *3 genotype.
In fact, these data lead to the addition of pharmacogenetic information in the clopi-
dogrel prescribing information.90 The updated prescribing information states that effec-
tiveness of clopidogrel depends on activation by CYP2C19, and CYP2C19 poor metabo-
lizers are at increased risk of cardiovascular events following acute coronary syndrome
(ACS) and PCI. In addition the insert states that tests are available to identify a patient’s
CYP2C19 genotype, and alternative treatment strategies should be considered in CYP2C19
poor metabolizers. The best strategy to manage patients with a CYP2C19 loss-of-function
allele is currently unclear. Several studies have assessed increasing the clopidogrel dose
in patients with a loss-of-function allele.91-94 These studies utilized platelet function to
assess clopidogrel responsiveness with different dosing strategies. The majority of studies
found that patients possessing only one loss-of-function allele achieved similar inhibition
of platelet aggregation with higher doses compared to CYP2C19*1 homozygotes on 75 mg/
day. However, the results are less consistent for patients who carry two loss-of-function
alleles. Thus, the most likely effective strategy for managing patients with CYP2C19 loss-
of-function alleles is changing them to an alternative P2Y12 inhibitor.
CLINICAL PEARL
Although increasing the clopidogrel dose to overcome treatment failure in
patients with CYP2C19 loss-of-function alleles may work in patients with
one loss-of-function allele, the most effective strategy is changing patients
to an alternative P2Y12 inhibitor.
The effect of the CYP2C19*17 gain-of-function allele on clopidogrel responsiveness

has also been studied. This allele has been associated with increased production of the
clopidogrel active metabolite, greater inhibition of platelet aggregation, and increased risk
of bleeding with clopidogrel therapy.95,96 However, a recent meta-analysis found no asso-
ciation between CYP2C19*17 genotype and cardiovascular event rates.88 While differences
in study design and patient population may explain the differences in results, another
explanation is that CYP2C19*2 and CYP2C19*17 are in linkage disequilibrium.97 Thus, the
CYP2C19*17 allele does not appear to occur on the same allele as CYP2C19*2. This makes it
likely that the effects of these variants are not independent of each other. In fact, when the
effects of CYP2C19*17 on platelet aggregation were adjusted for CYP2C19*2, the associa-
tion with CYP2C19*17 disappeared. Thus, the role of CYP2C19*17 on clopidogrel responsive-
ness needs further study before it can be incorporated into clinical genotyping.
In 2011, a study proposed that paraoxonase 1 (PON1) played a role in clopidogrel effi-
cacy.98 These authors demonstrated that PON1, a well-described hepatic esterase, played a
role in converting 2-oxo-clopidogrel to the active metabolite. They also demonstrated that
the PON1 192Q allele was associated with decreased clopidogrel active metabolite produc-
tion and increased occurrence of stent thrombosis. However, other groups were unable to
replicate the findings that the PON1 Q192R variant was associated with clopidogrel respon-
siveness or cardiovascular events.99 In addition, a meta-analysis also found no association
between PON1 Q192R and cardiovascular event risk.100 More recent studies suggest that
this PON1 polymorphism is associated with relative platelet inhibition from a mechanism
separate from clopidogrel.101,102 Thus, currently available information does not support the
use of PON1 genotyping for patients receiving clopidogrel.
It has been proposed that carboxylesterase 1 (CES1) is responsible for degrading up
to 85% of administered clopidogrel. Thus, genetic variation in CES1 has been studied for a
role in clopidogrel responsiveness. Three studies have demonstrated that genetic variants
in CES1 affect clopidogrel pharmacokinetics and pharmacodynamics.103-105 However, CES1 is
highly polymorphic and no studies to date have demonstrated an association between any
CES1 SNP and cardiovascular events with clopidogrel therapy. Additional data are neces-
sary before CES1 genotyping can be recommended for patients receiving clopidogrel.
Given the totality of evidence, expert consensus panels and guidelines have addressed
the use of pharmacogenetic testing with clopidogrel therapy. In 2010, the American College
of Cardiology and the American Heart Association Foundation issued a joint clinical alert
in response to the addition of genetic information to the clopidogrel labeling.106 Due
to the lack of outcomes data with routine genetic testing in large cohorts of patients,
they surmised that the available data were insufficient to recommend genotyping for all
patients prescribed clopidogrel. However, they further stated that CYP2C19 genotyping
may be considered in patients who are at moderate to high risk for poor cardiovascu-
lar outcomes, such as those undergoing elective high-risk PCI for extensive and/or very
complex coronary artery disease and others at the clinician’s discretion. In these patients,
alternative therapies (e.g., prasugrel, ticagrelor) can be considered. In addition, the guide-
lines state that platelet function testing may also be used in moderate- to high-risk
patients to assess clopidogrel responsiveness. The guidelines further suggest that large,
prospective, controlled trials will be required before widespread CYP2C19 genotyping with
clopidogrel therapy can be recommended. However, it will likely be several years before
these data are available.
Information on CYP2C19 genotyping with clopidogrel has been incorporated into the
recent guidelines for the management of patients with ACS undergoing PCI. The PCI guide-
lines suggest considering genetic testing to determine risk for inadequate platelet inhibition
with clopidogrel, but only in patients at high risk for poor clinical outcomes such as those
undergoing high-risk PCI procedures (i.e., elective PCI in patients with unprotected left
main, bifurcating left main, last patent coronary artery).107 Alternative anti-platelet therapy
is recommended for patients found to have the CYP2C19 loss-of-function genotype (Class
IIB, Level C). The PCI guidelines also state that there is no benefit for the routine use of
genetic testing; however, this is classified as the lowest level of recommendation based
on the lowest level of evidence (Class III, Level C). The guidelines for the management of
patients with non-ST-elevation myocardial infarction (NSTEMI) or unstable angina provide

similar recommendations and refer back specifically to the guidelines for PCI.108 The guide-
lines for STEMI also state that the data are insufficiently consistent to recommend routine
genotyping for clopidogrel therapy.109
In 2011, CPIC guidelines regarding the pharmacogenetics of clopidogrel were published
and then updated in 2013.110,111 The guidelines work under the assumption that genotype
information is already available. They recommend considering an alternative antiplate-
let agent (e.g., prasugrel, ticagrelor) in patients who possess at least one CYP2C19*2 or
*3 allele. However, they state that in patients who are heterozygous for CYP2C19*2 or *3,
there is wide variability in clopidogrel responsiveness. Thus, clinicians are encouraged to
take into account other factors that may place these patients at increased risk of a cardio-
vascular or bleeding event and use clinical judgment to individualize therapy. The authors
also summarize the challenges related to clinical genetic testing, namely the need for rapid
genotyping. In the setting of ACS or emergent PCI, genetic test results are unlikely to be
available at the time of antiplatelet initiation.
To address the need for rapid genotyping mentioned in the guidelines, two point-of-
care genotyping assays have been developed. The Verigene CYP2C19 XP system utilizes a
blood sample and provides the presence or absence of CYP2C19*2 or *3 in approximately 3
hours. The Spartan RX CYP2C19 system uses a buccal swab to determine CYP2C19*2 carrier
status (*1/*1, *1/*2, *2/*2) and gives results in approximately 60 minutes. A proof of concept
study, entitled RAPID GENE, for the Spartan assay was done.112 Patients undergoing PCI for
ACS or stable angina were randomly assigned to rapid genotype screening or conventional
treatment with subsequent genotyping for data analysis purposes only. Those random-
ized to rapid screening and carrying a CYP2C19*2 allele were treated with prasugrel 10 mg
daily, while everyone else was treated with clopidogrel 75 mg daily (including the entire
conventional arm). None of the CYP2C19*2 allele carriers in the rapid genotyping arm had
HOTPR, while 30% of the CYP2C19*2 allele carriers in the conventional group did. This
demonstrated that use of the Spartan assay prospectively decreased HOTPR in this popu-
lation and that this test could be used clinically.
Although these genetic tests provide more rapid genotype information, the lack of
randomized, controlled trial data for clopidogrel pharmacogenetics still inhibits routine
genetic testing. Until data from large prospective trials are available, CYP2C19 genotyping
for clopidogrel is unlikely to become a routine part of patient care.
Next-Generation P2Y12 Inhibitors

Both clopidogrel and ticagrelor have been recently approved. Prasugrel, like clopidogrel, is
administered as a prodrug and also irreversibly inhibits the platelet P2Y12 receptor. How-
ever, unlike clopidogrel, the vast majority of prasugrel is converted to the active metabolite,
and high concentrations are achieved. Ticagrelor is a novel, non-thienopyridine reversible
P2Y12 inhibitor that is administered as an active drug. Even though ticagrelor and prasu-
grel generally achieve high levels of platelet inhibition, some variability still exists. The
association between genetic variability in CYP450 enzymes (CYP2C19, CYP2C9, CYP2B6,

and CYP1A2) have not been consistently linked to prasugrel responsiveness.79,113-115 One
study showed an association between genetic variation in PEAR1 and prasugrel inhibition
of platelet aggregation, but these findings have not been replicated or an association
seen with clinical events.116 Similarly, no association between genetic variation in CYP450
enzymes and ticagrelor response has been seen.117-120 Given the lack of consistent asso-
ciation with genetic variation in response to these medications, routine genotyping is
not recommended. In addition, this also makes ticagrelor and prasugrel good alternative
choices to clopidogrel for patients with CYP2C19 loss-of-function alleles.
AGENTS FOR HYPERTENSION, HEART FAILURE, AND

ISCHEMIC HEART DISEASE
Research regarding the genetics of hypertension, heart failure, and ischemic heart disease
has been growing rapidly. Hypertension alone affects 80 million U.S. adults ≥20 years of
age, 38.3 million men and 41.7 million women.1 These multifaceted disease states likely
result from a combination of environmental and genetic effects. In addition, these disease
states are complicated because their etiology and time course is complex and varied.
Thus, there is not one single cause of elevated blood pressure or heart failure. A complex
network of genetic variation likely affects these disease states, and this genetic variation
is probably modulated by environmental factors such as diet, exercise, and lifestyle. Thus,
the underlying genetic mechanisms of the disease states themselves are complicated to
assess. This also makes understanding the pharmacogenetics of medications used to treat
these disease states complicated. However, interpatient variability has been noted with the
β-blockers, angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers
(ARBs), diuretics, and hydralazine. This interpatient variability has been well studied. This
section addresses the pharmacogenomic basis for these alterations as well as relevant
clinical data.
ß-Blockers
The primary site of action for β-blockers is the adrenergic system, specifically the beta-1 (β1)
and beta-2 (β2) receptors. These receptors are polymorphic, and genetic variation may result
in an altered treatment response. The effect of pharmacogenomics on β-blocker response
likely differs based on the pharmacologic characteristics of the agent. For instance, some
β-blockers are relatively selective for the β1 receptor (e.g., metoprolol) whereas others are
nonselective (e.g., propranolol). Moreover, certain agents may have α-adrenergic receptor
blockade (e.g., carvedilol) or intrinsic sympathomimetic activity (e.g., pindolol).
The gene encoding the ß1 receptor (ADRB1) has two well-studied polymorphisms
(Arg389Gly and Ser49Gly). Of note, frequencies of these polymorphisms appear to vary
based on ethnic background. For the ADRB1 gene, both the variant 389Gly and 49Gly
alleles occur more frequently in African Americans than Causasians.121 In vitro, these poly-
morphisms have been shown to affect the function of the receptor as well as its intracel-
lular signaling. Specifically, data demonstrate that the variant 389Arg and 49Ser alleles
result in increased in vitro activity and adrenergic signal transduction. This increased activ-
ity suggests that patients possessing an ADRB1 389Arg or 49Ser allele for ADRB1 would
have better response to β-blockade.121,122 Polymorphisms of ADRB1 have been well studied
in patients with hypertension and heart failure.
Response to β-blockers in hypertension is highly variable, with up to 60% of patients

on treatment not achieving adequate control with monotherapy.123,124 Genetic variation
may be a contributing factor. Clinical trials have validated in vitro hypotheses, and patients
with the 389Arg and 49Ser genotypes have been shown to exhibit a greater reduction of
blood pressure with β-blocker therapy.125-127 Specifically, associations have been seen with
metoprolol and atenolol. In addition, in a genetic substudy on the International Verapamil

SR/Trandolapril Study (INVEST), the Ser49Arg389 haplotype was associated with higher
mortality rates; however, the increased mortality risk was observed only in the verapamil
arm, not the β-blocker arm.127 Therefore, the data suggest that the Ser49Arg389 haplo-
type is not only associated with increased blood pressure lowering with β-blocker therapy,
but β-blocker therapy may also provide a protective effect against mortality risk with this
haplotype.
In heart failure, several studies have evaluated ADRB1 polymorphisms and response to
β-blocker therapy. Similar to the data in hypertension, it is patients with the 389Arg geno-
type that showed improved response, including a greater reduction in left ventricular diam-
eter and improvement in left ventricular ejection fraction.128,129 However, the association of
the 389Arg allele with clinical endpoints with β-blocker therapy has been inconsistent.121 In
the Beta-blocker Evaluation in Survival Trial (BEST), bucindolol was evaluated in patients
with NYHA class III and IV heart failure. When compared to standard of care, bucindolol
showed no overall survival benefit.130 When considering the hazard ratios for death in
prespecified subgroups, it was suggested that a difference in response was seen based on
race. African-American patients did not receive any benefit from bucindolol therapy, while
non–African-American patients had a significant reduction in mortality with bucindolol.
Given the differing frequency of polymorphisms in the β-adrenergic receptor genes by race,
genetic variation may have played a role in the differing response.
A genetic substudy of BEST found that the 389Arg allele was associated with a great-
er reduction in mortality and improved bucindolol response.131 Although some published
reports have been consistent with this study, contrary data have been published.129,132,133
A substudy of the Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart
Failure (MERIT-HF) found no association between heart failure outcomes or response to
β-blocker therapy (metoprolol) and the Arg389 genotype.133 The differences seen between
bucindolol and other β-blockers are likely due to the pharmacology of bucindolol. This is
a further reminder that we cannot assume a class effect exists when looking at pharmaco-
genetic studies. Each medication, especially β-blockers, has its own unique pharmacology
and thus must be evaluated separately.
CLINICAL PEARL
The differences in the pharmacology of ß-blockers lead to different genetic
associations with response. This is a further reminder that we cannot
assume that a class effect exists when looking at pharmacogenetic studies.
Bucindolol attempted to earn the U.S. Food and Drug Administration (FDA) approval
based on the results of the BEST substudy. The FDA, however, deemed that additional data
were needed. Only when or if bucindolol receives FDA approval can ADRB1 genotyping be
recommended for bucindolol therapy in this patient population.
The gene encoding the β2 receptor (ADRB2) has two polymorphisms of interest—
Gly16Arg and Glu27Gln. The ADRB2 16Arg and 27Gln alleles have been associated with
reduced sensitivity to isoproterenol (a β1 and β2 receptor agonist) in vitro.134 Similar to
the ADRB1 gene, the frequency of polymorphisms in ADRB2 differ based on race. The
ADRB2 16Arg allele frequency is 0.39 among Caucasians, 0.49 among African Americans,
and 0.51 among Chinese. The ADRB2 27Gln is more frequent in African Americans (0.81)
and Chinese (0.91) compared to Caucasians (0.75). Genetic variation in ADRB2 has been
assessed in ischemic heart disease, specifically in patients with ACS. One investigation
considered the effect of the ADRB1 Arg389Gly, ADRB1 Ser49Gly, ADRB2 Gly16Arg, and
ADRB2 Gly27Glu polymorphisms on survival in patients receiving β-blocker therapy after
ACS.135 This prospective cohort study found that patients who were prescribed β-blocker
therapy after ACS had different survival based on ADRB2 genotype. Patients homozygous
for both the 27Gln and 16Arg alleles, who received β-blocker therapy, had increased 3-year
mortality compared to other genotypes. No association between mortality and ADRB2
genotype was seen in non-β-blocker treated patients; however, the population of non-β-
blocker treated patients was small. This suggests that the polymorphisms are associated
specifically with β-blocker response and not ACS mortality in general. However, without
confirmatory studies, assessing ADRB2 genotype for β-blocker therapy in ACS cannot be
recommended.
It has also been hypothesized that β-blocker response is related to genetic variabil-
ity in drug-metabolizing enzymes. Because CYP2D6 metabolizes carvedilol, metoprolol,
propranolol, labetalol, and timolol and well known polymorphisms in the gene encoding
CYP2D6 lead to variable CYP2D6 activity, there is the potential for altered metabolism.
Altered metabolism could then lead to changes in the efficacy or safety of the β-blocker.
Studies have shown that patients classified as poor metabolizers based on CYP2D6 geno-
type have increased β-blocker concentrations. However, β-blockers have a wide therapeutic
window; the majority of studies found no association between variance in CYP2D6 geno-
type and response to β-blocker therapy.128
CLINICAL PEARL
Studies have shown increased ß-blocker concentrations in patients
classified as poor metabolizers based on CYP2D6 genotype. However,
since ß-blockers have a wide therapeutic window, the majority of studies
have found no association between variants in CYP2D6 genotype and
response to ß-blocker therapy.
Angiotensin-Converting Enzyme Inhibitors and Angiotensin II Receptor

Blockers
ACE inhibitors and ARBs are common agents used in the management of hypertension,
heart failure, and ischemic heart disease. Both agents mitigate the renin-angiotensin-
aldosterone system, albeit by different mechanisms. ACE inhibitors block the step that
prevents the conversion of angiotensin I to angiotensin II. Angiotensin II, through its interac-
tion with the angiotensin type 1 (AT1) receptor, is responsible for the detrimental effects in
cardiac disease such as vasoconstriction, excess aldosterone release, and water retention.
ARBs exert their effects by blocking angiotensin II at the AT1 receptor.
Polymorphisms in three genes are relevant to ACE inhibitor and ARB therapy. The
gene encoding angiotensinogen has a common polymorphism, Met235Thr, which has
been associated with higher angiotensin concentrations and elevated blood pressure.136
The gene encoding ACE also has a common insertion/deletion (I/D) polymorphism, which
occurs in approximately 27% of the population.137 The ACE I/D polymorphism has been
associated with ACE plasma levels. Lastly, variants in the gene encoding the AT1 receptor
(AGTR1) has been studied for association with ARB response.
The ACE I/D polymorphism is one of the most frequently studied polymorphisms in
pharmacogenetics and hypertension. Two large studies have evaluated the association
between this polymorphism and blood pressure response or other outcomes related to
ACE inhibitor therapy, specifically treatment with lisinopril and perindopril.138,139 Both stud-
ies found no association between response to ACE inhibitor therapy and ACE I/D genotype.
Data from these two very large, randomized, controlled trials provide strong evidence that
the ACE I/D polymorphism is not associated with treatment outcomes in hypertension.
One study assessed the role of this polymorphism in treatment with ACE inhibitors in
heart failure. The Genetic Risk Assessment of Cardiac Events (GRACE) study enrolled 479
predominantly Caucasian patients with systolic heart failure.140 They found that the ACE-D
allele was associated with an increased risk of events in the overall population. However,
when analyzed by ACE inhibitor dose, this effect was seen primarily in the low-dose group.
These data have not been further validated in other heart failure studies.
The angiotensinogen polymorphism, Met235Thr, has also been evaluated in hyper-
tension. Possession of the 235Thr allele has been associated with increased angioten-
sinogen levels and blood pressure response to ACE inhibitor therapy.141 However, differing
results have been found in other studies. One study found an association between the
angiotensinogen 235Thr allele and an increased risk of MI in patients receiving ACE inhibi-
tor therapy,142 whereas another study suggested that possession of this same allele was
associated with a lower risk of stroke.143 These discordant data make it difficult to draw a
firm conclusion related to the angiotensinogen Met235Thr polymorphisms. Other related
angiotensinogen polymorphisms have also been evaluated with variable results.
The final relevant gene directly related to ARB therapy is the gene encoding the AT1
receptor (AGTR1). The A1166C SNP has been evaluated related to losartan therapy. Stud-
ies have shown that the 1166C allele is associated with a greater reduction in blood pres-
sure, while other data support an association between the 1166A allele and a decrease in
myocardial stiffness.144,145 However, other studies have found no association between blood
pressure reduction with ARB and ACE inhibitor therapy and this SNP.146,147 Small studies
have also assessed the role of this polymorphism in the response to ARB therapy in heart
failure. Currently, genotyping for this polymorphism in patients with hypertension or heart
failure cannot be recommended.
Diuretics
The use of thiazide diuretics is a standard of care for individuals with uncomplicated hyper-
tension. Thiazide diuretics exert their effect by blocking sodium and chloride reabsorption
in the distal tubule. Several studies assessed adducin, a α/β heterodimeric protein, which
is a cytoskeletal protein that plays a role in cell signal transduction and is associated with
renal sodium reabsorption. Polymorphisms in the gene encoding the α-adducin subunit
(ADD1) have been well studied for their role in response to thiazides in hypertensive indi-
viduals with conflicting results, which limit the clinical utility of ADD1 genotyping to predict
response to diuretic therapy.148-150
In addition, polymorphisms in the genes encoding the β3 subunit of the G protein
gene (GNB3), endothelial nitric oxide synthase (NOS3), ACE, WNK lysine deficient protein
kinase 1 (WNK1), ADRB2, the γ subunit of the nonvoltage-gated sodium channel (SCNNIG),
NEDD4L, and YEATS4 have been found to make small contributions to thiazide-induced
changes in blood pressure.122,151 However, without multiple studies replicating the results
listed above, genotyping for these polymorphisms cannot currently be recommended for
patients receiving diuretic therapy.
Hydralazine/Isosorbide Dinitrate
The combination of hydralazine and isosorbide dinitrate has shown benefit for patients
with heart failure, with early data demonstrating a lower mortality compared to placebo.152
It is believed that the combination of hydralazine and nitrates is effective in heart failure
because nitrates serve as a nitric oxide donor that leads to venodilation, while hydralazine
vasodilates arterial smooth muscle. In addition, hydralazine may have antioxidant proper-
ties that decrease tolerance to nitrates in heart failure patients. However, when compared
to ACE inhibitor therapy, specifically enalapril, the ACE inhibitor was superior in reduc-
ing mortality in patients with heart failure.153 A subsequent subgroup analysis, however,
demonstrated that African-American patients did not have the improved benefit with
enalapril treatment.154 This finding led to the African American Heart Failure Trial (A-HeFT),
which evaluated the use of hydralazine and isosorbide dinitrate in self-identified African-
American patients with heart failure.155 A-HeFT demonstrated a substantial reduction in
mortality in African Americans treated with the hydralazine-nitrate combination when
added to standard therapy (ACE inhibitors and β-blockers).
Genetic variation is one of the possible underlying reasons for this difference in
response to hydralazine and nitrates between Caucasians and African Americans. The
Genetic Risk Assessment of Heart Failure in African Americans (GRAHF) study, a genetic
substudy of A-HeFT, was initiated to assess the genetic differences between these two
racial groups with respect to the development of heart failure and response to hydralazine
and nitrates.156 Patients were genotyped for a promoter region polymorphism (T−344C) in
the gene encoding aldosterone synthase (CYP11B2). Possession of the −344C allele has
been associated with increased aldosterone synthase activity, increased risk of hyperten-
sion, and left ventricular remodeling. In addition, the −344C allele is significantly more
prevalent in Caucasian heart failure patients. In this study of African-American patients,
the −344C allele was associated with poorer hospitalization-free survival and increased
mortality. African-American patients possessing a −344C allele did not receive the same
benefit from the hydralazine and nitrate treatment as those with −344TT genotype did
(which is the most prevalent genotype in African Americans).
Based on the mechanism of action of the hydralazine/nitrate combination and the
difference in prevalence of variants in NOS3 in African Americans compared to Caucasians,
it was hypothesized that NOS3 genotype may predict those who would benefit from this
combination therapy.157 Three NOS3 polymorphisms (−786 T/C promoter, intron 4a/4b,
and Glu298Asp) have been studied. The −786T, 4a, and 298Glu alleles are found more
frequently in African Americans than in Caucasians; however, only the Glu298Asp polymor-
phism influenced treatment outcomes, with the fixed dosed combination of hydralazine-
isosorbide dinitrate improving the composite score and quality of life in the Glu298Glu
subset only. These studies suggest that the CYP11B2 T-344C or the NOS3 Glu298Asp
polymorphisms may be associated with response to hydralazine and nitrates in African
Americans; but, until confirmatory evidence is published from a separate cohort, genotyp-
ing for these SNPs cannot be recommended. In addition, these genetic studies have been
completed only in African Americans. Therefore, it is unclear if SNPs in CYP11B2 or NOS3
are associated with response to hydralazine and nitrate combination therapy in Caucasians.
Recently, the International Consortium for Antihypertensive Pharmacogenomics Studies
(ICAPS) was formed. The goal of ICAPS is to advance the pharmacogenomics of antihyper-
tensive drugs by facilitating collaboration between research groups and by amassing large
sample sizes for GWAS meta-analyses, which increase the likelihood for genetic discover-
ies. This group has done several large-scale studies and is working toward finding novel
pharmacogenomic associations with antihypertensives.122 Large consortia, like this one, are
necessary to find smaller genetic effects or to look at combined effects of multiple genes.
AGENTS FOR DYSLIPIDEMIA

Interpatient variability has been noted with agents used to treat dyslipidemia. Several
studies have sought to characterize the source of this variation. This section addresses the
pharmacogenomic basis for these alterations as well as the relevant clinical data.
Statins
3-Hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins) are
used primarily for lowering low-density lipoprotein cholesterol (LDL-C) levels. In addition,
statins have been shown to decrease morbidity and mortality in patients with and without
known heart disease. However, there is substantial variability in LDL-C lowering with statin
therapy. All statins exert their effect via competitive inhibition of HMG-CoA reductase,
which is the rate-limiting enzyme involved in cholesterol synthesis. For most statins, the
organic anion transporting polypeptides (OATPs) are responsible for transporting the drug
into the hepatocyte where they exert their effect and are metabolized (Table 7-2). However,
the metabolic pathway of each statin is different. Some are metabolized by CYP3A4 or
CYP2C9, while other statins are metabolized by uridine 5′-diphospho-glucuronosyltrans-
ferase sub families (Table 7-2). Genes associated with both the pharmacokinetics and
pharmacodynamics of statins have been studied for their contribution to the variability in
the efficacy and safety of statin therapy.
Statin-Induced Myopathy
The pharmacogenetic interaction with statins that is most well documented has to do with
safety, not efficacy. This class of medications is generally well tolerated but can produce
myopathies, with symptoms ranging from mild myalgias to life-threatening rhabdomyoly-
sis. In clinical trials, the reported incidence of statin-associated myalgias is 3-5%, and the
greatest risk is associated with the use of high-dose statin therapy. Fatal rhabdomyolysis
is rare, occurring in an estimated 1.5 per 10 million prescriptions.158 Although this adverse
event is rare, statins are commonly prescribed medications with recent guidelines recom-
mending them for more patients.4 Therefore, given the volume of prescriptions, the total
number of adverse events with statin therapy is large. However, perhaps the biggest
concern with statin-associated myopathy is its effect on patient adherence to statin ther-
apy. Often patients who experience even mild myopathy discontinue therapy, contributing
to the approximately 50% adherence rate with statins after 1 year.
Table 7-2
Role of Metabolic Enzymes and Transporter Proteins in the
Pharmacokinetics of the Different Statins
Metabolizing CYP450
Statins Enzymes Active Metabolite Transporter Proteins
Atorvastatin 3A4, 3A5 Yes OATP1B1, MDR1

Fluvastatin 2C9 No OATP1B1, MDR1
Lovastatin 3A4, 3A5 Yes OATP1B1, MDR1
Pravastatin None No OATP1B1, MDR1
Rosuvastatin None* Yes OATP1B1
Simvastatin 3A4, 3A5, 2C8 Yes OATP1B1, MDR1
*Approximately 10% of rosuvastatin is metabolized by CYP2C.
MDR1, multidrug resistance gene; OATP, organic anion transporting polypeptide.
In addition to the more common form of statin-induced myopathy, an anti-HMG-CoA

reductase antibody-mediated necrotizing myopathy has been described. This myopathy
persists even after discontinuation of statin therapy and is quite rare, with an incidence of
approximately two per million patients annually.159
The mechanism underlying statin-associated myopathies is unknown but appears to
be at least partially related to increased statin concentrations.158,159 Statin concentrations
are affected by extensive first-pass metabolism involving uptake into hepatocytes and
metabolism by hepatic CYP450 enzymes. Genetic variants in these processes have been
associated with altered statin concentrations and risk for myopathy.159,160
The strongest genetic association with statin-induced myopathy has been docu-
mented with genes affecting the hepatic uptake of statins.159 One of the transporters
involved in statin uptake into hepatocytes is OATP, which is encoded by the SLCO1B1 gene.
OATPs or solute carrier organic anion (SLCO) transporters are vital for drug uptake into
tissues and organ systems. These transporters are found in the liver, intestine, and central
nervous system. All statins, except for fluvastatin, are transported by this mechanism into
hepatocytes.
The first study to demonstrate an association between SLCO1B1 genotype and myop-
athy risk with statin therapy was a genome-wide analysis in participants of the Study of
the Effectiveness of Additional Reductions in Cholesterol and Homocysteine (SEARCH).161
Over 300,000 variants were genotyped in 85 patients who developed confirmed myopa-
thy (cases) and 90 patients who did not develop myopathy (controls) during treatment
with simvastatin 80 mg/day. Only one variant was statistically, significantly associated
with statin-induced myopathy—rs4363657, a noncoding SNP located within the SLCO1B1
gene on chromosome 12. The rs4363657 SNP was in near complete linkage disequilibrium
with the nonsynonymous rs4149056 (T521C, Val174Ala) SNP. The odds ratio for myopathy
was 4.5 (95% CI: 2.6-7.7) with a single rs4149056 C allele and nearly 17 (95% CI: 4.7-61) with
the CC versus TT genotype. The authors also verified these initial findings in a separate
cohort of patients who received simvastatin 40 mg/day as part of the Heart Protection
Study. Although the rs4149056 SNP remained associated with statin-induced myopathy
(OR: 2.6, 95% CI: 1.3-5.0) in this new cohort of patients, the association was not as strong.
The functional consequence of the SNP discovered in the SEARCH study (rs4149056
T521C, Val174Ala) has been well studied. This SNP results in decreased transporter function
in vitro and medication clearance in vivo.162 These alterations result in decreased clearance
of the statin and increased systemic, and potentially muscle, concentrations. The SLCO1B1
haplotypes containing the 521C allele are SLCO1B1*5, *15 and *17.162 The magnitude of effect
on SLCO1B1 function is the same with all of these haplotypes. Given the strong evidence
seen in this study with simvastatin and the understanding of the functional consequence
of this SNP, this SLCO1B1 variant and others have been evaluated with other statins and in
multiple populations.
This SNP and these haplotypes have been studied in multiple geographically, racially,
and ethnically diverse groups; the frequency of the 521C allele is 5–20% in most populations.
Consistent with previous data, in a study of patients receiving atorvastatin, simvastatin, or
pravastatin, the SLCO1B1*5 allele was associated with increased adverse effects from statins,
defined as statin discontinuation for any side effect, myalgia, or creatinine kinase greater
than three times the upper limit of normal.163 The association between the SLCO1B1*5 allele
and statin-induced myopathy was further validated in several additional studies.159,164-166
However, data from two of these studies, which used strict biochemical definitions for
myopathy, suggest the association may be stronger for simvastatin than atorvastatin.165,166
In addition, no association was seen between SLCO1B1 SNPs and myalgia in patients receiv-
ing rosuvastatin.167 Although the association between this SLCO1B1 SNP and myopathy with
simvastatin therapy is quite strong, this does not translate as well to other statins. This
highlights the differences in pharmacokinetic parameters for each of these drugs.
CLINICAL PEARL
The role of SLCO1B1 genotype in statin-induced myopathy was discovered
through GWAS. This highlights the importance of this study design in
discovery of novel mechanisms for adverse events.
However, given the strength of data related to simvastatin myopathy and SLCO1B1
genotype, a set of CPIC guidelines were published in 2012 and updated in 2014.162,168
These guidelines do not make recommendations for when or whom to genotype. Their
recommendations are limited to simvastatin, for which the most data exist. Regardless
of genotype, the simvastatin 80-mg dose should be avoided. For SLCO1B1 C521T allele
heterozygotes (CT genotype), the guidelines recommend using a lower simvastatin dose
(<40 mg/day) or consideration of an alternative statin. For homozygous variant carriers
(CC genotype), either a low simvastatin dose or alternative therapy is recommended. They
specifically recommend pravastatin or rosuvastatin as alternative therapy.
Variants in other transporter genes have also been found to be associated with statin-
induced myopathy. An SNP in ABCC2 was associated with simvastatin discontinuation and
dose reduction.169 There is also a theoretical role for ABCB1 SNPs in statin myopathy. In
addition, some genetic variants in the CYP450 system have been associated with statin-
induced myopathy, but this relates specifically to the metabolic pathway of each statin.170
Beyond variation in the pharmacokinetics of statins, variants in genes responsible for
pharmacodynamics within muscle likely play a role in statin-induced myopathy. Glycine
amidinotransferase (GATM) is a mitochondrial enzyme that is the rate-limiting enzyme in

hepatic creatinine synthesis and a key component of statin-induced myopathy.159,170 The
association of this gene with statin-induced myopathy was discovered in lymphoblastoid
cell lines of statin-exposed individuals.171 The association with statin myopathy was then
validated in two separate cohorts of patients with an odds ratio of 0.6 for the combined
analysis. Other experiments performed suggest that GATM acts as a functional link
between statin-mediated cholesterol lowering and myopathy.
Other genes have also been evaluated for their role in the muscle response to statin
therapy. Specifically, variants in the ATP2B1 and RYR genes have been evaluated in vitro
and have been associated with statin-induced myopathy.159,170 ATP2B1 encodes for the
plasma membrane calcium transporting ATPase1 and plays a critical role in calcium
homeostasis. RYR encodes for the ryanodine receptors, which are involved in calcium
release from the sarcoplasmic reticulum. Other studies have found associations with
variants in DMPK (myotonin-protein kinase gene), ABCB1, ABCG2, and HTR (hydroxytryp-
tamine receptor) with statin-induced myopathy. Additional studies will be necessary to
validate these findings; however, this is an interesting addition to the pharmacogenetics
of statin-induced myopathy.
The role of genetics in antibody-mediated myopathy with statin therapy has been
evaluated, too.159,170,172 This form of myopathy is far less common but does persist even after
statins have been discontinued. The HLA Class II allele DRB1*11:01 was associated with this
unique form of myopathy; however, it is unclear if statin exposure is the sole trigger for this
disease state. Until further evidence is available, genotyping to predict this unique form of
statin myopathy is not recommended.
The body of evidence related to the pharmacogenetics of statin-induced myopathy
will continue to grow. However, currently SLCO1B1 genotyping as it relates to simvastatin
therapy is the only testing that can be recommended.
Statin Efficacy
The metabolic pathways for statins vary, but they all share a uniform mechanism of action.
A recent meta-analysis of GWAS was published.173 The authors performed two steps of
genome-wide analysis in two separate cohorts of patients from randomized, controlled
trials and observational studies of statin therapy. The first and second cohorts included
18,596 and 21,975 patients, respectively. Meta-analysis of the first cohort found three loci
with 13 SNPs that reached genome-wide significance (P < 5 × 10−8) for association with
LDL-C response to statin treatment. The three loci were in the genes encoding apolipo-
protein E (APOE), lipoprotein (a) (LPA), and the rapamycin-insensitive companion of mTOR
(RICTOR). The association with APOE and LPA loci persisted in the second cohort, and two
new loci (SORT1/CELSR2/PSRC1 and SLCO1B1) were detected. The authors also performed
a genome-wide conditional analysis of these polymorphisms to detect combined effects.
They found 14 SNPs that were independently associated with LDL-C response to statin
therapy including those from LPA, APOE, SLCO1B1, and SORT1/CELSR2/PSRC1. Those 14
SNPs explained approximately 5% of the variability in LDL-C response to statin therapy.
The majority of genes identified in this study were associated with statin efficacy previously.
Apolipoproteins are found on lipoprotein particles and facilitate cellular uptake of
these lipoproteins. APOE genetic variation has been well studied for its association with
statin response. There are three common APOE isoforms: ε2, ε3, and ε4. The ε2 allele has
been associated with greater LDL-C reduction with statin therapy in several different types
of studies (candidate gene and GWAS) with several different statins.170,173-179 However, these
results have been inconsistent.170,180 Some studies have proposed an association between
APOE genotype and risk of MI or death; however, these results have also been inconsis-
tent.181 In contrast, APOE polymorphisms are well known to affect baseline lipid levels prior
to initiating statin therapy.159,170,175,182 Thus, it may be difficult to distinguish between the
effect of APOE genotype on baseline lipid levels and statin response.

Lipoprotein (a) is an LDL-like particle that has been well studied for its role as an inde-
pendent predictor of heart disease. Variants in the gene (LPA) encoding the defining protein
of lipoprotein (a), apo (a), have also been well studied in LDL-C response to statin therapy.
Several GWAS have found a change in LDL-C response with statins with the rs10455872
SNP.159,173,176-178 Given the potential importance of LPA in CVD risk, there will be continued
evaluation of this gene.
The effect of SLCO1B1 genotype on statin adverse events has been extensively
discussed; however, variants in this gene are also associated with statin concentrations and
efficacy, and these associations appear to extend beyond simvastatin.183 Genetic analysis of
the T521C variant (rs4149056) found an association between LDL-C lowering with simvas-
tatin, rosuvastatin, and pravastatin.176,177,184,185 These genetic substudies of large clinical trials
found the SNP to be associated with only a small decrease in lipid lowering. This is likely
because the SNP causes decreased transporter activity and, thus, decreased intrahepatic
statin concentrations.
The SORT1/CELSR2/PSRC1 loci was associated with LDL-C response to statins in
several GWAS.177,186,187 The SORT1 gene encodes for sortilin 1 (Sort1) that functions as a
multi-ligand sorting receptor.188 It is believed that Sort1 alters plasma LDL-C and very
low-density lipoprotein (VLDL) particle levels by modulating hepatic VLDL secretion.
This mechanism was discovered after initial GWAS found the association between SORT1
genotype and LDL-C response to statins. Of note, SORT1 polymorphisms have also been
associated with coronary artery disease and MI risk independent of medication therapy in
GWAS.186,189,190 This highlights the value that GWAS provide for novel discovery.
Other genes have also been associated with LDL-C response to statin therapy. Given
that all statins interact with HMG-CoA reductase, the gene encoding it (HMGCR) has been
well studied in statin response. Single gene and GWAS have documented an association
between HMGCR genotype and lipid response to statin therapy.174,191,192 However, these
results have not been consistently replicated.159,170 There were similar results with the LDL-
receptor; although some studies have seen significant associations with statin-induced
LDL-C effects, they have not been consistent.170 A recent GWAS of pravastatin-treated
patients from multiple clinical trials discovered a new locus associated with coronary heart
disease reduction with pravastatin. The gene involved is DNAJC5B that encodes for DnaJ
homologue subfamily C member 5B, and the SNP is rs13279522. However, the functional
role of the protein and SNP in dyslipidemia and heart disease is unknown, especially given
that the investigators did not detect a difference in lipid levels with the polymorphism.
The authors hypothesize that this gene is potentially related to the pleiotropic effects of
statins, but this question remains unanswered.

Another well-studied gene for its role in statin response is the gene for kinesin-like
protein 6 (KIF6), a kinesin involved in intracellular transport. The data for KIF6 genotype and
its association with statin efficacy have also been inconsistent. Initial studies showed an
association between the KIF6 Trp719Arg polymorphism and clinical outcomes with statins.
Specifically, the 719Arg allele was associated with both an increased risk for coronary
events and a greater reduction in coronary event risk with pravastatin in the retrospective
analyses of the secondary prevention study in patients from three clinical trials.193 Similar
findings were observed in two other studies.194,195 Based on these data, a commercially
available assay for KIF6 genotyping was developed and marketed to predict risk for cardiac
events and response to statins. However, subsequent studies with rosuvastatin and simv-
astatin revealed no association between KIF6 genotype and either risk for coronary events
or response to statins.196,197 Potential contributors to the inconsistencies in the data with
KIF6 include variation among studies in the statin used, study population, and trial design.
Nonetheless, the clinical utility of using KIF6 genotype as a marker of vascular risk for
statin response is questionable given the inconsistencies in the data. This also highlights
the importance of understanding the underlying physiology of a gene and the contribution
of genetic variation to a disease or drug response. Both KIF6 and DNAJC5B were novel
discoveries from GWAS, but both need significant further study to understand their true
role in the disease.
Even though statins all share the same mechanism of action, each has different pharma-
cokinetic properties (Table 7-2). CYP3A4 plays an important role in the metabolism of lovas-
tatin, simvastatin, and atorvastatin, while fluvastatin is metabolized primarily by CYP2C9.198
Pravastatin and rosuvastatin are primarily eliminated as unchanged parent compounds in
the feces and the urine. Thus, variability in the genes encoding CYP450 enzymes does not
uniformly affect statins as a class. In addition, although studies have linked polymorphisms
in genes encoding CYP450 enzymes with changes in statin pharmacokinetic parameters,
the clinical meaning of these pharmacokinetic changes is unknown.159,170 Increased serum
concentrations of statins have been linked to adverse events, but they have not been
linked to increased lipid lowering or better efficacy. Therefore, studies assessing the effect
of CYP450 polymorphisms on statin lipid-lowering effects have been inconsistent. Currently,
there does not appear to be a role for CYP450 enzyme genotyping in patients receiving
statin therapy. However, knowledge of the genetic variability in a patient’s drug-metabolizing
enzymes could possibly allow clinicians to better predict the effect of drug interactions
on statin therapy. Theoretically, patients who do not express certain CYP450 enzymes or
express deficient enzymes may be more susceptible to drug interactions via enzyme inhibi-
tion with statin therapy. These drug interactions can lead to increased statin concentrations
and perhaps adverse events. Few data exist related to the role of CYP450 genetic variability
and statin drug interactions currently. Nevertheless, pharmacists could play a role in utilizing
genetic information related to drug-metabolizing enzymes to prevent clinically significant
consequences associated with statin drug interactions.
In contrast, drug transporters have been associated with statin efficacy. Both SLCO1B1
and ABCG2 genotypes have been associated with the lipid-lowering effect of multiple
statins. As stated previously, SLCO1B1 genotype is associated with LDL-C lowering with
statin therapy. Similarly a genetic variant, C421A, in the efflux transporter ABCG2 has been
associated with increased systemic exposure and lipid-lowering effects with rosuvas-
tatin.176,199 This is also likely due to increased hepatic exposure to the drug. Similar results
have been seen with ABCB1.175,179 The association with transporters and statin response
appears to be more consistent than those with variants in the CYP450 system.
Given the complexities of pharmacogenetic prediction of statin efficacy and safety,
DeGorter et al. conducted a study to assess the relationship between drug transporter
polymorphisms and interindividual variability in rosuvastatin and atorvastatin concentra-

tions.200 The authors found a greater than 45-fold variability in plasma rosuvastatin and
atorvastatin concentrations among individuals receiving the same dose. Increased rosuv-
astatin concentrations were seen in individuals with the SLCO1B1 521C allele, the ABCG2
421A allele, and increased age. SLCO1B1 and ABCG2 genotypes contributed to 82% of the
explainable variability seen in the model. In contrast, two SLCO1B1 polymorphisms and age
were associated with atorvastatin concentrations, with the genetic factors explaining only
38% of the variability. In addition, the authors found no association between plasma statin
concentration and achievement of LDL-C target. This further supports the idea that plasma
statin concentrations may not affect lipid lowering but contribute to risk of adverse events.
The authors then proposed dosing algorithms for atorvastatin and rosuvastatin with
the goal of obtaining plasma concentrations that remain lower than the 90th percentile,
assuming that 10% of individuals experience statin-related complications. In a group of 16
patients on 80 mg of atorvastatin daily, nine were predicted to exceed the 90th percentile
based on the algorithm. Of those 16 patients, only two of the nine patients predicted to
exceed the 90th percentile remained on atorvastatin 80 mg 1 year later, whereas all seven
predicted to tolerate the high dose were still on it 1 year later. These results were not
statistically significant, but it highlights the importance of incorporating multiple genes
and clinical factors into the clinical evaluation process. Although these results are promis-
ing, they should not be incorporated into clinical practice until they have been replicated.
The pharmacogenetic data assessing statin response is complicated for several
reasons. Statins differ in both their pharmacokinetic properties and pharmacodynamic
potency and, thus, data from a study on one statin cannot be extrapolated to the class as
highlighted by DeGorter and colleagues.200 In addition, much of the data available on statin
pharmacogenomics comes from genotyping patients from completed statin clinical trials.
Each trial has unique inclusion criteria and study end-points. Therefore, it is difficult to
replicate the pharmacogenetic interactions. Numerous genes probably mediate the role of
statins in lipid lowering. In addition, CHD is a complex disease state likely involving multiple
genetic pathways. In the future, a pharmacogenetic method of predicting statin response
will likely involve genotyping multiple polymorphisms in multiple genes. Until there are
data available for this technique, utilizing genotype to predict statin response cannot be
recommended.
Ezetimibe
Ezetimibe is also utilized for lowering LDL-C concentration. Ezetimibe lowers serum LDL-C
levels by blocking the Niemann-Pick C1-like 1 (NPC1L1) intestinal cholesterol transporter.
The first genetic association report for ezetimibe was in a resistant patient who was
found to have rare nonsynonymous NPC1L1 gene mutations.201 The gene was then further
sequenced in additional patients, and 140 SNPs and five I/D polymorphisms were identified.
Two studies have assessed the association between NPC1L1 genotype and LDL-C
response to ezetimibe. The first study found a haplotype, made up of three SNPs, to be
associated with percent LDL-C reduction from baseline.202 In fact, those patients possess-
ing at least one copy of the NPC1L1 haplotype alleles studied (1735C, 25342A, 27677T)
had smaller LDL-C reduction from baseline with ezetimibe (−23.6 ± 1.6% versus −35.9
± 4.0, p = 0.0054). The second study also used three NPC1L1 SNPs to create haplotype
groups; however, they utilized different SNPs from the previous study.203 They found that
possession of the haplotype −133A, −18A, and 1679G was associated with increased LDL-C
lowering. However, because each study found different NPC1L1 SNPs and haplotypes to be
associated with ezetimibe response, it is as yet unclear which polymorphism or haplotype
may actually be leading to the altered LDL-C lowering response. In addition, there were
impressive racial differences in the allele frequencies for the SNPs studied.
Other SNPs in NPC1L1 have been associated with baseline cholesterol absorption and
lipid profile.204,205 A recent study re-sequenced the exons of NPC1L1 in 7,364 patients with
CHD and 14,728 controls of varied ancestry. These authors identified 15 distinct variants.
Heterozygous carriers of inactivating mutations had a mean LDL-C level that was 12 mg/
dL lower than noncarriers, which was a statistically significant difference. In addition, carrier
status was associated with a relative reduction in CHD risk of 53%. This makes it difficult to
distinguish between the baseline and pharmacogenetic effects of these variants. Another
set of investigators discovered similar associations between NPC1L1 genotype and cardio-
vascular events. This association persisted after controlling for total cholesterol, LDL-C, and
other cardiovascular risk factors.206 At this time, regular genotyping for NPC1L1 polymor-
phisms to predict ezetimibe response cannot be recommended. In addition, as discussed
with statins, lipid homeostasis involves several pathways with many different genes. There-
fore, a polygenetic approach will likely be necessary to assess ezetimibe response.
ANTIARRHYTHMIC AGENTS
Antiarrhythmic agents in general have narrow therapeutic windows and are often highly
susceptible to drug-drug interactions. In addition, some arrhythmias can lead to death,
especially if they are not treated appropriately. All of this has led to significant study of the
pharmacogenomics of antiarrhythmic therapy to improve response and decrease adverse
events.
Procainamide
Procainamide is a class Ia antiarrhythmic used for the treatment of several different arrhyth-
mias. It is metabolized by N-acetyltransferases (NAT) to N-acetylprocainamide (NAPA),
an active metabolite that also possesses antiarrhythmic properties. NAT1 is consistently
expressed in most patients; however, NAT2 is variably expressed and plays a major role in
the production of NAPA. The majority of patients receiving procainamide therapy will devel-
op autoantibodies over time and possibly drug-induced lupus. An early study showed that
patients who had a slow-acetylator phenotype developed antinuclear antibodies earlier
than rapid acetylators.207 This was confirmed in a later study of patients who received long-
term procainamide therapy.208 However, these authors found that acetylator status was
not associated with the risk of developing drug-related lupus, only with increased antibody
formation. Also, in theory, patients with NAT2 genotypes associated with rapid acetylation
may have increased NAPA concentrations and thus increased anti-arrhythmic effects and
possible excessive QT prolongation. However, this has not been documented in the litera-
ture. Although NAT2 genotyping may predict the rate at which patients will develop auto-
antibodies with procainamide therapy, the clinical utility of this is unclear.
Propafenone
Propafenone is a class Ic antiarrhythmic medication used for the treatment of ventricular
arrhythmias. Propafenone exerts its effects by blocking the fast inward sodium current and
has some β-receptor blocking properties at higher concentrations. Propafenone is primar-

ily metabolized by CYP2D6, while CYP1A2 and CYP3A4 also contribute to its metabolism.
Genetic classification of CYP2D6 activity is complex; however, patients are generally clas-
sified as poor metabolizers, normal metabolizers, and ultra-rapid metabolizers. CYP2D6
functional status can be determined via genotyping or phenotyping. Approximately 5–10%
of Caucasians and African Americans are considered to be poor metabolizers and essen-
tially make no active CYP2D6.209 Patients who are classified as CYP2D6 poor metabolizers
have decreased propafenone clearance. This decrease in clearance leads to an increase in
propafenone serum concentrations, which has more β-blocking properties than its active
metabolites. The additional β-blockade seen in CYP2D6 poor metabolizers can lead to
clinically significant adverse events in patients with asthma due to effects on β2 receptors
in the lungs. Also, patients with paroxysmal atrial fibrillation classified as CYP2D6 poor
metabolizers have been found more likely to maintain normal sinus rhythm with propafe-
none than normal metabolizers.210 However, the data are inconsistent.
Another more recent study assessed the role of propafenone in the prevention of
tachyarrhythmias after cardiac surgery.211 These authors found a significant association
between propafenone pharmacokinetics and CYP2D6 metabolizer status based on geno-
type. However, CYP2D6 metabolizer status was not associated with arrhythmia risk in
patients receiving propafenone. Although it is well documented that CYP2D6 genotype,
and thus metabolizer status, does affect the pharmacokinetics of propafenone, it is unclear
if this translates into differences in clinical outcomes. For this reason, routine CYP2D6
genotyping cannot be recommended for patients receiving propafenone therapy.
Flecainide
Flecainide is also a class Ic antiarrhythmic used for both supraventricular and ventricular
arrhythmias. Flecainide undergoes hepatic and renal clearance. Like propafenone, CYP2D6
is the primary enzyme responsible for the metabolism of flecainide. Therapy with flecainide
is challenging because there is substantial interpatient variability in its pharmacokinetics,
and increased concentrations can lead to excessive widening of the QRS interval. Although
therapeutic drug monitoring can be done for flecainide, it is possible that assessing metab-
olizer status for CYP2D6 may help to predict a patient’s dose response to flecainide. Phar-
macokinetic studies with flecainide have used both genotyping and phenotyping methods
to classify CYP2D6 metabolizer status. Two of the studies, which used genotyping to
assess CYP2D6 metabolizer status, demonstrated an association between CYP2D6 geno-
type and flecainide pharmacokinetics, with poor metabolizers having decreased flecainide
clearance.212,213 Inclusion of CYP2D6 genotype, body weight, age, sex, and serum creatinine
explained approximately 50% of the variability in flecainide pharmacokinetics. However,
another study utilized phenotyping to classify CYP2D6 status and did not find an associa-
tion between CYP2D6 metabolizer status and flecainide pharmacokinetics.214 It is possible
that CYP2D6 contributes to some of the variability seen in flecainide dose response;
however, the data are inconsistent. At this time CYP2D6 genotyping or phenotyping cannot
be recommended to predict flecainide dosing.
Amiodarone
Amiodarone is a class III antiarrhythmic that is effective for both supraventricular and
ventricular arrhythmias. Amiodarone metabolism is extensive with involvement of both
phases I and II drug-metabolizing enzymes. CYP3A4 and CYP2C8 are known to play a key
role in amiodarone metabolism. In addition, amiodarone is a potent inhibitor of CYP3A4,

CYP1A2, CYP2C9, and CYP2D6 leading to multiple drug interactions. Although genetic
variation in any of these metabolic enzymes may lead to alterations in amiodarone clear-
ance or its propensity for drug interactions, there is a scarcity of data to support it. At this
time, genotyping for alterations in CYP450 isoenzymes to predict amiodarone response
cannot be recommended.
Digoxin
Digoxin exerts its effects by inhibiting sodium-potassium ATP and is used in both atrial
fibrillation and heart failure. Digoxin is an ABCB1 (also known as P-gp) substrate, and the
role of ABCB1 polymorphisms in digoxin pharmacokinetics has been well studied. In fact,
the first drug documented to be affected by ABCB1 polymorphisms was digoxin. The most
studied polymorphism is the C3435T SNP.215,216 Several studies have demonstrated an
association between digoxin pharmacokinetics and ABCB1 polymorphisms; however, the
results have been very inconsistent. Although ABCB1 is an important consideration for
drug interactions with digoxin; at this time, genotyping for ABCB1 polymorphisms to predict
digoxin pharmacokinetics cannot be recommended. It is also possible that polymorphisms
related to digoxin’s pharmacodynamics (i.e., sodium-potassium ATPase, which is digoxin’s
site of action) might affect digoxin response.209 However, there are currently no data to
support this.
PRO-ARRHYTHMIC EFFECTS OF MEDICATION

The pro-arrhythmic effects of medications (both those used to treat arrhythmias and those
used for other indications) have been well studied. Interest in these rare adverse events is
warranted because they can be life threatening and require a significant amount of patient
monitoring. Proarrhythmia is generally defined as the worsening of the arrhythmia being
treated or generation of a new arrhythmia with medication therapy.217,218 Genetic studies
have focused on drug-induced increases in the QT interval and drug-induced torsade de
pointes. The focus on genetic factors associated with drug-induced prolonged QT inter-
vals is due to the knowledge gained from studying and evaluating congenital long QT
syndromes. In addition, the association between prolonged QT interval and torsade de
pointes has been well documented.
The QT interval on an electrocardiogram represents the action potentials of ventricu-
lar myocytes. The ventricular action potential is made up of several currents produced by
different ion channels. The action potential is prolonged when there is either increased
inward current or decreased outward current. The heart has significant built-in redundancy
because several ion channels participate in the ventricular action potential. This redun-
dancy is termed repolarization reserve. Thus, variation in one ion channel will not neces-
sarily lead to an increase in the QT interval. A combination of factors is generally neces-
sary for patients to exhibit both congenital and drug-induced long QT syndromes. Several
clinical factors, beyond genetic variation in ion channels, have been associated with an
increased risk of drug-induced long QT syndromes, and it is thought that these factors
affect the heart’s repolarization reserve.217 These factors include but are not limited to
hypokalemia, recent conversion of atrial fibrillation, advanced heart disease, and female
gender.170
Factors influencing the pharmacokinetics of medications may also affect the risk of
drug-induced long QT syndrome. The risk of drug-induced long QT syndrome may be
increased if the clearance of a drug is decreased via either a drug interaction or genetic
variants in hepatic enzyme systems. Pharmacists should be particularly vigilant in monitor-
ing for drug interactions with medications known to prolong the QT interval. In addition, if
genetic variability in hepatic enzyme systems for a patient is known, this should be consid-
ered as well.
Polymorphisms in several of the genes encoding ion channels associated with ventricu-
lar action potential have been studied for their role in congenital long QT syndromes
because of their possible effect on repolarization reserve.218,219 Mutations found in genes
encoding potassium voltage-gated channels (KCNQ1, KCNH2, KCNE1, KCNE2) and in a sodium
voltage-gated channel (SCN5A) have been associated with risk for congenital long QT
syndrome.218,220 Because the ion channels have been associated with congenital long QT
syndrome, they are a logical starting place for assessing drug-induced long QT syndrome.
Medications can prolong the QT interval by blocking the ion channel pore, inducing confor-
mational changes in the ion channel pore, and/or decreasing production of the proteins
encoding the ion channels. The amino acid structure of KCNH2 appears to make this ion
channel pore particularly susceptible to drug blockade. Polymorphisms in the gene encod-
ing KCNH2 may affect its susceptibility to drug binding.219 Polymorphisms in SCN5A may
also contribute to the risk of drug-induced long QT syndrome.219 Rare variants in these chan-
nels (KCNE1, KCNE2, and SCN5A) have all been associated with risk of drug-induced torsade
de pointes.218,220,221 In addition, recent studies using next-generation sequencing are verifying
the role of rare variants in the risk for drug-induced torsades de pointes.222 However, clini-
cians should remember that studying drug-induced torsades de pointes is difficult to do.
There is no standard phenotype to characterize patients with prolonged QT or drug-induced
torsade de pointes, and rare variants require large populations to discover and confirm. Prior
to utilizing pharmacogenetic information to predict proarrhythmic risk with medications, a
genetic variant must be validated in separate, distinct populations. In addition, an under-
standing of the biology and physiology of the polymorphism is necessary.
Currently using genetic information to predict pro-arrhythmia risk cannot be recom-
mended. The volume of knowledge on this topic is growing and with validated genetic
markers, genotyping may be clinically useful in the future. However, utilizing genetic infor-
mation to predict risk of drug-induced long QT syndrome will likely be complex due to the
repolarization reserve. It is unlikely that polymorphisms in a single gene or a single clinical
risk factor will be sufficient to predict risk. Therefore, a pharmacogenomic approach to
predicting risk for drug-induced long QT syndrome will be necessary where multiple genes
are assessed along with clinical factors.
SUMMARY
The study of pharmacogenomics in the area of cardiology is growing exponentially. Data
are currently available to support prospective genotyping for warfarin therapy. In addi-
tion, well-characterized genetic data are available for bucindolol, statins, and clopidogrel.
It is only a matter of time before this information is used clinically. Furthermore, with the
continuing growth of literature in this area, our clinical use of genetics in cardiovascular
drug therapy is likely to expand.
REFERENCES
1. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart Disease and Stroke Statistics—2016 Update Report
from the American Heart Association. Circulation. December 2015:CIR.0000000000000350.
doi:10.1161/CIR.0000000000000350.
2. James PA, Oparil S, Carter BL, et al. 2014 Evidence-Based Guideline for the Management of High
Blood Pressure in Adults: Report from the panel members appointed to the Eighth Joint National
Committee (JNC 8). JAMA. 2014;311(5):507-520. doi:10.1001/jama.2013.284427.
3. Yancy CW, Jessup M, Bozkurt B, et al. 2013 ACCF/AHA Guideline for the Management of Heart Failure:
A Report of the American College of Cardiology Foundation/American Heart Association Task Force
on Practice Guidelines. Circulation. 2013;128(16):e240-e327. doi:10.1161/CIR.0b013e31829e8776.
4. Stone NJ, Robinson J, Lichtenstein AH, et al. 2013 ACC/AHA Guideline on the Treatment of Blood
Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults: A Report of the American
College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation.
November 2013:01.cir.0000437738.63853.7a. doi:10.1161/01.cir.0000437738.63853.7a.
5. Scott SA, Khasawneh R, Peter I, et al. Combined CYP2C9, VKORC1 and CYP4F2 frequencies among
racial and ethnic groups. Pharmacogenomics. 2010;11(6):781-791. doi:10.2217/pgs.10.49.
6. Scordo MG, Pengo V, Spina E, et al. Influence of CYP2C9 and CYP2C19 genetic polymorphisms on
warfarin maintenance dose and metabolic clearance. Clin Pharmacol Ther. 2002;72(6):702-710.
7. Takahashi H, Kashima T, Nomizo Y, et al. Metabolism of warfarin enantiomers in Japanese patients
with heart disease having different CYP2C9 and CYP2C19 genotypes. Clin Pharmacol Ther.
1998;63(5):519-528. doi:10.1016/S0009-9236(98)90103-5.
8. Lindh JD, Holm L, Andersson ML, et al. Influence of CYP2C9 genotype on warfarin dose require-
ments—a systematic review and meta-analysis. Eur J Clin Pharmacol. 2009;65(4):365-375. doi:10.1007/
s00228-008-0584-5.
9. Higashi MK, Veenstra DL, Kondo LM, et al. Association between CYP2C9 genetic variants and anticoag-
ulation-related outcomes during warfarin therapy. JAMA. 2002;287(13):1690-1698.
10. Sanderson S, Emery J, Higgins J. CYP2C9 gene variants, drug dose, and bleeding risk in warfarin-
treated patients: a HuGEnet systematic review and meta-analysis. Genet Med Off J Am Coll Med
Genet. 2005;7(2):97-104. doi:10.109701.GIM.0000153664.65759.CF.
11. Dickmann LJ, Rettie AE, Kneller MB, et al. Identification and functional characterization of a new
CYP2C9 variant (CYP2C9*5) expressed among African Americans. Mol Pharmacol. 2001;60(2):382-387.
doi:10.1124/mol.60.2.382.
12. Allabi AC, Gala J-L, Horsmans Y. CYP2C9, CYP2C19, ABCB1 (MDR1) genetic polymorphisms and phenyt-
oin metabolism in a black Beninese population. Pharmacogenet Genomics. 2005;15(11):779-786.
13. Allabi AC, Gala JL, Horsmans Y, et al. Functional impact of CYP2C95, CYP2C96, CYP2C98, and
CYP2C911 in vivo among black Africans. Clin Pharmacol Ther. 2004;76(2):113-118.
14. Liu Y, Jeong H, Takahashi H, et al. Decreased warfarin clearance associated with the CYP2C9 R150H
(*8) polymorphism. Clin Pharmacol Ther. 2012;91(4):660-665. doi:10.1038/clpt.2011.269.
15. Blaisdell J, Jorge-Nebert LF, Coulter S, et al. Discovery of new potentially defective alleles of human
CYP2C9. Pharmacogenetics. 2004;14(8):527-537.
16. Perera M, Gamazon E, Cavallari L, et al. The missing association: sequencing-based discovery of novel
SNPs in VKORC1 and CYP2C9 that affect warfarin dose in African Americans. Clin Pharmacol Ther.
2011;89(3):408-415. doi:10.1038/clpt.2010.322.
17. Cavallari LH, Langaee TY, Momary KM, et al. Genetic and clinical predictors of warfarin dose require-
ments in African Americans. Clin Pharmacol Ther. 2010;87(4):459-464.
18. Limdi NA, Arnett DK, Goldstein JA, et al. Influence of CYP2C9 and VKORC1 on warfarin dose, anti-
coagulation attainment and maintenance among European-Americans and African-Americans. Phar-
macogenomics. 2008;9(5):511-526.
19. Margaglione M, Colaizzo D, D’Andrea G, et al. Genetic modulation of oral anticoagulation with warfa-
rin. Thromb Haemost. 2000;84(5):775-778.

20. Limdi NA, McGwin G, Goldstein JA, et al. Influence of CYP2C9 and VKORC1 1173C/T genotype on the
risk of hemorrhagic complications in African-American and European-American patients on warfarin.
Clin Pharmacol Ther. 2008;83(2):312-321.
warfarin dose. N Engl J Med. 2005;352(22):2285-2293.
22. Huang S-W, Chen H-S, Wang X-Q, et al. Validation of VKORC1 and CYP2C9 genotypes on inter-
individual warfarin maintenance dose: a prospective study in Chinese patients. Pharmacogenet
Genomics. 2009;19(3):226-234. doi:10.1097/FPC.0b013e328326e0c7.
23. Wadelius M, Chen LY, Lindh JD, et al. The largest prospective warfarin-treated cohort supports genetic
forecasting. Blood. 2009;113(4):784-792.
24. Limdi NA, Wadelius M, Cavallari L, et al. Warfarin pharmacogenetics: a single VKORC1 polymorphism
is predictive of dose across 3 racial groups. Blood. 2010;115(18):3827-3834. doi:10.1182/blood-2009-12-

255992.
25. Gage BF, Eby C, Milligan PE, et al. Use of pharmacogenetics and clinical factors to predict the mainte-
nance dose of warfarin. Thromb Haemost. 2004;91(1):87-94.
26. Consortium IWP, Klein TE, Altman RB, et al. Estimation of the warfarin dose with clinical and pharma-
cogenetic data. N Engl J Med. 2009;360(8):753-764.
27. McDonald MG, Rieder MJ, Nakano M, et al. CYP4F2 Is a vitamin K1 oxidase: an explanation for altered
warfarin dose in carriers of the V433M variant. Mol Pharmacol. 2009;75(6):1337-1346. doi:10.1124/
mol.109.054833.
28. Caldwell MD, Awad T, Johnson JA, et al. CYP4F2 genetic variant alters required warfarin dose. Blood.
2008;111(8):4106-4112. doi:10.1182/blood-2007-11-122010.
29. Pautas E, Moreau C, Gouin-Thibault I, et al. Genetic factors (VKORC1, CYP2C9, EPHX1, and CYP4F2)
are predictor variables for warfarin response in very elderly, frail inpatients. Clin Pharmacol Ther.
2010;87(1):57-64. doi:10.1038/clpt.2009.178.
30. Choi JR, Kim J-O, Kang DR, et al. Proposal of pharmacogenetics-based warfarin dosing algorithm in
Korean patients. J Hum Genet. 2011;56(4):290-295. doi:10.1038/jhg.2011.4.
31. Takeuchi F, McGinnis R, Bourgeois S, et al. A genome-wide association study confirms VKORC1,
CYP2C9, and CYP4F2 as principal genetic determinants of warfarin dose. PLoS Genet.
2009;5(3):e1000433.
32. Cha P-C, Mushiroda T, Takahashi A, et al. Genome-wide association study identifies genetic determi-
nants of warfarin responsiveness for Japanese. Hum Mol Genet. 2010;19(23):4735-4744. doi:10.1093/
hmg/ddq389.
33. Suriapranata IM, Tjong WY, Wang T, et al. Genetic factors associated with patient-specific warfarin
dose in ethnic Indonesians. BMC Med Genet. 2011;12:80. doi:10.1186/1471-2350-12-80.
34. Shahin MHA, Khalifa SI, Gong Y, et al. Genetic and nongenetic factors associated with warfarin
dose requirements in Egyptian patients. Pharmacogenet Genomics. 2011;21(3):130-135. doi:10.1097/
FPC.0b013e3283436b86.
35. Biss TT, Avery PJ, Brandão LR, et al. VKORC1 and CYP2C9 genotype and patient characteristics
explain a large proportion of the variability in warfarin dose requirement among children. Blood.
2012;119(3):868-873. doi:10.1182/blood-2011-08-372722.
36. King CR, Deych E, Milligan P, et al. Gamma-glutamyl carboxylase and its influence on warfarin dose.
Thromb Haemost. 2010;104(4):750-754. doi:10.1160/TH09-11-0763.
37. Wallin R, Hutson SM, Cain D, et al. A molecular mechanism for genetic warfarin resistance in the rat.
FASEB J. September 2001;15:2542-2544. doi:10.1096/fj.01-0337fje.
38. Voora D, Koboldt DC, King CR, et al. A polymorphism in the VKORC1 regulator calumenin predicts
higher warfarin dose requirements in African Americans. Clin Pharmacol Ther. 2010;87(4):445-451.
doi:10.1038/clpt.2009.291.
39. Caldwell MD, Berg RL, Zhang KQ, et al. Evaluation of genetic factors for warfarin dose prediction. Clin
Med Res. 2007;5(1):8-16.
40. Sconce EA, Khan TI, Wynne HA, et al. The impact of CYP2C9 and VKORC1 genetic polymorphism and
patient characteristics upon warfarin dose requirements: proposal for a new dosing regimen. Blood.
2005;106(7):2329-2333.
41. Sagreiya H, Sagrieya H, Berube C, et al. Extending and evaluating a warfarin dosing algorithm that
includes CYP4F2 and pooled rare variants of CYP2C9. Pharmacogenet Genomics. 2010;20(7):407-413.
doi:10.1097/FPC.0b013e328338bac2.
42. Schelleman H, Chen J, Chen Z, et al. Dosing algorithms to predict warfarin maintenance dose
in Caucasians and African Americans. Clin Pharmacol Ther. 2008;84(3):332-339. doi:10.1038/
clpt.2008.101.
43. Owen RP, Altman RB, Klein TE. PharmGKB and the International Warfarin Pharmacogenetics Consor-
tium: the changing role for pharmacogenomic databases and single-drug pharmacogenetics. Hum
Mutat. 2008;29(4):456-460. doi:10.1002/humu.20731.
44. Squibb B-M. Warfarin (Coumadin®) package insert. Princeton, NJ; August 2007.
45. Roper N, Storer B, Bona R, et al. Validation and comparison of pharmacogenetics-based warfa-
rin dosing algorithms for application of pharmacogenetic testing. J Mol Diagn. 2010;12(3):283-291.
doi:10.2353/jmoldx.2010.090110.
46. Shin J, Cao D. Comparison of warfarin pharmacogenetic dosing algorithms in a racially diverse large
cohort. Pharmacogenomics. 2010;12(1):125-134. doi:10.2217/pgs.10.168.
47. Langley MR, Booker JK, Evans JP, et al. Validation of clinical testing for warfarin sensitivity. J Mol
Diagn. 2009;11(3):216-225. doi:10.2353/jmoldx.2009.080123.
48. Kimmel SE, French B, Kasner SE, et al. A pharmacogenetic versus a clinical algorithm for warfarin
dosing. N Engl J Med. 2013;369(24):2283-2293. doi:10.1056/NEJMoa1310669.
49. Pirmohamed M, Burnside G, Eriksson N, et al. A randomized trial of genotype-guided dosing of warfa-
rin. N Engl J Med. 2013;369(24):2294-2303. doi:10.1056/NEJMoa1311386.
50. Verhoef TI, Ragia G, de Boer A, et al. A randomized trial of genotype-guided dosing of acenocoumarol
and phenprocoumon. N Engl J Med. 2013;369(24):2304-2312. doi:10.1056/NEJMoa1311388.
51. Ageno W, Gallus AS, Wittkowsky A, et al. Oral anticoagulant therapy: antithrombotic therapy and
prevention of thrombosis, 9th ed.: American College of Chest Physicians Evidence-Based Clinical
Practice Guidelines. Chest. 2012;141(2 Suppl):e44S-88S.
52. Johnson JA, Gong L, Whirl-Carrillo M, et al. Clinical pharmacogenetics implementation consor-
tium guidelines for CYP2C9 and VKORC1 genotypes and warfarin dosing. Clin Pharmacol Ther.
2011;90(4):625-629. doi:10.1038/clpt.2011.185.
53. Do EJ, Lenzini P, Eby CS, et al. Genetics informatics trial (GIFT) of warfarin to prevent deep vein throm-
bosis (DVT): rationale and study design. Pharmacogenomics J. 2012;12(5):417-424. doi:10.1038/tpj.2011.18.
54. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation.
N Engl J Med. 2009;361(12):1139-1151.
55. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of anti-
platelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ.
2002;324(7329):71-86.
56. Faraday N, Yanek LR, Mathias R, et al. Heritability of platelet responsiveness to aspirin in activation
pathways directly and indirectly related to cyclooxygenase-1. Circulation. 2007;115(19):2490-2496.
57. Shen H, Herzog W, Drolet M, et al. Aspirin resistance in healthy drug-naive men versus women (from
the Heredity and Phenotype Intervention Heart Study). Am J Cardiol. 2009;104(4):606-612. doi:10.1016/j.
amjcard.2009.04.027.
58. Maree AO, Curtin RJ, Chubb A, et al. Cyclooxygenase-1 haplotype modulates platelet response to
aspirin. J Thromb Haemost. 2005;3(10):2340-2345.
59. Lepantalo A, Mikkelsson J, Resendiz JC, et al. Polymorphisms of COX-1 and GPVI associate with the
antiplatelet effect of aspirin in coronary artery disease patients. Thromb Haemost. 2006;95(2):253-
259.
60. Hillarp A, Palmqvist B, Lethagen S, et al. Mutations within the cyclooxygenase-1 gene in aspirin non-
responders with recurrence of stroke. Thromb Res. 2003;112(5-6):275-283.
61. Clappers N, Oijen MG van, Sundaresan S, et al. The C50T polymorphism of the cyclooxygenase-1
gene and the risk of thrombotic events during low-dose therapy with acetyl salicylic acid. Thromb
Haemost. 2008;100(1):70-75.
62. Faraday N, Yanek LR, Yang XP, et al. Identification of a specific intronic PEAR1 gene variant associated
with greater platelet aggregability and protein expression. Blood. 2011;118(12):3367-3375. doi:10.1182/
blood-2010-11-320788.
63. Lewis JP, Ryan K, O’Connell JR, et al. Genetic variation in PEAR1 is associated with platelet aggrega-
tion and cardiovascular outcomes. Circ Cardiovasc Genet. 2013;6(2):184-192. doi:10.1161/CIRCGENET-

ICS.111.964627.
64. Gurbel PA, Bliden KP, Hiatt BL, et al. Clopidogrel for coronary stenting response variability, drug
resistance, and the effect of pretreatment platelet reactivity. Circulation. 2003;107(23):2908-2913.
doi:10.1161/01.CIR.0000072771.11429.83.
65. Gurbel PA, Cummings CC, Bell CR, et al. Onset and extent of platelet inhibition by clopidogrel loading
in patients undergoing elective coronary stenting: the Plavix reduction of new thrombus occurrence
(PRONTO) trial. Am Heart J. 2003;145(2):239-247. doi:10.1067/mhj.2003.109.
66. Quinn MJ, Topol EJ. Common variations in platelet glycoproteins: pharmacogenomic implications.
67. Bura A, Bachelot-Loza C, Ali FD, et al. Role of the P2Y12 gene polymorphism in platelet responsive-
ness to clopidogrel in healthy subjects. J Thromb Haemost. 2006;4(9):2096-2097.

68. Cuisset T, Frere C, Quilici J, et al. Role of the T744C polymorphism of the P2Y12 gene on platelet
response to a 600-mg loading dose of clopidogrel in 597 patients with non-ST-segment elevation
acute coronary syndrome. Thromb Res. 2007;120(6):893-899.
69. Simon T, Verstuyft C, Mary-Krause M, et al. Genetic determinants of response to clopidogrel and
cardiovascular events. N Engl J Med. 2009;360(4):363-375.
70. Ameyaw MM, Regateiro F, Li T, et al. MDR1 pharmacogenetics: frequency of the C3435T mutation in
exon 26 is significantly influenced by ethnicity. Pharmacogenetics. 2001;11(3):217-221.
71. Taubert D, von Beckerath N, Grimberg G, et al. Impact of P-glycoprotein on clopidogrel absorption.
Clin Pharmacol Ther. 2006;80(5):486-501.
72. Mega JL, Close SL, Wiviott SD, et al. Genetic variants in ABCB1 and CYP2C19 and cardiovascular
outcomes after treatment with clopidogrel and prasugrel in the TRITON-TIMI 38 trial: a pharmaco-
genetic analysis. Lancet. 2010;376(9749):1312-1319.
antiplatelet effect and clinical efficacy of clopidogrel therapy. JAMA. 2009;302(8):849-857.
74. Luo M, Li J, Xu X, et al. ABCB1 C3435T polymorphism and risk of adverse clinical events in clopidogrel
treated patients: A meta-analysis. Thromb Res. 2012;129(6):754-759. doi:10.1016/j.thromres.2011.12.003.
75. Su J, Xu J, Li X, et al. ABCB1 C3435T polymorphism and response to clopidogrel treatment in coro-
nary artery disease (CAD) patients: a meta-analysis. PLoS ONE. 2012;7(10):e46366. doi:10.1371/journal.
pone.0046366.
76. Fisch AS, Perry CG, Stephens SH, et al. Pharmacogenomics of anti-platelet and anti-coagulation ther-
apy. Curr Cardiol Rep. 2013;15(7):1-12. doi:10.1007/s11886-013-0381-3.
77. Morais SM de, Wilkinson GR, Blaisdell J, et al. The major genetic defect responsible for the polymor-
phism of S-mephenytoin metabolism in humans. J Biol Chem. 1994;269(22):15419-15422.
78. Sim SC, Risinger C, Dahl ML, et al. A common novel CYP2C19 gene variant causes ultrarapid drug
metabolism relevant for the drug response to proton pump inhibitors and antidepressants. Clin Phar-
macol Ther. 2006;79(1):103-113.
79. Brandt JT, Close SL, Iturria SJ, et al. Common polymorphisms of CYP2C19 and CYP2C9 affect
the pharmacokinetic and pharmacodynamic response to clopidogrel but not prasugrel. J Thromb
Haemost. 2007;5(12):2429-2436. doi:10.1111/j.1538-7836.2007.02775.x.
80. Kim KA, Park PW, Hong SJ, et al. The effect of CYP2C19 polymorphism on the pharmacokinetics and
pharmacodynamics of clopidogrel: a possible mechanism for clopidogrel resistance. Clin Pharmacol
Ther. 2008;84(2):236-242.
81. Hulot JS, Collet JP, Cayla G, et al. CYP2C19 but not PON1 genetic variants influence clopidogrel phar-
macokinetics, pharmacodynamics, and clinical efficacy in post-myocardial infarction patients. Circ
Interv. 2011;4(5):422-428.
82. Collet J-P, Hulot J-S, Anzaha G, et al. High doses of clopidogrel to overcome genetic resistance: the
randomized crossover CLOVIS-2 (Clopidogrel and Response Variability Investigation Study 2). JACC
Cardiovasc Interv. 2011;4(4):392-402. doi:10.1016/j.jcin.2011.03.002.
83. Mega JL, Close SL, Wiviott SD, et al. Cytochrome p-450 polymorphisms and response to clopidogrel.
N Engl J Med. 2009;360(4):354-362.
84. Collet JP, Hulot JS, Pena A, et al. Cytochrome P450 2C19 polymorphism in young patients treated
with clopidogrel after myocardial infarction: a cohort study. Lancet. 2009;373(9660):309-317.
85. Paré G, Mehta SR, Yusuf S, et al. Effects of CYP2C19 genotype on outcomes of clopidogrel treatment.
N Engl J Med. 2010;363(18):1704-1714. doi:10.1056/NEJMoa1008410.
86. Mega JL, Simon T, Collet JP, et al. Reduced-function CYP2C19 genotype and risk of adverse clinical
outcomes among patients treated with clopidogrel predominantly for PCI: a meta-analysis. JAMA.
2010;304(16):1821-1830.
87. Sorich MJ, Rowland A, McKinnon RA, et al. CYP2C19 genotype has a greater effect on adverse cardio-
vascular outcomes following percutaneous coronary intervention and in Asian populations treated
with clopidogrel: a meta-analysis. Circ Cardiovasc Genet. 2014;7(6):895-902. doi:10.1161/CIRCGENET-
ICS.114.000669.
88. Bauer T, Bouman HJ, van Werkum JW, et al. Impact of CYP2C19 variant genotypes on clinical efficacy
of antiplatelet treatment with clopidogrel: systematic review and meta-analysis. BMJ. 2011;343:d4588.
doi:10.1136/bmj.d4588.
89. Zabalza M, Subirana I, Sala J, et al. Meta-analyses of the association between cytochrome CYP2C19
loss- and gain-of-function polymorphisms and cardiovascular outcomes in patients with coronary
artery disease treated with clopidogrel. Heart. 2012;98(2):100-108. doi:10.1136/hrt.2011.227652.
90. Partnership B-MSP. Clopidogrel Plavix® package insert. Bridgewater, NJ; July 2015.
91. Mega JL, Hochholzer W, Frelinger AL III, et al. Dosing clopidogrel based on CYP2C19 genotype and the
effect on platelet reactivity in patients with stable cardiovascular disease. JAMA. 2011;306(20):2221-
2228.
92. Simon T, Bhatt DL, Bergougnan L, et al. Genetic polymorphisms and the impact of a higher clopido-
grel dose regimen on active metabolite exposure and antiplatelet response in healthy subjects. Clin
Pharmacol Ther. 2011;90(2):287-295.
93. Cuisset T, Quilici J, Cohen W, et al. Usefulness of high clopidogrel maintenance dose according to
CYP2C19 genotypes in clopidogrel low responders undergoing coronary stenting for non ST elevation
acute coronary syndrome. Am J Cardiol. 2011;108(6):760-765.
94. Bonello L, Armero S, Mokhtar OA, et al. Clopidogrel loading dose adjustment according to platelet
reactivity monitoring in patients carrying the 2C19*2 loss of function polymorphism. J Am Coll Cardiol.
2010;56(20):1630-1636.
95. Sibbing D, Koch W, Gebhard D, et al. Cytochrome 2C19*17 allelic variant, platelet aggregation, bleeding
events, and stent thrombosis in clopidogrel-treated patients with coronary stent placement. Circula-
tion. 2010;121(4):512-518.
96. Frere C, Cuisset T, Gaborit B, et al. The CYP2C19*17 allele is associated with better platelet response
to clopidogrel in patients admitted for non-ST acute coronary syndrome. J Thromb Haemost.
2009;7(8):1409-1411.
97. Lewis JP, Stephens SH, Horenstein RB, et al. The CYP2C19*17 variant is not independently associated
with clopidogrel response. J Thromb Haemost. 2013;11(9):1640-1646. doi:10.1111/jth.12342.
98. Bouman HJ, Schomig E, van Werkum JW, et al. Paraoxonase-1 is a major determinant of clopidogrel
efficacy. Nat Med. 2011;17(1):110-116.
99. Lewis JP, Fisch AS, Ryan K, et al. Paraoxonase 1 (PON1) gene variants are not associated with clopido-
grel response. Clin Pharmacol Ther. 2011;90(4):568-574.
100. Mega JL, Close SL, Wiviott SD, et al. PON1 Q192R genetic variant and response to clopidogrel and
prasugrel: pharmacokinetics, pharmacodynamics, and a meta-analysis of clinical outcomes. J Thromb
Thrombolysis. November 2015:1-10. doi:10.1007/s11239-015-1264-9.
101. Li X, Zhang L, Chen X, et al. PON1 Q192R genotype influences clopidogrel responsiveness by rela-
tive platelet inhibition instead of on-treatment platelet reactivity. Thromb Res. 2013;132(4):444-449.
doi:10.1016/j.thromres.2013.08.004.
102. PLOS ONE: Paraoxonase 1 gene polymorphism does not affect clopidogrel response variability
but is associated with clinical outcome after PCI. Available at: http://journals.plos.org/plosone/
article?id=10.1371/journal.pone.0052779. Accessed January 28, 2016.
103. Zhu H-J, Wang X, Gawronski BE, et al. Carboxylesterase 1 as a determinant of clopidogrel metabolism
and activation. J Pharmacol Exp Ther. 2013;344(3):665-672. doi:10.1124/jpet.112.201640.
104. Lewis JP, Horenstein RB, Ryan K, et al. The functional G143E variant of carboxylesterase 1 is associated
with increased clopidogrel active metabolite levels and greater clopidogrel response. Pharmacogenet
Genomics. 2013;23(1):1-8. doi:10.1097/FPC.0b013e32835aa8a2.
105. Tarkiainen EK, Holmberg MT, Tornio A, et al. Carboxylesterase 1 c.428G>A single nucleotide variation
increases the antiplatelet effects of clopidogrel by reducing its hydrolysis in humans. Clin Pharmacol
Ther. 2015;97(6):650-658. doi:10.1002/cpt.101.
106. Holmes DR Jr, Dehmer GJ, Kaul S, et al. ACCF/AHA clopidogrel clinical alert: approaches to the FDA
“boxed warning”: a report of the American College of Cardiology Foundation Task Force on Clinical
Expert Consensus Documents and the American Heart Association. Circulation. June 2010.
107. Members WC, Levine GN, Bates ER, et al. 2011 ACCF/AHA/SCAI Guideline for percutaneous coronary
intervention: a report of the American College of Cardiology Foundation/American Heart Association
Task Force on Practice Guidelines and the Society for Cardiovascular Angiography and Interventions.
Circulation. 2011;124(23):e574-e651.
108. Amsterdam EA, Wenger NK, Brindis RG, et al. 2014 AHA/ACC Guideline for the management of
patients with non−ST-elevation acute coronary syndromes: a report of the American College
of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol.

2014;64(24):e139-e228. doi:10.1016/j.jacc.2014.09.017.
109. American College of Emergency Physicians, Society for Cardiovascular Angiography and Interventions,
O’Gara PT, et al. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction:
a report of the American College of Cardiology Foundation/American Heart Association Task Force on
Practice Guidelines. J Am Coll Cardiol. 2013;61(4):e78-e140. doi:10.1016/j.jacc.2012.11.019.
110. Scott SA, Sangkuhl K, Stein CM, et al. Clinical pharmacogenetics implementation consortium
guidelines for CYP2C19 genotype and clopidogrel therapy: 2013 Update. Clin Pharmacol Ther.
2013;94(3):317-323. doi:10.1038/clpt.2013.105.
111. Scott SA, Sangkuhl K, Gardner EE, et al. Clinical pharmacogenetics implementation consortium guide-
lines for cytochrome P450-2C19 (CYP2C19) genotype and clopidogrel therapy. Clin Pharmacol Ther.
2011;90(2):328-332.
112. Roberts JD, Wells GA, Le May MR, et al. Point-of-care genetic testing for personalisation of anti-
platelet treatment (RAPID GENE): a prospective, randomised, proof-of-concept trial. Lancet.
2012;379(9827):1705-1711. doi:10.1016/S0140-6736(12)60161-5.
113. Mega JL, Close SL, Wiviott SD, et al. Cytochrome P450 genetic polymorphisms and the response
to prasugrel: relationship to pharmacokinetic, pharmacodynamic, and clinical outcomes. Circulation.
2009;119(19):2553-2560.
114. Franken CC, Kaiser AFC, Krüger JC, et al. Cytochrome P450 2B6 and 2C9 genotype polymorphism—a
possible cause of prasugrel low responsiveness. Thromb Haemost. 2013;110(1):131-140. doi:10.1160/TH13-
01-0021.
115. Cuisset T, Loosveld M, Morange PE, et al. CYP2C19*2 and *17 alleles have a significant impact on
platelet response and bleeding risk in patients treated with prasugrel after acute coronary syndrome.
JACC Cardiovasc Interv. 2012;5(12):1280-1287. doi:10.1016/j.jcin.2012.07.015.
116. Xiang Q, Cui Y, Zhao X, et al. Identification of PEAR1 SNPs and their influences on the variation in
prasugrel pharmacodynamics. Pharmacogenomics. 2013;14(10):1179-1189. doi:10.2217/pgs.13.108.
117. Tantry US, Bliden KP, Wei C, et al. First analysis of the relation between CYP2C19 genotype and
pharmacodynamics in patients treated with ticagrelor versus clopidogrel: the ONSET/OFFSET and
RESPOND genotype studies. Circ Cardiovasc Genet. 2010;3(6):556-566. doi:10.1161/CIRCGENETICS.
110.958561.
118. Wallentin L, James S, Storey RF, et al. Effect of CYP2C19 and ABCB1 single nucleotide polymorphisms
on outcomes of treatment with ticagrelor versus clopidogrel for acute coronary syndromes: a genetic
substudy of the PLATO trial. Lancet. 2010;376(9749):1320-1328.
119. Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopidogrel in patients with acute coronary
syndromes. N Engl J Med. 2009;361(11):1045-1057. doi:10.1056/NEJMoa0904327.
120. Varenhorst C, Eriksson N, Johansson Å, et al. Effect of genetic variations on ticagrelor plasma levels
and clinical outcomes. Eur Heart J. 2015;36(29):1901-1912. doi:10.1093/eurheartj/ehv116.
121. Taylor MR, Sun AY, Davis G, et al. Race, common genetic variation, and therapeutic response disparities
in heart failure. JACC Heart Fail. 2014;2(6):561-572. doi:10.1016/j.jchf.2014.06.010.
122. Arwood MJ, Cavallari LH, Duarte JD. Pharmacogenomics of hypertension and heart disease. Curr
Hypertens Rep. 2015;17(9):586. doi:10.1007/s11906-015-0586-5.
123. Materson BJ, Reda DJ, Cushman WC, et al. Single-drug therapy for hypertension in men. A comparison
of six antihypertensive agents with placebo. The Department of Veterans Affairs Cooperative Study
Group on Antihypertensive Agents. N Engl J Med. 1993;328(13):914-921.

124. Comparison of propranolol and hydrochlorothiazide for the initial treatment of hypertension. II.
Results of long-term therapy. Veterans Administration Cooperative Study Group on Antihypertensive
Agents. JAMA. 1982;248(16):2004-2011.
125. Johnson JA, Zineh I, Puckett BJ, et al. Beta 1-adrenergic receptor polymorphisms and antihypertensive
response to metoprolol. Clin Pharmacol Ther. 2003;74(1):44-52.
126. Liu J, Liu ZQ, Yu BN, et al. Beta1-adrenergic receptor polymorphisms influence the response to meto-
prolol monotherapy in patients with essential hypertension. Clin Pharmacol Ther. 2006;80(1):23-32.
127. Pacanowski M, Gong Y, Cooper-DeHoff R, et al. ß-adrenergic receptor gene polymorphisms and
ß-blocker treatment outcomes in hypertension. Clin Pharmacol Ther. 2008;84(6):715-721. doi:10.1038/
clpt.2008.139.
128. Terra SG, Hamilton KK, Pauly DF, et al. Beta1-adrenergic receptor polymorphisms and left ventricular
remodeling changes in response to beta-blocker therapy. Pharmacogenet Genomics. 2005;15(4):227-

234.
129. Perez JM, Rathz DA, Petrashevskaya NN, et al. Beta 1-adrenergic receptor polymorphisms confer differ-
ential function and predisposition to heart failure. Nat Med. 2003;9(10):1300-1305.
130. The Beta-Blocker Evaluation of Survival Trial Investigators. A trial of the beta-blocker bucindolol
in patients with advanced chronic heart failure. N Engl J Med. 2001;344(22):1659-1667. doi:10.1056/
NEJM200105313442202.
131. Liggett SB, Mialet-Perez J, Thaneemit-Chen S, et al. A polymorphism within a conserved beta(1)-
adrenergic receptor motif alters cardiac function and beta-blocker response in human heart failure.
Proc Natl Acad Sci U S A. 2006;103(30):11288-11293.
132. Fiuzat M, Neely ML, Starr AZ, et al. Association between adrenergic receptor genotypes and beta-
blocker dose in heart failure patients: analysis from the HF-ACTION DNA substudy. Eur J Heart Fail.
2013;15(3):258-266. doi:10.1093/eurjhf/hfs175.
133. Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial
in Congestive Heart Failure (MERIT-HF). Lancet. 1999;353(9169):2001-2007.
134. Khalaila JM, Elami A, Caraco Y. Interaction between beta2 adrenergic receptor polymorphisms
determines the extent of isoproterenol-induced vasodilatation ex vivo. Pharmacogenet Genomics.
2007;17(10):803-811.
135. Lanfear DE, Jones PG, Marsh S, et al. Beta2-adrenergic receptor genotype and survival among
patients receiving beta-blocker therapy after an acute coronary syndrome. JAMA J Am Med Assoc.
2005;294(12):1526-1533.
136. Winkelmann BR, Russ AP, Nauck M, et al. Angiotensinogen M235T polymorphism is associated with
plasma angiotensinogen and cardiovascular disease. Am Heart J. 1999;137(4 Pt 1):698-705.
137. Rigat B, Hubert C, Alhenc-Gelas F, et al. An insertion/deletion polymorphism in the angiotensin
I-converting enzyme gene accounting for half the variance of serum enzyme levels. J Clin Invest.
1990;86(4):1343-1346. doi:10.1172/JCI114844.
138. Harrap SB, Tzourio C, Cambien F, et al. The ACE gene I/D polymorphism is not associated with the
blood pressure and cardiovascular benefits of ACE inhibition. Hypertension. 2003;42(3):297-303.
139. Arnett DK, Davis BR, Ford CE, et al. Pharmacogenetic association of the angiotensin-converting
enzyme insertion/deletion polymorphism on blood pressure and cardiovascular risk in relation to anti-
hypertensive treatment: the Genetics of Hypertension-Associated Treatment (GenHAT) study. Circula-
tion. 2005;111(25):3374-3383.
140. McNamara DM, Holubkov R, Postava L, et al. Pharmacogenetic interactions between angiotensin-
converting enzyme inhibitor therapy and the angiotensin-converting enzyme deletion polymorphism
in patients with congestive heart failure. J Am Coll Cardiol. 2004;44(10):2019-2026. doi:10.1016/j.
jacc.2004.08.048.
141. Hingorani AD, Jia H, Stevens PA, et al. Renin-angiotensin system gene polymorphisms influence blood
pressure and the response to angiotensin converting enzyme inhibition. J Hypertens. 1995;13(12 Pt
2):1602-1609.
142. Schelleman H, Klungel OH, Witteman JC, et al. Angiotensinogen M235T polymorphism and the risk of
myocardial infarction and stroke among hypertensive patients on ACE-inhibitors or beta-blockers. Eur
J Hum Genet. 2007;15(4):478-484.
143. Bis JC, Smith NL, Psaty BM, et al. Angiotensinogen Met235Thr polymorphism, angiotensin-converting
enzyme inhibitor therapy, and the risk of nonfatal stroke or myocardial infarction in hypertensive
patients. Am J Hypertens. 2003;16(12):1011-1017.
144. Miller JA, Thai K, Scholey JW. Angiotensin II type 1 receptor gene polymorphism predicts response to
losartan and angiotensin II. Kidney Int. 1999;56(6):2173-2180.
145. Diez J, Laviades C, Orbe J, et al. The A1166C polymorphism of the AT1 receptor gene is associated with
collagen type I synthesis and myocardial stiffness in hypertensives. J Hypertens. 2003;21(11):2085-
2092.
146. Kurland L, Melhus H, Karlsson J, et al. Angiotensin converting enzyme gene polymorphism predicts
blood pressure response to angiotensin II receptor type 1 antagonist treatment in hypertensive
patients. J Hypertens. 2001;19(10):1783-1787.
147. Brunner M, Cooper-DeHoff RM, Gong Y, et al. Factors influencing blood pressure response to trandol-
april add-on therapy in patients taking verapamil SR (from the International Verapamil SR/Trandol-
april [INVEST] Study). Am J Cardiol. 2007;99(11):1549-1554. doi:10.1016/j.amjcard.2007.01.029.
148. Cusi D, Barlassina C, Azzani T, et al. Polymorphisms of alpha-adducin and salt sensitivity in patients
with essential hypertension. Lancet. 1997;349(9062):1353-1357.
149. Psaty BM, Smith NL, Heckbert SR, et al. Diuretic therapy, the alpha-adducin gene variant, and the risk
of myocardial infarction or stroke in persons with treated hypertension. JAMA. 2002;287(13):1680-1689.
150. Gerhard T, Gong Y, Beitelshees AL, et al. Alpha-adducin polymorphism associated with increased risk
of adverse cardiovascular outcomes: results from GENEtic Substudy of the INternational VErapamil
SR-trandolapril STudy (INVEST-GENES). Am Heart J. 2008;156(2):397-404.
151. Vormfelde SV, Burckhardt G, Zirk A, et al. Pharmacogenomics of diuretic drugs: data on rare mono-
genic disorders and on polymorphisms and requirements for further research. Pharmacogenomics.
2003;4(6):701-734. doi:10.1517/phgs.4.6.701.22817.
152. Cohn JN, Archibald DG, Ziesche S, et al. Effect of vasodilator therapy on mortality in chronic
congestive heart failure. Results of a Veterans Administration Cooperative Study. N Engl J Med.
1986;314(24):1547-1552.
153. Cohn JN, Johnson G, Ziesche S, et al. A comparison of enalapril with hydralazine-isosorbide dinitrate
in the treatment of chronic congestive heart failure. N Engl J Med. 1991;325(5):303-310.
154. Carson P, Ziesche S, Johnson G, et al. Racial differences in response to therapy for heart failure:
analysis of the vasodilator-heart failure trials. Vasodilator-Heart Failure Trial Study Group. J Card Fail.
1999;5(3):178-187.
155. Taylor AL, Ziesche S, Yancy C, et al. Combination of isosorbide dinitrate and hydralazine in blacks with
heart failure. N Engl J Med. 2004;351(20):2049-2057.
156. McNamara DM, Tam SW, Sabolinski ML, et al. Aldosterone synthase promoter polymorphism predicts
outcome in African Americans with heart failure: results from the A-HeFT Trial. J Am Coll Cardiol.
2006;48(6):1277-1282.
157. McNamara DM, Tam SW, Sabolinski ML, et al. Endothelial nitric oxide synthase (NOS3) polymorphisms
in African Americans with heart failure: results from the A-HeFT trial. J Card Fail. 2009;15(3):191-198.
158. Ghatak A, Faheem O, Thompson PD. The genetics of statin-induced myopathy. Atherosclerosis.
2010;210(2):337-343.
159. Gryn SE, Hegele RA. Pharmacogenomics, lipid disorders, and treatment options. Clin Pharmacol Ther.
2014;96(1):36-47. doi:10.1038/clpt.2014.82.
160. Maggo SD, Kennedy MA, Clark DW. Clinical implications of pharmacogenetic variation on the effects
of statins. Drug Saf. 2011;34(1):1-19.
161. SEARCH Collaborative Group, Link E, Parish S, et al. SLCO1B1 variants and statin-induced myopathy—a
genome-wide study. N Engl J Med. 2008;359(8):789-799. doi:10.1056/NEJMoa0801936.
162. Ramsey LB, Johnson SG, Caudle KE, et al. The Clinical Pharmacogenetics Implementation Consor-
tium Guideline for SLCO1B1 and simvastatin-induced myopathy: 2014 Update. Clin Pharmacol Ther.
2014;96(4):423-428. doi:10.1038/clpt.2014.125.
163. Voora D, Shah SH, Spasojevic I, et al. The SLCO1B1*5 genetic variant is associated with statin-induced
side effects. J Am Coll Cardiol. 2009;54(17):1609-1616.
164. Donnelly LA, Doney AS, Tavendale R, et al. Common nonsynonymous substitutions in SLCO1B1 predis-
pose to statin intolerance in routinely treated individuals with type 2 diabetes: a go-DARTS study. Clin
Pharmacol Ther. 2011;89(2):210-216.
165. Brunham LR, Lansberg PJ, Zhang L, et al. Differential effect of the rs4149056 variant in SLCO1B1 on
myopathy associated with simvastatin and atorvastatin. Pharmacogenomics J. 2012;12(3):233-237.
doi:10.1038/tpj.2010.92.
166. Carr DF, O’Meara H, Jorgensen AL, et al. SLCO1B1 Genetic variant associated with statin-induced
myopathy: a proof-of-concept study using the clinical practice research datalink. Clin Pharmacol Ther.
2013;94(6):695-701. doi:10.1038/clpt.2013.161.
167. Danik JS, Chasman DI, MacFadyen JG, et al. Lack of association between SLCO1B1 polymorphisms
and clinical myalgia following rosuvastatin therapy. Am Heart J. 2013;165(6):1008-1014. doi:10.1016/j.
ahj.2013.01.025.
168. Wilke RA, Ramsey LB, Johnson SG, et al. The Clinical Pharmacogenomics Implementation Consortium:
CPIC Guideline for SLCO1B1 and simvastatin-induced myopathy. Clin Pharmacol Ther. 2012;92(1):112-117.
doi:10.1038/clpt.2012.57.
169. Becker ML, Elens LLFS, Visser LE, et al. Genetic variation in the ABCC2 gene is associated with dose
decreases or switches to other cholesterol-lowering drugs during simvastatin and atorvastatin thera-
py. Pharmacogenomics J. 2013;13(3):251-256. doi:10.1038/tpj.2011.59.
170. Shahabi P, Dubé M-P. Cardiovascular pharmacogenomics; state of current knowledge and implemen-
tation in practice. Int J Cardiol. 2015;184:772-795. doi:10.1016/j.ijcard.2015.02.025.
171. Mangravite LM, Engelhardt BE, Medina MW, et al. A statin-dependent QTL for GATM expression is
associated with statin-induced myopathy. Nature. 2013;502(7471):377-380. doi:10.1038/nature12508.
172. Patel J, Superko HR, Martin SS, et al. Genetic and immunologic susceptibility to statin-related myopa-
thy. Atherosclerosis. 2015;240(1):260-271. doi:10.1016/j.atherosclerosis.2015.03.025.
173. Postmus I, Trompet S, Deshmukh HA, et al. Pharmacogenetic meta-analysis of genome-wide associa-
tion studies of LDL cholesterol response to statins. Nat Commun. 2014;5. doi:10.1038/ncomms6068.
174. Thompson JF, Hyde CL, Wood LS, et al. Comprehensive whole-genome and candidate gene analysis
for response to statin therapy in the Treating to New Targets (TNT) cohort. Circ Genet. 2009;2(2):173-
181.
175. Mega JL, Morrow DA, Brown A, et al. Identification of genetic variants associated with response to
statin therapy. Arterioscler Thromb Vasc Biol. 2009;29(9):1310-1315. doi:10.1161/ATVBAHA.109.188474.
176. Chasman DI, Giulianini F, MacFadyen J, et al. Genetic determinants of statin-induced low-density lipo-
protein cholesterol reduction the justification for the use of statins in prevention: an intervention trial
evaluating rosuvastatin (JUPITER) trial. Circ Cardiovasc Genet. 2012;5(2):257-264. doi:10.1161/CIRCGE-
NETICS.111.961144.
177. Hopewell JC, Parish S, Offer A, et al. Impact of common genetic variation on response to simvastatin
therapy among 18,705 participants in the Heart Protection Study. Eur Heart J. 2013;34(13):982-992.
doi:10.1093/eurheartj/ehs344.
178. Deshmukh HA, Colhoun HM, Johnson T, et al. Genome-wide association study of genetic determi-
nants of LDL-c response to atorvastatin therapy: importance of Lp(a). J Lipid Res. 2012;53(5):1000-1011.
doi:10.1194/jlr.P021113.
179. Thompson JF, Man M, Johnson KJ, et al. An association study of 43 SNPs in 16 candidate genes with
atorvastatin response. Pharmacogenomics J. 2005;5(6):352-358.
180. Chasman DI, Posada D, Subrahmanyan L, et al. Pharmacogenetic study of statin therapy and choles-
terol reduction. JAMA. 2004;291(23):2821-2827.
181. Gerdes LU, Gerdes C, Kervinen K, et al. The apolipoprotein epsilon4 allele determines prognosis and
the effect on prognosis of simvastatin in survivors of myocardial infarction: a substudy of the Scandi-
navian simvastatin survival study. Circulation. 2000;101(12):1366-1371.
182. Trompet S, de Craen AJM, Postmus I, et al. Replication of LDL GWAs hits in PROSPER/PHASE as vali-
dation for future (pharmaco)genetic analyses. BMC Med Genet. 2011;12:131. doi:10.1186/1471-2350-12-131.
183. Birmingham BK, Bujac SR, Elsby R, et al. Impact of ABCG2 and SLCO1B1 polymorphisms on pharma-
cokinetics of rosuvastatin, atorvastatin and simvastatin acid in Caucasian and Asian subjects: a class
effect? Eur J Clin Pharmacol. 2015;71(3):341-355. doi:10.1007/s00228-014-1801-z.
184. Akao H, Polisecki E, Kajinami K, et al. Genetic variation at the SLCO1B1 gene locus and low
density lipoprotein cholesterol lowering response to pravastatin in the elderly. Atherosclerosis.
2012;220(2):413-417. doi:10.1016/j.atherosclerosis.2011.09.028.
185. Meyer zu Schwabedissen HE, Albers M, Baumeister SE, et al. Function-impairing polymorphisms of the
hepatic uptake transporter SLCO1B1 modify the therapeutic efficacy of statins in a population-based
cohort. Pharmacogenet Genomics. 2015;25(1):8-18. doi:10.1097/FPC.0000000000000098.
186. Teslovich TM, Musunuru K, Smith AV, et al. Biological, clinical and population relevance of 95 loci for
blood lipids. Nature. 2010;466(7307):707-713.
187. Shim H, Chasman DI, Smith JD, et al. A multivariate genome-wide association analysis of
10 LDL subfractions, and their response to statin treatment, in 1868 Caucasians. PLoS ONE.
2015;10(4):e0120758. doi:10.1371/journal.pone.0120758.
188. Musunuru K, Strong A, Frank-Kamenetsky M, et al. From noncoding variant to phenotype via SORT1 at
the 1p13 cholesterol locus. Nature. 2010;466(7307):714-719. doi:10.1038/nature09266.
189. Samani NJ, Erdmann J, Hall AS, et al. Genomewide association analysis of coronary artery disease. N
Engl J Med. 2007;357(5):443-453. doi:10.1056/NEJMoa072366.
190. Kathiresan S, Voight BF, Purcell S, et al. Genome-wide association of early-onset myocardial infarc-
tion with single nucleotide polymorphisms and copy number variants. Nat Genet. 2009;41(3):334-341.
doi:10.1038/ng.327.
191. Donnelly LA, Doney AS, Dannfald J, et al. A paucimorphic variant in the HMG-CoA reductase gene is
associated with lipid-lowering response to statin treatment in diabetes: a GoDARTS study. Pharmaco-
genet Genomics. 2008;18(12):1021-1026.
192. Krauss RM, Mangravite LM, Smith JD, et al. Variation in the 3-hydroxyl-3-methylglutaryl coenzyme a
reductase gene is associated with racial differences in low-density lipoprotein cholesterol response to
simvastatin treatment. Circulation. 2008;117(12):1537-1544. doi:10.1161/CIRCULATIONAHA.107.708388.
193. Iakoubova OA, Tong CH, Rowland CM, et al. Association of the Trp719Arg polymorphism in kinesin-like
protein 6 with myocardial infarction and coronary heart disease in 2 prospective trials: the CARE and
WOSCOPS trials. J Am Coll Cardiol. 2008;51(4):435-443.
194. Iakoubova OA, Sabatine MS, Rowland CM, et al. Polymorphism in KIF6 gene and benefit from statins
after acute coronary syndromes: results from the PROVE IT-TIMI 22 study. J Am Coll Cardiol.
2008;51(4):449-455.
195. Iakoubova OA, Robertson M, Tong CH, et al. KIF6 Trp719Arg polymorphism and the effect of
statin therapy in elderly patients: results from the PROSPER study. Eur J Cardiovasc Prev Rehabil.
2010;17(4):455-461.
196. Ridker PM, MacFadyen JG, Glynn RJ, et al. Kinesin-like protein 6 (KIF6) polymorphism and the efficacy
of rosuvastatin in primary prevention. Circ Genet. 2011;4(3):312-317.
197. Hopewell JC, Parish S, Clarke R, et al. No impact of KIF6 genotype on vascular risk and statin response
among 18,348 randomized patients in the heart protection study. J Am Coll Cardiol. 2011;57(20):2000-
2007.
198. Neuvonen PJ, Niemi M, Backman JT. Drug interactions with lipid-lowering drugs: mechanisms and clini-
cal relevance. Clin Pharmacol Ther. 2006;80(6):565-581.
199. Tomlinson B, Hu M, Lee VWY, et al. ABCG2 polymorphism is associated with the low-density lipo-
protein cholesterol response to rosuvastatin. Clin Pharmacol Ther. 2010;87(5):558-562. doi:10.1038/
clpt.2009.232.
200. DeGorter MK, Tirona RG, Schwarz UI, et al. Clinical and pharmacogenetic predictors of circulating ator-
vastatin and rosuvastatin concentrations in routine clinical care. Circ Cardiovasc Genet. 2013;6(4):400-
408. doi:10.1161/CIRCGENETICS.113.000099.
201. Wang J, Williams CM, Hegele RA. Compound heterozygosity for two non-synonymous polymorphisms
in NPC1L1 in a non-responder to ezetimibe. Clin Genet. 2005;67(2):175-177.
202. Hegele RA, Guy J, Ban MR, et al. NPC1L1 haplotype is associated with inter-individual variation in plas-
ma low-density lipoprotein response to ezetimibe. Lipids Health Dis. 2005;4:16.
203. Simon JS, Karnoub MC, Devlin DJ, et al. Sequence variation in NPC1L1 and association with improved
LDL-cholesterol lowering in response to ezetimibe treatment. Genomics. 2005;86(6):648-656.
204. Kim DS, Burt AA, Ranchalis JE, et al. Novel gene-by-environment interactions: APOB and NPC1L1 vari-
ants affect the relationship between dietary and total plasma cholesterol. J Lipid Res. 2013;54(5):1512-
1520. doi:10.1194/jlr.P035238.
205. Stitziel NO, Won HH, Morrison AC, et al. Inactivating mutations in NPC1L1 and protection from coronary
heart disease. N Engl J Med. 2014;371(22):2072-2082. doi:10.1056/NEJMoa1405386.
206. Muendlein A, Leiherer A, Saely CH, et al. Common single nucleotide polymorphisms at the
NPC1L1 gene locus significantly predict cardiovascular risk in coronary patients. Atherosclerosis.
2015;242(1):340-345. doi:10.1016/j.atherosclerosis.2015.07.011.
207. Woosley RL, Drayer DE, Reidenberg MM, et al. Effect of acetylator phenotype on the rate at which
procainamide induces antinuclear antibodies and the lupus syndrome. N Engl J Med. 1978;298(21):1157-
1159.
208. Mongey AB, Sim E, Risch A, et al. Acetylation status is associated with serological changes but not
clinically significant disease in patients receiving procainamide. J Rheumatol. 1999;26(8):1721-1726.

209. Darbar D, Roden DM. Pharmacogenetics of antiarrhythmic therapy. Expert Opin Pharmacother.
2006;7(12):1583-1590.
210. Jazwinska-Tarnawska E, Orzechowska-Juzwenko K, Niewinski P, et al. The influence of CYP2D6 poly-
morphism on the antiarrhythmic efficacy of propafenone in patients with paroxysmal atrial fibrillation
during 3 months propafenone prophylactic treatment. Int J Clin Pharmacol Ther. 2001;39(7):288-292.
211. Morike K, Kivisto KT, Schaeffeler E, et al. Propafenone for the prevention of atrial tachyarrhythmias
after cardiac surgery: a randomized, double-blind placebo-controlled trial. Clin Pharmacol Ther.
2008;84(1):104-110.
212. Lim KS, Cho JY, Jang IJ, et al. Pharmacokinetic interaction of flecainide and paroxetine in relation to
the CYP2D6*10 allele in healthy Korean subjects. Br J Clin Pharmacol. 2008;66(5):660-666.
213. Doki K, Homma M, Kuga K, et al. Effect of CYP2D6 genotype on flecainide pharmacokinetics in Japa-
nese patients with supraventricular tachyarrhythmia. Eur J Clin Pharmacol. 2006;62(11):919-926.

214. Tenneze L, Tarral E, Ducloux N, et al. Pharmacokinetics and electrocardiographic effects of a new
controlled-release form of flecainide acetate: comparison with the standard form and influence of
the CYP2D6 polymorphism. Clin Pharmacol Ther. 2002;72(2):112-122.
215. Chowbay B, Li H, David M, et al. Meta-analysis of the influence of MDR1 C3435T polymorphism on
digoxin pharmacokinetics and MDR1 gene expression. Br J Clin Pharmacol. 2005;60(2):159-171.
216. Leschziner GD, Andrew T, Pirmohamed M, et al. ABCB1 genotype and PGP expression, function and
therapeutic drug response: a critical review and recommendations for future research. Pharmaco-
genomics J. 2007;7(3):154-179.
217. Roden DM. Proarrhythmia as a pharmacogenomic entity: a critical review and formulation of a unifying
hypothesis. Cardiovasc Res. 2005;67(3):419-425.
218. Behr ER, Roden D. Drug-induced arrhythmia: pharmacogenomic prescribing? Eur Heart J. 2013;34(2):89-
95. doi:10.1093/eurheartj/ehs351.
219. Kannankeril PJ. Understanding drug-induced torsades de pointes: a genetic stance. Expert Opin Drug
Saf. 2008;7(3):231-239.
220. Paulussen ADC, Gilissen RAHJ, Armstrong M, et al. Genetic variations of KCNQ1, KCNH2, SCN5A, KCNE1,
and KCNE2 in drug-induced long QT syndrome patients. J Mol Med. 2004;82(3):182-188. doi:10.1007/
s00109-003-0522-z.
221. Abbott GW. KCNE genetics and pharmacogenomics in cardiac arrhythmias: much ado about nothing?
Expert Rev Clin Pharmacol. 2013;6(1):49-60. doi:10.1586/ecp.12.76.
222. Ramirez AH, Shaffer CM, Delaney JT, et al. Novel rare variants in congenital cardiac arrhythmia
genes are frequent in drug-induced torsades de pointes. Pharmacogenomics J. 2013;13(4):325-329.
doi:10.1038/tpj.2012.14.
CHAPTER
8
Hematology/Oncology
Pharmacogenomics
LEARNING OBJECTIVES
After completing this chapter, the reader Describe key differences in the
should be able to information that would be obtained
by comparing mRNA and microRNA
Define terms necessary for understanding
expression profiling in cancer
concepts relevant to cancer
pharmacogenomics.
pharmacogenomics.
Describe three areas that define cancer
pharmacogenomics. KEY DEFINITIONS
Distinguish between pharmacodynamic
and pharmacokinetic considerations in CANCER—a disease of abnormal cellular
cancer pharmacogenomics. proliferation in which cells have invaded
tissue locally or at locations distant from
Identify specific genetic variants their site of origin.
that impact the use of cancer
chemotherapeutic agents. CANCER CHEMOPREVENTION—drug-based
interventions that may be useful in
Describe the three distinct areas of preventing cancer onset for individuals at
cancer prevention that may be targeted increased risk for cancer development.
by cancer chemopreventive measures.
CANCER CHEMOTHERAPY—drug-based
Describe the difference between variant therapeutics that may be used to treat
forms of cancers and pharmacogenomics specific types of cancer.
relevant to different forms of cancer.
CANDIDATE GENE STUDY—genetic
Describe how differences between association study performed to assess
hereditary variants and somatic the association of specific gene variants
mutations are determined. (e.g., in a candidate gene) with a
Give examples of how differences phenotype of interest such as an adverse
between hereditary variants and somatic drug response.
mutations can be used to optimize CARCINOMA—cancers of epithelial tissue
cancer therapy. origin.
Explain a critical difference between COPY NUMBER VARIATION (CNV)—regions
technologies used to analyze mRNA of DNA or genes that can vary in the
expression and genetic variants found in number of copies present (e.g., CYP2D6).
DNA.
257
DEOXYRIBONUCLEIC ACID (DNA)—a self-replicating material present in nearly all living

organisms as the main constituent of chromosomes.
DIFFERENTIATION—in most tissues, the process that involves the progression from a
progenitor cell to cells with distinct characteristics associated with the functional role of
the tissue.
EPIGENETICS—modifications of genetic material that are associated with changes in gene
expression but do not alter the base pair sequence of DNA.
GENOME-WIDE ASSOCIATION STUDIES (GWAS)—studies performed to comprehensively
examine genetic variations present in DNA (i.e., a person’s whole genome) for association
with a specific phenotype such as a specific response to drug treatment.
INDEL—insertion/deletion of a segment of DNA (i.e., ≥2 base pairs).
LEUKEMIA—a group of cancers derived from blood-forming cells.
MESSENGER RNA (mRNA)—the ribonucleic acid polymer derived from DNA through the
process of transcription providing the code for protein production through translation.
MICROARRAY—a nucleotide hybridization-based technology used to measure characteristics
of nucleic acids, such as the specific presence of thousands of distinct cellular mRNAs,
or to measure specific variants of DNA.
MicroRNA (miRNA)—a large class of genomic short RNA molecules that are noncoding and
act to alter gene expression, whose expression in cancers may serve as diagnostic
biomarkers, prognostic indicators, and predictors of therapeutic outcomes.
MULTIDRUG RESISTANCE—resistance to drug action usually in reference to the action of cell
membrane transport proteins that export drugs from the intracellular compartment to
outside the cell, preventing the drug from acting on its intracellular target.
OLIGONUCLEOTIDE—a stretch of nucleotides ranging up to approximately 50 base pairs.
ONCOGENES—genes associated with cancer development that have normal cellular
counterparts or proto-oncogenes.
ONCOLOGY (CANCER BIOLOGY)—the study of cancer often with an emphasis on cancer
development and therapy.
P-GLYCOPROTEIN—(ABCB1; multiple drug resistance 1) an efflux pump present in many
cell types. In some cancers, p-glycoprotein is overexpressed and responsible for cancer
chemotherapeutic drug resistance due to the actions of p-glycoprotein pumping the drug
out of the cancer cell precluding interactions with the intracellular therapeutic target.
POLYMERASE CHAIN REACTION—a technique used to amplify segments of DNA that most
commonly use thermal stable DNA polymerases.
POLYMORPHIC VARIANTS—differences in the DNA between two individuals that may account
for the genetic basis of phenotype. Such variations may include SNPs, indels, and CNVs.
RIBONUCLEIC ACID (RNA)—a nucleic acid present in all living cells.
SARCOMAS—a group of cancers originating from cells of mesodermal origin, such as bone
and other connective tissues.
SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—genetic variants between individuals present at
a single, distinct base pair location in DNA.
Chapter 8 • Hematology/Oncology Pharmacogenomics 259
SPLICE SITE VARIANTS—variants of mRNA due to differential processing of introns and

exons from the immature mRNA transcript.
THIOPURINE METHYLTRANSFERASE DEFICIENCY—a deficiency in thiopurine methyltransferase
activity due to genetic variants in the thiopurine methyltransferase gene with reduced
ability to metabolize the purine analogs, such as mercaptopurine and thioguanine.

TUMOR SUPPRESSOR GENES—genes coding for proteins that deter cancer development
through regulation of cell growth and division.
INTRODUCTION
Pharmacogenomics has the potential to significantly enhance optimization of cancer thera-
peutics. Because cancer is frequently a terminal disease, it is critical to utilize the most
effective therapeutic regimen to treat it and enact a cure under the best circumstances.
Many classes of cancer chemotherapeutic agents have been developed, largely over
the past seven decades. These classes include alkylating agents, purine and pyrimidine
analogs, folic acid antimetabolites, topoisomerase inhibitors, taxanes, and hormonal modu-
lators. Specific protocols that utilize these agents are continually optimized to provide the
most effective treatments for the different types of cancers ranging from hematological
cancers to solid cancers, including sarcomas and carcinomas. The advent of cancer pharma-
cogenomics has heavily influenced therapeutics for all cancer chemotherapeutic classes.
Additionally, our increasing recognition of molecular lesions associated with cancers, a key
area of cancer pharmacogenomics, is greatly assisting the development of novel targeted
cancer therapeutics.
Our understanding of the role that heredity plays in both the recognition of cancer
syndromes and the biology of drug action is increasing. Cancer is a disease known to
involve numerous changes to the cellular genome. These changes include modifications to
both deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) expression. Thus, cancer phar-
macogenomics incorporates an understanding of hereditary factors that alter genes asso-
ciated with cancer predisposition as well as genetic changes that occur in the multistage
process of carcinogenesis, including alterations in the expression of genes that affect drug
action (Figure 8-1). Methods have been developed for both research and clinical use (e.g.,
microarrays) to examine genetic variants from a patient’s DNA. Microarrays have also been
developed that allow the assessment of changes in messenger RNA (mRNA) or microRNA
expression present in cancer cells. Information combined from each of these different
types of genetic assays can now be utilized to provide more precise cancer therapy for
each patient individually.
CASE STUDY-PHARMACOGENOMIC
CONSIDERATIONS FOR TAMOXIFEN THERAPY
Laura has lived in St. Louis, Missouri, her entire life; after many years of searching for
the right career, she graduated from pharmacy school in 2004. Since receiving her
degree, she has worked in retail pharmacy. One of the driving reasons for her interest in
a healthcare profession stemmed from her mother’s health problems. Her mother died in
Patient’s Genome
[DNA and RNA]
Hereditary
Factors Affecting
Genetic
Drug Action Changes
Cancer in the
Predisposing
Genes (?)
Patient’s
Cancer Cells
(Somatic Mutations)
(Germline Variants)
FIGURE 8-1 Areas of genomics contributing to cancer pharmacogenomics (e.g., analyses

of both DNA and RNA). Analysis of RNA expression can include messenger RNAs, long
noncoding RNAs, and microRNAs. Hereditary factors (i.e., germline variants) have a
significant role in determining both the effectiveness and toxicities associated with cancer
chemotherapy. Currently, studies are evaluating the role of cancer-predisposing genes
associated with cancer syndromes and their impact on cancer chemotherapy as a subset of
hereditary factors. In addition, genetic changes (e.g., somatic mutations, alterations in gene
expression) that occur in the patient’s cancer cells during the progression of the cancer may
have a significant impact in both determining the effectiveness of drug therapy as well as
providing information useful for targeting drug therapy.
1998 after a lengthy battle with breast cancer. Laura’s grandmother, of Ashkenazi Jewish
descent, also died of breast cancer, which made Laura concerned about hereditary links
her family might have to breast cancer. This became of paramount concern when she too
was diagnosed with breast cancer. On mammography, a radiodense mass was observed
in her left breast. The mass was biopsied and evaluated to identify a stage IIa infiltrating
lobular carcinoma. A lumpectomy was performed to remove the cancerous mass from
Laura’s left breast, and all axillary lymph nodes were removed for analyses.
The breast cancer was found to be estrogen and progesterone receptor-positive without
amplification of ErbB2, and a metastatic workup was negative. Results from an Oncotype
DX© Breast Cancer Assay on a histological sample of Laura’s breast cancer resulted in a
low recurrence score (<18). Based on these findings, Laura was considered to be a good
candidate for tamoxifen therapy. Given Laura’s family history, her medical oncologist
thought more information was necessary to be confident that Laura did not have other
factors affecting the success of tamoxifen therapy. Laura’s mother did not respond to
tamoxifen therapy for treatment of her breast cancer.
Questions
1. What known hereditary gene variants may be contributing factors to the development
of Laura’s breast cancer? Which of these factors may be consistent with Laura’s family
history?
2. What pharmacokinetic/pharmacogenomic considerations may affect the efficacy of tamoxi-

fen therapy? How could screening for gene variants be used to optimize therapeutic strate-
gies for the chemotherapy used to treat Laura’s breast cancer?
3. What considerations may be relevant in recognizing a role for endoxifen compared to tamoxi-
fen for breast cancer treatment among individuals who do not have active CYP2D6?
Background
Because genetic considerations in cancer therapy include both hereditary factors as well as
mutations that occur during cancer progression, individualized drug therapy for cancer is a
model example of the increasing impact of pharmacogenomics. Additionally, a growing area
of cancer pharmacogenomics addresses cancer prevention. Individuals with a recognized
genetic predisposition to cancer or that engage in high-risk activities for cancer may be
strong candidates for cancer-prevention approaches. These measures may include drug-
based strategies that define the emerging field of chemoprevention. The field of pharma-
cogenomics offers exceptional promise for cancer therapy, in both optimizing therapeutic
strategies as well as reducing the side effects associated with cancer chemotherapeutics.
These side effects are often based on the patient’s hereditary genetics as well as genetic
considerations of the cancer. The ability to correlate pharmacogenomic characteristics of
a cancer and an individual’s specific DNA profile with responsiveness to a specific drug
provides an “individualized” and more precise approach to pharmacotherapy that is signifi-
cantly impacting pharmacy practice. This chapter provides an overview of pharmacogenom-
ics used in pharmacy of key relevance to cancer chemoprevention and therapy.
General Considerations in Cancer Pharmacogenomics

Pharmacokinetics in Cancer Pharmacogenomics—Drug Metabolism
As with other areas of pharmacogenomics, important considerations for the pharmaco-
genomics of cancer therapeutic agents can generally be divided into two broad classifica-
tions: (1) pharmacokinetics, the disposition of the drug in the body (especially in relation-
ship to drug transport and metabolic activity in the body that may alter drug action); and
(2) pharmacodynamics, the action of the drug on the body (e.g., the action of a drug on a
specific drug receptor). Pharmacodynamic considerations important to cancer pharmaco-
genomics will be discussed in the following section. Pharmacokinetic considerations in phar-
macogenomics were critical in the early observations of adverse drug reactions that led to
an understanding of genetic contributions to drug activity. Pharmacokinetic considerations
relate primarily to metabolic actions that facilitate drug inactivation and excretion. How-
ever, drug metabolism may also have an important impact on the activation of prodrugs
(e.g., the metabolism of tamoxifen to 4-hydroxytamoxifen and endoxifen).
Many different enzymes are involved in the metabolism of drugs used in cancer
chemotherapy. One of the most pervasive families of metabolic enzymes is the cyto-
chrome P450s (CYPs).1 In humans, 57 different CYP genes have been identified.2 Substrates
for different CYP members vary, and almost all drugs are metabolized by more than a
single CYP. Polymorphic variants of the CYPs are extensive and involve virtually all possible
forms of genetic variants including single nucleotide polymorphisms (SNPs), indels, and
copy number variations (CNVs).
Because numerous variants of CYPs have been identified, phenotypes based on
these variants have been organized into the more succinct designations of poor, inter-
mediate, extensive, and ultrarapid metabolizers. For example, numerous variants of
CYP2D6 have been identified such that the metabolic profile for different CYP2D6 vari-
ants themselves have been classified into groups of poor, intermediate, extensive, and
ultrarapid metabolizers.3 CYPs may also have an important role in metabolizing prodrugs
to their active form, such as the conversion of tamoxifen to endoxifen, which involves the
action of multiple CYPs, including CYP2D6. Many other classes of drug-metabolizing
enzymes have highly significant roles in cancer pharmacogenomics. These include UDP
glucuronosyltransferases, thiopurine methyltransferases, N-acetyl transferases, and
cytidine deaminase to name a few. A complete listing of metabolic enzymes important
in drug metabolism is beyond the scope of this chapter (see Chapter 4), although many
important examples are featured throughout. It is important to recognize that numerous
polymorphic variants of drug-metabolizing enzymes exist, and many play a vital role in
cancer pharmacogenomics.
Pharmacodynamics in Cancer Pharmacogenomics—Drug Targets
Pharmacological considerations for most drugs address the action of the drug on the drug
target. Many drug targets are proteins (i.e., gene products) with specialized cellular activ-
ity. The genes associated with the drug target in humans may exist in variant forms, which
could result in the production of gene products with different amino acid sequences. Such
differences in amino acid sequences of these gene products can have pronounced func-
tional consequences impacting the efficacy of drug action. This paradigm of drug action
regarding pharmacodynamics applies to cancer chemotherapeutic agents. Importantly, our
understanding of drug targets is a key consideration in the design of candidate gene stud-
ies, where polymorphic variants in the accepted drug target gene are logical candidates for
investigating associations for differences in drug action among individuals.
An example of pharmacodynamic considerations in cancer pharmacogenomics
includes the target of 5-fluorouracil (5-FU), thymidylate synthase (TYMS). As will be
discussed, TYMS possesses a number of well-established polymorphic variants.4 Although
it seems reasonable to consider that variants of TYMS might affect the actions of 5-FU,
this has not proven to be decisive clinically. However, the action of folic acid inhibitors (e.g.,
methotrexate), which indirectly affect TYMS, may be affected by TYMS variants.5 In many
instances, candidate gene analyses comparing target gene variants with drug action have
failed to show meaningful associations to explain differences in drug action. As a result
of increased knowledge of genetic variants, genome-wide association studies (GWAS) are
being performed to identify variants that may account for differences in drug action among
individuals. In these studies, genetic variants are tested without choosing a candidate gene
or genes. GWAS has resulted in many meaningful and unexpected variants accounting for
individual differences in drug action. Therefore, association studies for determining genetic
variants affecting drug action do not necessarily follow the sensibility of pharmacodynamic
considerations in cancer pharmacogenomics studies.
Drug Efflux Pumps
Drug transport pumps represent a special case with potential relevance to either the
pharmacokinetics or pharmacodynamics of cancer pharmacogenomics, depending on the
context of drug action. The plasma cell membrane contains pumps that allow the transfer
of small molecules into (i.e., uptake pumps) and out of (i.e., efflux pumps) the cell. These
pumps control the transfer of a vast array of molecules, from small ions such as sodium to
relatively large cancer therapeutic agents (e.g., docetaxel). Numerous pumps (or transport-
ers) may be selectively expressed in different cell types; however, the large array of pumps
is categorized into families based on activity and similarity of the encoded proteins.
Of particular importance to cancer therapy are efflux pumps that remove cancer
therapeutic agents from the inside of the cancer cell, preventing the drug from reaching
its therapeutic target. Therefore, drug efflux pumps present a special case regarding phar-
macodynamic and pharmacokinetic considerations in pharmacogenomics. Because efflux
pumps actively remove drugs from cells, an activity important to drug disposition, these
proteins may have a critical role in the pharmacokinetics of drug action. However, these
proteins may also be the target of drug action, whereby pharmacodynamic considerations
are relevant. Examples of efflux pumps with activity important to cancer therapy include
the mitoxantrone resistance protein and P-glycoprotein (P-gp).
MITROXANTRONE RESISTANCE PROTEIN. The mitoxantrone resistance protein (MXR or

ABCG2) is also referred to as the breast cancer resistance protein. It is an efflux trans-
porter that confers resistance to important classes of cancer chemotherapeutic agents that
include the anthracyclines, anthracenediones (e.g., mitoxantrone), and camptothecins such
as topotecan and irinotecan.6 MXR has been implicated in therapeutic resistance for acute
myelogenous leukemia and breast cancer.7 Although many variants of MXR have been
identified, further studies are necessary to determine what therapeutic consequence these
variants have on the outcome of cancer therapy.
MULTIPLE DRUG RESISTANCE 1 (P-GP). Among the most extensively investigated efflux
pumps associated with resistance to cancer therapy is P-gp (MDR1 or ABCB1). P-gp was
first identified in cancer cells as an over-expressed protein that is believed to be impor-
tant in the resistance of many cancers to chemotherapeutic agents. Drugs that may be
pumped from cells by P-gp include anthracyclines, calcium-channel blockers, digoxin,
paclitaxel, and vinca alkaloids.8 Many variants of P-gp have been identified, and a few
may have important effects on the activity of this protein. Hoffmeyer et al. examined the
DNA sequence from 21 individuals to identify 15 polymorphic variants.9 Of these variants,
one designated C3435T was found to have a significant impact on the expression of P-gp.
Individuals homozygous (i.e., two copies of the same allele) for the T form of P-gp were
found to have reduced duodenal protein expression levels of P-gp, which was believed to
be responsible for increased drug plasma protein levels. In this case, P-gp variants could
have an important impact on pharmacokinetic properties related to drug administration.
However, modification of cancer therapy based on the presence of specific genetic variants
of P-gp has not been translated into contemporary pharmacotherapy.
PHARMACOGENOMICS OF CANCER CHEMOTHERAPEUTIC

AGENTS
Many different classes of cancer chemotherapies are available with a broad scope of
therapeutic applications for the many different forms of cancer. As found throughout phar-
macogenomics, many different genetic polymorphisms are known that affect the action of
cancer chemotherapies. Classes of cancer chemotherapeutic agents known to be affected
by genetic variants include purine and pyrimidine analogs, folic acid antimetabolites, topoi-
somerase I inhibitors, and selective hormone receptor modulators. Some key examples of
cancer therapies associated with relevant genetic variants are listed in Table 8-1.
Purine Analogs: 6-Mercaptopurine and 6-Thioguanine

Purine analogs such as mercaptopurine and thioguanine are used in the treatment of
acute lymphoblastic leukemia (ALL). Mercaptopurine and thioguanine are activated to form
thioguanine nucleotides that are incorporated into DNA and RNA, causing DNA damage
and cytotoxicity. A major metabolic pathway involved in the inactivation of these drugs
is S-methylation catalyzed by thiopurine S-methyltransferase (TPMT) using S-adenosyl-
L-methionine as the S-methyl donor. Although metabolism of mercaptopurine and thio-
guanine by TPMT is a major route of their inactivation, the physiologic function of TPMT
is uncertain.
Multiple genetic variations of the TPMT gene have been identified.10 However, at least
three variants are of sufficient prevalence to have a clinical impact. They are designated
TPMT *2, *3A, and *3C. The *2 variant is due to a G238C polymorphism that results in the
substitution of an alanine with a proline at amino acid 80 and a 100-fold decrease in
enzyme activity. The *3A variant has been ascribed to multiple differences in a single gene
that includes G460A and A719G variants resulting in the substitution of an alanine with a
threonine and a tyrosine with a cysteine at amino acids 154 and 240, respectively. The *3C
variant is representative of the A719G variant only of *3A. As with the *2 variant, *3A and
*3C variants have greatly reduced TPMT activity. Because of the severe toxicity associated
with reduced mercaptopurine or thioguanine clearance in individuals harboring these TPMT
variants, screening for these variants prior to treatment with these purine analogs is now
commonplace.
An important pharmacogenomic consideration in the use of purine analogs is that
reduced dosing levels due to the presence of TPMT variants does not reduce therapeutic
efficacy, but greatly reduces toxicity associated with the use of purine analogs (reviewed in
10). At least 15 other variants of TPMT have been identified but at lower frequencies than
Table 8-1
Pharmacogenomics of Cancer Chemotherapeutic Agents
Cancer
Chemotherapeutic
Drug Class Agent Gene(s)
Alkylating agents Cyclophosphamide CYP2B6, CYP2C9, CYP2C19, CYP3A4
Folic acid antimetabolites Methotrexate DHFR, TYMS, SLCO1B1

Hormone modulators Tamoxifen CYP2D6
Purine antimetabolites Mercaptopurine TPMT

Thioguanine TPMT
Pyrimidine antimetabolites Gemcitibine Cytidine deaminase
5-Fluorouracil DPYD, TYMS
Topoisomerase inhibitors Irinotecan UGT1A1
TPMT *2, *3A, and *3C. Variations among ethnic groups are extensive10 and may necessitate
further characterization of additional TPMT variants for therapeutic considerations among
specific ethnic groups.
CLINICAL PEARL
Treatment with thiopurines can produce severe toxicities. Because

opportunities to achieve therapeutic efficacy are limited, recognizing
factors affecting therapy are critical and can impact thiopurine dosing. For
example, reductions in 6-mercaptopurine dosing levels may be instituted
for a single TPMT variant form (i.e., those individuals heterozygous
that have a TPMT variant with reduced activity). More extreme dosage
reductions are instituted for those individuals with two TPMT variants
having reduced activity (i.e., homozygous for reduced activity TPMT
variants). In these individuals, dose reductions of up to 90% the standard
dosing levels may be warranted.11
Pyrimidine Analogs: 5-Fluorouracil and Gemcitabine

Pyrimidine analogs are used in the treatment of a variety of cancers, including colorectal
and gastic cancers. The drug 5-FU is a pyrimidine analog that acts as a uracil analog and
functions, in part, by inhibiting TYMS, which results in decreased thymidine production
and inhibition of DNA synthesis. 5-FU itself is not active but requires metabolism through
pyrimidine salvage pathways to produce the active forms.
TYMS is expressed in both normal and cancer cells and is a key enzyme involved in the
synthesis of thymidylate. Because TYMS is the key target of 5-FU activity, genetic variants
of the TYMS gene have been investigated at length.4 Polymorphic variants affecting TYMS
expression have been identified with variations in the 5' and 3' untranslated regions of the
TYMS gene.12 For the 5' untranslated region, 3 copies of a 28 base pair tandem repeat have
been described.12 For the 3' untranslated region, a 6 base pair deletion has been described
that results in reduced TYMS gene expression.12 However, the impact of these variants on
TYMS gene expression and patient response to 5-FU have shown conflicting results.13 In
contrast, TYMS variants may affect response to folic acid antimetabolite therapies.
A key enzyme involved in pyrimidine catabolism is dihydropyrimidine dehydrogenase
(DPYD), which is important in the breakdown of 5-FU. It is estimated that approximately
5-10% of the population possess a deficiency in DPYD enzyme activity. DPYD is a complex
gene in which many genetic variants have been identified. A splice site variant, designated
IVS14+1G>A, results in a deficiency of DPYD with resultant 5-FU associated toxicity, as
reported by Van Kuilenburg et al.14 In this study, 28% of patients were found to harbor the
IVS14+1G>A variant.14 Because of its high prevalence, screening for the IVS14+1G>A DPYD
variant in patients before 5-FU administration was recommended.14 Further, the DPYD
variants c.1236G>A/HapB3 and c.1679T>G have been reported as clinically relevant predic-
tors of fluoropyrimidine-associated toxicity where screening for these variants, as well as
DPYD*2A and c.2846A>T variants, is recommended to improve the safety of treating cancer
patients with fluoropyrimidines.15
Gemcitabine is a deoxycytidine analog with a broad spectrum of antitumor activ-
ity, including hard-to-treat cancers such as pancreatic and non-small cell lung cancer.
Inactivation of gemcitabine occurs by metabolism to 2′, 2′-difluorodeoxyuridine through
the actions of cytidine deaminase. Sugiyama et al. examined the pharmacokinetics of

gemcitabine in a study of 256 Japanese patients receiving a 30-minute intravenous infu-
sion of 800 or 1,000 mg/m2 gemcitabine.16 It was found that individuals with a G208A vari-
ant of cytidine deaminase (designated CDA*3), which has a threonine in place of an alanine
at amino acid 70, had decreased clearance of gemcitabine and an increased incidence of
neutropenia when coadministered with 5-FU. In fact, a patient homozygous for the CDA*3
variant suffered life-threatening toxicity after gemcitabine treatment.17 The CDA 79A>C
variant has also been identified as a marker of gemcitabine-based toxicity with recommen-
dations for polymorphism testing prior to chemotherapy.18
Folic Acid Antimetabolites: Methotrexate
Folic acid antimetabolites were among the first cancer chemotherapies with broad use in
cancer treatment including breast, head and neck, gastrointestinal, and lung cancers as
well as osteosarcomas and ALL. A primary action of these agents is the blockade of dihy-
drofolate reductase (DHFR). Inhibition of the conversion of folic acid to tetrahydrofolate
results in cell cycle arrest due to inhibition of DNA, RNA, and protein synthesis. Variants in
TYMS have also been suggested to affect the efficacy of folic acid antimetabolites. Vari-
ants of SLCO1B1 also affect methotrexate activity. Both pharmacodynamic and pharma-
cokinetic considerations are believed to contribute to the pharmacogenomics of folic acid
antimetabolites.
Few variants of the DHFR gene have been identified. In a laboratory setting, cells
exposed to methotrexate have developed mutations in the DHFR gene that confer resis-
tance to methotrexate.19 However, these mutations have not been linked to the use of
methotrexate in a clinical setting. Similarly, genetic amplifications of the DHFR gene have
been produced in cell lines cultured to develop methotrexate resistance.20,21 Currently, it is
unclear whether or not variants of DHFR affecting enzyme activity or levels of DHFR have
a significant impact on the action of folic acid antimetabolites in cancer treatment.
As with pyrimidine analogs, variants in TYMS are important in the pharmacogenomics
of folic acid antimetabolites. Anti-cancer drugs such as methotrexate inhibit this enzyme in
cancer cells to interfere with cancer cell nucleic acid metabolism. As discussed previously, an
indel of two or three 28 base pair repeats has been identified in the promoter of the TYMS
gene.12 The presence of a triplet of the 28 base pair repeat is associated with higher levels
of TYMS expression.4 Krajinovic et al. examined the variability in response to methotrexate
in children treated for ALL based on the presence of these variants.5 Among 205 children
treated with methotrexate, those homozygous for the triple 28 base pair repeat were found
to have a worse outcome than children with other TYMS gene variants.5 Based on this obser-
vation, the investigators concluded that “genotyping of thymidylate synthase might make it
possible to individualize treatment for patients with acute lymphoblastic leukemia.”5
The SLCO1B1 gene encodes for the OATP1B1 liver-specific, transmembrane receptor
that mediates the uptake of both endogenous and drug compounds. An example of a drug
that is affected by OATP1B1 includes methotrexate, which is transported from the blood
into liver cells. Variants in the SLCO1B1 gene are associated with a protein that has impaired
transporter function. Ramsey et al report that SLCO1B1 genetic variants account for 10.7%
of the variability observed in methotrexate clearance.22
Topoisomerase I Inhibitors: Irinotecan
Irinotecan is used for the treatment of metastatic colorectal and lung cancers. It is a semi-
synthetic derivative of camptothecin that exhibits anti-cancer activity through inhibition of
topoisomerase I, which is necessary for DNA replication. Biotransformation of irinotecan

to SN-38 by carboxylesterases is required for the potent anti-tumor activity of irinotecan.
From a pharmacogenomic standpoint, polymorphic variants of carboxylesterases would be
expected to affect the efficacy of irinotecan. However, variants affecting SN-38 production
have not been described. Instead, the inactivation of SN-38 by glucuronidation has phar-
macogenomic consequences.
UDP glucuronosyltransferases are a family of enzymes that are responsible for the
glucuronidation of drugs, steroids, bilirubin, and other fat-soluble molecules to form excre-
table metabolites. SN-38 is a substrate for UDP glucuronosyltransferase 1A1 (UGT1A1).
Variants of the UGT1A1 gene are associated with Crigler-Najjar and Gilbert syndromes.
In addition, variants are associated with decreased SN-38 excretion. Ratain’s group first
reported the altered metabolism of SN-38 in liver cells from individuals with a variant
in the promoter region of the UGT1A1 gene.23 This variant, designated UGT1A1*28, has an
additional TA repeat in the promoter region of the UGT1A1 gene (i.e., 7 TAs compared to
6) that results in decreased UGT1A1 liver expression. The reduced levels of UGT1A1 lead to
decreased SN-38 hepatic excretion and toxicity due to elevated SN-38 levels. Interestingly,
this variant is also commonly associated with Gilbert’s syndrome, which is manifested as
a mild form of hyperbilirubinemia.24,25 For treatment of metastatic colorectal cancer, Li et
al. found that UGT1A1*28 and UGT1A1*6 variants were predictors of irinotecan-associated
severe diarrhea, with no association of UGT1A1 polymorphisms for severe neutropenia in a
Chinese patient population.26
Selective Estrogen Receptor Modulators: Tamoxifen and Raloxifene

Tamoxifen and raloxifene are selective estrogen receptor modulators (SERMs) that act as
estrogen antagonists in estrogen receptor-positive breast cancers.27 Many breast cancers
express estrogen receptor alpha that in the presence of estrogen acts to promote breast
cancer cell growth. Inhibition of estrogen receptor activation reduces the proliferative activ-
ity of breast cancer cells. In this regard, SERMs block the interaction of estrogens with the
estrogen receptor, thereby reducing breast cancer cell proliferation. Tamoxifen is prescribed
for the treatment of premenopausal women with estrogen receptor-positive breast cancer,
whereas raloxifene may be prescribed for treatment in postmenopausal patients.27
As pointed out previously, the CYPs are a large family of enzymes responsible for the
oxidative metabolism of numerous xenobiotics, including many cancer chemotherapeutic
agents.1 One of the most polymorphic forms of the CYPs is CYP2D6, which is also one of
the most important in the metabolism of tamoxifen to one of its active forms, 4-hydroxy-
tamoxifen. However, polymorphic variants of CYP2D6 exist that code for proteins, which do
not efficiently metabolize tamoxifen to 4-hydroxytamoxifen.28 One of the more prevalent
variants with reduced activity is designated CYP2D6*4, present in 17-21% of the Caucasian
population. CYP2D6*4 occurs from a G to A variation present in the first base pair of exon
4 of the CYP2D6 gene, resulting in a splice site shift and the generation of a premature
termination codon. Individuals homozygous for CYP2D6*4 lack CYP2D6 activity, and, thus,
individuals homozygous for CYP2D6*4 are deemed poor metabolizers. Although conflict-
ing results are observed when comparing clinical studies on the efficacy of tamoxifen and
CYP2D6 variants, in 2006 the U.S. Food and Drug Administration recommended an update
on the tamoxifen label to reflect an increased recurrence in breast cancer patients who are
CYP2D6 poor metabolizers.29
CLINICAL PEARL
Metabolism of drugs by metabolic enzymes such as CYPs is often
associated with drug excretion, where impairment of activity may be
associated with increased levels of parent drug and increased toxicity. As
discussed, CYP variants can have a pronounced effect on drug excretion.

Alternatively, as in the case of tamoxifen, therapeutic efficacy of a drug
may be dictated by metabolism to the active form by metabolic enzymes,
which may also be affected by the presence of specific polymorphic
variants. Therefore, drug action resulting from genetic variants is not
always predictable and requires a careful pharmacological assessment
before the full physiologic effects of these variants are accurately
understood.
Taxanes: Paclitaxel and Docetaxel

Taxanes are cancer therapeutic agents that act through promoting and stabilizing the
assembly of microtubule formation by preventing depolymerization. Taxanes include pacli-
taxel, which is a dipertene plant derivative from the needles and bark of the western yew
tree, and docetaxel, a semi-synthetic analog of paclitaxel that is derived from the European
yew tree. Taxanes have broad efficacy for the treatment of many forms of cancer. Phar-
macokinetic considerations have been a focus of pharmacogenomic studies evaluating the
cancer chemotherapeutic efficacy of the taxanes.
Metabolism of docetaxel by CYPs is considered a major route of elimination.1 For exam-
ple, CYP3A4 is known to metabolize docetaxel to its inactive hydroxylated forms. Thus, high
CYP3A4 activity would be expected to result in reduced therapeutic efficacy of docetaxel.
In support of this concept, Engels et al. found that inhibition of CYP3A4 using ketoconazole
decreased the clearance of docetaxel.30 Clinically, although many CYP3A4 variants have
been recognized, there is currently limited support that these variants affect docetaxel
activity in cancer treatment. Because of interindividual diversity in paclitaxel pharmacoki-
netics and the identification of a CYP3A4 haplotype associated with paclitaxel pharmaco-
kinetics, applications of pharmacogenomics to optimizing paclitaxel therapy are advancing.31
Alkylating Agents: Cyclophosphamide
Cyclophosphamide is an alkylating agent that is used primarily for the treatment of
lymphomas and breast cancer. As with tamoxifen, cyclophosphamide is a prodrug that
requires metabolism by CYPs to produce the active form of the drug.1 Multiple CYP forms
are able to activate cyclophosphamide, including CYP 2B6 and 2C19.1 Studies are ongoing
to determine the effect of variants in these CYP genes on the efficacy of cyclophospha-
mide therapy.
PHARMACOGENOMICS OF CANCER CHEMOPREVENTION

Genetic variants exist that are associated with a predisposition to cancer. Development of
multiple cancers, an early age of onset, and the development of cancers among relatives
may all be suggestive of hereditary factors contributing to cancer predisposition.32 A goal
of pharmacogenomics includes the institution of preventive measures to reduce the likeli-
hood of developing cancers, which may be greatly improved by performing genetic screens
for variants associated with cancer predisposition. In addition, cancer chemopreventive
measures may be prudent among individuals that engage in high-risk activities for cancer
development. Strategies for cancer prevention can be divided in three distinct areas: primary,
secondary, and tertiary prevention.
Primary cancer prevention involves preventive measures among individuals not
currently diagnosed with cancer but who may be involved in high-risk activities (e.g.,
working with radiation) or who have a genetic predisposition to cancer development. An

example of a promising cancer primary chemopreventive intervention includes the use of
nonsteroidal anti-inflammatory drugs (NSAIDs) for the prevention of colorectal cancer
development.33 Colorectal cancer is the third leading cause of cancer cases and deaths
in the United States.34 Results from epidemiologic studies support that regular aspirin
intake is associated with a reduced risk of colorectal cancer development.35 Other NSAIDs
have also shown efficacy in prevention of colorectal cancer.36 It is anticipated that genetic
factors will play a significant role in determining the efficacy of colorectal cancer preven-
tive measures, which in turn will open the field of chemopreventive pharmacogenomics in
the primary prevention setting.36
Secondary prevention involves measures among individuals diagnosed with premalig-
nant lesions to reduce the chance that these lesions progress to more advanced stages of
disease. An example of secondary prevention may include the institution of chemopreven-
tive measures on the identification of preneoplastic lesions in colon cancer (e.g., polyps).37
Tertiary prevention measures are taken by individuals who have been treated for cancer
to prevent further cancer development. An example of tertiary prevention is the use of
tamoxifen to prevent tumor development in a contralateral breast in high-risk individuals.
Pharmacogenomic considerations that apply to tamoxifen therapy for breast cancer are
also relevant to tamoxifen chemoprevention of breast cancer. The breast cancer preven-
tive potential of tamoxifen in women with breast cancer predisposing genetic variants in
BRCA1 or BRCA2 was studied for contralateral breast cancer development.38 In this study,
tamoxifen treatment was found to reduce the risk of contralateral breast cancer by 50%.38
In addition to colon and breast cancer, chemoprevention of cancer is being investigated
for head and neck, lung, bladder, prostate, skin, and cervical cancer.39 With an increased
understanding of cancer genetics, the ability to estimate a risk to cancer development
will become an increasingly important element of healthcare. In the future of pharmacy
practice, it is likely that cancer pharmacogenomics will include greater considerations for
cancer chemopreventive measures.
CANCER GENETICS AND PHARMACOGENOMICS

Cancer is a genomic disease. During cancer initiation and progression, genetic modifica-
tions (i.e., mutations) develop within the disease. In the early stages of cancer, mutations
occur in a select subpopulation of cells (i.e., the nascent cancer cell) that may facilitate
the progression of the cancer. In many instances, the molecular abnormalities caused by
the cytogenetic changes can be targeted therapeutically (Table 8-2). Key examples include
the treatment of breast cancers over-expressing an amplified Her-2/neu gene (e.g., treated
with trastuzumab40 and pertuzumab41) and chronic myelogenous leukemia associated with
translocations resulting in a novel fusion gene Bcr-Abl (e.g., treated with imatinib42 and
dasatinib43). Treatment of lung cancers harboring gene mutations in the epidermal growth
factor receptor (EGFR) have been targeted with tyrosine kinase inhibitors such as gefitinib,
erlotinib, and afatinib.44 Melanomas with V600E mutations in BRAF have also been selec-
Table 8-2
Examples of Genetic Modifications in Cancers and Targeted Therapies
Cancer Type Genetic Modification Targeted Therapy
Hematologic cancers
Acute myeloid leukemia FLT3 mutations quizartinib (clinical trials)
Acute promyelocytic leukemia PML:RARa - t(15;27) all-trans retinoic acid
Chronic myelogenous leukemia BCR:ABL fusion imatinib, dasatinib
Carcinomas
Breast cancer ErbB2 amplification trastuzumab, pertuzumab

Colorectal cancer EGFR & KRAS mutations cetuximab, panitumumab
Lung cancer EGFR mutations erlotinib, gefitinib
Melanoma BRAF mutations vemurafenib, dabrafenib
tively targeted (e.g., vemurafenib45 and dabrafenib46). Mutations found in cancer cells are
not heritable as they would likely be detrimental to the development of the organism.
However, cancer syndromes involving genetic factors that may predispose an individual to
cancer development will be described later. In an increasing number of cases, drug therapy
for a specific cancer targets the proteins that are associated with gene mutations. As such,
genetic considerations in cancer pharmacogenomics are to a large degree distinct from
pharmacogenomic considerations associated with inborn genetic variants.
Genetic Changes Occurring During Cancer Progression

Hematologic Malignancies
Hematologic cancers are those malignancies that affect the blood-forming cells and are
distinct from solid cancers in many respects, including the therapeutic regimens used to
treat these diseases. Examples of hematologic malignancies include ALL, acute myeloid
leukemia (AML) and the subtype acute promyelocytic leukemia, chronic lymphocytic leuke-
mia, and chronic myelogenous leukemia. Many hematologic malignancies are associated
with gene translocations that have critical roles in the development and maintenance of
the cancerous phenotype.47 In some instances, the gene product associated with these
translocations has been exploited as a therapeutic target.
ACUTE LYMPHOBLASTIC LEUKEMIA. ALL is the most common childhood cancer.48 Multiple
genetic lesions that are associated with ALL have proven useful in determining disease
prognosis. Historically, these genetic lesions have been determined cytogenetically. For
example, a chromosomal translocation that occurs in ALL is t(12;21), which is associated
with good prognosis. However, not all genetic changes in ALL are easily identifiable and
some patients experience treatment failure or disease relapse after therapy or may experi-
ence severe treatment-related toxicities.49
Significant progress has occurred in the treatment of ALL over the past four decades.
As discussed previously, antimetabolites such as 6-mercaptopurine and methotrexate may
be used in ALL treatment. Thus, the pharmacogenomic considerations that apply to these
drugs must be considered in ALL treatment. For example, variants in TPMT, as discussed,
should be assessed to avoid severe toxicity associated with 6-mercaptopurine among indi-
viduals carrying less active variants of this gene.
ACUTE MYELOID LEUKEMIA. AML is more common than ALL and occurs primarily in adults.
AML is a heterogeneous group of leukemia that results from clonal transformation of

hematopoietic precursors through the acquisition of chromosomal rearrangements and
multiple gene mutations.50 In addition to cytogenetic considerations, dysplastic features,
the patient’s preceding history, and age are important prognostic factors for AML.
As with most leukemia, AML can be characterized by genetic modifications present in
the disease. An example of a genetic change present in approximately one-third of AML
patients are mutations in the FLT3 gene.51 FLT3 (fms-like tyrosine kinase 3) is a recep-
tor tyrosine kinase expressed in hematopoietic stem/progenitor cells. Mutations of FLT3
in AML patients include internal tandem duplications and point mutations that lead to
a constitutively active receptor.51 Presence of these mutations in AML patients has been
associated with a higher relapse rate to conventional therapies. Tyrosine kinase inhibitors
targeting FLT3 have been designed and are in clinical trials for AML treatment. However,
the development of resistance has proven to be a challenge to successful FLT3 inhibitor
therapy.51
Acute promyelocytic leukemia (APL) is a subtype of AML that is cytogenetically clas-
sified by a balanced reciprocal translocation between chromosomes 15 and 17, resulting in
the fusion between the promyelocytic leukemia gene and the retinoic acid receptor alpha
gene.52 The complete remission rate for APL has greatly increased since the use of all-
trans retinoic acid (ATRA) was introduced for the treatment of APL due to its action as a
differentiating agent. However, relapse following ATRA treatment still occurs such that it
must be combined with other chemotherapies to optimize response. Interestingly, altered
expression of CYP2D6, which is important in retinoid metabolism, may have an impact in
relapse to ATRA treatment.53
CLINICAL PEARL
Prior to the use of ATRA, the long-term response rate for APL treatment
was lower than 10%. Although relapse can occur with ATRA, the addition
of ATRA to APL treatment regimens has increased the number of long-
term survivors to nearly 80%.54 ATRA therapy for APL represents a
strategy based on rational drug design that resulted from a heightened
understanding of molecular lesions associated with specific cancers.
CHRONIC LYMPHOCYTIC LEUKEMIA. Chronic lymphocytic leukemia (CLL) is a blood cancer

of B-lymphocyte origin that occurs primarily in adults. Prognostic groups of CLL are based
on genetic changes.55 One of the more common genetic modifications associated with CLL
is a deletion located in the long arm of chromosome 13. The specific genetic elements
contributing to CLL from chromosome 13 deletion are not known. However, because this
likely represents a deletion of a tumor suppressor gene and the absence of a target, thera-
peutic interventions based on this molecular lesion may not lend themselves to targeted
therapeutics.
CHRONIC MYELOGENOUS LEUKEMIA. As with other leukemia, chronic myelogenous leuke-

mia (CML) is cytogenetically classified by a reciprocal translocation. In CML, this transloca-
tion occurs between chromosomes 9 and 22 involving a fusion of the ABL proto-oncogene
from chromosome 9 with the breakpoint cluster region of chromosome 22. This chromo-
somal fusion is referred to as the Philadelphia chromosome. Upon translocation, the regu-
lation of ABL expression is altered resulting in oncogenic character. More specifically, the
ABL protein has tyrosine kinase activity that alters cell cycle regulation. A breakthrough
in CML treatment occurred with the discovery of imatinib, a comparatively selective ABL
tyrosine kinase inhibitor.56 The success of imatinib in CML treatment has prompted the
discovery of other tyrosine kinase inhibitors for cancer therapy, including advanced agents
for the treatment of CML, and has served as a paradigm for rational drug design.42,57
Carcinomas
Early models of carcinoma development illustrated that cancer proceeds in multiple
stages, contributing to the progression of the disease.58 In these models, cancer develop-
ment begins with the initiating lesion, representing the first stage in cancer development,
which is genetic in origin. During the initiation, mutations occur that can facilitate cancer
development. The subsequent promotion stage of cancer development may also involve
carcinogenic mutations. However, promotion is primarily associated with changes in gene
expression that further facilitate cancer development. The progression stage of cancer
development is associated with the production of the key genetic changes that lead to
malignancy. Note that genetic lesions associated with sarcoma progression have also been
identified that may have important therapeutic implications. Because sarcomas are less
prevalent compared to carcinomas, they are not covered here. The numbers of genetic
lesions identified that are associated with the stages of carcinoma development are
growing rapidly, facilitating the parallel discovery of novel therapeutic targets. A few key
examples are provided here.
BREAST CANCER. The identification of BRCA1 and BRCA2 gene variants associated with
familial breast cancer represents a historical breakthrough in medical genetics. Breast
cancer is the most prominent form of cancer to develop among women in the United
States.34 Among these cancers, approximately 10% are associated with mutations in BRCA1
or BRCA2.59 As noted in the pharmacogenomics of cancer chemoprevention, tamoxifen use
was found to reduce the risk of contralateral breast cancer in women with predisposing
variants in the BRCA1 or BRCA2 genes.60
CLINICAL PEARL
Ethnic heritage may provide useful information regarding inherited
genetic variants. For example, women of Ashkenazi Jewish descent have
a 21-30% presence of founder mutations in BRCA1 or BRCA2 among
those diagnosed with breast cancer before 50 years of age compared to
an approximately 6.1% rate among non-Jewish women.61 Thus, ethnicity
may provide clues to assist the identification of inherited variants that can
impact therapeutic decisions.
Gene amplifications observed in cancers may result in the overexpression of proteins

that contribute to disease pathology. An example is the ErbB2 protein (also referred to as
Her2/neu) in breast cancer. The ErbB2 gene codes for a protein that belongs to the same
family of proteins as the EGFR (i.e., ErbB1). Overexpression of ErbB2 is associated with
a poorer prognosis in breast cancer. Monoclonal antibody therapies (e.g., trastuzumab)
targeted at the ErbB2 protein can be effective in the treatment of breast cancers over-
expressing ErbB2. Analysis of breast cancers for ErbB2 overexpression has become a standard
of practice to optimize therapies for breast cancer using monoclonal antibody therapies.
COLORECTAL CANCER. Genomic changes associated with the progression of colorectal

cancer have been extensively investigated.62 Among the lesions identified, activating
mutations in the KRAS gene are present in approximately 40% of these cancers and likely
represent an early event (i.e., initiating lesion) in colorectal cancer progression. The pres-
ence of activating mutations in the KRAS oncogene of colorectal cancers has been found
to influence the efficacy of EGFR-targeted therapies.63 For example, cetuximab, an anti-
EGFR antibody, has been used effectively in the treatment of advanced colorectal cancer.
Lievre et al. reported a 0% response to cetuximab among 24 patients with KRAS muta-
tions in their cancer compared to a 40% response among 65 patients without these muta-
tions (P <0.001).64 These results, combined with results from previous studies, strongly
suggest that the status of KRAS mutations should be considered in cetuximab treatment
for advanced colorectal cancer.64
PROSTATE CANCER. Gene fusions and chromosomal rearrangements, such as those

associated with hematologic malignancies, have not been investigated as extensively in
carcinomas. A number of molecular lesions have been identified with significant frequency
in prostate cancer. An example is the TMPRSS2/ETS family gene fusions.65 These gene
fusions involve the androgen-regulated TMPRSS2 gene located on chromosome 21, which
codes for a transmembrane serine protease. The ETS (erythroblastosis virus E26 trans-
formation sequence) family members that are found fused with TMPRSS2 include ERG,
ETV1, or ETV4, which are transcription factors with transforming potential (i.e., change cells
to possess cancer-associated characteristics). Because the TMPRSS2 gene is androgen-
sensitive, current hormonal ablation therapies may act in part through downregulation of
the fusion product activity. TMPRSS2/ETS translocations in prostate cancer have important
diagnostic and clinical therapy implications.65
MELANOMA. Melanocytes are cells derived from the neural crest during embryogenesis
that produce the pigment melanin and are the cellular origin for the development of
melanoma. More than half of malignant melanomas have identifiable oncogenic lesions,
predominantly in BRAF and NRAS genes. For example, Kim et al. performed a meta-
analysis of BRAF mutations and their association with clinicopathological characteristics
and found V600-mutated BRAF (e.g., V600E, V600K, and V600R) in primary melanomas
to be present in 40.3% of the Caucasian populations examined, which was significantly
higher than that found in Asian populations at 19.5%.66 BRAF inhibitors developed for the
treatment of melanoma include vemurafenib and dabrafenib.67 Initial clinical studies with
vemurafenib in patients with V600E melanomas were promising therapeutically, reporting
a partial or complete cancer regression in the majority of patients.68 Similarly, dabrafenib
was developed as an inhibitor of mutated BRAF and was found to significantly improve
progression-free survival for treatment of metastatic melanoma.69 Interestingly, the mito-
gen-activated extracellular signal-regulated kinase (MEK) inhibitor trametinib has also

shown efficacy for the treatment of metastatic melanoma, which may involve the action
of pathways downstream from mutated BRAF.70 Studies investigating the frequency of
melanomas with mutations in NRAS report a wide variation with a mean of 26% of mela-
nomas.71 Remarkably, melanomas harboring mutations in both NRAS and BRAF genes in
the same tumor are uncommon. The development of therapies targeting activated NRAS
in melanomas has proven to be challenging in comparison to treatments for melanomas
with mutated BRAF.
GENETICS OF CANCER SYNDROMES. Familial connections to cancer were recognized long

before precise genetic variants were identified as contributing factors to these hereditary
diseases. Of particular importance are cancer-preventive measures for individuals predis-
posed to cancer syndromes. This will likely lead to the dispensing of drugs that is impor-
tant for cancer chemoprevention. Especially relevant is the recognition that COX inhibitors
are preventive for colorectal cancer. Daily aspirin supplementation is recognized as reduc-
ing colorectal cancer development. Moreover, the influence of these genetic contributors
in cancer development is now being incorporated therapeutically. Examples of cancers with
known familial contributing factors are listed in Table 8-3.
GENETIC SCREENING METHODS USED IN CANCER

PHARMACOGENOMICS
The base sequence of DNA refers to the consecutive order of guanine (G), adenine (A),
thymine (T), and cytosine (C) bases, which defines DNA as the genetic code. Under ideal
conditions, the complete DNA sequence for each person would be available for pharma-
cogenomic studies because this is the most comprehensive information for assessing poly-
morphic variants (e.g., by next-generation sequencing [NGS]). For example, DNA sequence
information allows the assessment of all forms of genetic variants including SNPs, indels,
and CNVs. Currently, this can be achieved effectively using NGS methods. Thus, for pharma-
cogenomic studies, DNA sequence information would be utilized as it contains all genetic
contributors relevant for assessing pharmacogenomic endpoints. However, the technologi-
cal complexity, time, and cost of a comprehensive sequencing of each patient’s DNA are
somewhat prohibitive for routine clinical performance. Thus, the less comprehensive, but
technically more feasible, use of microarray-based nucleic acid analysis continues to have
value in standard of practice.
Table 8-3
Cancer Syndromes and Associated Predisposing Genes
Cancer Syndrome Genes Associated with Syndrome
Familial breast and ovarian cancer BRCA1, BRCA2

Hereditary non-polyposis colorectal cancer MLH1, MSH2, MSH6, PMS1, PMS2
Multiple endocrine neoplasia MEN 1, MEN 2A/B
Li-Fraumeni syndrome TP53
Retinoblastoma RB1
Because microarray-based technologies are advancing for nucleic acid analysis in the
clinical setting, a basic understanding of the use of these methods is important. Different
types of microarrays are available for the analyses of nucleic acids. These technologies can
be used to analyze either DNA or RNA. DNA microarray technology relies on the hybridiza-
tion of complementary pieces of DNA present in relatively short stretches of DNA (e.g.,
approximately 25 base pairs = oligonucleotide). Over this stretch of DNA sequence, hybrid-
ization can be optimized so that only regions of DNA that match perfectly will hybridize. It
is this precise hybridization of complementary segments of nucleic acids present as part of
the microarray system with DNA, mRNA, or microRNA derived from a patient sample that
forms the basis of microarray analysis. Thousands to hundreds of thousands of different
oligonucleotides representing different genetic variants can be assayed in a single micro-
array analysis. Microarrays may also be used to determine cellular levels of mRNA being
expressed, which is referred to as gene expression profiling. Similar methods can be used
to screen microRNA expression, which is an emerging area in cancer pharmacogenomics.
Gene Expression Profiling

Gene expression profiling is used to assess the aberrant expression of genes associated
with a person’s cancer. Determination of mRNA expression in cancers is an increasingly
important means of profiling cancers and can be effective in identifying subgroups of
cancers. For the microarray analysis of mRNA, samples are typically required from both
normal and diseased tissue. This allows the identification of mRNA expression differences
that facilitate an understanding of abnormal mRNA expression occurring in cancer cells.
This information may guide an understanding of the pathological basis of the disease as
well as facilitate the identification of novel therapeutic targets. The ability to use gene
expression profiling to optimize cancer therapy, in part, distinguishes cancer pharmaco-
genomics from pharmacogenomics applied to other disease states. The profiling of genes
abnormally expressed in breast cancer is a key example of the clinical use of gene expres-
sion profiling.
Oncotype DX® assays are gene expression profiling platforms that can be performed
to help determine therapeutic strategies for breast, colon, and prostate cancer. The Onco-
type DX® Breast Cancer Assay is a gene expression-based test of 21 genes used to quan-
tify the likelihood of breast cancer recurrence in women with early-stage disease. The
genes analyzed include 16 breast cancer-related genes (e.g., Her2/neu, estrogen receptor,
progesterone receptor) and five reference genes.72 Currently, this test is recommended for
women with node-negative, estrogen receptor-positive invasive breast cancer, and post-
menopausal women with node-positive, hormone receptor-positive invasive breast cancer.
Results from the assay are provided in the form of a recurrence score. From clinical studies
performed using the Oncotype DX® Breast Cancer Assay, it was determined that “the likeli-
hood of distant recurrence at 10 years increases continuously with an increase in the assay
recurrence score result.”
CLINICAL PEARL
Breast cancer samples used for the Oncotype DX® Breast Cancer Assay
are provided as paraffin blocks derived from histological samples. A
multigene expression assay is performed, resulting in the assignment
of a recurrence score. A key question addressed is whether or not

chemotherapy in addition to tamoxifen treatment may be beneficial. The
likelihood of distant recurrence at 10 years increases with the recurrence
score value. A 6.8% rate of distant recurrence at 10 years is associated
with a recurrence score less than 18, suggesting these patients derive
minimal benefit from additional chemotherapy. In contrast, a 30.5% rate
is associated with a score of greater than or equal to 31, where substantial
benefit may be gained from additional chemotherapy.
Other gene expression profiling platforms include the more recently developed
Oncotype DX® Colon Cancer and Oncotype DX® Prostate Cancer Assays. Similar to the
Oncotype DX® Breast Cancer Assay, the Oncotype DX® Colon Cancer Assay provides a
recurrence score value that may be useful in developing a treatment plan for the individual
colon cancer patient.73 The Oncotype DX® Prostate Cancer Assay builds on clinically useful
information provided by other parameters, such as Gleason score and prostate-specific
antigen that may be useful in helping to distinguish indolent and aggressive disease.74
Results from these assays allow the inclusion of more informed treatment decisions based
on pharmacogenomic information.
Gene expression profiling is also being used to estimate the response of patients to
therapy. For example, these methods are being developed to assess the characteristics and
to optimize the treatment of hematological malignancies.75 To date, treatments for leuke-
mia are largely based on genotypic characteristics of the leukemic blast, clinical features
of the patient, and the initial therapeutic response.76 In one example, a study was carried
out to assess the utility of gene expression profiling to identify (at initial diagnosis) children
who may have a poor therapeutic response.77 In this study, it was concluded that altered
expression of early blast regression genes may be helpful in identifying patients who may
be at risk for inferior responses to treatment.77 Examples from multiple cancers illustrate
the increasing utility of gene expression profiling in cancer pharmacogenomics to improve
cancer therapy. Gene expression profiling for cancers is becoming a standard of practice in
precision oncology.
Screening for Genetic Variation in Cancer Susceptibility and Drug

Response
Microarrays can be used to determine inherited genetic variations in a person’s DNA (e.g.,
DNA obtained from a buccal swab or sputum sample). Such an analysis can be performed to
assess the many variants as described in the section Pharmacogenomics of Cancer Chemo-
therapeutic Agents. The majority of these microarray assays examine the presence of known
SNPs. Alternatively, DNA from a tumor sample can be analyzed by microarrays to determine
mutations that have arisen in a specific cancer. In fact, genome-wide profiling may be used
to identify genetic changes in specific types of cancer to determine consistency of genetic
changes in these cancers. For example, using microarray-based techniques to examine
genetic alterations in ALL, genetic changes in key cellular pathways including lymphoid
differentiation, cell cycle regulation, and drug responsiveness have been identified.78 The
analysis of variations in DNA may greatly facilitate the identification of strategies to individ-
ualize cancer therapy. Key examples include the treatment of breast cancers overexpressing
an amplified Her-2/neu gene.40 An advantage of microarray analyses of DNA is the ability to
analyze thousands of informative genetic changes from a single sample.
MicroRNAs IN CANCER PHARMACOGENOMICS

The expression of microRNAs (miRNAs) in cancers represents a novel consideration for
utilizing genomic information to optimize cancer drug therapy. MiRNAs are short RNA
molecules, approximately 22 nucleotides in length that are noncoding and act to alter
gene expression through destabilizing mRNA or altering translation activity.79 The activity
of multiple mRNAs may be affected by a single miRNA. Methods similar to those discussed
for gene expression profiling can be used to profile expression of miRNAs.
MiRNAs active in cancer are referred to as oncomiRs and may have oncogenic or tumor
suppressor activity.79 OncomiRs have been associated with many different forms of cancer
including colon, lung, pancreatic, prostate, and breast cancer.79 In certain cancers, such as
breast cancer, miRNAs have been identified that serve as diagnostic biomarkers and prog-
nostic indicators as well as predictors of therapeutic outcomes.80 It is expected that miRNA
expression may provide valuable information for response to chemotherapy, but a compre-
hensive evaluation of it is beyond the scope of this chapter. However, the impact of miRNAs
on tamoxifen in breast cancer serves as a useful paradigm for this rapidly expanding field.
As an example, Rodriguez-Gonzalez et al. have shown that higher expression levels of miR-
30c may serve as an independent predictor of benefit to tamoxifen treatment in patients
with advanced breast cancer.81 Both miR-221 and miR-222 have been associated with resis-
tance to tamoxifen therapy in breast cancer.82 Functionally, elevated expression of miR-221
and miR-222 can act to down-regulate estrogen receptor alpha expression where these
miRNAs may have a significant role in development of estrogen receptor alpha-negative
breast cancer and, therefore, resistance to tamoxifen therapy.
Other studies have shown that no single miRNA serves as an accurate predictor of
adjuvant tamoxifen as a monotherapy.83 Instead, comprehensive profiling of miRNAs may
be necessary to provide useful information to guide therapeutic choices. Thus, miRNA
profiling in cancer is emerging as a component of a comprehensive effort to identify phar-
macogenomically relevant biomarkers that provide additional critical information for the
efficacy of specific treatment regimens based on the hallmarks of an individual’s cancer.
SUMMARY
Cancer pharmacogenomics includes both distinct and traditional elements of pharma-
cogenomics. From a traditional standpoint, polymorphic variants are known that affect
both pharmacokinetic and pharmacodynamic aspects of pharmacogenomics in cancer
chemotherapy. Unique to cancer pharmacogenomics is the determination of specific muta-
tions associated with an individual’s cancer. In addition, through the use of gene expression
analyses, characteristics of each independent cancer can be determined, which will allow
for highly personalized cancer therapeutics. An emerging area of cancer pharmacogenom-
ics is the identification of genetic variants associated with cancer development that may be
modulated by chemopreventive agents. These agents hold strong potential as preventive
measures for specific cancers. The field of chemoprevention is poised to become a major
component of pharmacy practice. These unique areas of pharmacogenomics combine with
traditional pharmacogenomics to illustrate the enormous potential of cancer pharmaco-
genomics in pharmacy practice.
ACKNOWLEDGMENTS
The author is grateful for the insightful advice provided by Drs. Mark Holdsworth, PharmD,
Debra MacKenzie, PhD, and Ian Rabinowitz, MD, in the preparation of this chapter.
REFERENCES
1. Rodriguez-Antona C, Gomez A, Karlgren M, et al. Molecular genetics and epigenetics of the cyto-
chrome P450 gene family and its relevance for cancer risk and treatment. Hum Genet. 2010;127:1-17.
2. Nelson DR, Zeldin DC, Hoffman SMG, et al. Comparison of cytochrome P450 (CYP) genes from the
mouse and human genomes, including nomenclature recommendations for genes, pseudogenes and
alternative-splice variants. Pharmacogenetics. 2004;14:1-18.
3. Zhou S-F. Polymorphism of human cytochrome P450 2D6 and its clinical significance: Part I. Clin
Pharmacokinet. 2009;48:689-723 (Part II. 761-804).
4. Marsh S. Thymidylate synthase pharmacogenetics. Invest New Drugs. 2005;23:533-537.
5. Krajinovic M, Costea I, Chiasson S. Polymorphism of the thymidylate synthase gene and outcome of
acute lymphoblastic leukemia. Lancet. 2002;359:1033-1034.
6. Noguchi K, Katayama K, Sugimoto Y. Human ABC transporter ABCG2/BCRP expression in chemo-
resistance: basic and clinical perspectives for molecular therapeutics. Pharmacogenomics Pers Med.
2014;7:53-64. doi: 10.2147/PGPM.S38295.
7. Hardwick LJ, Velamakanni S, van Veen HW. The emerging pharmacotherapeutic significance of the
breast cancer resistance protein (ABCG2). Br J Pharmacol. 2007;151:163-174.
8. Bruhn O, Cascorbi I. Polymorphisms of the drug transporters ABCB1, ABCG2, ABCC2, and ABCC3
and their impact on drug bioavailability and clinical relevance. Expert Opin Drug Metab Toxicol.
2014;10:1337-1354.
9. Hoffmeyer S, Burk O, von Richter O, et al. Functional polymorphisms of the human multidrug resis-
tance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression
and activity in vivo. Proc Natl Acad Sci USA. 2000;97:3473-3478.
10. Chouchana L, Narjoz C, Roche D, et al. Interindividual variability in TPMT enzyme activity: 10 years of
experience with thiopurine pharmacogenetics and therapeutic drug monitoring. Pharmacogenomics.
2014;15:745-757.
11. Eichelbaum M, Ingelman-Sundberg M, Evans WE. Pharmacogenomics and individualized drug therapy.
Annu Rev Med. 2006;57:119-137.
12. Nief N, Le Morvan V, Robert J. Involvement of gene polymorphisms of thymidylate synthase in gene
expression, protein activity, and anticancer drug cytotoxicity using the NCI-60 panel. Eur J Cancer.
2007;43:955-962.
13. Lurje G, Manegold PC, Ning Y, et al. Thymidylate synthase gene variations: predictive and prognostic
markers. Mol Cancer Ther. 2009;8:1000-1007.
14. Van Kuilenburg AB, Meinsma R, Zoetekouw L, et al. High prevalence of the IVS14+1G>A mutation in
the dihydropyrimidine dehydrogenase gene of patients with severe 5-fluorouracil-associated toxicity.
Pharmacogenetics. 2002;12:555-558.
15. Meulendijks D, Hendricks LM, Sonke GS, et al. Clinical relevance of DPYD variants c.1679T>G, c.
1236G>A/HapB3, and c. 1601G>A as predictors of severe fluoropyrimidine-associated toxicity: a system-
atic review and meta-analysis of individual patient data. Lancet Oncol. 2015;16:1639-1650.
16. Sugiyama E, Kaniwa N, Kim S-R, et al. Pharmacokinetics of gemcitibine in Japanese cancer patients:
the impact of a cytidine deaminase polymorphism. J Clin Oncol. 2007;25:32-42.
17. Ueno H, Kaniwa N, Okusaka T, et al. Homozygous CDA*3 is a major cause of life-threatening toxicities
in gemcitibine-treated Japanese cancer patients. Br J Cancer. 2009;100:870-873.
18. Ding X, Chen W, Fan H, et al. Cytidine deaminase polymorphism predicts toxicity of gemcitabine-based
chemotherapy. Gene. 2015;559:31-37.
19. Blakley RL, Sorrentino BP. In vitro mutations in dihydrofolate reductase that confer resistance to
methotrexate: potential for clinical application. Hum Mutat. 1998;11:259-263.
20. Singer MJ, Mesner LD, Friedman CL, et al. Amplification of the human dihydrofolate reductase gene
via double minutes is initiated by chromosome breaks. Proc Nat Acad Sci USA. 2000;97:7921-7926.
21. Banerjee D, Mayer-Kuchuk P, Capiaux G, et al. Novel aspects of resistance to drugs targeted to dihy-
drofolate reductase and thymidylate synthase. Biochim Biophys Acta. 2002;1587:164-173.
22. Ramsey LB, Bruun GH, Yang W, et al. Rare versus common variants in pharmacogenetics: SLCO1B1
variation and methotrexate disposition. Genome Res. 2012;22:1-8.
23. Iyer L, King CD, Whitington PF, et al. Genetic predisposition to the metabolism of irinotecan (CPT-11).
Role of uridine diphosphate glucuronosyltransferase isoform 1A1 in the glucuronidation of its active
metabolite (SN-38) in human liver microsomes. J Clin Invest. 1998;101:847-854.
24. Bosma PJ, Chowdhury JR, Bakker C, et al. The genetic basis of the reduced expression of bilirubin
UDP-glucuronosyltransferase 1 in Gilbert’s syndrome. N Engl J Med. 1995;333:1171-1175.
25. Strassburg CP. Pharmacogenetics of Gilbert’s syndrome. Pharmacogenomics. 2008;9:703-715.
26. Li M, Wang Z, Gao J, et al. Clinical significance of UGT1A1 gene polymorphisms on irinotecan-based
regimens as the treatment in metastatic colorectal cancer. Onco Targets Ther. 2014;7:1653-1661.
27. Swaby RF, Sharma CGN, Jordan VC. SERMs for the treatment and prevention of breast cancer. Rev
Endocr Metab Disord. 2007;8:229-239.
28. Goetz MP, Kamal A, Ames MM. Tamoxifen pharmacogenomics: the role of CYP2D6 as a predictor of
drug response. Clin Pharmacol Ther. 2008;83:160-166.
29. Dezentje VO, Guchelaar H-J, Nortier JWR, et al. Clinical implications of CYP2D6 genotyping in tamoxi-
fen treatment for breast cancer. Clin Cancer Res. 2009;15:15-21.
30. Engels FK, Ten Tije AJ, Baker SD, et al. Effect of cytochrome P450 3A4 inhibition on the pharmaco-
kinetics of docetaxel. Clin Pharmacol Ther. 2004;75:448-454.
31. Mielke S. Individualized pharmacotherapy with paclitaxel. Curr Opin Oncol. 2007;19:586-589.
32. Foulkes WD. Inherited susceptibility to common cancers. New Engl J Med. 2008;359:2143-2153.
33. Ulrich CM, Bigler J, Potter JD. Non-steroidal anti-inflammatory drugs for cancer prevention: promise,
perils, and pharmacogenetics. Nat Rev Cancer. 2006;6:130-140.
34. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin. 2015;65:5-29.
35. Chan AT, Giovannucci EL, Meyerhardt JA, et al. Aspirin dose and duration of use and risk of colorectal
cancer in men. Gastroenterology. 2008;134:21-28.
36. Cross IT, Poole EM, Ulrich CM. A review of gene-drug interactions for nonsteroidal anti-inflammatory
drug use in preventing colorectal neoplasia. Pharmacogenomics J. 2008;8:237-247.
37. Sinicrope FA, Gill S. Role of cyclooxygenase-2 in colorectal cancer. Cancer Metastasis Rev.
2004;23:63-75.
38. Metcalfe K, Lynch HT, Ghadirian P, et al. Contralateral breast cancer in BRCA1 and BRCA2 mutation
carriers. J Clin Oncol. 2004;22:2328-2335.
39. Tsao AS, Kim ES, Hong WK. Chemoprevention of cancer. CA Cancer J Clin. 2004;54:150-180.
40. Pegram MD, Konecny G, Slamon DJ. The molecular and cellular biology of HER2/neu gene ampli-
fication/overexpression and the clinical development of herceptin (trastuzumab) therapy for breast
cancer. Cancer Treat Res. 2000;103:57-75.
41. Moya-Horno I, Cortes J. The expanding role of pertuzumab in the treatment of HER2-positive breast
cancer. Breast Cancer: Targets and Therapy. 2015;7:125-132.
42. Druker BJ. Perspectives on the development of a molecularly targeted agent. Cancer Cell. 2002;1:31-36.
43. McCormack PL, Keam SJ. Dasatinib: A review of its use in the treatment of chronic myeloid leukemia
and Philadelphia chromosome-positive acute lymphoblastic leukemia. Drugs. 2011;71:1771-1795.
44. Roviello G. The distinctive nature of adenocarcinoma of the lung. OncoTargets Ther. 2015;8:2399-
2406.
45. Ravnan MC, Matalka MS. Vemurafenib in patients with BRAF V600E mutation-positive advanced
melanoma. Clin Ther. 2012;34:1474-1486.
46. Hauschild A, Grob J-J, Demidov LV, et al. Dabrafenib in BRAF-mutated metastatic melanoma: a multi-
center, open-label, phase 3 randomized controlled trial. Lancet. 2012;380:358-365.
47. Nambiar M, Kari V, Raghavan SC. Chromosomal translocations in cancer. Biochim Biophys Acta.
2008;1786:139-152.
48. Ansari M, Krajinovic M. Pharmacogenomics in cancer treatment defining genetic bases for inter-individual
differences in responses to chemotherapy. Curr Opin Pediatr. 2007;19:15-22.
49. Cunningham L, Aplenc R. Pharmacogenetics of acute lymphoblastic leukemia treatment response.
Exper Opin Pharmacother. 2007;8:2519-2531.
50. Rubnitz JE, Gibson B, Smith FO. Acute myeloid leukemia. Pediatr Clin North Am. 2008;55:21-51.
51. Grunwald MR, Levis MJ. FLT3 inhibitors for acute myeloid leukemia: a review of their efficacy and
mechanisms of resistance. Int J Hematol. 2013;97:683-694.

52. Wang ZY, Chen Z. Acute promyelocytic leukemia: from highly fatal to highly curable. Blood.
2008;111:2505-2515.
53. Quere R, Baudet A, Cassinat B, et al. Pharmacogenomic analysis of acute promyelocytic leukemia
cells highlights CYP26 cytochrome metabolism in differential all-trans retinoic acid sensitivity. Blood.
2007;109:4450-4460.
54. Lengfelder E, Saussele S, Weisser A, et al. Treatment concepts of acute promyelocytic leukemia. Crit
Rev Oncol Hematol. 2005;56:261-274.
55. Döhner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and survival in chronic lymphocytic
leukemia. N Engl J Med. 2000:343;1910-1916.
56. Druker BJ, Lydon NB. Lessons learned from the development of an ABL tyrosine kinase inhibitor for
chronic myelogenous leukemia. J Clin Invest. 2000;105:3-7.
57. Eck MJ, Manley PW. The interplay of structural information and functional studies in kinase drug
design: insights from BCR-ABL. Curr Opin Cell Biol. 2009;21:288-295.
58. Kemp CJ. Multistep skin cancer in mice as a model to study the evolution of cancer cells. Semin
Cancer Biol. 2005;15:460-473.
59. Olopade OI, Grushko TA, Nanda R, et al. Advances in breast cancer: pathways to personalized medi-
cine. Clin Cancer Res. 2008;14:7988-7999.
60. Narod SA, Brunet JS, Ghadirian P, et al. Tamoxifen and risk of contralateral breast cancer in BRCA1 and
BRCA2 mutation carriers: a case-control study. Hereditary Breast Cancer Clinical Study Group. Lancet.
2000;356:1876-1881.
61. Rubenstein W. Hereditary breast cancer in Jews. Familial Cancer. 2004;3:249-257.
62. Kinzler KW, Vogelstein B. Lessons from hereditary colorectal cancer. Cell. 1996;87:159-170.
63. Balko JM, Black EP. A gene expression predictor of response to EGFR-targeted therapy stratifies
progression-free survival to cetuximab in KRAS wild-type metastatic colorectal cancer. BMC Cancer.
2009;9:145.
64. Lievre A, Bachet J-B, Boige V, et al. KRAS mutations as an independent prognostic factor in patients
with advanced colorectal cancer treated with cetuximab. J Clin Oncol. 2008;26:374-379.
65. Rubin MA, Maher CA, Chinnaiyan AM. Common gene rearrangements in prostate cancer. J Clin Oncol.
2011;29:3659-3668.
66. Kim SY, Kim SN, Hahn HJ, et al. Meta-analysis of BRAF mutations and clinicopathologic characteristics
in primary melanoma. J Am Acad Dermatol. 2015;72:1036-1046.
67. Atkinson V. Medical management of malignant melanoma. Aust Prescr. 2015;38:74-78.
68. Flaherty KT, Puzanov I, Kim KB, et al. Inhibition of mutated, activated BRAF in metastatic melanoma.
New Engl J Med. 2010;363:809-819.
69. Hauschild A, Grob JJ, Demidov LV, et al. Dabrafenib in BRAF-mutated metastatic melanoma: a multi-
centre, open-label, phase 3 randomised controlled trial. Lancet. 2012;380:358-365.
70. Lugowska I, Kosela-Paterczyk H, Kozak K, et al. Trametinib: a MEK inhibitor for management of meta-
static melanoma. OncoTarg Ther. 2015;8:2251-2259.
71. Platz A, Egyhazi S, Ringborg U, et al. Human cutaneous melanoma; a review of NRAS and BRAF muta-
tion frequencies in relation to histogenetic subclass and body site. Mol Oncol. 2008;1:395-405.
72. Paik S, Tang G, Shak S, et al. Gene expression and benefit of chemotherapy in women with node-
negative, estrogen receptor-positive breast cancer. J Clin Oncol. 2006;23:3726-3734.
73. Srivastava G, Renfro LA, Behrens RJ, et al. Prospective multicenter study of the impact of Oncotype
DX colon cancer assay results on treatment recommendations in stage II colon cancer patients.
Oncologist. 2014;19:492-497.
74. Knezevic D, Goddard AD, Natraj N, et al. Analytical validation of the Oncotype DX prostate cancer
assay—a clinical RT-PCR assay optimized for prostate needle biopsies. BMC Genomics. 2013;14:690.
75. Bacher U, Kohlmann A, Haferlach T. Gene expression profiling for diagnosis and therapy in acute
leukaemia and other haematological malignancies. Cancer Treat Rev. 2010;36:637-646.
76. Bhojwani D, Moskowitz N, Raetz EA, et al. Potential of gene expression profiling in the management of
childhood acute lymphoblastic leukemia. Pediatric Drugs. 2007;9:149-156.
77. Bhojwani D, Kang H, Menezes RX, et al. Gene expression signatures predictive of early response and
outcome in high risk childhood acute lymphoblastic leukemia: A Children’s Oncology Group Study. J
Clin Oncol. 2008;26:4376-4384.
78. Mullighan CG, Dowing JR. Genome-wide profiling of genetic alterations in acute lymphoblastic leuke-
mia: recent insights and future directions. Leukemia. 2009;23:1209-1218.

79. Mishra S, Yadav T, Rani V. Exploring miRNA based approaches in cancer diagnostics and therapeutics.
Crit Rev Oncol Hematol. 2015 Oct 8. pii: S1040-8428(15)30052-4. doi: 10.1016/j.critrevonc.2015.10.003.
[Epub ahead of print].
80. Bertoli G, Cava C, Castiglioni I. MicroRNAs: new biomarkers for diagnosis, prognosis, therapy prediction,
and therapeutic tools for breast cancer. Theranostics. 2015;5:1122-1143.
81. Rodriguez-Gonzalez FG, Sieuwerts AM, Smid M, et al. MicroRNA-30c expression level is an indepen-
dent predictor of clinical benefit of endocrine therapy in advanced estrogen receptor positive breast
cancer. Breast Cancer Res Treat. 2011;127:43-51.
82. Zhao JJ, Lin J, Yang H, et al. MicroRNA-221/222 negatively regulates estrogen receptor alpha and is
associated with tamoxifen resistance in breast cancer. J Biol Chem. 2008;283:31079-31086.
83. Lyng MB, Laenkholm AV, Sokilde R, et al. Global microRNA expression profiling of high-risk ER+
breast cancers from patients receiving adjuvant tamoxifen mono-therapy: a DBCG study. PLoS One.
2012;7(5):e36170. doi: 10.1371/journal.pone.0036170. Epub 2012 May 18.
CHAPTER
9
Central Nervous System

After completing this chapter, the reader ATYPICAL ANTIPSYCHOTIC—second-
should be able to generation antipsychotics; primary
pharmacological property is coupling of
Describe the utility of pharmacogenomics
dopamine 2 antagonist with serotonin 2A
in the treatment of various psychiatric
antagonist.
disorders including major depressive
disorder, schizophrenia, bipolar disorder, OPIOIDS—compounds, both natural and
and attention-deficit hyperactivity synthetic, with morphine-like actions that
disorder. are mu opioid receptor agonists used for
the treatment of mild-to-severe, acute,
Discuss the application of
and chronic pain.
pharmacogenomics in optimizing the
treatment of various neurological REMISSION—treatment of depression that
disorders including Alzheimer’s disease, results in removal of essentially all
pain, and epilepsy. symptoms.
Discuss potential barriers to the use of RESPONSE—treatment of depression
pharmacogenomics in the treatment of that results in ≥50% improvement in
psychiatric and neurological disorders. symptoms.
TYPICAL ANTIPSYCHOTIC—first-generation
antipsychotics; primary pharmacological
property of dopamine 2 antagonist that
is responsible for their antipsychotic
efficacy but also for many of their side
effects.
283
INTRODUCTION
The human brain is one of the body’s most complex organs. This complexity makes the
treatment of disorders of the central nervous system (CNS) very challenging. A patient’s
response to the many medications used to treat various brain disorders can be highly
variable. Given the recent abundance of literature investigating the genetics behind these
results, it can be challenging to apply this material systematically to each patient. The
literature is lacking large, controlled studies specifically designed to assess drug selec-
tion and dosing versus current standard of care. These types of studies can be difficult
to complete due to the multidimensional biological causes of neuropsychiatric illnesses
and the limited knowledge regarding the mechanism of action of the medications used
to treat them.
Currently, drug labeling and consensus guidelines are in place to assist with the
potential clinical utility of pharmacogenomic testing. The U.S. Food and Drug Adminis-
tration has 40 neuropsychiatric medications with pharmacogenomic information in their
labeling materials (see Table 9-1).1-43 With the added knowledge from these test results,
clinicians are able to optimize the risk-benefit ratios of current available treatments. The
Clinical Pharmacogenomics Implementation Consortium publishes the most up-to-date
drug guidelines and addresses some barriers to the implementation of pharmacogenomics
tests into clinical practice. Guidelines are available for the following neuropsychiatric medi-
cations: tricyclic antidepressants (TCAs), selective serotonin reuptake inhibitors (SSRIs),
carbamazepine, and phenytoin. The website offers a unique interface to enter the patient’s
genotype for specific dosing recommendations of his or her prescribed medications.
CLINICAL PEARL
The translation of pharmacogenomic testing into clinical practice has
several limitations in CNS disorders, including
●● A lack of clear relationships between the serum concentrations of many
neuropsychiatric drugs and response to treatment.
●● The wide therapeutic ranges for many neuropsychiatric medications.
●● The lack of clinical guidelines and limited number of well-designed

trials investigating the use of genomic testing in methods that could be
translated into clinical practice.
●● The use of multiple medications to treat neuropsychiatric illnesses,
which can complicate the understanding of how and when to utilize
available testing options.
●● The diverse ethnicities of the patients with CNS disorders.
Chapter 9 • Central Nervous System 285
Table 9-1
Medications with Pharmacogenomic Biomarkers in Drug Labeling
Medication Biomarker Implications for Practice
Amitriptyline CYP2D6 PMs: May have higher than expected plasma

concentrations of TCAs when given
usual doses, and the increase in plasma
concentration range can be large.
Aripiprazole CYP2D6 PMs: Initial dose should be reduced by ½ of
the usual dose.
PMs: 80% increase in aripiprazole exposure
and about a 30% decrease in exposure to
the active metabolite compared to EMs.
Aripiprazole Abilify Aristada™ CYP2D6 If CYP modulators are added for >2 weeks:
Strong CYP3A4 inhibitor and poor
metabolizer of CYP2D6, dose should be
reduced to 441 mg from 662 mg or 882
mg. No dosage adjustment is necessary in
patients taking 441 mg if the 441 mg dose is
tolerated.
Strong CYP2D6 inhibitor and poor metabolizer
of 2D6, reduce the dose of Aristada to the
next lower strength. No additional dose
reduction is required if patient is on the 441
mg dose and it is tolerated.
Aripiprazole Abilify Maintena™ CYP2D6 PMs and those taking a CYP3A4 inhibitor for
greater than 14 days: Adjusted dose of 200
mg.
PMs: Adjusted dose of 300 mg.
Atomoxetine CYP2D6 PMs: 10-fold higher AUC and 5-fold higher
Cmax to a given dose of atomoxetine
compared with EMs; higher rates of some
adverse effects.
Children/adolescents up to 70 kg body weight:
Initiation dose should be 0.5 mg/kg/day and
only increased to the usual start dose of
1.2 mg/kg/day if symptoms fail to be
improved after 4 weeks, and the initial dose
is well tolerated.
Children/adolescents >70 kg body weight:
Initial dose should be 40 mg/day and only
increased to the usual target dose of
80 mg/day if symptoms fail to be improved
after 4 weeks, and the initial dose is well
tolerated.
Brexpiprazole CYP2D6 In a CYP2D6 poor metabolizer, administer
1/2 of the usual dose. In a known CYP2D6
poor metabolizer taking a strong/moderate
CYP3A4 inhibitor as well, administer 1/4 of
the usual dose.
Brivaracetam CYP2C19 CYP2C19 poor metabolizers may require a

dose reduction.

Carbamazepine HLA-B*15:02 Chinese ancestry: Strong association between

risk of developing SJS/TEN and the
presence of HLA-B*15:02.
Patients with ancestry in genetically at-risk
populations should be screened for the
presence of HLA-B*15:02 prior to initiating
treatment with carbamazepine.
HLA-A*31:01 In those of European, Korean, and Japanese
ancestry, there is a moderate association
in the risk of developing hypersensitivity
reactions and the presence of HLA-A*31:01
in patients on carbamazepine. Risks
and benefits should be weighed before
considering starting carbmazepine in
those who are known to be positive for
HLA-A*31:01.
Citalopram CYP2C19 Maximum dose should be limited to
20 mg/day in patients who are CYP2C19
PMs due to risk of QT prolongation.
CYP2D6 Citalopram steady state levels were not
significantly different in poor metabolizers
and extensive metabolizers of CYP2D6.
Clobazam CYP2C19 PMs: Levels of n-desmethylclobazam, the
active metabolite of clobazam, increases;
starting dose should be 5 mg/day, with a
slow titration period.
Clomipramine CYP2D6 PMs: Higher than expected plasma
concentrations of TCAs when given usual
doses. Depending on the fraction of drug
metabolized by CYP2D6, the increase in
plasma concentration may be small or quite
large (8-fold increase in plasma AUC of the
TCAs).
Clozapine CYP2D6 PMs: May develop higher than expected
plasma concentrations of clozapine when
given usual doses.
Desipramine CYP2D6 PMs: Higher than expected plasma
doses; the increase in plasma concentration
may be small or quite large (8-fold increase
in plasma AUC of the TCAs).
Dextromethorphan and CYP2D6 PMs: Increase in adverse effects of quinidine.
quinidine
Diazepam (rectal gel CYP2C19 Marked interindividual variability in the
formulation only, not the oral clearance of diazepam reported, likely
tablet) attributable to variability of CYP2C19.

Divalproex sodium POLG Contraindicated in individuals with
known mitochondrial disorders caused
by mutations in mitochondrial DNA
polymerase gamma and POLG-related
disorders in children under 2 years of age.

Doxepin CYP2D6 PMs: CYP2D6 may have higher plasma levels

than normal subjects.
CYP2C19 CYP2C19 poor metabolizers may have higher
plasma levels than normal subjects
Duloxetine CYP2D6 Administration of a potent CYP1A2 inhibitor
to a known poor metabolizer of CYP2D6,
resulted in a 6-fold increase in AUC and
Cmax of duloxetine.
Escitalopram CYP2D6 Drugs metabolized by CYP2D6 are unlikely

to have clinically significant effects on
escitalopram metabolism.
CYP2C19 Poor metabolizers of CYP2C19 after a
20 mg dose of escitalopram had the same
QtcF interval as healthy subjects
at supratherapeutic 30 mg doses.
Eteplirsen DMD Indicated for the treatment of Duchenne
Muscular Dystrophy (DMD) in patients who
have a confirmed mutation of the DMD gene
that’s amendable to exon 51 skipping.
Fluoxetine CYP2D6 PMs: Metabolized S-fluoxetine at a slower
rate and achieved higher concentrations
of S-fluoxetine. Compared to normal
metabolizers, the total sum at steady-state
of the plasma concentrations of the four
active enantiomers was not significantly
greater among poor metabolizers. The
net pharmacodynamics activities were
essentially the same for PMs and normal
metabolizers.
Fluoxetine and olanzapine CYP2D6 Same as fluoxetine.
combination
Fluvoxamine CYP2D6 In vivo study of fluvoxamine single-dose
pharmacokinetics in 13 PMs demonstrated
altered pharmacokinetic properties
compared to 16 EMs. Mean Cmax, AUC, and
t½ were increased by 52%, 200%, and
62%, respectively.
Galantamine CYP2D6 PMs: No change in dosage required
Iloperidone CYP2D6 PMs: Dose should be reduced by ½. Observed
mean elimination half-lives for iloperidone,
P88 and P95 in EMs are 18, 26, and 23 h,
respectively; and in PMs are 33, 37, and
31 h, respectively.
Imipramine CYP2D6 PMs: Higher than expected plasma

plasma concentration may be small, or quite
TCAs).

Lacosamide CYP2C19 No clinically relevant differences in the

pharmacokinetics of lacosamide between
CYP2C19 poor and extensive metabolizers.
Modafinal CYP2D6 PMs: Levels of CYP2D6 substrates may be
increased by coadministration of modafinil.
Nefazodone CYP2D6 PMs: Treatment adjustment not required
based on CYP2D6 status.
Nortriptyline CYP2D6 PMs: Higher than expected plasma
TCAs).
Oxcarbazepine HLA-B*15:02 Chinese ancestry: Strong association between
risk of developing SJS/TEN and the
presence of HLA-B*15:02.
Patients with ancestry in genetically at-risk
populations should be screened for the
presence of HLA-B*15:02 prior to initiating
treatment with oxcarbazepine.
Paroxetine CYP2D6 Only information about inhibition of CYP2D6.
Perphenazine CYP2D6 PMs: Will metabolize perphenazine more
slowly and will experience higher
concentrations compared with EMs.
Phenytoin HLA-B*1502 Chinese ancestry: Strong association between
the risk of developing SJS/TEN and the
presence of HLA-B*1502 in patients with
carbamazepine. Limited evidence suggests
the polymorphism is also a risk for the
development of SJS/TEN in patients of
Asian ancestry taking other antiepileptic
medications associated with SJS/TEN,
including phenytoin.
CYP2C9 and Patients with unusually low levels of
CYP2C19 phenytoin may be hypermetabolizers and
those with unusually high levels may have
variant CYP2C9 and CYP2C19 alleles.
Pimozide CYP2D6 PMs: Exhibit higher pimozide concentrations
than EMs. Time to achieve steady state
concentrations is expected to be longer in
PMs because of the prolonged t½. Alternate
dosing strategies are recommended in PMs.
Children: At doses >0.05 mg/kg/day, CYP2D6
genotyping should be performed. In PMs,
doses should not exceed 0.05 mg/kg/day,
and doses should not be increased earlier

than 14 days.
Adult dosing: In doses above 4 mg/day,
CYP2D6 genotyping should be performed.
In PMs, doses should not exceed 4 mg/day,
and doses should not be increased earlier
than 14 days.

Protriptyline CYP2D6 PMs: Higher than expected plasma

doses; the increase in plasma concentration
may be small or quite large (8-fold increase
in plasma AUC of the TCAs).
Risperidone CYP2D6 EMs: Convert risperidone rapidly to
9-hydroxyrisperidone; whereas PMs
convert it much more slowly.
Tetrabenazine CYP2D6 Doses above 50 mg should not be given
without CYP2D6 genotyping.
PMs: Will have substantially higher levels of
the primary drug metabolites than EMs.
PMs: Maximum recommended total daily dose
is 50 mg, and the maximum recommended
single dose is 25 mg.
EMs: Maximum recommended total daily dose
is 100 mg, and the maximum recommended
single dose is 37.5 mg.
Thioridazine CYP2D6 Reduced CYP2D6 activity: Would be expected
to augment the prolongation of the QTc
interval associated with thioridazine and
may increase the risk of serious, potentially
fatal cardiac arrhythmias. Thioridazine is
contraindicated in individuals with a known
genetic defect leading to reduced levels of
activity of CYP2D6.
Trimipramine CYP2D6 PMs: Higher than expected plasma
TCAs).
Valproic acid UCD Contraindicated in patients with known urea
cycle disorders.
Prior to the initiation of valproate therapy,
evaluation for UCD should be considered
in the following patients: A history of
unexplained encephalopathy or coma,
encephalopathy associated with a protein
load, pregnancy-related postpartum
encephalopathy, unexplained mental
retardation, or history of elevated plasma
ammonia or glutamine, cyclical vomiting
and lethargy, episodic extreme irritability,
ataxia, low BUN or protein avoidance, a
family history of UCD or a family history
of unexplained infant deaths (particularly

males), or other signs or symptoms of UCD.
Venlafaxine CYP2D6 Total concentrations of active metabolites
were similar between PMs and EMs.
No dosage adjustment is required when
venlafaxine is coadministered with a
CYP2D6 inhibitor.

Vortioxetine CYP2D6 PMs: Maximum recommended dose is

10 mg/day.
AUC, area under the curve; BUN, blood urea nitrogen; Cmax, maximum drug plasma concentration; CYP, cytochrome 450;
EM, extensive metabolizer; PM, poor metabolizer; SJS, Steven-Johnson syndrome; TCA, tricyclic antidepressant;
TEN, toxic epidermal necrolysis; t½, half-life; UCD, urea cycle disorder.
Source: See references 1-43 for more information.
CASE STUDY—MAJOR DEPRESSION

A 35-year old patient is experiencing his third episode of major depression, with
symptoms including lack of energy and loss of hope for the future. He wonders if he has
any worth to society and wants to hurt himself. Additional complaints of loss of appetite,
decreased concentration, an inability to sleep, and a lack of motivation are voiced. One
previous hospitalization for depression and suicidal ideations is noted.
Questions
1. What genes or drug targets are currently being researched for drug response in major depres-
sion?
2. What discussions are being debated in the current literature about the association of treatment
response with selective serotonin reuptake inhibitors (SSRIs) and known genetic differences?
3. How can pharmacogenomics affect a patient’s response to antipsychotics?
ANTIDEPRESSANTS
According to the National Comorbidity Survey Replication, 16.2% of the population studied
had a history of major depressive disorder and only 65–70% of patients with this diagnosis
improved with drug therapy.44 In double-blind efficacy trials, initial treatment with anti-
depressants appears to lead to remission in only 35–47% of patients.45,46 A meta-analysis
of 315 studies compared the efficacy of the TCAs to the SSRIs and found similar response
rates in the treatment of depression and dysthymia.47 With the discovery of the newer
antidepressants, higher response and remission rates may be expected.
One of the most studied genes affecting the treatment of depression is the serotonin
transporter gene (SLC6A4), which is located on chromosome 17q.48,49 SLC6A4 is a protein
structure made up of 12 transmembrane helices with an extracellular loop between helices
three and four. This transporter is responsible for the reuptake of 5HT into the presynaptic
neurons.
A well-characterized variant of this gene is a functional polymorphism located in the

5′ promoter region, 5HT transporter gene-linked polymorphic region (5-HTTLPR).50,51 The
polymorphism consists of a repetitive region containing 16 imperfect repeat units of 22
base pairs (bp) located ~1,000bp upstream of the transcriptional start site. It is polymorphic
because of an insertion/deletion of units 6–8, which produces a short (s) allele that is 44bp
shorter than the long (l) allele.50,51 Although originally thought to be bi-allelic, very long
and extra-long alleles have been identified in a small percentage of Japanese and African
Americans (<5%).52
Extensive research, including several meta-analyses, has focused on the pharmaco-
genomic variability of SLC6A4 on antidepressant response to SSRIs. In all of the studies
of unipolar or bipolar depressed Caucasian patients, at least nominally significant associa-
tions could be found between the long variant and antidepressant response.53-63 This was
different with an Asian population, in which several studies found an association with the
s allele and better outcomes with SSRI antidepressants.64-70 The s allele is present in 50%
of Caucasians and 75% of Asians.52 This ultimately leads to very few Asians who would
present with a homozygous l allele, a finding that could explain the difference in associa-
tions with response and 5-HTTLPR.
One meta-analysis of 15 studies concluded that patients who are homozygous for the l
allele and are of European ancestry have a more consistent therapeutic response to SSRI
treatment, while an additional meta-analysis (n = 28 various ethnicities) concluded that
there is no significant effect on the transporter promoter length polymorphism on the rates
of antidepressant response. The authors stated that there was substantial unexplained
heterogeneity of effect sizes across the studies, suggesting additional interacting factors
could contribute to an association between 5-HTTLPR and response to SSRI treatment in
some cases.71,72
CLINICAL PEARL
The serotonin transporter gene is an important and highly studied gene
in terms of antidepressant response. Statistically significant associations
have been found between variations in this gene and antidepressant
response. Currently, point-of-care testing is not being completed for this
polymorphism.
Additional polymorphisms within the SLC6A4 gene could also account for some of
the observed variation in drug response. An association of a single nucleotide polymor-
phism (SNP), rs25531, which is located upstream of the 5-HTTLPR, has been reported with
response to fluoxetine treatment.62 It has been proposed that rs25531 and 5-HTTLPR are
in linkage disequilibrium with each other (r2 = 0.75).62 In the presence of the G allele of
rs25531, the l allele of 5-HTTLPR seems to be associated with reduced drug response.
Likewise in the presence of the A allele of rs25531, the s allele of the 5-HTTLPR seems to
be associated with reduced drug response.62 The influences of this and other SNPs in the
serotonin transporter protein gene need to be considered when utilizing this information
to determine drug response.
Polymorphisms in the genes that code for the various serotonin receptors have also
been studied with regard to their role in altering the efficacy of various antidepressants.
The 5-HT 1A receptor has one SNP located in the promoter region, C-1019G, which has
shown a trend association between the G allele and various antidepressant response
(p = 0.049).73 An additional study demonstrated a combined effect when measuring the
C-1019G SNP with a variation on the SLC6A4 (5-HTTLPR) gene in 130 patients treated
with an SSRI over 12 weeks. It was determined that patients with the “risk genotype” (s/s-
G/G) were more likely to not achieve remission (p = 0.009).74 The SNP within the 5-HT 2A
receptor gene, rs7997910, demonstrated an 18% absolute risk of having no response to
treatment compared to those homozygous for the other allele in a study of 1,953 patients
treated with citalopram. The A allele of rs7997910 was six times more frequent in white
than black participants, with treatment being less effective in the black participants.75
Although additional studies regarding other serotonin receptors have been completed,
they failed to demonstrate an association between genetic variation and the effectiveness
of antidepressants.76
Studies of additional areas of interest—to a lesser extent compared to the serotonin
transporter and receptors—determined an association between efficacy and the anti-
depressants, including the g-protein coupled receptors, tryptophan hydroxylase I,
monoamine oxidase, dopamine receptor genes, noradrenergic receptor gene, nitric
oxide, angiotensin-converting enzyme, interleukin-1 beta, stress hormone system, and
phosphodiesterase. Routine use of these polymorphic genes in practice will require the
completion of larger studies in the future.
To better understand the utility of this information evolving with antidepressant
responsiveness and pharmacokinetic pharmacogenomics, the Evaluation of Genomic
Applications in Practice and Prevention (EGAPP) Working Group reviewed it. This group
investigated the evidence for genetic testing of CYP450 polymorphisms in the manage-
ment of patients with nonpsychotic depression treated with SSRIs.77 The group determined
evidence was lacking in the use of CYP450 genotyping as a guide to SSRI management
for patients with nonpsychotic depression. No prospective studies of CYP450 genotyping
have been conducted that would help guide treatment and measurement of subsequent
clinical outcomes. The investigators concluded evidence was lacking for incorporation of
any of the CYP450 genotyping tests into guidelines for clinical practice. Realizing that
CYP450 genotyping may be only one of several genetic variants affecting response to
antidepressant treatment, the EGAPP recommended that more research be conducted in
this area that combines pharmacokinetics and pharmacodynamics. The investigators called
for extreme caution in interpreting any CYP genotyping results until prospective evidence-
based research provides data to prove an unequivocal association between variations in
CYP450 genotypes and SSRI outcomes in the treatment of major depressive disorder.78
Additional studies have also shown similar results to the finding of the EGAPP group.79
A more recent review of the literature provided extensive dosing guidelines for SSRIs
with CYP2D6 and CYP2C19 genotypes. The guidelines suggest potential benefits of iden-
tifying patients who are at an increased risk of experiencing adverse drug reactions or
therapeutic failure as candidates for CYP450 genotyping.80
CLINICAL PEARL
Obtaining CYP450 enzyme status for a patient may provide a benefit to
identifying those patients who are at an increased risk of experiencing
adverse drug reactions or therapeutic failure.
ANTIPSYCHOTICS
The Epidemiologic Catchment Area Study estimated that 0.6–1.9% of the population suffers
from schizophrenia.81 Efficacy studies have demonstrated a response rate to antipsychotics
of 60–70% for patients with schizophrenia, although the Clinical Antipsychotic Trials of
Intervention Effectiveness demonstrated a much poorer clinical outcome with the atypi-
cal antipsychotics used in general clinical practice.82,83 Given these statistics, researchers
have focused on investigating the various genetic factors that can alter response to these
medications. Pharmacogenomic studies have focused on atypical antipsychotic agents with
emphasis on identifying candidate genes coding for receptors or metabolizing enzymes
identified in preclinical and clinical studies. Recent pharmacogenomic studies have also
investigated neurotransmitter disposition, second messenger systems, and peripheral
neurotransmitters.84
The primary treatment for schizophrenia is antipsychotics, even though variation in
treatment response is considerable with these medications.85 The typical antipsychot-
ics work by blockade of the dopamine 2 (D2) receptors, while the atypical antipsychotics
are also antagonists at 5-HT2A and 2C receptors in addition to D2 receptors.86,87 Many
researchers have studied the D2, 5-HT2A, and 2C receptors in hopes of finding an associa-
tion with antipsychotic efficacy. These studies are difficult to complete, however, due to
differing patient samples and illness characteristics, antipsychotic medication types, treat-
ment durations, adherence, adjuvant treatments, and outcome phenotypes.
CLINICAL PEARL
The D2, 5-HT2A, and 2C receptors are currently being investigated for
an association between genetic variability and treatment response in
schizophrenia.
Adverse Effects: Weight Gain, Tardive Dyskinesia, and Agranulocytosis

Antipsychotics can induce substantial weight gain that can lead to diabetes, lipid abnor-
malities, metabolic syndrome, and psychological distress.88 Treatment emergent weight
gain varies within the class of antipsychotics, with several agents causing severe weight
gain and others limited weight gain. The propensity to develop weight gain depends on not
only the medication that individuals are taking, but also on their genetic factors.88
Weight Gain and HTR2C Gene
The C allele of rs3813929 polymorphism (−759C/T) has been consistently associated with
weight gain in patients taking antipsychotic medications. A meta-analysis of the associa-
tion between this polymorphism and weight gain concluded that the −795T allele is associ-
ated with less weight gain.89 In clinical practice, if testing is available, antipsychotic drugs
with demonstrated propensity to cause an increase in weight should be avoided in carriers
of the protective C allele.
Tardive Dyskinesia
Tardive dyskinesia (TD) is a debilitating motor disorder characterized by hyperkinetic invol-
untary, repetitive, and purposeless movements predominantly of the orofacial region. The
movements can include chewing, tongue protrusion, lip smacking, puckering and pursing of
the lips, and rapid eye blinking. Choreoathetotic movements of the limb and trunk region
have also been observed.90 TD develops in 20–30% of chronic schizophrenic patients
on long-term treatment with typical antipsychotic drugs.91 Genetic factors (e.g., drug-
metabolizing enzymes, neurotransmitters, oxidative stress pathway genes) and nongenetic
factors (e.g., advanced age, presence of early extrapyramidal symptoms, female gender,
African-American ethnicity, organic brain dysfunction, smoking, duration of exposure to

typical antipsychotic drugs) are both thought to play a role in TD pathogenesis.92
Additionally, TD has been associated with genes in the D2 and D3 receptors and the
catechol-o-methyltransferase (COMT) enzyme. Several studies have demonstrated that
the minor T allele of the taq 1A polymorphism, rs1800497, appears to be protective against
TD while the rs6280 SNP of the D3 receptor has shown inconsistent results.93-95 A func-
tional polymorphism that codes for a substitution of methionine (met) for valine (val) at
codon 158 has also been reported to have an association with TD.96 Other pathways that
have been investigated with varying results include the accumulation of free radicals or
oxidative stress, 5-HT2A receptors, and CYP2D6 enzyme.97-99
Clozapine-Induced Agranulocytosis
Clozapine-induced agranulocytosis (CA) occurs in approximately 1% of patients taking this
medication; however, its mechanism is poorly understood. The relatively rarity of CA signifi-
cantly limits the collection of the large data samples required to conduct robust genomic
tests, determining the predictability of the potential for this adverse effect. Several mecha-
nisms have been studied, and one resulted in a commercially available product.
From 2002 to 2007, PGxHealth (pgxhealth.com) completed a genome-wide associa-
tion study (GWAS) that demonstrated HLA-DQB1 667G>C was a risk for CA, with a 16.9-
fold increased risk for carriers of this marker. In 2007, a commercial test was marketed to
determine whether a patient was at a lower risk or higher risk for CA. One shortcoming of
the test was a 21.5% sensitivity, which has been a limiting factor in its application in clinical
settings.100,101
CLINICAL PEARL
Antipsychotic–associated weight gain is a serious consequence for
morbidity and mortality for patients taking antipsychotic medications.
Pharmacogenomic approaches have allowed for detection of more than
300 possible candidate genes for this adverse effect. Given the variable
histories of prior medication exposure and adherence of patients treated
with these medications, it has been difficult to identify genetic effects on
this complex phenotype. Clinicians should monitor a patient’s body mass
index, total fat mass, blood glucose, and insulin levels for development of
obesity.102
ANTI-EPILEPTICS/MOOD STABILIZERS
Bipolar disorder (BD) is a chronic and often severe psychiatric disorder characterized by
alternating manic and depressive episodes. It affects approximately 1% of the general
population.103 The gold standard of treatment for BD is lithium. The response to lithium
is variable, with about 30% of subjects being excellent responders and 25% being non-
responders.104 Recently, other anticonvulsants such as VPA and carbamazepine have been
used as primary prophylactic agents. Response rates with these agents are similar to that
of lithium.105
Lithium-responsive BD is linked to several unique characteristics including euphoric

manias, positive family history, few comorbidities, and symptom-free intervals between
episodes.106 Pharmacogenomic association studies have focused on selected genes. Stud-
ies that investigated the mechanism of action of lithium and predication of efficacy via
SNP pharmacogenomic testing have had only modest success.106 Future GWAS may prove
to be a better design to determine the combination of genetic variations, which may detect
lithium response.106
STIMULANTS
Attention-deficit hyperactivity disorder (ADHD) is one of the most common neurobehav-
ioral disorders of childhood and can also persist through adolescence and into adulthood.
Chronic levels of inattention, impulsive hyperactivity, or both characterize the disorder
to the extent that daily functioning is compromised.107 A few predictors of medication
response exist in the treatment of ADHD. Without this type of information, treatment is
often determined empirically, similar to many other CNS disorders.108
Preliminary studies have suggested that candidate genes involved in the catechol-
amine pathway could influence individual responses to ADHD treatment. Results from
several studies are contradictory, and the nature, magnitude, and direction of the genetic
effects remains unclear. In addition, most studies focus on the use of methylphenidate but
many other medications are currently used in the treatment of ADHD. Although recently, an
increasing number of studies have assessed gene-gene and gene-environment association
approaches, neuroimaging, and pharmacokinetic properties.
In addition to the various stimulants used to treat ADHD, a nonstimulant medication
atomoxetine is also approved for the treatment of ADHD. Clinical development trials
demonstrated that CYP2D6 metabolizer status influenced dosing titration and ultimately
approved dosing limits of this drug. A recent meta-analysis describes the outcome data from
several atomoxetine trials.109 Subjects who were CYP2D6 poor metabolizers had greater
symptom improvement than extensive metabolizers, which was likely due to the higher
drug concentrations in the blood and CNS. However, higher rates of appetite suppression
and insomnia were also reported more often in the poor metabolizer group. This group also
experienced greater increases in medication-related pulse and blood pressure changes.110
NEUROLOGIC DISORDERS
Multiple Sclerosis
Multiple sclerosis (MS) is a demyelinating disease of the CNS with inflammatory and
degenerative components.111 Candidate genes for human leukocyte antigen (HLA) and inter-
feron receptor polymorphisms have been studied to determine their effect on a patient’s
response to therapy. The HLA studies demonstrated that allelic variation has a correlation
with autoantibody production but not with treatment response. Interferon receptor studies
have demonstrated conflicting results. Only one completed GWAS found promising results,
but it was limited because it examined only SNPs detected by the author’s microarray.112-115
Alzheimer’s Disease
Alzheimer’s disease (AD) is a nonreversible, progressive dementia manifested by gradual
deterioration in cognition and behavior. Approximately 10–15% of the direct costs for treat-
ing dementia can be attributed to pharmacological treatment, while only 10–20% of the
patients are moderate responders to conventional antidementia drugs, with questionable
cost-effectiveness.116-118 The exact etiology of AD is unknown. Genetic factors have been

linked to errors in protein synthesis, which result in the formation of abnormal proteins
involved in the pathogenesis of AD.119 Mutations in the presenilin-1, amyloid precursor
protein, and presenilin-2 lead to an increase in the accumulation of amyloid beta (Aβ) in
the brain. An increase in Aβ results in oxidative stress, neuronal destruction, and the clinical
syndrome of AD. Apolipoprotein (APO) E has also been identified as a strong risk factor for
late-onset AD. Individuals who are carriers of two or more APOE4 alleles have an earlier
onset of AD (approximately 6 years) compared with noncarriers.119
Numerous studies have been conducted to determine the influence of the APOE geno-
type on drug response in AD. In the monogenic-related studies, the APOE 4/4 carriers are
the poorest responders to medication. In trigenic-related studies, the APOE 4/4 (preseni-
lin-1 and -2) are the best responders in those patients carrying the 331222-, 341222-, and
441112-genomic profiles. The worst responders in all genomic clusters are those patients
with the 441122+ genotype. These results demonstrate the deleterious effect of the APOE
4/4 genotype on AD, in sporadic and familial late-onset AD, and therapeutics in combina-
tion with the other AD-related genes.120-122
Acetylcholinesterase Inhibitors
The current treatment mainstay for AD is the cholinesterase inhibitors, which are metabo-
lized via the CYP450 pathway. Poor metabolizers and ultra-rapid metabolizers are the
poorest responders to drug treatment, whereas the extensive and intermediate metabo-
lizers are the best responders.120-122 In light of the emerging data, the determination of
response to the current AD medication treatment could depend on the interaction of genes
involved in the medication’s metabolism and those genes associated with AD pathogen-
esis.120-122
Additionally, the gene encoding for choline transferase, which encodes the major
catalytic enzyme of the cholinergic pathway, is associated with increased response to
acetylcholinesterase inhibitors (AChEIs). A SNP in the promoter region of choline O-
acetyltransferase, rs733722, accounts for 6% of the variance in response to AChEIs.123
CLINICAL PEARL
Larger population studies are still needed to confirm the predictive effects
of polymorphisms in various genes on Alzheimer patient response to
AChEIs: AChE, butyrylcholinesterase, choline AChE, and paraoxonase.
Parkinson’s Disease
Parkinson’s disease is a progressive neurodegenerative disease characterized by continuous
dopaminergic cell loss in the nigrostriatal system at a rate of approximately 10% per year as
confirmed by positron-emission tomography studies.124 Recently, evidence highlighting the
importance of environmental factors and their interaction with genetics has prompted a
number of association studies on the role of gene polymorphisms in the risk of Parkinson’s
disease.125 Although many studies have been conducted to discover the risk of developing
Parkinson’s disease, little has been done in terms of associations between drug response
and a particular genotype (see Table 9-2).
Epilepsy
Epilepsy is characterized by the presence of many different types of seizures that vary
in severity, appearance, cause, consequence, and management. Each year, approximately
125,000 new epilepsy cases occur in the United States.127 The treatment goal for epilepsy
is complete elimination of seizures with no side effects and an optimal quality of life.128
The anti-epileptic drug (AED) phenytoin is metabolized via CYP2C9, which is known
to have several genetic polymorphisms.129 Two of the known SNPs, CYP2C9*21 (Arg144Cys)
and CYP2C9*3 (Ile359Leu), have been shown to increase the risk of intoxication with medi-
um-range doses of phenytoin in homozygous and heterozygous carriers.130-137 Phenytoin is
also metabolized via CYP2C19. Two mutant CYP2C19m1 and/or CYP219m2 alleles have been
found to be associated with reduced phenytoin metabolism.134,138,139
Unlike phenytoin, conflicting data exist regarding the influence of genetic polymor-
phisms in CYP isoenzymes on phenobarbital metabolism.140 Phenobarbital is mainly
metabolized via CYP2C19. One group of researchers demonstrated a moderate decrease
in clearance of phenobarbital in CYP2C19*2/*2 and *2/*3 genotypes relative to those with
CYP2C19 *1/ *1; however, this report was not confirmed in a subsequent study.141,142
A third AED, VPA, was the focus of a study examining the role of CYP2C9 polymor-
phisms in the biotransformation of VPA into its hepatotoxic and inactive metabolites. The
homozygous and heterozygous genotypes of alleles CYP2C9*2 and CYP2C9*3 were associ-
Table 9-2
Associations Between Pharmacotherapy for Parkinson’s Disease and
Genetic Polymorphisms
L-dopa with COMT and dopamine receptors—a SNP at nucleotide 1947 that encodes for low or high
activity genotype is associated with changes in response to both tolcapone and L-dopa; DRD2
gene may play a pivotal role in more frequent motor complications of long-term L-dopa use;
associations of late-onset hallucinations with the C allele of the taq1A polymorphism 10.5 kb of
DRD2; and similar results obtained in DRD1, DRD2, DRD3, and DRD4 in patients with PD with or
without chronic visual hallucinations
COMT inhibitors with COMT and glucuronosyltransferase (UGT) 1A9—COMT haplotype appears to
have little influence on the development of L-dopa-induced dyskinesias
Selegiline with CYP2B6—CYP2B6*18 and other CYP2B6 defect variant alleles may be potential
biomarkers for altered selegiline biodisposition; currently there are no recommendations for
selegiline use and CYP2B6 pharmacogenetics
Pramipexole with DRD3—no significant associations exist between DRD3 (MscI polymorphism)
and DRD4 (120-bp tandem duplication polymorphism in the promoter region) polymorphisms
and the phenomenon of “sleep attacks”
COMT, catechol-O-methyltransferase; PD, Parkinson’s disease; SNP, single nucleotide polymorphism.
Source: See reference 126 for more information.
ated with a decrease in the formation of both these metabolites. The consequences for the
drug’s hepatotoxicity and teratogenicity have not been determined so far.143,144
In addition to the CYP isoenzyme genotypes, P-glycoprotein (P-gp) variations can
alter the intracerebral penetration of several AEDs such as carbamazepine, phenytoin,
phenobarbital, VPA, gabapentin, topiramate, lamotrigine, and felbamate; P-gp is encoded
in humans by the multidrug-resistant (MDR) 1 gene (designated ABCB1).144-147 The ABCB1

gene is polymorphic, and seven retrospective case-control studies have been published
on the issue of genetic risk factors for pharmacoresistance in epilepsy associated with the
multi-drug resistance transporter.148-154 Four of the case reports demonstrated a positive
response, finding an association of genetic polymorphisms with AED pharmacoresistance,
while three showed negative results.148-154 A more recent study found no association of
several polymorphisms within ABCB1 and any of the outcome measures (time to first
seizure after starting drug therapy, time to 12-month remission, or time to drug withdrawal
due to unacceptable adverse effects or to lack of seizure control).155 Additional studies
are needed in this area with stratification for specific epilepsy syndromes and the use of a
unified definition of pharmacoresistance.156,157
Polymorphisms in drug targets of AEDs are another area where association studies
with pharmacoresistance have begun. There is a published study that demonstrates an
association between variants in the α-subunit of a voltage-gated sodium channel with
differing response to carbamazepine and phenytoin.158 Further studies are needed in
this area for each AED before this information can be utilized prior to prescribing these
medications.
Few studies have been completed thus far that have demonstrated an association
between adverse drug reactions and polymorphisms with regard to AEDs. Cutaneous
adverse effects of drugs are rare, but there have been isolated case reports of familial
observations of Stevens-Johnson syndrome (SJS) and the drug reaction with eosinophilia
and systemic symptoms (DRESS) syndrome with the AEDs.159-161 HLA genotyping in a recent
study of a Han Chinese population that developed SJS after exposure to carbamazepine
revealed HLA-B*15:02 alleles in all patients.162 A replication study showed that only 33% of
patients had the HLA-B*15:02 allele with SJS and carbamazepine, although this study was
done only in Asian ancestry patients.163 The HLA region may contain important genes for
SJS, but it is not a universal marker because it is most likely linked to ethnic background.161
Pain
Opioids are mu opioid receptor (OPRM1) agonists used for the treatment of mild-to-severe,
acute, and chronic pain.164 They are characterized as having a narrow therapeutic index,
with the most serious toxicity being respiratory depression.164 Many factors may play a role
in balancing pain control without causing respiratory depression/sedation including the
patient characteristics, the patient’s perception of pain, the severity of pain, the duration
of pain, the opioid drug and dosing regimen, and the patient’s genetic makeup. The deter-
mination of the role of pharmacogenomics in opioid pain control is difficult because the
various experimental pain stimuli used in studies are different from the numerous clinical
pain responses. In addition, the identification of genes involved in pain control is only now
being fully elucidated.
OPRM1 is the primary binding site for opioid drugs. There are 100 variants in the OPRM1
identified, with more than 20 producing amino acid changes with a frequency of greater
than 1%.165,166 The most commonly identified SNP is A118G (frequency of 2–48%, ethnicity
dependent), which leads to a loss of putative N-glycosylation site in the extracellular
receptor region.166 Several studies have investigated the role of this polymorphism on the
efficacy and/or dosage requirements of opioids. The results have demonstrated the poly-
morphism may cause a decreased opioid effect and an increase in opioid dosage required
for pain control. These results are thought to be opioid drug and response specific, owing to
the contribution of environmental factors and other genes affecting opioid responses.167-171
The metabolism of opioid medications through the CYP450 system has been more
extensively studied with regard to genetic variability. A significant correlation between
CYP2D6 phenotype and the ability to metabolize weaker opioids (codeine, dihydroco-
deine, oxycodone, hydrocodone, and tramadol) to their more potent hydroxyl metabo-
lites (morphine, dihydromorphine, oxymorphone, and hydromorphone) has been estab-
lished.172-200 The importance of pharmacogenomics in opioid pain control has contributed
greatly to our understanding of the variability in drug response and toxicity seen with
these agents. However, large, ethnically diverse studies with standardized protocols will be
needed before this knowledge can be translated into routine clinical practice.201
Alcohol, Nicotine, and Drug Addiction Treatments

One of the most promising areas of research in pharmacogenomics and addiction
treatment is gene OPRM1 A118G (rs561720) polymorphism. Three randomized, placebo-
controlled studies have shown that individuals with a G allele display longer abstinence
and greater reduction in the positive effects of alcohol when taking naltrexone.202-204 Addi-
tional studies are needed to test and develop adaptive approaches to treatment based on
genotype information.205,206
CLINICAL PEARL
Further research is needed to determine how genetic information in the
treatment of addiction disorders will be incorporated into clinical practice
given the controversy regarding the implementation of genetic testing
for complex phenotypes. Additional prospective studies and pragmatic
clinical trials evaluating both the use of genetic testing in a clinical setting
and its effect on treatment outcome could further prospective testing in
addiction medicine prior to implementation.
SUMMARY
With the growing number of medications used to treat all of the varying disorders of the
CNS, prescribing the most beneficial medication for each patient is becoming increasingly
more difficult. Numerous genetic factors contribute to the CNS disorders, and many more
may ultimately lead to a medication’s response or failure. In addition, these medications
cause a staggering number of adverse drug events.

Several factors affect the implementation of neurologic and psychiatric pharmaco-
genomics. Initially, the cost of testing was the largest barrier to implementation of pharma-
cogenomics into clinical practice; but as genotyping technologies decrease in cost, it may
be a cost-effective strategy to incorporate into pharmacotherapy because it can improve
outcomes and reduce adverse effects. Another concern with the implementation of phar-
macogenomic testing is the lack of understanding about the implications of such test-
ing.206 Although pharmacogenomic markers in the treatment of neurology and psychiatry
are unlikely to attain perfect sensitivity and specificity, they are still beneficial and can be
meaningful in informing a clinical decision, clarifying a prognosis, and guiding the develop-
ment of a clinical treatment plan.
CLINICAL PEARLS
●● Poor metabolizers: Examples of extreme adverse effects can be found in
the literature. With the ability to predict poor metabolizers of CYP2D6,
the clinician would be able to minimize adverse effects by choosing
either a medication that is metabolized by an alternate enzyme or by
modifying the dosing strategy for the drug metabolized via CYP2D6.
●● CYP1A2: This is an inducible enzyme. When patients who smoke
cigarettes are treated with olanzapine and clozapine on an inpatient unit
where smoking is restricted, the blood levels can decrease on release
when smoking is reinitiated. If a patient’s CYP1A2 activity can be
determined prior to discharge, proper dosing of these medications could
be used to prevent relapse.
●● Appropriate antidepressant selection for patients: The trial and error
method of selecting antidepressant medications can take its toll on a
patient and clinician. Prior to initiation of the medications, a patient’s
genetic susceptibility to response and potential adverse effects should
be determined in order to speed the process of recovery.
●● HLA-*B-15:02: A determination should be made for the susceptibility
of patients of Asian ancestry to develop SJS with initiation of
carbamazepine. This determination of HLA-*B-15:02 allele status
prior to the initiation of carbamazepine allows for a decrease in the
potentially life-threatening adverse effects.
REFERENCES
1. Amitriptyline package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=1e6d2c80-
fbc8-444e-bdd3-6a91fe1b95bd. Accessed October 18, 2015.
2. Aripiprazole package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=569f1c32-fb6f-
c34f-0894-4b1e7b744a8a. Accessed October 18, 2015.
3. Aripiprazole Aristada package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.
cfm?setid=17a8d11b-73b0-4833-a0b4-cf1ef85edefb. Accessed March 21, 2017.
4. Aripiprazole Maintena package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.
cfm?setid=17a8d11b-73b0-4833-a0b4-cf1ef85edefb. Accessed October 18, 2015.
5. Atomoxetine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=47137273-b9c1-
4cb4-99ca-2dab9690b560. Accessed October 18, 2015.
6. Brexpiprazole package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=2d301358-
6291-4ec1-bd87-37b4ad9bd850. Accessed March 21, 2017.

7. Brivaracetam package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=3cf2f439-
0e97-443e-8e33-25ecef616f6c. Accessed March 21, 2017.
8. Carbamazepine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=c13bc0b8-
7900-4ef4-98ed-e1315a08d95d. Accessed October 18, 2015.
9. Citalopram package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=6daeb45c-451d-
b135-bf8f-2d6dff4b6b01. Accessed October 18, 2015.
10. Clobazam package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=de03bd69-2dca-

459c-93b4-541fd3e9571c. Accessed October 18, 2015.
11. Clomipramine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=d7cee3fa-
d05c-4702-8f75-f446627bdb49. Accessed October 18, 2015.
12. Clozapine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=d5c8a456-6f3c-
4963-b321-4ed746f690e4. Accessed October 18, 2015.
13. Desipramine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=ba02a95d-

d82e-4a13-90b4-a219abc0249a. Accessed October 18, 2015.
14. Dextromethorphan and quinidine package insert. https://www.nlm.nih.gov/medlineplus/druginfo/
meds/a611048.html. Accessed October 18, 2015.
15. Diazepam package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7733052d-177b-
49da-949e-4d950bd2afeb. Accessed October 18, 2015.
16. Divalproex sodium package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=f911748c-
fb3a-fe1a-ab2a-4b40455e05ef. Accessed October 18, 2015.
17. Doxepin package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=aa69c63c-ced6-
4676-a16b-554f1af7d210. Accessed October 18, 2015.
18. Duloxetine package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0a541d20-
5466-433b-a104-40a7b2296076. Accessed March 21, 2017.
19. Escitalopram package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=2db02476-
325f-ee47-a7e5-b5aed56c364c. Accessed March 21, 2017.
20. Eteplirsen package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=33bff678-7829-
479e-9110-b8e33a0bc0aa. Accessed March 21, 2017.
21. Fluoxetine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=aa69c63c-ced6-
4676-a16b-5. Accessed October 18, 2015.
22. Fluoxetine and olanzapine combination package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.
cfm?setid=6b28c424-0b7e-4b75-b090-f116b113554e54f1af7d210. Accessed October 18, 2015.
23. Fluvoxamine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=9c16f853-1a0b-
4ce3-8e17-e2bf666c4292. Accessed October 18, 2015.
24. Galantamine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=bbdaf2f2-14e6-
45c4-b807-5b22955444b0. Accessed October 18, 2015.
25. Iloperidone package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=43452bf8-76e7-
47a9-a5d8-41fe84d061f0. Accessed October 18, 2015.
26. Imipramine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=2f7d5e7f-79aa-
4f3a-b9a6-758d9f3ce74d. Accessed October 18, 2015.
27. Lacosamide package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=f7ea3ce6-
4ca7-4405-a011-48e7d71360a5. Accessed March 21, 2017.
28. Modafinil package insert. http://dailymed.nlm.nih.gov/dailymed/search.cfm?labeltype=all&query=Moda
finil&pagesize=20&page=1. Accessed October 18, 2015.
29. Nefazodone package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b1d149db-ad43-
4f3f-aef1-fb0395ba4191. Accessed October 18, 2015.
30. Nortriptyline package insert. http://dailymed.nlm.nih.gov/dailymed/search.cfm?labeltype=all&query=N
ORTRIPTYLINE+HYDROCHLORIDE&pagesize=20&page=1. Accessed October 18, 2015.
31. Oxcarbazepine package insert. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=17325a80-
fb9c-4a83-b4b4-98e0b999d852. Accessed March 21, 2017.
32. Paroxetine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=89dd7e24-85fc-
4152-89ea-47ec2b48a1ed. Accessed October 18, 2015.
33. Perphenazine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=5fbfe9da-
26e8-4705-98f3-42acd3d7b439. Accessed October 18, 2015.
34. Phenytoin package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7e467c02-49a0-
4b62-b537-430fdfa4f10e. Accessed October 18, 2015.

35. Pimozide package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=70b079e2-a1f7-
4a93-8685-d60a4d7c1280. Accessed October 18, 2015.
36. Protriptyline package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=700abc58-
9362-4ef5-9d7a-dd3c4d364d0a. Accessed October 18, 2015.
37. Risperidone package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=1c3250f3-d291-
4280-97d5-da3bb171e2b6. Accessed October 18, 2015.
38. Tetrabenazine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=5867189c-

17bd-4e70-a079-c4730f9950d1. Accessed October 18, 2015.
39. Thioridazine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=9c4bedb4-
2d59-4fcd-aad7-fce988cd96d8. Accessed October 18, 2015.
40. Trimipramine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=3e1e4157-bc15-
4e2a-bb2a-ff41b2a8f905. Accessed October 18, 2015.
41. Valproic acid package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=9d3b0fc6-

c69f-4110-a606-c0b5d690b5e5. Accessed October 18, 2015.
42. Venlafaxine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=8fba17da-573a-
425b-b1d7-7ddeeef3c957. Accessed October 18, 2015.
43. Vortioxetine package insert. http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=4b0700c9-b417-
4c3a-b36f-de461e125bd3. Accessed October 18, 2015.
44. Kessler RC, Berglund P, Demler O. The epidemiology of major depressive disorders: results from the
National Comorbidity Survey replication (NCS-R). JAMA. 2003;289:3095-3105.
45. Thase ME, Entsuah AR, Rudolph RL. Remission rates during treatment with venlafaxine or selective
serotonin reuptake inhibitors. Br J Psychiatry. 2001;178:234-241.
46. Mulrow CP, Williams JW Jr, Trivedi M, et al. Treatment of depression-newer pharmacotherapies. Evid
Rep Technol. 1999;7:1-4.
47. Thase ME, Haight BR, Richard N, et al. Remission rates following antidepressant therapy with bupropi-
on or selective serotonin reuptake inhibitors: a meta-analysis of original data from seven randomized
controlled trials. J Clin Psychiatry. 2005;66:974-981.
48. Ramamoorthy S, Leibach FH, Mahesh VB, et al. Partial purification and characterization of the human
placental serotonin transporter. Placenta. 1993;14:449-461.
49. Lesch KP, Wolozin BL, Estler HC, et al. Isolation of a cDNA encoding the human brain serotonin trans-
porter. J Neural Transm Gen Sect. 1993;91:67-72.
50. Lesch KP, Bengel D, Heils A, et al. Association of anxiety-related traits with a polymorphism in the
serotonin transporter gene regulatory region. Science. 1996;274:1527-1531.
51. Heils A, Teufel A, Petri S, et al. Allelic variation of human serotonin transporter gene expression. J
Neurochem. 1996;66:2621-2624.
52. Gelernter J, Cubells JF, Kidd JR, et al. Population studies of polymorphisms of the serotonin trans-
porter protein gene. Am J Med Genet. 1999;88:61-66.
53. Smeraldi E, Zanardi R, Benedetti F, et al. Polymorphism within the promoter of the serotonin trans-
porter gene and antidepressant efficacy of fluvoxamine. Mol Psychiatry. 1998;3:508-511.
54. Zanardi R, Benedetti F, Di Bella D, et al. Efficacy of paroxetine in depression is influenced by a func-
tional polymorphism within the promoter of the serotonin transporter gene. J Clin Psychopharmacol.
2000;20:105-107.
55. Pollock BG, Ferrell RE, Mulsant BH, et al. Allelic variation in the serotonin transporter promoter
affects onset of paroxetine treatment response in late-life depression. Neuropsychopharmacology.
2000;35:587-590.
56. Arias B, Catalan R, Gastro C, et al. 5-HTTLPR polymorphism of the serotonin transporter gene predicts
nonremission in major depression patients treated with citalopram in a 12-weeks follow up study. J
Clin Psychopharmacol. 2003;23:563-567.
57. Zanardi R, Serretti A, Rossini D, et al. Factors affecting fluvoxamine antidepressant activity; influence
of pindolol and 5-HTTLPR in delusional and nondelusional depression. Biol Psychiatry. 2001;50:323-
330.
58. Rausch JL, Johnson ME, Fei YJ, et al. Initial conditions of serotonin transporter kinetics and genotype:
influence on SSRI treatment trial outcome. Biol Psychiatry. 2001;51:723-732.
59. Joyce PR, Mulder RT, Luty SE, et al. Age-dependent antidepressant pharmacogenomics: polymor-
phisms of the serotonin transporter and G protein beta3 subunit as predictors of response to fluox-
etine and nortriptyline. Int J Neuropsychopharmacol. 2003;6:339-346.

60. Durham LK, Webb SM, Milos PM, et al. The serotonin transporter polymorphism, 5HTTLPR, is associ-
ated with a faster response time to sertraline in an elderly population with major depressive disorder.
Psychopharmacology. 2004;174:525-529.
61. Murphy GM Jr, Hollander SB, Rodrigues HE, et al. Effects of the serotonin transporter gene promoter
polymorphism on mirtazapine and paroxetine efficacy and adverse events in geriatric major depres-
sion. Arch Gen Psychiatry. 2004;61:1163-1169.
62. Serretti A, Cusin C, Rossini D, et al. Further evidence of a combined effect of SERTPR and TPH on
SSRIs response in mood disorders. Am J Med Genet B Neuropsychiatr Genet. 2004;129:36-40.
63. Kraft JB, Slager SL, McGrath PJ, et al. Sequence analysis of the serotonin transporter and associations
with antidepressant response. Biol Psychiatry. 2005;58:374-381.
64. Kim DK, Lim SW, Lee S, et al. A variable number of tandem repeats in the serotonin transporter gene
does not affect the antidepressant response to fluvoxamine. Psychiatry Res. 2002;111:253-259.
65. Ito K, Yoshida K, Sato K, et al. A variable number of tandem repeats in the serotonin transporter gene
does not affect the antidepressant response to fluvoxamine. Psychiatry Res. 2002;111:235-239.
66. Yoshida K, Ito K, Sato K, et al. Influence of the serotonin transporter gene-linked polymorphic region
on the antidepressant response to fluvoxamine in Japanese depressed patients. Prog Neuropsycho-
pharmacol Biol Psychiatry. 2002;26:383-386.
67. Yu YW, Tsai SJ, Chen TJ, et al. Association study of the serotonin transporter promoter polymorphism
and symptomatology and antidepressant response in major depressive disorders. Mol Psychiatry.
2002;7:1115-1119.
68. Lee MS, Lee HY, Lee HJ, et al. Serotonin transporter promoter gene polymorphism and long-term
outcome of antidepressant treatment. Psychiatr Genet. 2004;14:111-115.
69. Kato M, Ikenaga Y, Wakeno M, et al. Controlled clinical comparison of paroxetine and fluvoxamine
considering the serotonin transporter promoter polymorphism. Int Clin Psychopharmacol. 2005;20:151-
156.
70. Kim H, Lim SW, Seonwoo K, et al. Monoamine transporter gene polymorphisms and antidepressant
responses in Koreans with late-life depression. JAMA. 2006;296:1609-1618.
71. Taylor MJ, Sen J, Bhagwagar Z. Antidepressant response and the serotonin transporter gene-linked
polymorphic region. Biol Psychiatry. 2010;68:536-543.
72. Kato M, Serretti A. Review and meta-analysis of antidepressant pharmacogenetic findings in major
depressive disorder. Mol Psychiatry. 2010:15:473-500.
73. Lemonde S, Du L, Bakish D, et al. Association of the C(-1019)G 5-HT 1A functional promoter polymor-
phism with antidepressant response. Int J Neuropsychopharmacology. 2004;7:501-506.
74. Arias B, Catalan R, Gastro C, et al. Evidence for a combined genetic effect of the 5-HT (1A) receptor
and serotonin transporter genes in the clinical outcome of major depressive patients treated with
citalopram. J Psychopharmacol. 2005;19:166-172.
75. McMahon FJ, Buervenich S, Charney D, et al. Variation in the gene encoding the serotonin 2A receptor
is associated with outcome of antidepressant treatment. Am J Hum Genet. 2006;78:804-814.
76. Wu WH, Huo SJ, Cheng CY, et al. Association study of the 5-HT (6) receptor polymorphism (C267T)
and symptomatology and antidepressant response in major depressive disorders. Neuropsychobiology.
2001;44:172-175.
77. Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group. Recom-
mendations from the EGAPP Working Group: testing for cytochrome P450 polymorphisms in adults
with nonpsychotic depression treated with selective serotonin reuptake inhibitors. Genetics Medicine.
2007;9:819-825.
78. Thakur M, Grossman I, McCrory DC, et al. Review of evidence for genetic testing for CYP450 polymor-
phisms in management of patients with nonpsychotic depression with selective serotonin reuptake
inhibitors. Genetics Medicine. 2007;9:826-835.
79. Peters EJ, Slager SL, Kraft JB, et al. Pharmacokinetic genes do no influence response or tolerance to
citalopram in the STAR*D sample. PLOS One. 2008;3:e1872.
80. Hicks JR, Bishop JR, Sangkuhl K, et al. Clinical Pharmacogenomics Implementation Consortium (CPIC)
Guideline for CYP2D6 and CYP2C19 genotypes and dosing of selective serotonin reuptake inhibitors.
Clin Pharmacol Ther. 2015;98:127-134.
81. Jones P, Buckley P. Schizophrenia. London: Mosby; 2006.
82. Bishop JR, Ellingrod VL. Neuropsychiatric pharmacogenomics: moving toward a comprehensive under-
standing of predicting risks and response. Pharmacogenomics. 2004;5:463-477.

83. Lieberman JA, Stroup TS, McEvoy JP, et al. Effectiveness of antipsychotic drugs in patients with chron-
ic schizophrenia. N Engl J Med. 2005;353:1209-1223.
84. Leckband SG, Bishop JR, Ellingrod V. Pharmacogenomics in psychiatry. J Pharm Pract. 2007;20:252-
264.
85. Moore TA, Buchanan RW, Buckley PF, et al. The Texas Medication Algorithm Project antipsychotic
algorithm for schizophrenia: 2006 update. J Clin Psychiatry. 2007;68:1751-1762.
86. Kapur S, Mamo D. Half a century of antipsychotics and still a central role for dopamine D2 receptors.
Prog Neuropsychopharmacol Biol Psychiatry. 2003;27:1081-1090.
87. Meltzer HY, Li Z, Kaneda Y, et al. Serotonin receptors: their key role in drugs to treat schizophrenia.
Prog Neuropsychopharmacol Biol Psychiatry. 2003;27:1159-1172.
88. Muller DJ, Kennedy JL. Genetics of antipsychotic treatment emergent weight gain in schizophrenia.
Pharmacogenomics. 2006;7:863-887.
89. De Luca V, Mueller DJ, de Bartolomeis A, et al. Association of the HTR2C gene and antipsychotic
induced weight gain: a meta-analysis. Int J Neuropsychopharmacol. 2007;10:697-704.
90. Casey DE. Tardive dyskinesia. West J Med. 1990;153:535-541.
91. Yassa R, Jeste DV. Gender differences in tardive dyskinesia: a critical review of the literature. Schizophr
Bull. 1992;18:701-715.
92. Thelma BK, Srivastava V, Tiwari AK. Genetic underpinnings of tardive dyskinesia: passing the baton to
pharmacogenetics. Pharmacogenomics. 2008;9:1285-1306.
93. Neville MJ, Johnstone EC, Walton RT. Identification and characterization of ANKKI: a novel kinase gene
closely linked to DRD2 on chromosome band 11.q23.1. Hum Mutat. 2004;23(6):540-545.
94. Zai CC, Romano-Silva MA, Hwang R, et al. Genetic study of eight AKT1 gene polymorphism and their
interaction with DRD2 gene polymorphisms in tardive dyskinesia. Schizophr Res. 2008;106:248-252.
95. Tsai HT, North KE, West SL, et al. The DRD3 rs6280 polymorphism and prevalence of tardive dyskine-
sia: a meta-analysis. Am J Med Genet B Neuropsychiatr Genet. 2010;153B:57-66.
96. Mannisto PT, Ulnianen I, Lundstrom K, et al. Characteristics of catechol-o-methyltransferase (COMT)
and properties of selective COMT inhibitors. Prog Drug Res. 1992;39:291-350.
97. Hori H, Ohmori O, Shinkai T, et al. Manganese superoxide dismutase gene polymorphism and schizo-
phrenia: relation to tardive dyskinesia. J Neuropsychopharmacol. 2000;23:170-177.
98. Lerer B, Segman RH, Tan EC, et al. Combined analysis of 635 patients confirms an age-related asso-
ciation of the serotonin 2A receptor gene with tardive dyskinesia and specificity for the non-orofacial
subtype. Int J Neuropsychopharmacol. 2005;8:411-425.
99. Patsopoulos NA, Ntzani EE, Zintzaras E, et al. CYP2D6 polymorphisms and the risk of tardive dyskine-
sia in schizophrenia: a meta-analysis. Pharmacogenet Genomics. 2005;15:151-158.
100. Dettling M, Cascorbi I, Opgen Rhein C, et al. Clozapine-induced agranulocytosis in schizophrenic
Caucasians: confirming clues for associations with human leukocyte class I and II antigens. Pharmaco-
genomics J. 2007;7:325-332.
101. Athanasious MC, Dettling MC, Cascorbi I, et al. Candidate gene analysis identifies a polymorphism in
HLA-DQB1 associated with clozapine-induced agranulocytosis. J Clin Psychiatry. 2-11;72:458-463.
102. Panariello F, De Luca V, de Bartolomies A. Weight gain, schizophrenia and antipsychotics: new findings
from animal model and pharmacogenomic studies. Schizophr Res Treat. Volume 2011 (2011); Article ID
459284:16.
103. Akiskal HS, Bougeois ML, Angst J, et al. Reevaluating the prevalence of and diagnostic composition
within the broad clinical spectrum of bipolar disorders. J Affect Disord. 2000;59:5-30.
104. Rybakowski JK, Chlopocka-Wozniak M, Suwalska A, et al. The prophylactic effect of long-term
lithium administration in bipolar patients entering treatment in the 1970s and 1980s. Bipolar Disord.
2001;3:63-67.
105. Kasper S, Stamenkovic M, Letmaier M, et al. Atypical antipsychotics in mood disorders. Int Clin Psycho-
pharmacol. 2002;3:1-10.
106. McCarthy MJ, Leckband SG, Kelsoe JR. Pharmacogenetics of lithium response in bipolar disorder.
Pharmacogenomics. 2010;11:1439-1465.
107. Centers for Disease Control and Prevention: What is Attention-Deficit/Hyperactivity Disorder? Avail-
able at: http://www.cdc.gov/ncbddd/adhd/what.htm. Accessed October 30, 2008.
108. Stein MA, McGough JJ. The pharmacogenomic era: promise for personalizing attention deficit hyper-
activity disorder therapy. Child Adolesc Psychiatric Clin N Am. 2008;17:475-490.
109. Michelson D, Read HA, Ruff D, et al. CYP2D6 and clinical response to atomoxetine in children and
adolescents with ADHD. J Am Acad Child Adolesc Psychiatry. 2007;46:242-251.
110. Bruxel EM, Akutagava-Martins GC, Salatino-Oliveria R, et al. ADHD pharmacogenetics across the life
cycle: new findings and perspectives. Am J Med Genet. Part B 2014;165B:263-282.
111. Trapp BD. Pathogenesis of multiple sclerosis: the eyes only see what the mind is prepared to compre-
hend. Ann Neurol. 2004;55:455-457.
112. Comabella M, Fernandez-Arquero M, Rio J, et al. HLA class I and II alleles and response to treatment
with interferon-beta in relapsing-remitting multiple sclerosis. J Neuroimmunol. 2009;210:116-119.
113. Domanski P, Colamonicic OR. The type-I interferon receptor. The long and short of it. Cytokine
Growth Factor Rev. 1996;7:143-151.
114. Sriram U, Barcellos LF, Villoslada P, et al. Pharmacogenomic analysis of interferon receptor polymor-
phisms in multiple sclerosis. Genes Immun. 2003;4:137-152.
115. Byun E, Caillier SJ, Montalban X, et al. Genome-wide pharmacogenomic analysis of the response to
interferon beta therapy in multiple sclerosis. Arch Neurol. 2008;65:337-344.
116. Cacabelos R, Alvarez XA, Lombardi V, et al. Pharmacological treatment of Alzheimer disease: from
psychotropic drugs and cholinesterase inhibitors to pharmacogenomics. Drugs Today. 2000;36:415-499.
117. Giacobini E. Cholinesterase in human brain: the effect of cholinesterase inhibitors on Alzheimer’s
disease, related disorders. In: Giacobini E, Pepeu G, eds. The Brain Cholinergic System in Health and
Disease. Abingdon, Oxon: Informa Healthcare; 2006:235-264.
118. Loveman E, Green C, Kirby J, et al. The clinical and cost-effectiveness of donepezil, rivastigmine,
galantamine and memantine for Alzheimer’s disease. Health Technol Assess. 2006;10:1-176.
119. Kamboh MI. Molecular genetics of late-onset Alzheimer’s disease. Am Hum Genet. 2004;68:381-404.
120. Cacabelos R. Pharmacogenomics in Alzheimer’s disease. Mini Rev Med Chem. 2002;2:59-84.
121. Cacabelos R. Pharmacogenomics for the treatment of dementia. Ann Med. 2002;34:357-379.
122. Cacabelos R. The application of functional genomics to Alzheimer’s disease. Pharmacogenomics.
2003;4:587-621.
123. Cacabelos R. Pharmacogenomics and therapeutic prospects in dementia. Eur Arch Psychiatry Clin
Neuosci. 2008;258:28-47.
124. Leender KL. Pathophysiology of movement disorders studied using PET. J Neural Transm Suppl.
1997;50:39-46.
125. Gilgun-Sherki Y, Djaldetti R, Melamed E, et al. Polymorphism in candidate genes: implications for the
risk and treatment of idiopathic Parkinson’s disease. Pharmacogenomics J. 2004;4:291-306.
126. Wang J, Si YM, Liu ZL, et al. Cholecystokinin, cholecystokinin-A receptor and cholecystokinin-B
receptor gene polymorphisms in Parkinson’s disease. Pharmacogenetics. 2003;13:365-369.
127. Sander JW. The epidemiology of epilepsy revisited. Curr Opin Neurol. 2003;16:165-170.
128. Vickrey BG, Hays RD, Rausch R, et al. Quality of life of epilepsy surgery patients as compared with
outpatients with hypertension, diabetes, heart disease, and/or depressive symptoms. Epilepsia.
1994;35:597-607.
129. Lee Cr, Goldstein JA, Pieper JA. Cytochrome P450 2C9 polymorphisms: a comprehensive review of the
in-vitro and human data. Pharmacogenetics. 2002;12:251-263.
130. Rettie AE, Wienkers LC, Gonzalez FJ, et al. Impaired S-warfarin metabolism catalyzed by the R144C
allelic variant of CYP2C9. Pharmacogenetics. 1994;4:39-42.
131. Haining RL, Hunter AP, Veronese ME, et al. Allelic variants of human cytochrome P450 2C9: baculovirus-
mediated expression, purification, structural characterization, substrate stereoselectivity, and prochiral
selectivity of the wild-type and 1359L mutant forms. Arch Biochem Biophys. 1996;353:447-458.
132. Brandolese R, Scordo MG, Spina E, et al. Severe phenytoin intoxication in a subject homozygous for
CYP2C9*3. Clin Pharmacol Ther. 2001;70:391-394.
133. Kidd RS, Curry TB, Gallagher S, et al. Identification of a null allele of CYP2C9 in an African-American
exhibiting toxic phenytoin. Pharmacogenetics. 2001;11:803-808.
134. Mamiya K, Ieiri I, Shimamoto J, et al. The effects of genetic polymorphisms of CYP2C9 and CYP2C19
on phenytoin metabolism in Japanese adult patients with epilepsy: studies in stereoselective hydrox-
ylation and population pharmacokinetics. Epilepsia. 1998;39:1317-1323.
135. Soga Y, Nishimura F, Ohtsuka Y, et al. CYP2C9 polymorphisms, phenytoin metabolism and gingival
overgrowth in epileptic subjects. Life Sci. 2004;74:827-834.
136. Schwarz UI. Clinical relevance of genetic polymorphisms in the human CYP2C9 gene. Eur J Clin Invest.
2003;33:23-30.
137. Van der Weide J, Steijins LSW, van Weelden JM, et al. The effect of genetic polymorphism of cyto-
chrome P450 CYP2C9 on phenytoin dose requirement. Pharmacogenetics. 2001;11:287-291.
138. Xiao ZS, Goldenstein JA, Xie HG, et al. Differences in the incidence of the CYP2C19 polymorphism
affecting the S-mephenytoin phenotype in Chinese Han and Bai populations and identification of a
new rare CYP2C19 mutant allele. J Pharmacol Exp Ther. 1997;281:604-609.
139. Odani A, Hashimoto Y, Otsuki Y, et al. Genetic polymorphism of the CYP2C subfamily and its effect
on the pharmacokinetics of phenytoin in Japanese patients with epilepsy. Clin Pharmacol Ther.
1997;62:282-287.
140. Kwan P, Brodie MJ. Phenobarbital for the treatment of epilepsy in the 21st century: a critical review.
Epilepsia. 2004;45:1141-1149.
141. Mamiya K, Hadama A, Yukawa E, et al. CYP2C19 polymorphism effect on phenobarbital. Pharmaco-
kinetics in Japanese patients with epilepsy: analysis by population pharmacokinetics. Eur J Clin Phar-
macol. 2000;55:821-825.
142. Hadama A, Ieri I, Morita T, et al. P-hydroxylation of phenobarbital: relationship to (S)-mephenytoin
hydroxylation (CYP2C19) polymorphism. Ther Drug Monit. 2001;23:115-118.
143. Ho PC, Abbott FS, Zanger UM, et al. Influence of CYP2C9 genotypes on the formation of a hepato-
toxic metabolite of valproic acid in human liver microsomes. Pharmacogenomics J. 2003;3:335-342.
144. Loscher W, Potschka H. Role of multidrug transporter in pharmacoresistance to antiepileptic drugs.
145. Sills GJ, Kwan P, Butler E, et al. P-glycoprotein mediated efflux of antiepileptic drugs: preliminary
studies in MDR1 knockout mice. Epilepsy Behav. 2002;3:427-432.
146. Dean M, Rzhetsky A, Allikmets R. The human ATP-binding cassette (ABC) transporter superfamily.
Genome Res. 2001;11:1156-1166.
147. HUGO gene Nomenclature Committee (online). Available at: http://www.gene.ucl.ac.uk/nomenclature/
Accessed October 28, 2008.
148. Siddiqui A, Kerb R, Weale ME, et al. Association of multidrug resistance in epilepsy with a polymor-
phism in the drug-transporter gene ABCB1. N Engl J Med. 2003;348:1442-1448.
149. Zimprich F, Sunder-Plassman R, Stogmann E, et al. Association of an ABCB1 gene haplotype with
pharmacoresistance in temporal lobe epilepsy. Neurology. 2004;63:1087-1089.
150. Hung CC, Tai JJ, Lin CJ, et al. Complex haplotypic effects of the ABCB1 gene on epilepsy treatment
response. Pharmacogenomics. 2005;6:411-417.
151. Seo T, Ishitsu T, Ueda N, et al. ABCB1 polymorphisms influence the response to antiepileptic drugs in
Japanese epilepsy patients. Pharmacogenomics. 2006;7:551-561.
152. Tan NCK, Heron SE, Scheffer IE, et al. Failure to confirm association of a polymorphism in ABCB1 with
multidrug resistant epilepsy. Neurology. 2004;63:1090-1092.
153. Sills GJ, Mohanraj R, Butler E, et al. Lack of association between the C3435T polymorphism in the
human multidrug resistance (MDR1) gene and response to antiepileptic drug treatment. Epilepsia.
2005;46:643-647.
154. Kim YO, Kim MK, Woo YJ, et al. Single nucleotide polymorphisms in the multidrug resistance 1 gene in
Korean epileptics. Seizure 2006;15:67-72.
155. Leschziner G, Jorgensen AI, Andrew T, et al. Clinical factors and ABCB1 polymorphisms in prediction of
antiepileptic drug response: a prospective cohort study. Lancet Neuro. 2006;5:668-676.
156. Ott J. Association of genetic loci. Neurology. 2004;63:955-958.
157. Cardon LR, Bell JI. Association study designs for complex diseases. Nat Rev Genet. 2001;2:91-99.
158. Tate SK, Depondt C, Sisodiya SM, et al. Genetic predictors of the maximum doses patients receive
during clinical use of the anti-epileptic drugs carbamazepine and phenytoin. Proc Natl Acad Sci USA.
2005;102:5507-5512.
159. Gennis MA, Vemuri R, Burns EA, et al. Familial occurrence of hypersensitivity to phenytoin. Am J Med.
1991;91:631-634.
160. Fischer PR, Shigeoka AO. Familial occurrence of Stevens-Johnson syndrome. Am J Dis Child.
1983;137:914-916.
161. Foujeau J-C, Stern RS. Severe cutaneous adverse reactions to drugs. N Engl J Med. 1994;331:19:1272-
1285.
162. Wen-Hung C, Shuen-iu H, Hong-Shang E, et al. A marker for Stevens-Johnson syndrome. Nature.
2004;428:486.
163. Lonjou C, Thomas L, Borot N, et al. A marker for Stevens-Johnson syndrome: ethnicity matters. Phar-
macogenomics J. 2006;6:265-268.
164. Somogyi AA, Barratt DT, Coller JK. Pharmacogenetics of opioids. Clin Pharmacol Ther. 2007;81:429-
444.
165. Ikeda K, Ide S, Han W, et al. How individual sensitivity to opiates can be predicted by gene analyses.
Trends Pharmacol Sci. 2005;26:311-317.
166. Lotsch J, Geisslinger G. Are μ-opioid receptor polymorphisms important for clinical opioid therapy?
Trends Mol Med. 2005;11:82-89.
167. Cararco Y, Maroz Y, Davidson E. Variability in alfentanil analgesia may be attributed to polymorphism
in the mu-opioid receptor. Clin Pharmacol Ther. 2001;39:63.
168. Klepstad P, Rakvag TT, Kaasa S, et al. The 118A>G polymorphism in the human μ-opioid receptor gene
may increase morphine requirements in patients with pain caused by malignant disease. Acta Anaes-
thesiol Scand. 2004;79:316-324.

169. Coulbault L, Beaussier M, Verstuyft C, et al. Environmental and genetic factors associated with
morphine response in the postoperative period. Clin Pharmacol Ther. 2006;79:316-324.
170. Hirota T, Ieiri I, Takane H, et al. Sequence variability and candidate gene analysis in two cancer patients
with complex clinical outcomes during morphine therapy. Drug Metab Dispos. 2003;31:677-680.
171. Lotsch J, Zimmerman M, Darimont J, et al. Does the A118G polymorphism at the μ-opioid receptor
gene protect against morphine-6-glucuronide toxicity? Anesthesiology. 2002;97:814-819.
172. Sindrup SH, Brosen K, Bjerring P, et al. Codeine increases pain thresholds to copper vapor laser stimuli
in extensive but not poor metabolizers of sparteine. Clin Pharmacol Ther. 1991;48:686-693.
173. Yue QY, Alm C, Svensson JO, et al. Codeine O-demethylation co-segregates with polymorphic
debrisoquine hydroxylation. Br J Clin Pharmacol. 1989;28:639-645.
174. Chen ZR, Somogyi AA, Bochner F. Polymorphic-O-demethylation of codeine. Lancet. 1988;2:914-915.
175. Chen ZR, Somogyi AA, Reynolds G, et al. Disposition and metabolism of codeine after single and
chronic doses in one and seven extensive metabolizers. Br J Clin Pharmacol. 1991;31:381-390.
176. Yue QY, Alm C, Svensson JO, et al. Quantification of O- and N-demethylated and the glucuronidated
metabolites of codeine relative to the debrisoquine metabolic ratio in urine in ultrarapid, rapid, and
poor debrisoquine hydroxylators. Ther Drug Monit. 1997;19:539-542.
177. Williams DG, Patel A, Howard RF. Pharmacogenetics of codeine metabolism in an urban population of
children and its implications for analgesic reliability. Br J Anaesth. 2002;89:839-845.
178. Heiskanen T, Olkkola KT, Kalso E. Effects of blocking CYP2D6 on the pharmacokinetics and pharma-
codynamics of oxycodone. Clin Pharmacol Ther. 1998;64:603-611.
179. Otton SV, Schadel M, Cheung SW, et al. CYP2D6 phenotype determines the metabolic conversion of
hydrocodone to hydromorphone. Clin Pharmacol Ther. 1993;4:463-472.
180. Fromm MF, Hofmann U, Griese EU, et al. Dihydrocodeine: a new opioid substrate for the polymorphic
CYP2D6 in humans. Clin Pharmacol Ther. 1995;58:374-382.
181. Poulsen L, Arendt-Nielsen L, Brosen K, et al. The hypoalgesic effect of tramadol in relation to
CYP2D6. Clin Pharmacol Ther. 1996;60:636-644.
182. Borlak J, Hermann R, Erb K, et al. A rapid and simple CYP2D6 genotyping assay-case study with anal-
gesic tramadol. Metabolism. 2003;52:1439-1443.
183. Filegert F, Kurth B, Gohler K. The effects of tramadol on static and dynamic pupillometry in healthy
subjects—the relationship between pharmacodynamics, pharmacokinetics and CYP2D6 metabolizer
status. Eur J Pharmacol. 2005;61:257-266.
184. Slanar O, Nobilis M, Kventina J, et al. Miotic action of tramadol is determined by CYP2D6 genotype.
Physiol Res. 2007;56:129-136.
185. Slanar O, Nobilis M, Kventina J, et al. CYP2D6 polymorphism, tramadol pharmacokinetics and pupillary
response. Eur J Clin Pharmacol. 2006;62:75-76.
186. Pedersen RS, Damkier P, Brosen K. Tramadol as a new probe for cytochrome P450 2D6 phenotyping:
a population study. Clin Pharmacol Ther. 2005;77:458-467.
187. Eap CB, Broly F, Mino A, et al. Cytochrome P450 2D6 genotype and methadone steady-state concen-
trations. J Clin Psychopharmacol. 2001;21:229-234.
188. Coller JK, Joergensen C, Foster DJ, et al. Lack of influence of CYP2D6 genotype on the clearance of
(R)-, (S)- and racemic-methadone. Int J Clin Pharmacol Ther. 2007;45:410-417.
189. Sindrup SH, Arendt-Nielsen L, Brosen K, et al. The effect of quinidine on the analgesic effect of
codeine. Eur J Pharmacol. 1992;42:587-591.

190. Desmeules J, Gascon MP, Dayer P, et al. Impact of environmental and genetic factors on codeine
analgesia. Eur J Clin Pharmcol. 1991;41:23-26.
191. Enggaard TP, Poulsen L, Arendt-Nielson L, et al. The analgesic effect of tramadol after intravenous
injection in healthy volunteers in relation to CYP2D6. Anesth Analg. 2006;102:146-150.
192. Poulsen L, Brosen K, Arendt-Nielson L, et al. Codeine and morphine in extensive and poor metabo-
lizers of aparteine: pharmacokinetics, analgesic effect and side effects. Eur J Clin Pharmacol.
1996;51:289-295.
193. Eckhardt K, Li S, Ammons S, et al. Same incidence of adverse drug events after codeine administration
irrespective of the genetically determined differences in morphine formation. Pain. 1998;76:27-33.
194. Persson K, Sjostrom S, Sigurdardottir I, et al. Patient-controlled analgesia (PCA) with codeine for post-
operative pain relief in ten extensive metabolisers and one poor metaboliser of dextromethorphan. Br
J Clin Pharmacol. 1995;39:182-186.
195. Poulsen L, Riishede L, Brosen K, et al. Codeine in post-operative pain: study of the influence of spar-
teine phenotype and serum concentrations of morphine and morphine-6-glucuronide. Eur J Clin
Pharmacol. 1998;54:451-454.
196. Maddocks I, Somogyi A, Abbott F, et al. Attenuation of morphine-induced delirium in palliative care by
substitution with infusion of oxycodone. J Pain Symptom Manage. 1996;12:182-189.
197. Stamer UM, Stuber F. Impact of CYP2D6 genotype on postoperative tramadol analgesia. Pain.
2003;105:231-238.
198. Caraco Y, Sheller J, Wood AJ. Pharmacogenetic determination of the effects of codeine and predic-
tion of drug interactions. J Pharmacol Exp Ther. 1996;278:1165-1174.
199. Mikus G, Trausch B, Rodewald C, et al. Effect of codeine on gastrointestinal motility in relation to
CYP2D6 phenotype. Clin Pharmacol Ther. 1997;61:459-466.
200. Hasselstrom J, Yue QY, Sawe J. The effect of codeine on gastrointestinal transit in extensive and poor
metabolisers of debrisoquine. Eur J Clin Pharmacol. 1997;53:145-148.
201. Contopoulos-loannidis DG, Alexious GA, Gouvias TC, et al. An empirical evaluation of multifarious
outcomes in pharmacogenetics: a beta-2 adrenoceptor gene polymorphisms in asthma treatment.
Pharmacogenet Genom. 2006;16:705-711.
202. Anton RF, Oroszi G, O’Malley S, et al. An evaluation of mu-opioid receptor (OPRM1) as a predictor of
naltrexone response in the treatment of alcohol dependence: results from the combined pharmaco-
therapies and behavioral interventions for alcohol dependence (COMBINE) study. Arch Gen Psychiatry.
2008;65:135-144.
203. Kim SG, Kim CM, Choi SW, et al. A micro opioid receptor gene polymorphism (A118G) and naltrex-
one treatment response in adherent Korean alcohol-dependent patients. Psychopharmacology.
2009;201:611-618.
204. Oslin DW, Berrettini W, Kranzler HR, et al. A functional polymorphism of the mu-opioid receptor gene
is associated with naltrexone response in alcohol-dependent patients. Neuropsychopharmacology.
2003;28:1546-1552.
205. De Leon J. Pharmacogenomics: the promise of personalized medicine for CNS disorders. Neuro-
psychopharmacology. 2009;34:159-172.
206. Ray LA, Hutchinson KE. Effects of naltrexone on alcohol sensitivity and genetic moderators of medi-
cation response: a double-blind placebo-controlled study. Arch Gen Psychiatry. 2007;64:1069-1077.
CHAPTER
10
Infectious Diseases
Edward Grace, PharmD, BCPS (AQ-ID), AAHIVP

and Christopher Farrell, PhD

After completing this chapter, the reader DEOXYRIBONUCLEIC ACID (DNA)—a self-
should be able to replicating material present in nearly all
Describe the role of genetics in
of chromosomes.
identifying various infectious diseases
agents in the clinical setting. POLYMORPHISMS—existence of a gene in
several allelic forms in a single species
Explain the variability in anti-infective
or among humans.
pharmacokinetics and pharmacodynamics
based on differences in individual and SINGLE NUCLEOTIDE POLYMORPHISM
organism genetics. (SNP)—a variant DNA sequence in which
the purine or pyrimidine base
Apply the understanding of genetics and
(e.g., cytosine) of a single nucleotide has
pharmacogenomics in optimizing anti-
been replaced by another such base
infective therapy.
(e.g., thymine).
Formulate an appropriate plan to monitor
patient outcomes and adverse effects
based on pharmacogenomics factors.
309
INTRODUCTION
Genetics play a large role in infectious diseases management as it enables us to identify
specific infectious agents, optimize drug therapy, and prevent unwanted anti-infective
adverse effects. To maximize understanding of the role genetics play in infectious diseases,
clinicians must comprehend the genetic variability among patients and also among various
infectious organisms. In addition, pharmacists must understand the relationship between

human genetics and the genetics of the organisms that infect humans. In this chapter, we
will discuss how human genetics play a role in the response of an individual to bacterial,
fungal, and viral infections with regard to efficacy and adverse effects. Additionally, we will
discuss how genetic differences and similarities among specific organisms enable us to
rapidly detect infectious organisms in addition to optimizing anti-infective therapy.
CASE STUDY
A 52-year-old Caucasian male recently tested positive for human immunodeficiency
virus (HIV) and has not started antiretroviral therapy at this time. He presents to
the infectious diseases clinic for followup on his HIV care and a new complaint of
oropharyngeal candidiasis. In addition to HIV, the patient has a history of hypertension,
diabetes (type II), hyperlipidemia, and one episode of Pneumocystis jirovecii pneumonia
(formerly known as Pneumocystis carinii pneumonia or PCP) 5 years ago. The patient
notes he was previously tested for glucose-6-phosphate dehydrogenase (G6PD)
deficiency and was not deficient. The patient is currently on lisinopril 10 mg PO QHS and
atenolol 25 mg PO daily for his hypertension; metformin 1,000 mg BID; sitagliptin 100
mg PO daily; and pravastatin 40 mg PO QHS.
Labs obtained prior to this visit:
HIV-Ab: positive
HBeAg: negative
HIV RNA: 384,500 copies/mL
CD4+ count: 121 cells/mm3
CD4+ %: 11%
HLA-B*5701: positive
Genotype Tropism Test: dual-mixed tropic virus
Genotype: mutations present include K103N, M184V, and L210W
Questions
1. Given this information, what would be an appropriate combination therapy for the treatment
of the patient’s HIV?
2. What would be an appropriate agent for the prevention of PCP re-infection in this patient?
BACTERIA
Role of Genetics in Identification of Bacteria and Bacterial Resistance
Culturing the site of infection including wounds, blood, sputum, and cerebral spinal fluid
has been the most commonly used technique to identify bacterial organisms based on
individual characteristics of organisms with regard to certain dyes, growth media, and other
chemical agents. However, culturing bacteria in media to later identify them takes 3 to 7
days for rapidly growing organisms while some slower replicating bacteria may take up to
Chapter 10 • Infectious Diseases 311
several weeks to identify. A delay of days to weeks needed to identify an organism often
precludes clinicians from using specific antibiotics, which are effective against the limited
bacteria without using broad-spectrum agents that are associated with increased adverse
effects and bacterial resistance. It was not until 1983 that a group of scientists conceived
of using a technique called polymerase chain reaction (PCR), a type of nucleic acid ampli-
fication test (NAAT) to identify specific bacteria.1 In 1993, the founder of PCR, Kary B. Mullis,
PhD, was awarded the Nobel Prize in chemistry for this discovery.2 PCR identifies a bacte-
rium by amplifying the highly conserved sequence of deoxyribonucleic acid (DNA) encoding
for certain proteins such as the 16S ribosomal ribonucleic acid (rRNA) subunit that is unique
to each species of bacteria.3,4
Current PCR methods also detect unique DNA sequences that encode for specific
toxins, transporters, and antigens specific to certain bacterial species. Early methods of
performing PCR were labor intensive and required specialized equipment that were not
readily available to most clinical microbiology laboratories. Over the past two decades,
PCR has become a widely used clinical tool due to improvements in the time and materials
needed to perform such tests in the clinical setting. Today, PCR is used as a rapid test for
identifying various types of bacteria such as sexually transmitted organisms (e.g., Neisseria
gonorrhoeae, Chlamydia trachomatis) in addition to other virulent organisms such as Clos-
tridium difficile, Staphylococcus aureus (e.g., methicillin-resistant Staphylococcus aureus
[MRSA]), and enterococcus species (e.g., vancomycin-resistant enterococcus [VRE] species).5
Most often, bacteria are identified through PCR by amplification of 16S and/or 18S
rRNA. However, some bacteria are identified by targeting specific genes in the organ-
ism, which encode for resistance to certain antibiotics. Currently, two virulent organisms
can be identified in blood samples using real-time PCR (RT-PCR), which provides results
within 1 hour. MRSA can be identified from blood samples by targeting the mecA gene
that encodes for expression of an alternative penicillin-binding protein known as PBP2A,
which does not bind to most b-lactam antibiotics.4 Additionally, VRE can be identified by
PCR through targeting of the vanA genes that cause resistance by alteration of the cross-
linking of peptidoglycan in the cell wall, thus preventing vancomycin from binding to its
target side.4
The same technology can be applied to other sites of infection such as sputum and
cerebral spinal fluid. Mycobacterium tuberculosis (TB), which has been shown to be a diffi-
cult organism to grow, can also be identified through RT-PCR. Using this technique, organ-
isms that have the rifampin-resistant allele can be detected, which results in the resistance
to rifampin and possible resistance to other anti-TB agents. Organisms such as Clostridium
difficile can be identified in addition to detecting the NAP1/BI/027 gene that is more viru-
lent and possesses greater resistance to standard antibiotic therapy. Similarly, Escherichia
coli O157:H7 variants can be identified and treated accordingly given its significant level of
resistance to antibiotic therapy.6
In recent years, NAATs have evolved from PCR (including RT-PCR) to other less
labor-intensive and more specific methods such as next-generation sequencing (NGS),
loop-mediated amplification (LAMP), transcription-mediated amplification, nucleic acid

sequence-based amplification (NASBA), strand displacement amplification (SDA), multi-
plex/microarray, pyrosequencing, and fluorescence in situ hybridization (FISH). NGS can’t
only identify suspected bacterial pathogens such as MRSA and VRE, but it can also be used
to identify any organism present in a clinical sample by a process called metagenomics
assembly. This process is where random nucleotide fragments are aligned/matched with
known reference sequences to identify a variety of organisms.7 With this process, scientists
can identify organisms that had not been previously identified or known to cause a certain
type of infection. This use of NGS with metagenomics assembly (also known as the Shot-
gun approach) has helped scientists identify new bacteria and viruses not previously clas-
sified, such as the West Nile virus and the Middle East respiratory syndrome coronavirus.
In addition to NGS, multiplex/microarray assays have rapidly evolved in the field of

diagnostics in infectious diseases due to the faster turnaround times compared to the NGS
(<4 hours compared to 8 hours). One such test marketed as BioFire Diagnostics BCID has
been introduced into the clinical setting in recent years.8 This platform detects up to 27
organisms, including various gram-positive organisms such as MRSA and VRE as well as
gram-negative organisms such as carbapenem-resistant enterobacteriaceae in addition to
fungi such as Candida species.
Another platform known as Prove-It Sepsis is designed to identify over 60 bacterial
pathogens (including MRSA) and up to 13 fungal pathogens in <4 hours.9 Such platforms,
in addition to other techniques such as the SeptiFast and Hyplex BloodScreen Multiplex
PCR systems, have an overall sensitivity of 90–100% for various organisms. This form of
testing enables clinicians to identify pathogenic organisms in a few hours after a patient
presents for medical care, thereby decreasing the need for empiric anti-infective therapy
that can increase organism resistance, adverse effects, and cost to the patient and health-
care system.10
CLINICAL PEARL
The use of NAAT in identification of bacteria has become one of the most
common tests performed on patients admitted to hospitals. Patients are
often screened for the presence and colonization of MRSA in the nares in
addition to testing blood samples for the presence of MRSA.
Host Response to Infection and Antibiotics

Genetic testing has not only helped clinicians identify bacteria and their resistance patterns
in patients, it has also helped to determine a patient’s immune response to infection. Tran-
scriptomic analysis of a host’s response to infection can help differentiate patients with
active versus latent TB infections using the transcriptional profile of circulating white blood
cells, which differ based on the activity of TB.11 This method has enabled clinicians to deter-
mine a patient’s response to anti-TB agents in as little as a couple of weeks compared to
months when the progress was measured only by improvement in imaging studies. Several
human polymorphisms in the genetic locus for leukotriene A4 hydrolase (LTA4H) were
found to affect the individual’s inflammatory response to TB infection. The LTA4H gene
encodes for the leukotriene-modifying enzyme that converts leukotriene A4 to leuko-
triene B4. Individuals with TB meningitis who had a polymorphism in LTA4H benefited from
corticosteroid adjunctive therapy, while those without a polymorphism were harmed by the
addition of adjunctive corticosteroid therapy.12,13

In addition to determining an individual’s response to an infection or antibiotic agents,
genetic testing can help to predict the likelihood of antibiotic-related adverse effects
in certain patients. The use of genetic testing in identifying patients with G6PD defi-
ciency enables clinicians to avoid the use of certain antibiotics such as sulfamethoxazole-
trimethoprim, dapsone, and nitrofurantoin thus decreasing the risk of hemolytic anemia
associated with these medications.14,15 Patients with G6PD deficiency have been linked to
genetic polymorphisms with single base pair substitutions on the gene that is located on
the X chromosome. Males and homozygous females with these polymorphisms are at the
highest risk of red blood cell hemolysis and anemia when taking the above-mentioned
antibiotics compared to heterozygous females. Affected individuals collectively account for

over 400 million individuals worldwide.
G6PD deficiency is linked to red blood cell hemolysis and severe anemia because
G6PD is the only source of nicotinamide adenine dinucleotide phosphate hydrogen
(NADPH) and reduced glutathione (GSH) that acts as an antioxidant in red blood cells.
NADPH plays a major role in protecting human red blood cells and hemoglobin from oxida-
tive stress and subsequent damage. During infection, G6PD activity rises to counteract the
increase oxidative stress caused by infection. In patients with G6PD deficiency, suboptimal
levels of G6PD are unable to overcome the oxidative stress in red blood cells resulting
in significant hemolysis. Individuals should be tested for G6PD deficiency before being
prescribed sulfamethoxazole-trimethoprim or dapsone. This is especially important for
patients who will be on prolonged sulfamethoxazole-trimethoprim therapy such as those
with acquired immunodeficiency syndrome (AIDS) who need prophylaxis against PCP and
toxoplasmosis. In certain studies, the sulfamethoxazole hypersensitivity reaction has been
linked to a single nucleotide polymorphism (SNP) in glutamate cysteine ligase catalytic
(GCLC) subunit coding gene; however, the use of the GCLC gene test is not yet available
in clinical practice.14,16
CLINICAL PEARL
All patients being considered for long-term therapy with sulfamethoxazole-
trimethoprim or dapsone should be screened for G6PD deficiency prior to
initiation of these antibiotics in order to avoid severe hemolytic anemia if
they are G6PD-deficient.
Metabolomics and Transporters Affecting Antibiotics

Various antibiotics are activated and/or metabolized through the cytochrome P-450
system. In recent years, discovery of polymorphisms in the genes that encode for various
CYP-450 isozymes has led to the identification of individuals who are considered slow or
rapid metabolizers of certain medications due to either down- or up-regulation of CYP-450
encoding genes, respectively. Several CYP-450 enzymes, including CYP-450 3A4, 2C9,
2C19, and 2D6, are major isozymes in the activation, bioconversion, and/or deactivation of
many medications that are used in the clinical setting.
Perhaps one of the most significant polymorphisms identified in the CYP-450 system
related to anti-infectives has been the CYP3A4*22 variant.17 This variant has a frequency
of 0.08 in Caucasians and 0.04 in Asian and African populations. Individuals with this
polymorphism exhibit significantly lower CYP3A4 expression and have been shown to
have supratherapeutic serum concentrations of antibiotics metabolized through this path-
way.18,19 Examples of drugs affected by this polymorphism include rifampin, isoniazid, and
clarithromycin. Polymorphisms affecting CYP2C9 may also result in altered concentrations
of antibiotics such as rifampin, metronidazole, and sulfamethoxazole-trimethoprim.20 Of
note, sulfamethoxazole (a component of sulfamethoxazole-trimethoprim) requires activa-

tion by CYP2C9 when taken orally and; therefore, polymorphisms such as CYP2C9*2 and
CYP2C9*3 may decrease the efficacy of sulfamethoxazole-trimethoprim by reducing sulfa-
methoxazole activation.
Intracellular transport pumps have been shown to play a significant role in regulating
the intracellular concentration of many medications, including antibiotics. When upregu-

lated, these pumps increase the cell’s capability to efflux the medications from within
the cells thus removing them from their site of action. The most extensively studied
efflux pump is the P-glycoprotein (P-gp) transporter. P-gp is encoded by a gene known
as ABCB1, which is a member of the ATP-binding cassette (ABC) transporters family.21,22
Several antibiotics including minocycline, azithromycin, linezolid, levofloxacin, cipro-
floxacin, daptomycin, and rifampin are known substrates for P-gp and are affected by
P-gp activation and upregulation. Of note, some antibiotics can act as both inducers and
substrates of ABCB1 resulting in increased excretion of the antibiotics from the cell and/
or body.23 Such antibiotics would benefit from the addition of an inhibitor of the specific
efflux transporter; however, no specific inhibitors are currently approved that do not have
other medicinal effects.
FUNGI
Role of Genetics in Identification of Fungi
As previously discussed in the bacteria section, multiplex/microarray systems are currently
used in the clinical setting to identify fungal infections from blood, sputum, and other
infection sites. In addition, the U.S. Food and Drug Administration (FDA) has approved other
new technologies such as peptide nucleic acid-FISH (PNA-FISH) for detecting various
fungi and bacteria that may be present in blood cultures.24,25 PNA is a DNA analog with
synthetic peptide backbone that is fluorescently labelled and binds specific DNA and/or
RNA sequences from bacteria or fungi.25 A procedure is currently under development that
will identify both fungi and bacteria in blood cultures under the name QuickFISH BC.26 This
platform will also detect and identify various Candida species in a single blood culture.
Metabolomics and Transporters Affecting Antifungals

Triazole antifungals such as fluconazole work by inhibiting fungal 14α-sterol demethylase
(also known as CYP51A1), which is responsible for converting lanosterol to ergosterol.
Triazoles are potent inhibitors and substrates for various CYP-450 enzymes including 3A4,
2C9, 2C19, and 2B6. Many of these CYP-450 enzymes are expressed in different levels in
individuals based on the presence of specific polymorphisms that may increase or decrease
their activity—for example, fluconazole is a potent inhibitor of several CYP-450 enzymes,
which may result in decreased metabolism of other medications, leading to higher serum
concentrations without affecting the levels of fluconazole in serum.27
On the other hand, voriconazole is both a substrate and inhibitor of CYP 2C19, 3A4,
and 3A5.27 Voriconazole levels can be greatly altered in patients with genetic polymor-
phisms affecting CYP-450 enzymes expression in addition to voriconazole affecting the
metabolism of other drugs through its effects on CYP enzymes.28 This finding necessitates
the measurement of voriconazole serum levels in patients with severe infections because
serum levels may be greatly affected by the presence of various CYP450 polymorphisms.
Similar to voriconazole, itraconazole is both a substrate and inhibitor of CYP3A4 and,

therefore, prone to variations in serum levels dependent on the presence/absence of
certain polymorphisms that affect CYP3A4.27 However, itraconazole is also a substrate
and inhibitor of P-gp in addition to being an inhibitor of multidrug-resistant protein 3 and
breast cancer resistant protein (BCRP), which are important for the efflux of medications
from within cells.29 Posaconazole, like fluconazole, is not a substrate of CYP450 enzymes;
however, it is an inhibitor of CYP3A4 and can affect the levels of other medications that
are substrates for this enzyme.30
CLINICAL PEARL
Triazoles antifungals interact with many medications due to their
inhibitory action on the CYP-450 system. Significant and serious drug-
drug interactions can still exist despite the absence of any polymorphisms.
VIRUSES
HIV and hepatitis C virus (HCV) are among the two most studied and common viral infec-
tions worldwide. HIV and HCV currently infect over 35 million and 150 million individuals
worldwide, respectively. In this section, we will focus on the genetically based diagnostic
testing for HIV and HCV in addition to other tests commonly used in the clinical setting to
optimize therapy and/or decrease the incidence of adverse drug reactions related to HIV
and HCV treatments.
Human Immunodeficiency Virus

HIV Immunity
HIV is a retrovirus that is dependent on human CD4+ receptors and one or more types
of chemokine co-receptors (CCR5 or CXCR4) for attachment and entry into human CD4+
cells.31 These viruses are often selective for CCR5 during early infectivity and are more
likely to be transmitted to other individuals (CCR5 tropic virus); however, over time, HIV
subpopulations can shift leading to favor viral attachment to the CXCR4 co-receptor
instead of the CCR5 co-receptor.31 Of note, some HIV viruses can bind to either CCR5 or
CXCR4 at any given time (dual tropic virus).
In the mid-1990s, researchers observed that certain individuals were resistant to HIV
infection despite multiple exposures to the virus. These individuals were later identified as
having defects in the CCR5 expression due to a delta-32 mutation (deletion of a 32 base
pairs) on the CCR5 gene, which is located on human chromosome 3.32 A defective gene
results in the absence of a functional CCR5 co-receptor on the host’s CD4+ cell surface,
causing immunity against R5-tropic virus (in individuals homozygous for delta-32 mutation)
or decreased CCR5 co-receptors and slower progression of HIV (in individuals heterozygous
for delta-32 mutation). Individuals who are homozygous for the delta-32 mutation and
thus have no functional CCR5 are generally considered resistant to HIV infection because
R5-tropic virus is the most common form of transmitted HIV. Approximately 1% of Cauca-
sians from European descent are homozygous for the delta-32 mutation and are healthy
and do not suffer from any unique diseases compared to other individuals who lack this
mutation.32 Individuals who are heterozygous for this mutation and thus have one func-
tional copy of the CCR5 gene can be infected with HIV; however, when infected, HIV infec-
tion progresses more slowly and often results in fewer complications compared to those
with two functioning copies of CCR5 gene.31,33
However, the delta-32 mutation does not provide any immunity or protection against
CXCR4-tropic HIV tropic viruses. Based on these findings, the antiretroviral drug maraviroc
was developed to block the CCR5 co-receptors that HIV uses for binding. Because maravi-
roc selectively binds to CCR5 co-receptors and not to CXCR4 co-receptors, it is essential
for patients to be tested for HIV viral tropism prior to initiating maraviroc therapy.34 Individ-
uals with X4-, dual-, or mixed-tropic virus (combination of two of more types of HIV tropic
viruses) will not benefit from maraviroc because HIV can enter the CD4+ cells through the
use of CXCR4 co-receptors, regardless of the presence of maraviroc on CCR5 co-receptors.
Of note, some individuals may initially show viral suppression using maraviroc but later fail
therapy. Although individuals infected with HIV are more likely to harbor R5-tropic virus
early in infection, some individuals may harbor X4-, dual-, or mixed-tropic virus either as a
major or a minor species. Persons who harbor a minor species of a non-R5-tropic virus will
eventually fail therapy due to selective pressure of non-R5-tropic virus that will become
the major population in a given person. To date, no CXCR4 co-receptor inhibitor has been
approved by the FDA due to significant cardiotoxic effects seen in Phase I and II trials.
Tropism testing is essential for all individuals who are being considered for maraviroc to
ensure prolonged efficacy of the agent.
Individuals who lack delta-32 mutations may also be resistant to HIV infection through
another genetic mutation.35 Researchers have observed that some Africans have duplica-
tions of the gene encoding for the chemokine (C-C motif) ligand 3-like 1 protein (CCL3L1),
which is a ligand to the CCR5 co-receptor for HIV R5-tropic viruses. Increases in CCL3L1
levels may also block HIV infection in these particular individuals.36
Currently, two methods are used for testing HIV tropism in individuals considered for
maraviroc and/or failing maraviroc-based therapy: phenotypic and genotypic assays.37,38
The phenotypic assays utilize cell lines that express CD4+ receptors in addition to either
CCR5 or CXCR4 co-receptors. Detection of infected cells is observed using reporter genes
that express a luminescent signal such as luciferase. One such assay is the Enhanced
Sensitivity Trofile Assay (ESTA, Monogram Biosciences Inc., USA), which utilizes pseudo
virus U87 cell lines that exhibit either CCR5 or CXCR4 co-receptors.37 The current genera-
tion of ESTA is able to differentiate R5-, R4-, dual-, or mixed-tropic HIV virus. Despite
the inability of older generations of ESTA to detect minor populations of non-R5 tropic
virus, the current generation of ESTA is able to detect minor populations of non-R5-tropic
viruses that account for ≥0.3% of the total HIV population given the HIV viral load is ≥1,000
copies/mL. The phenotypic ESTA assay is considered the gold standard for HIV tropism
testing, despite the significant cost and turnaround time.38
In comparison to phenotypic tropism tests, genotypic tropism assays are cheaper and
have a shorter turnaround time. Genotypic tropism tests work by ultra-deep sequencing the
V3 loop of glycoprotein-120 (gp120) that contains the main determinates of HIV tropism.37
The presence of positively charged amino acids on the V3 loop, resulting in a positive
charge of greater than +5, predicts the presence of X4-tropic virus. Following analysis of
the V3-loop, interpretation of amino acid sequences is required for establishing the pres-
ence of X4-tropism. Use of various bioinformatics databases such as Geno2Pheno allows
for confirmation of the genotype results based on the sequence of amino acids identified
in the V3 loop. The Illumina (Illumina Inc., USA), PacBio RS (Pacific Biosciences, USA), and
Ion Torrent (Ion Torrent/Life Technologies, USA) are examples of commercially available HIV
genotypic tropism.
HIV RNA Testing
HIV diagnosis is made by use of two sequential tests to confirm the presence or absence
of HIV in a sample. The first step entails the use of either an enzyme-linked immunosor-
bent assay (ELISA) or a NAAT or HIV RNA testing in early infection, followed by an HIV
immunoassay to confirm the results of the initial test.39 NAATs are able to detect the HIV
genome in both a quantitative and qualitative fashion, thus providing the number of virus
copies per milliliter of blood (also known as HIV RNA or viral load) and indicating the pres-
ence of the virus in a patient’s blood or sputum sample much sooner than any other test.
However, due to the NAAT costs, it is often used to screen a large number of patients at
once to determine if one or more individuals within the cohort has HIV infection.40 This
involves testing a pool of individuals together; if a test is positive for the given pool, then
each individual in that given pool is tested by confirmatory testing with an immunoassay
test. NAATs are primarily used in screening blood donors for HIV and in neonates born to
HIV-positive mothers. Additionally, HIV RNA testing is routinely performed on HIV-positive
patients to ensure viral suppression by antiretrovirals and to detect “blips” in viral load,
which are often an early indicator of emerging HIV resistance to antiretroviral therapy in a
given patient. Most HIV RNA tests currently used can detect as few as 15–25 copies of HIV
virus per milliliter of blood.
CLINICAL PEARL
All patients with HIV are routinely tested for HIV RNA viral load, regardless
of whether they are on antiretroviral therapy or not, because increases
in HIV RNA are the earliest warning signs of possible underlying HIV
resistance to a given regimen.
Identifying Antiretroviral Hypersensitivity

In addition to playing an important role in identifying patients with HIV infection and their
viral load, genetic testing has also been an extremely important tool in identifying patients
who may be at risk of developing a severe and possible life-threatening reaction to the
antiretroviral abacavir (Ziagen® or ABC).41 Approximately 5–8% of individuals will develop
a hypersensitivity reaction (HSR) to ABC within 5–10 days of first exposure, which can be
fatal if ABC is not immediately discontinued.42 Signs and symptoms associated with this
specific HSR include fever, rash, diarrhea, abdominal pain, cough, dyspnea, and/or fatigue.
This HSR is believed to be due to the carbovir monophosphate (a metabolite of abacavir)
that may bind directly to the major histocompatibility complex (MHC) molecule on the
cell surface to elicit T-cell activation.43 Specific highly conserved loci with the MHC region
including HLA-B*5701, HLA-DR7, and HLA-DQ3 are believed to play a major role in predict-
ing abacavir HSR in patients. The presence of HLA-B*5701, HLA-DR7, and HLA-DQ3 alleles
have shown to have a positive-predictive value of 100% and a negative-predictive value of

97% for ABC HSR.41 However, in one study, 2% of patients who experienced an ABC HSR
tested negative for HLA-B*5701, HLA-DR7, and HLA-DQ3. In a subsequent study that used
HLA-B*5701 alone to identify patients who are at risk of ABC HSR, 38.8% of individuals
with a positive HLA-B*5701 allele tolerated ABC for 6 weeks without exhibiting any signs
or symptoms of ABC HSR.
Currently, the guidelines for the management of HIV in adults recommend the use of
HLA-B*5701 testing in all patients who are being considered for ABC therapy.41 However,
no commercial test is available for the detection of either HLA-DR7 or HLA-DQ3. It is
important for clinicians to understand that HLA-B*5701 testing can result in both false-
positive and false-negative results if used as a screening test. False-positive tests are
likely due to the presence of HLA-B*5701 without the presence of HLA-DR7 and/or HLA-
DQ3, which are needed, in addition to HLA-B*5701 to cause an HSR. Patients need to be
educated on the classical signs and symptoms of ABC HSR and advised to discontinue
the entire antiretroviral therapy and/or seek immediate medical attention if an HSR to
ABC is suspected.
CLINICAL PEARL
Individuals who are HLA-B*5701 positive or have a history of ABC HSR
should never be placed on ABC-containing regimen again during their
lifetime due to the risk of serious complications and possible death.
HIV Resistance Testing

Newly diagnosed patients with HIV may harbor resistance to antiretrovirals despite never
being exposed to antiretrovirals.39 In the United States, it is estimated that 15% of newly
diagnosed patients with HIV harbor at least one HIV mutation that confers some level of
resistance to antiretrovirals.44 Approximately 8% of newly diagnosed patients with HIV
harbor resistance to at least one antiretroviral agent despite being antiretroviral naïve. One
of the most commonly transmitted HIV mutations is the K103N (a substitution of a base
pair at position 103) mutation, which results in resistance to all first-generation non-nucle-
oside reverse transcriptase inhibitors (nNRTIs) such as efavirenz (EFV) and nevirapine (NVP).
This occurs through the alteration of the p66 pocket where these medications bind.44
Despite high-level resistance to first-generation nNRTIs, second-generation nNRTIs can
continue to exert antiretroviral activity despite the presence of K103N mutation. In addition,
HIV resistance has shown to increase in nonadherent individuals (<95% adherence) despite
being on antiretrovirals.
In light of these observations, it is recommended that all newly diagnosed patients with
HIV be tested for antiretroviral resistance in addition to individuals who are failing combina-
tion antiretroviral therapy (cART).39 Two types of assays are used to detect HIV resistance
to antiretrovirals: phenotype- and genotype-based assays. Phenotype assays measure the
ability of the HIV virus isolated from a given individual to grow in the presence of various
concentrations of antiretrovirals. In addition, reverse transcriptase, protease, and integrase
gene sequences derived from a patient’s HIV RNA are used to create a pseudo virus that
will express the patient’s specific HIV genes. The interaction of the pseudo virus with various
concentrations of antiretrovirals enables laboratories to determine if therapeutic concentra-
tions of a given antiretroviral can inhibit the specific patient’s HIV multiplication. Phenotypic
testing is both time consuming (2–3 weeks) and costly. However, a benefit of phenotypic
testing is the ability to determine the magnitude of susceptibility/resistance of HIV to a
specific antiretroviral drug. Phenotypic testing has shown the most benefit in antiretroviral-
experienced patients who have been on several different regimens and thus harbor a large
number of mutations, which result in resistance to most antiretrovirals.
On the other hand, genotype assays use population-based Sanger sequencing (PBSS)
or NGS to identify the presence of specific mutations in the viral DNA that encode for
reverse transcriptase, protease, and integrase.45,46 After detection of the specific mutations,
a database is used to correlate the identified mutations with specific resistance to antiretro-
virals. Genotypic assays have the benefit of having a faster turnaround time and are cheaper
than phenotypic assays. However, genotypic assays fail to identify the level of resistance
of HIV to specific antiretrovirals in the given patient because it is based only on popula-
tion data. Databases such as the RD-TRePS, EuResist Prediction Engine, Geno2Pheno,
and Stanford HIV database can be used by both researchers and clinicians to determine
possible antiretroviral resistance patterns based on identified HIV mutations for a specific
patient.47-49 Currently, ViroSeq HIV-1 Genotyping System (Abbott Molecular, USA) is the
only FDA-cleared in vitro genotypic HIV resistance test. Some institutions and laboratories
have switched to using in-house genotypic methods that employ systems such as the Ion
Torrent as mentioned above.
Both genotypic and phenotypic assays suffer from common drawbacks such as their
inability to detect minor HIV populations (subpopulations of HIV with a unique resis-
tance pattern, which accounts for less than 10–20% of HIV population in a given patient)
that may harbor antiretroviral resistance.46 Due to the higher likelihood of missing minor
populations that harbor such resistance, it is recommended to perform these tests when
a patient’s viral load is at least 1,000 copies/mL. However, in some circumstances, despite
a higher likelihood of overlooking a minor population, phenotypic and genotypic assays
can be conducted in patients with a viral load of >200 copies/mL.39 In addition, because
both assays target specific sections of DNA that encode for resistance, mutations in other
areas of the DNA that may result in novel resistance to antiretrovirals may not be detected.
Lastly, HIV resistance testing is recommended while a patient is on a failing cART or soon
after discontinuation of the regimen (within 4 weeks of discontinuation). Rapid testing of
a patient’s HIV resistance will enable “detection” of the resistant HIV population for the
current cART before reverting to wild-type virus as the dominant virus form, which is often
antiretroviral susceptible or harbors limited resistance.39
CLINICAL PEARL
When assessing a patient’s HIV resistance test, clinicians must review
all the prior HIV resistance testing that has been conducted on the
given patient despite how long ago the test was performed. Possible
sequestering of a resistant subpopulation that was previously detectable
can now be hidden as a subpopulation accounting for <20% of the viral
population in a given patient.
Metabolomics and Transporters Affecting Viruses

Most antiretrovirals are metabolized by various CYP-450 enzymes including 3A4, 2B6, and
2B1. Generally, nucleoside reverse transcriptase inhibitors and nNRTIs are metabolized by
CYP2B6 and CYP3A4, while protease inhibitors (PIs) are mainly metabolized by CYP3A4.50
Polymorphisms in the genes, which encode for these CYP-450 enzymes, have been shown
to alter antiretroviral concentrations in patients with HIV leading to either increased toxicity
or increased treatment failures.
EFV, an nNRTI, has been the most studied antiretroviral with regard to pharmaco-
genomics and altered metabolism. EFV is metabolized through CYP2B6, which has been
shown to have an altered expression based on the presence or absence of SNPs at posi-
tions 516 and 983 of the CYP2B6 gene.50,51 The presence of a SNP at position 516 results in
the glutamine-histidine amino acid change. The presence of either of the polymorphisms
has been associated with both higher EFV blood concentrations and the presence of
central nervous system adverse effects.51 The same phenomena has been observed with
other drugs such as NVP where patients with a SNP on the 516 position of CYP2B6 showed
elevated serum concentrations of NVP and an increased risk of hepatotoxicity. Polymor-
phisms in CYP3A5 and CYP3A4 have been associated with increased activity of CYP3A5
and CYP3A4, respectively, leading to rapid metabolism and subtherapeutic concentrations
of several PIs.50
Contributing to the complexity of the pharmacogenomic interactions with antiretro-
viral drugs in the human body is the fact that several intracellular transporters may also
play a significant role in the effectiveness of antiretroviral delivery into human CD4 cells.
Among these transporters are P-gp, BCRP, and organic anion transporters (OAT).52-54 Genes
controlling the expression of these transporters have exhibited SNPs leading to various
degrees of expression of the related transporter. For example, maraviroc is transported into
hepatocytes for metabolism via OATP1B1, which is a carrier protein encoded by the SLCO1B1
gene.52 The presence of a polymorphism on the SLCO1B1 gene at position 512 can result in
significant loss of OATP1B1 function and expression, leading to increased maraviroc levels
in plasma that may lead to increased drug toxicity. Due to the factors affecting antiretro-
viral metabolism and transport within the body and the limited evidence, HIV patients are
not routinely tested for polymorphisms in the CYP-450 system and transporter systems
(such as P-gp and OATP1B1) but mainly restricted to clinical trials at this time.53,54
Hepatitis C
HCV is one of the most common chronic viral infections with more than 150 million individ-
uals infected worldwide. HCV infection is present in 3% of the world’s population and 2% of
the U.S. population.55,56 HCV is a 50-nm enveloped virus whose RNA encodes a polyprotein
precursor that is approximately 3,100 amino acids in length. This polyprotein is processed
by both human and viral proteases into 10 proteins.55 Three of the 10 proteins are known
as structural proteins (core, E1, and E2), and seven are nonstructural proteins (p7, NS2, NS3,
NS4A, NS4B, NS5A, and NS5B).55 The presence of two hypervariable regions in the E2
protein that encodes for envelope glycopeptides lack proofreading capabilities and, thus,
are changed to a proline with a high rate of mutation that results in new viral variants.55
As a result of genetic variability, HCV is classified into seven subtypes known as
genotypes 1–7; each genotype differs from the other genotypes by at least 30% of the
RNA code. Within the seven genotypes, over 60 subtypes have been identified that differ
genetically from each other by approximately 20%.56 Numerous other quasispecies exist in
patients, which differ genetically by approximately 10%. In the United States, genotypes 1a,
1b, 2, and 3 are the most common; genotypes 1a and 1b account for the majority of cases.56
HCV Diagnostic Testing

Due to the large number of patients with HCV infections, screening of individuals who
are high risk is essential in detecting those who need HCV therapy. Detection of high-risk
patients helps prevent HCV complications such as liver cirrhosis, steatosis, and hepato-
cellular carcinoma. Similar to HIV, HCV can be detected in patients based on two types of
assays: ELISA and NAAT.57-60 Current ELISA testing utilizes antigens from the core protein
in addition to NS3, NS4, and NS5 to detect the presence of antibodies (anti-HCV) to these
proteins in an individual.57,58 ELISA testing can be used to identify HCV infection after 8–10
weeks following initial exposure to the HCV virus because anti-HCV must be present in
serum to detect exposure to HCV. One form of ELISA uses rapid immunochromatographic
assays for detecting anti-HCV. This assay has proven to be >99% sensitive so it is recom-
mended for screening individuals for HCV infection.59
NAATs are often used to detect and quantify HCV in blood/serum by use of PCR,
branched-DNA signal amplification, and transcription-medicated amplification.60 Currently,
semi-automated and fully-automated systems are being used in clinical settings to iden-
tify patients with HCV infection and determine the concentration of virus present in a
given individual’s blood/serum. Examples of such systems include Amplicor HCV Monitor
(Roche, USA), COBAS Amplicor HCV Monitor v2.0 (Roche, USA), Abbott RealTime HCV
Assay (Abbott Diagnostics, USA), and Versant HCV 1.0 Assay K-PCR (Siemens Healthcare
Diagnostics, USA). NAAT assays should also be used as a confirmatory test in individuals
who test positive for HCV using an ELISA-based assay.
After detecting the presence of HCV in a given individual, clinicians should determine
the genotype(s) of HCV before initiating treatment for the HCV infection. Certain geno-
types such as 1a, 1b, and 4 may not respond to certain therapies as well as other geno-
types (see section below).60 HCV genotype can be determined in clinical practice by use
of real-time PCR/qPCR with genotype-specific primers or automated reverse hybridiza-
tion to analyze the 5′ region of the HCV genome and the core region, which are thought
to be the genome’s most conserved regions.60 Newer technology such as the microarray
assay containing genotype and subtype specific nucleotides can be used to detect specific
sequences of the NS5B region on the genome.61
Genomics in Natural Disease Progression of HCV
In 2009, a genome-wide association study identified several polymorphisms near the inter-
leukin-28B (IL28B or IFNL3) gene that predicted an HCV patient’s risk for development of
liver inflammation, fibrosis, and cirrhosis.62-64 The GG variant of rs8099917 has been associ-
ated with a higher incidence of spontaneous viral clearance and decreased chance of liver
inflammation, fibrosis, and cirrhosis compared to those who carry the CT and TT variants
in patients with HCV genotypes 1 and 2.62,63 In addition the TT variant rs1279860 carried
similar protective effects for patients with genotype 3 HCV, while the CC and CT variants
of rs1279860 place individuals at higher risk for HCV-related complications.62
The patatin like phospholipase domain containing 3 (PNPLA3) gene is located on chro-
mosome 22 and encodes for a protein responsible for lipid metabolism in hepatocytes and
to a lesser degree in adipose and cutaneous tissues.65 Genetic variations in the PNPLA3
gene due to substitution of an isoleucine with a methionine at residue 148 has been asso-
ciated with an increased risk for and severity of hepatic steatosis.65 PNPLA3 homozygous
individuals with the GG variant may have an accelerated progression of steatosis as well
as an increased risk for liver fibrosis and hepatocellular carcinoma.66 The exact mechanism
behind worsening outcomes of patients who are homozygous for this variant is not fully
understood but may be due to the inability or decreased ability to hydrolyze glycerol lipids,
which leads to steatosis, fibrosis, and eventual hepatocellular carcinoma. Certain ethnici-
ties, including Hispanics and Mediterranean populations, show a higher proportion of indi-
viduals with a homozygous GG variant.66
Genomics and HCV Response to Therapy

IL28B gene variants have been reported to be an important factor in determining treat-
ment success of patients on interferon- and ribavirin-based therapies.62 IL28B variant CC
at rs12979860 has been shown to have a significantly higher treatment success rate. Even
after the introduction of first-generation, direct-acting antivirals (DAA) such as boceprevir
and telaprevir, CC variant individuals have shown a more favorable outcome from therapy
using DAA in combination with interferon and ribavirin therapy compared to those who
possess the CT or TT variants.62,64,67 In addition, individuals with IL28B polymorphism
possessing the TT variant for rs8099917 and genotype-1 HCV have experienced a favorable
outcome to interferon and ribavirin therapy too.67
Based on the most current HCV treatment guidelines published by the American Acad-
emy for Study of Liver Diseases (AASLD) and the Infectious Diseases Society of America
(IDSA), interferon-based therapy is no longer considered first-line therapy for treatment-
naïve patients with HCV infections with genotypes -1, -2, -4, -5, and -6.68 Interferon- and
ribavirin-based therapy can be considered as an alternative therapy for some patients with
genotypes such as genotypes 4-6. With the availability of various DAAs that have shown
higher treatment success rates compared to conventional interferon- and ribavirin-based
therapies, clinicians generally replaced interferon-based therapy with DAA-based therapy.
The latter has been associated with fewer adverse effects and higher chances of viral
suppression following the completion of therapy. Thus, the role of IL28B polymorphisms
will likely play a smaller role in determining treatment success rates given the advent of
DAAs.62,63,67
Pharmacogenomics also can play a role in predicting the likelihood of specific adverse
drug effects in specific patients. The inosine triphosphate pyrophosphatase (ITPA) gene,
which is located on the human chromosome 20 and encodes for an ITPase enzyme, plays
a major role in the conversion of inosine triphosphate to inosine monophosphate thus
affecting the probability of a patient experiencing ribavirin-related anemia.68 ITPA gene is
involved in limiting the amount of nucleotides within a cell as to prevent DNA replication
and RNA transcription errors. The presence of one of the variants (rs1127354 and rs7270101)
has been associated with a decreased incidence and severity of anemia in patients on
ribavirin therapy.69,70 In addition, these variants have been associated with a lower level of
ribavirin in plasma (explaining the decreased incidence of anemia).71 However, neither vari-
ant has been associated with higher treatment failure rates despite lower serum concen-
trations of ribavirin observed in this population.69-71
HCV Resistance to Therapy
Despite the availability of methods to sequence the HCV genome to identify possible poly-
morphisms, which result in resistance to DAAs such as the TaqMAMA and NGS assays, little
is known about the incidence and risk of developing HCV resistance to newly approved
DAAs.72 Mutations in HCV genome at the R155 and/or D168 amino acid positions have
been shown to result in resistance to all first-generation DAA and some second-generation
agents such as simeprevir.73 Although other mutations such as the A156S are selected only
for resistance to first-generation DAA, presence of this mutation did not result in resis-
tance with second-generation DAAs.73 Interestingly, use of first-generation DAA resulted
in a substitution at the R155 position and 36 position in HCV genotype-1a viruses, where
the same DAAs resulted in mutations at positions 156, 36, and 54 in patients infected with
HCV genotype 1b virus. Thus, resistance testing in patients with HCV may be based on the
genotype the individual harbors because mutations that confer resistance may be specific
to individual genotypes. Some patients with HCV genotype 1a have harbored the Q80K
polymorphism within the viral genome, resulting in significant resistance to the second-
generation DAA simeprevir.72,74 The presence of the Q80K polymorphism has shown to
decrease the susceptibility of HCV to simeprevir by approximately 10-fold, which results in
reduced treatment success rates with this agent.74

The current AASLD/IDSA guidelines for the treatment of HCV infection recommend
routine testing of patients infected with HCV genotype 1a for the Q80K polymorphism.68
However, simeprevir (regardless of dose) is not recommended in genotype 1a patients with
evidence of cirrhosis if the Q80K polymorphism is present.68
SUMMARY
Genetics currently play a major role in the identification of infections, the host’s response
to infection and anti-infectives, and the infecting organisms’ response to the host’s immune
system. In bacterial infections, genetic-based testing has enabled clinicians to identify
bacterial organisms causing infections several days earlier than traditional culturing of the
organism. In addition, resistance testing using PCR-based testing has enabled clinicians
to initiate appropriate antibiotics within hours of a patient being admitted to the hospital.
Despite the recent availability of newer tests to identify suspected fungi and/or virus-
es, this area remains a growing field of both clinical and research interest. Certain viruses
such as HIV and HCV have become a prime example of how genetic testing can play a
major role in identifying the presence or absence of infection as well as determining the
genotype of the virus, a host’s response to infection, and the viruses’ response to specific
antiviral medications.
Despite the long-time availability of genetic-based testing in the diagnosis and
management of HIV, this field continues to grow and provides clinicians with more detailed
and specific information regarding HIV resistance (e.g., resistance to the smaller subpopu-
lations present in patients, which has been impossible to detect until recent years). The
availability of such sensitive and specific tests will aid clinicians in selecting the most
appropriate antiretroviral regimen for a patient without concern for the resistance in the
smaller subpopulations, leading to longer lasting and more effective therapy. Genetics
and pharmacogenomics will continue to be important in the management of patients with
various infections. They will also target the infecting organisms with a precision never seen
before in the history of medicine, which will lead to individualized medicine in patients with
infectious diseases.
REFERENCES
1. Mullis K. The unusual origin of the polymerase chain reaction. Sci Am. 1990;262(4):56-61, 64-65.
2. Wieslander L. The PCR method and directed mutagenesis. What significance do discoveries by chem-
istry Nobel Prize winners have for medicine? Lakartidningen. 1993;90(46):4088-4091.
3. Huttunen R, Syrjanen J, Vuento R, et al. Current concepts in the diagnosis of blood stream infections.
Are novel molecular methods useful in clinical practice? Intern J Infect Dis. 2013;17:934-938.
4. Buehler S, Madison B, Snyder S, et al. Effectiveness of practices to increase timeliness of providing
targeted therapy for inpatients with bloodstream infections: a laboratory medicine best practices
systematic review and meta-analysis. Clin Microbiol Rev. 2016;29(1):59-103.
5. Book M, Lehmann L, Zhang X, et al. Monitoring infection: from blood culture to polymerase chain
reaction (PCR). Best Pract Res Clin Anaesthesiol. 2013;27(2):279-288.
6. Collins D, Elliott B, Riley T. Molecular methods for detecting and typing of Clostridium difficile. Pathol-
ogy. 2015;47(3):211-218.
7. Loonen A, Wolffs P, Bruggeman C, et al. Developments for improved diagnosis of bacterial blood-
stream infections. Eur J Clin Microbiol Infect Dis. 2014;33(10):1687-1702.
8. Biofire. bioMérieux. Available at: http://www.biofiredx.com. Accessed December 26, 2015.
9. Prov-it Sepsis Sepsis StripArray information. MOBIDIAG. Available at: http://old.mobidiag.com/Prod-
ucts/ProveittradeSepsis/ProveittradeSepsisStripArray/tabid/242/Default.aspx. Accessed December 26,

2015.
10. Warhurst G, Dunn G, Chadwick P, et al. Rapid detection of health-care−associated bloodstream infec-
tion in critical care using multipathogen real-time polymerase chain reaction technology: a diagnostic
accuracy study and systematic review. Health Technol Assess. 2015;19(35):1-142.
11. Benjak A, Sala C, Hartkoorn RC. Whole-transcriptome sequencing for high-resolution transcriptomic
analysis in Mycobacterium tuberculosis. Methods Mol Biol. 2015;1285:17-30.
12. Brancusi F, Farrar J, Heemskerk D. Tuberculous meningitis in adults: a review of a decade of develop-
ments focusing on prognostic factors for outcome. Future Microbiol. 2012;7(9):1101-1116.
13. Tobin D, Roca F, Oh S, et al. Host genotype-specific therapies can optimize the inflammatory response
to mycobacterial infections. Cell. 2012;148(3):434-446.
14. Luzzatto L, Seneca E. G6PD deficiency: a classic example of pharmacogenetics with on-going clinical
implications. Br J Haematol. 2014;164(4):469-480.
15. Ma M, Woo M, McLeod H. Genetic basis of drug metabolism. Am J Health-Syst Pharm. 2002;59(21)
:2061-2069.
16. Rahman I, MacNee W. Oxidative stress and regulation of glutathione in lung inflammation. Eur Respir
J. 2000;16(3):534-554.
17. Elens L, van Gelder T, Hesselink DA, et al. CYP3A4*22: promising newly identified CYP3A4 variant
allele for personalizing pharmacotherapy. Pharmacogenomics. 2013;14(1):47-62.
18. Kurzawski M, Dąbrowska J, Dziewanowski K, et al. CYP3A5 and CYP3A4, but not ABCB1 polymorphisms
affect tacrolimus dose-adjusted trough concentrations in kidney transplant recipients. Pharmaco-
genomics. 2014;15:179-188.
19. Provenzani A, Santeusanio A, Mathis E, et al. Pharmacogenetic considerations for optimizing tacroli-
mus dosing in liver and kidney transplant patients. World J Gastroenterol. 2013;19:9156-9173.
20. Hesselink D, Bouamar R, Elens L, et al The role of pharmacogenetics in the disposition of and
response to tacrolimus in solid organ transplantation. Clin Pharmacokinet. 2014;53(2):123-139.
21. Zhou SF. Structure, function and regulation of P-glycoprotein and its clinical relevance in drug dispo-
sition. Xenobiotica. 2008;38(7-8):802-832.
22. Pal D, Mitra A. MDR- and CYP3A4-mediated drug-drug interactions. J Neuroimmune Pharmacol.
2006;1(3):323-339.
23. Collu F, Cascella M. Multidrug resistance and efflux pumps: insights from molecular dynamics simula-
tions. Curr Top Med Chem. 2013;13(24):3165-3183.
24. Stender H. PNA FISH: an intelligent stain for rapid diagnosis of infectious diseases. Expert Rev Mol
Diagn. 2003;3(5):649-655.
25. Forrest G. PNA FISH: present and future impact on patient management. Expert Rev Mol Diagn.
2007;7(3):231-236.
26. Abdelhamed A, Zhang S, Watkins T, et al. Multicenter evaluation of Candida QuickFISH BC for identifi-
cation of Candida species directly from blood culture bottles. J Clin Microbiol. 2015;53(5):1672-1676.
27. Niwa T, Imagawa Y, Yamazaki H. Drug interactions between nine antifungal agents and drugs metabo-
lized by human cytochromes P450. Curr Drug Metab. 2014;15(7):651-679.
28. He H, Sun J, Ren X, et al. Effects of CYP3A4 polymorphisms on the plasma concentration of voricon-
azole. Eur J Clin Microbiol Infect Dis. 2015;34(4):811-819.
29. Gupta A, Unadkat J, Mao Q. Interactions of azole antifungal agents with the human breast cancer
resistance protein (BCRP). J Pharm Sci. 2007;96(12):3226-3235.

30. Ashbee H, Gilleece M. Has the era of individualised medicine arrived for antifungals? A review of anti-
fungal pharmacogenomics. Bone Marrow Transplant. 2012;47(7):881-894.
31. Blanpain C, Libert F, Vassart G, et al. CCR5 and HIV infection. Receptors Channels. 2002;8(1):19-31.
32. Mueller A, Strange P. The chemokine receptor, CCR5. Int J Biochem Cell Biol. 2004;36(1):35-38.
33. Friedrich B, Dziuba N, Li G, et al. Host factors mediating HIV-1 replication. Virus Res. 2011;161(2):101-114.
34. Maeda K, Das D, Nakata H, et al. CCR5 inhibitors: emergence, success, and challenges. Expert Opin
Emerg Drugs. 2012;17(2):135-145.

35. Pilotti E, Bianchi M, De Maria A, et al. HTLV-1/-2 and HIV-1 co-infections: retroviral interference on host
immune status. Front Microbiol. 2013;4:372.
36. Nakajima T, Kaur G, Mehra N, et al. HIV-1/AIDS susceptibility and copy number variation in CCL3L1,
a gene encoding a natural ligand for HIV-1 co-receptor CCR5. Cytogenet Genome Res. 2008;123(1-
4):156-160.
37. Lin N, Kuritzkes D. Tropism testing in the clinical management of HIV-1 infection. Curr Opin HIV AIDS.
2009;4(6):481-487.
38. Obermeier M, Symons J, Wensing A. HIV population genotypic tropism testing and its clinical signifi-
cance. Curr Opin HIV AIDS. 2012;7(5):470-477.
39. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. U.S. Depart-
ment of Health and Human Services. Available at: https://aidsinfo.nih.gov/guidelines/html/1/adult-
and-adolescent-treatment-guidelines/0. Accessed December 17, 2015.
40. Patel P, Bennett B, Sullivan T, et al.; CDC AHI Study Group. Rapid HIV screening: missed opportunities
for HIV diagnosis and prevention. J Clin Virol. 2012;54(1):42-47.
41. Mallal S, Nolan D, Witt C, et al. Association between presence of HLA-B*5701, HLA-DR7, and HLA-DQ3
and hypersensitivity to HIV-1 reverse-transcriptase inhibitor abacavir. Lancet. 2002;359(9308):727-732.
42. Hughes A, Mosteller M, Bansal A, et al. Association of genetic variations in HLA-B region with hyper-
sensitivity to abacavir in some, but not all, populations. Pharmacogenomics. 2004;(2):203-211.
43. Watson M, Pimenta J, Spreen W, et al. HLA-B*5701 and abacavir hypersensitivity. Pharmacogenetics.
2004;14(11):783-784.
44. Eron J. Managing antiretroviral therapy: changing regimens, resistance testing, and the risks from
structured treatment interruptions. J Infect Dis. 2008;197(Suppl 3):S261-271.
45. Fonager J, Larsson J, Hussing C, et al. Identification of minority resistance mutations in the HIV-1 inte-
grase coding region using next generation sequencing. J Clin Virol. 2015;73:95-100.
46. Van Laethem K, Theys K, Vandamme A. HIV-1 genotypic drug resistance testing: digging deep, reaching
wide? Curr Opin Virol. 2015;14:16-23.
47. EuResist prediction system. Available at: http://engine.euresist.org/. Accessed December 17, 2015.
48. Geno2Pheno. Available at: http://www.geno2pheno.org/. Accessed December 17, 2015.
49. Stanford HIV Database. Stanford University. Available at: http://hivdb.stanford.edu/. Accessed Decem-
ber 17, 2015.
50. Aceti A, Gianserra L, Lambiase L, et al. Pharmacogenetics as a tool to tailor antiretroviral therapy: A
review. World J Virol. 2015;4(3):198-208.
51. Asensi V, Collazos J, Valle-Garay E. Can antiretroviral therapy be tailored to each human immunodefi-
ciency virus-infected individual? Role of pharmacogenomics. World J Virol. 2015;4(3):169-177.
52. Hashiguchi Y, Hamada A, Shinohara T, et al. Role of P-glycoprotein in the efflux of raltegravir from
human intestinal cells and CD4+ T-cells as an interaction target for anti-HIV agents. Biochem Biophys
Res Commun. 2013;439(2):221-227.
53. Bousquet L, Pruvost A, Guyot A, et al. Combination of tenofovir and emtricitabine plus efavirenz:
in vitro modulation of ABC transporter and intracellular drug accumulation. Antimicrob Agents
Chemother. 2009;53(3):896-902.
54. Neumanova Z, Cerveny L, Ceckova M, et al. Role of ABCB1, ABCG2, ABCC2 and ABCC5 transporters in
placental passage of zidovudine. Biopharm Drug Dispos. 2015;10.1002/bdd.1993.
55. Khullar V, Firpi R. Hepatitis C cirrhosis: New perspectives for diagnosis and treatment. World J Hepatol.
2015;7(14):1843-1855.
56. Dan C, Moses-Eisenstein M, Valdiserri R. Viral hepatitis: New U.S. screening recommendations, assess-
ment tools, and treatments. Am J Nurs. 2015;115(7):26-35.
57. Saludes V, González V, Planas R, et al. Tools for the diagnosis of hepatitis C virus infection and hepat-
ic fibrosis staging. World J Gastroenterol. 2014;20(13):3431-3442.
58. Pondé R. Enzyme-linked immunosorbent/chemiluminescence assays, recombinant immunoblot assays

and nucleic acid tests in the diagnosis of HCV infection. Eur J Clin Microbiol Infect Dis. 2013;32(8):985-988.
59. Liu J, Yu W, Liu S. Positive ratio of specific antibodies to F protein in serum samples from chronic
HCV-infected patients using an enzyme-linked immunosorbent assay: systematic review and meta-
analysis. Eur J Gastroenterol Hepatol. 2013;25(10):1152-1158.
60. Ferreira-Gonzalez A, Shiffman M. Use of diagnostic testing for managing hepatitis C virus infection.
Semin Liver Dis. 2004;24(Suppl 2):9-18.
61. Waldron P, Holodniy M. MicroRNA and hepatitis C virus-challenges in investigation and translation: a
review of the literature. Diagn Microbiol Infect Dis. 2014;80(1):1-12.
62. Chinnaswamy S. Genetic variants at the IFNL3 locus and their association with hepatitis C virus infec-
tions reveal novel insights into host-virus interactions. J Interferon Cytokine Res. 2014;34(7):479-497.
63. Matsuura K, Watanabe T, Tanaka Y. Role of IL28B for chronic hepatitis C treatment toward personal-
ized medicine. J Gastroenterol Hepatol. 2014;29(2):241-249.
64. Saito T, Ueno Y. Transmission of hepatitis C virus: self-limiting hepatitis or chronic hepatitis? World J
Gastroenterol. 2013;19(41):6957-6961.
65. Dongiovanni P, Donati B, Fares R, et al. PNPLA3 I148M polymorphism and progressive liver disease.
World J Gastroenterol. 2013;19(41):6969-6978.
66. Geng N, Xin Y, Xia H, et al. Association of PNPLA3 I148M variant with chronic viral hepatitis, auto-
immune liver diseases and outcomes of liver transplantation. Hepat Mon. 2015 Apr;15(4):e26459.
67. Lee T, Tillmann H, Patel K. Individualized therapy for hepatitis C infection: focus on the interleukin-
28B polymorphism in directing therapy. Mol Diagn Ther. 2014;18(1):25-38.
68. HCV Guidance: Recommendations for Testing, Managing, and Treating Hepatitis C. American Academy
for the Study of Liver Diseases (AASLD). Available at: http://www.hcvguidelines.org/. Accessed Decem-
ber 17, 2015.
69. Cariani E, Villa E, Rota C, et al. Translating pharmacogenetics into clinical practice: interleukin (IL) 28B
and inosine triphosphatase (ITPA) polymorphisms in hepatitis C virus (HCV) infection. Clin Chem Lab
Med. 2011;49(8):1247-1256.
70. Stättermayer A, Scherzer T, Beinhardt S, et al. Review article: genetic factors that modify the outcome
of viral hepatitis. Aliment Pharmacol Ther. 2014;39(10):1059-1070.
71. Hsu C, Kao J. Genomic variation-guided management in chronic hepatitis C. Expert Rev Gastroenterol
Hepatol. 2012;6(4):497-506.
72. Ahmed A, Felmlee D. Mechanisms of hepatitis C viral resistance to direct acting antivirals. Viruses.
2015;7(12):6716-6729.
73. Fourati S, Pawlotsky J. Virologic tools for HCV drug resistance testing. Viruses. 2015;7(12):6346-6359.
74. Costilla V, Mathur N, Gutierrez J. Mechanisms of virologic failure with direct-acting antivirals in hepati-
tis C and strategies for retreatment. Clin Liver Dis. 2015;19(4):641-656.
CHAPTER
11
Respiratory Diseases
Samit Shah, PhD, RPh, MBA and David Ha, PharmD

After completing this chapter, the reader ALLELE—one of a pair of genes on a specific
should be able to location of a chromosome that controls
the same trait.
Describe the genetic variants that can
affect the response to treatment with BINDING MOTIF—a short, highly conserved
b2-agonists, inhaled corticosteroids, region in a protein sequence that is
leukotriene modifiers, and theophylline involved in binding.
in asthmatic patients. BIOFILM—structural layer of microorganisms
Discuss the pros and cons of genetic adhered to a surface and held together
testing to guide asthma therapy. and protected by a polymeric matrix
formed from secreted or extracellular
Explain the role of the cystic fibrosis
substances.
transmembrane conductance regulator
(CFTR) in the pathophysiology of BRONCHIECTASIS—localized and irreversible
pulmonary disease in cystic fibrosis. dilation of the bronchiole or bronchi due
to airway inflammation and obstruction.
Describe the various classes of
mutations found in CFTR and their CHEMOATTRACTANT—inorganic or organic
effect on CFTR expression and function. substance that induces or influences cell
migration.
Differentiate between therapeutic
agents that help manage cystic fibrosis FORCED EXPIRATORY VOLUME IN 1 SECOND
symptoms and disease-modifying agents (FEV1)—the volume of air that can be
that modulate CFTR activity. forced out in the first second after taking
a deep breath.
Describe how genetic testing is used
to identify cystic fibrosis patients who INTERPATIENT VARIABILITY—differences in
will respond to Kalydeco (ivacaftor) and response to drug(s) between patients.
Orkambi (lumacaftor/ivacaftor). ISOFLAVONE—naturally occurring organic
compounds derived almost exclusively
from the bean family.
Note: Significant contributions from Dr. Emily Weidman-Evans and Dr. W. Greg Leader as authors of earlier
versions of this chapter, published in the previous edition, are gratefully acknowledged.
327
NASAL POTENTIAL DIFFERENCE (NPD)—a measurement of the voltage across the nasal
epithelial cell determined by using a high impedance voltmeter between two electrodes,
one placed on the inside and one on the outside of the epithelia. The nasal potential
difference characterizes the voltage created by the secretion or absorption of chloride
ions that are abnormal in cystic fibrosis.

OPSONOPHAGOCYTOSIS—the process by which a foreign body (usually microbial) is
identified by complement or antigens to phagocytes, resulting in a more efficient energy
dependent endocytosis (or engulfment) of the microbe by the phagocytic cell (usually a
neutrophil or macrophage).
PEAK EXPIRATORY FLOW (PEF)—the maximum airflow during a forced expiration beginning
with the lungs fully inflated.
UBIQUITINATION—degradation of a protein caused by modification of the protein by covalent
attachment of ubiquitin, a regulatory protein, which directs the protein to proteasomes
for degradation.
INTRODUCTION
Asthma and cystic fibrosis (CF) are two very different conditions that have major effects
on the respiratory system. Asthma is characterized by chronic inflammation of the airways
and bronchoconstriction, resulting from both multiple genetic and environmental factors.
CF is a genetic condition that also affects other organ systems, caused by a mutation in
the gene encoding for a single ion channel. In light of the differences between the two
conditions, the pharmacogenomic research that has been applied to the conditions also
differs. In asthma, interpatient variability in response to the available treatments has led to
retrospective pharmacogenetic analyses to identify genetic differences affecting responses
to currently accepted therapy. The identification of genetic variants that can impact the
efficacy and safety of treatment might then allow for rational changes in drug therapy
before an exacerbation or adverse effect occurs. In contrast, with CF, genetic mutations
that affect the severity of the disease have been identified, and new therapies are being
prospectively investigated that will address the physiologic effects of the mutations at the
level of the affected protein. This research has resulted in the development and approval
of two mutation-specific therapies for CF. In spite of the differences in approach, the end
result of the pharmacogenomic research in both of these conditions is improved patient
health and clinical outcomes, which is the ultimate goal of all healthcare providers.
CASE STUDY—ASTHMA
R.P. is an 8-year-old African-American male admitted to the emergency room for acute
shortness of breath and wheezing. He is diagnosed with an asthma exacerbation, his third
episode in the past 12 months. His current medications include fluticasone/salmeterol
(dry powder inhaler) 500 mcg/50 mcg twice daily, montelukast 5 mg daily, and albuterol
(metered-dose inhaler) 1–2 puffs as needed. At his last admission, he was sent home
with a 10-day course of prednisone and started on montelukast; his dose of fluticasone/
salmeterol was increased from 250 mcg/50 mcg. In the 24 hours prior to this admission,
R.P.’s mother states that he probably used his albuterol 10 times, and that it “just didn’t
Chapter 11 • Respiratory Diseases 329
seem to work.” Furthermore, you note that his FEV1 is currently only 50% of expected,
and that it is not reversing with the administration of albuterol via nebulizer.
Questions
1. Polymorphisms found in the β2AR gene can influence how an individual responds to β2-
agonists. What will be the likely effect of a Thr164Ile polymorphism on how an individual
responds to salmeterol?
2. The presence of which single nucleotide polymorphism (SNP) in the promoter region of the
LTA4H gene could result in an asthma patient responding poorly to treatment with montelu-
kast?
CASE STUDY—CYSTIC FIBROSIS

A 5-month-old infant girl Aria is brought to an emergency room for a rapidly spreading
diaper rash and chronic cough according to her parents. The attending pediatrician
notices salt crystals on Aria’s skin and suspects that the baby could be suffering from CF.
A sweat test is ordered, and the results confirm that Aria suffers from CF. Genotyping tests
are conducted, and they show that Aria is homozygous for the ∆F508 mutation.
Questions
1. Is Aria likely to be prescribed ivacaftor for her condition?
2. If pancreatic enzyme supplementation is prescribed for Aria, will pancreatic enzyme replace-
ment help correct the underlying genetic defect that causes CF?
3. Can any therapeutic options available for Aria help modulate the activity of CFTR?
4. A new investigational drug that acts as a potentiator is shown to have positive effects in
cells and animals with Class III mutations only. Is Aria likely to benefit from this drug if it is
approved for treating CF?
ASTHMA
Introduction
Asthma is a chronic inflammatory airway disease. In the United States, asthma afflicts over
22 million people, of which more than 6 million are children, making it one of the most
common chronic childhood diseases.1 The natural history of asthma, though variable, is
classically characterized by wheezing, cough, and shortness of breath. Patients with asthma
typically exhibit episodic symptoms that occur following exposure to specific triggers like
allergens, exercise, or cold air.
The foundation of asthma pathophysiology is inflammation. In studies of allergic
asthma, inhaled allergens are encountered in the airways by antigen-presenting cells of
the immune system resulting in the production of cytokines and allergen-specific IgE,
which attaches to mast cells in the airways. Airway mast cells exposed to a trigger rapidly
degranulate releasing histamine, leukotrienes, and cytokines that cause bronchospasm,
edema, and airflow obstruction leading to asthma symptoms. Approximately 6 to 9 hours
after allergen incitement, in the late phase reaction, eosinophils, T lymphocytes, basophils,
neutrophils, and macrophages are activated resulting in enhanced inflammation and bron-
chial hyperresponsiveness.2
Asthma therapies seek to attenuate this inflammatory response and subsequent
physiologic events like bronchospasm. Inhaled corticosteroids (ICS), leukotriene (LT) modi-
fiers, and an anti-IgE antibody target the inflammatory response. Bronchospasm is targeted
by theophylline and β2-receptor agonists, which can be divided into short-acting agonists
(SABA) and long-acting agonists (LABA).
Pharmacotherapy of asthma is applied in a stepwise fashion and is predicated on a
patient’s age, disease severity, and response to therapy. At initial diagnosis, disease sever-
ity is assessed with spirometry, symptom frequency and timing, and interference with daily
activities. Appropriate therapy is initiated based on severity classification. As outlined in
Table 11-1, all patients with asthma receive a SABA and, in addition, may receive daily main-
tenance medication depending on severity classification. Therapy may be stepped up or
down based on clinical response.1
Genetics
It has long been established that there is a genetic link in asthma, involving interactions of
multiple genes and a variety of environmental factors, including allergens and infections.2
Independent genome-wide screens have found regions of linkage with asthma susceptibil-
ity and severity on chromosomes 5, 6, 11, 12, 13, 16, 17, and 19. It is only logical, then, to begin
pharmacogenomic research based on differences in these genetic regions.
In spite of the well-accepted treatment guidelines for this condition, a large degree
of interpatient variability in treatment response was noted in drug studies. Fortunately, we
do have a relatively good grasp on the precise mechanisms of a number of drugs that are
being used to treat asthma. Therefore, identifying potential genetic targets that involve
the receptors or mediators affected by the drugs is the most straightforward approach. To
date, specific genes have been identified that affect bronchodilation directly, the produc-
tion and action of drug receptors, and a number of steps in the inflammatory process
(e.g., T-cell proliferation and recruitment, macrophage recruitment, mast cell proliferation,
Table 11-1
2007 National Asthma Education and Prevention Program Asthma
Pharmacotherapy Recommendations1
Severity Treatment
Classification Step Preferred Therapy Alternative Therapy
Intermittent Step 1 SABA as needed

Mild persistent Step 2 Low-dose ICS Cromolyn, LTRA or theophylline
Moderate Step 3 Low-dose ICS + LABA or Low-dose ICS + either LTRA,
persistent medium-dose ICS theophylline or zileuton
Severe persistent Step 4 Medium-dose ICS + Medium-dose ICS + either LTRA,
LABA theophylline or zileuton
Severe persistent Step 5 High-dose ICS + LABA Consider omalizumab for patients
who have allergies

Severe persistent Step 6 High-dose ICS + LABA + Consider omalizumab for patients
oral corticosteroid who have allergies
ICS, inhaled corticosteroid; LABA, long-acting beta-2 receptor agonist; LTRA, leukotriene receptor antagonist; SABA,
short-acting beta-2 agonist.
antigen presentation, inhibition of cytokine activity).3 The genetic variances that impact
four common classes of drugs used in asthma are presented below.
β2-Agonists
Beta-2 adrenergic receptor (β2AR) agonists are agents that bind to and activate β2 adrener-
gic receptors. The β2AR is a 413 amino acid protein that belongs to the G-protein coupled
receptor family and consists of seven transmembrane segments.4 β2ARs are
widely distributed throughout the body including cardiac, respiratory, and endocrine tissues.
administered via the inhalational route to limit their effects to the respiratory tract.
Their activation in bronchial smooth muscle stimulates cyclic adenosine monophosphate
(cAMP) production, which leads to a decrease in intracellular calcium and subsequent
smooth muscle relaxation and bronchodilation.5,6
β2-agonists are divided into two distinct clinical categories based on duration of action
and consequent clinical use: short-acting and long-acting β2-agonists (SABAs and LABAs).
SABAs, including albuterol, levalbuterol, and pirbuterol, provide immediate, short-term
relief of asthma symptoms. SABAs are prescribed to all patients with asthma, regardless of
severity and used on an as-needed basis for control of acute symptoms. LABAs, including
salmeterol and formoterol, provide long-term relief of asthma symptoms. LABAs are dosed
twice daily as maintenance therapy and are indicated for patients with moderate-to-severe
disease in addition to an inhaled corticosteroid.1
Interpatient Variability
The most clinically relevant concern regarding β2-agonists is the increased risk of life-
threatening and fatal asthma exacerbations with LABA use, prompting a black box warning
from the U.S. Food and Drug Administration (FDA). Recognition of this risk occurred after
a postmarketing study of nearly 25,000 patients showed an increased absolute number
of deaths in patients receiving salmeterol versus albuterol in addition to usual asthma
therapy.1,7 These findings, however, were not statistically significant and follow-up studies
could not reproduce this risk.1,8,9 Therefore, a large trial assessing the safety of salmeterol,
the Salmeterol Multicenter Asthma Research Trial (SMART), was undertaken and an interim
analysis showed an increase in asthma-related death, particularly in African Americans.10 A
subsequent meta-analysis including these patients corroborated these findings.11 Despite
these findings, the benefit of LABA therapy, particularly with concurrent inhaled cortico-
steroid therapy, was deemed to merit continued recommendation of LABAs for asthma. It
has been postulated that concurrent therapy with an inhaled corticosteroid may obviate
this risk, although definitive studies are lacking.1
CLINICAL PEARL
It was theorized at the time that the disparity seen in African Americans
in the SMART trial was due to either more severe underlying disease
and/or the fact that the baseline ICS use in this subpopulation was
lower.
Genetic Variants Affecting Treatment

The high degree of interpatient variability in response to β2-agonists, as well as the possible
relationship between treatment and race exhibited in SMART, leads to questions regard-
ing a possible genetic cause of the variability. Table 11-2 shows the three nonsynonymous
coding region SNPs in the β2AR gene, which is located on chromosome 5q31-32, that could
contribute to some of the demonstrated differences in individual responses to β2-agonists

in asthma and other respiratory diseases.12,13 These SNPs are substitutions of glycine for
arginine at the 16th amino acid (Arg16Gly), of glutamate for glutamine at the 27th amino
acid (Gln27Glu), and of isoleucine for threonine at the 164th amino acid (Thr164Ile). In addi-
tion to these three SNPs on the ADRB2 gene, polymorphisms on a few other genes have
also been identified that may influence the response β2-agonists. All of these polymor-
phisms occur with the same frequency in both asthmatic and nonasthmatic patients, and
thus none of them can be used for predicting risk for asthma.12,14 Genome-wide association
studies (GWAS) have also not identified an association between polymorphisms on the
β2AR gene and risk for asthma.15
ARG16GLY (285 A>G). Three meta-analyses have shown that this relatively common poly-
morphism is not associated with asthma diagnosis. However, some evidence suggests that
this polymorphism may be associated with nocturnal asthma and a more severe asthma
phenotype.17,18 Several studies have suggested an association between the Arg16Gly poly-
morphism and response to SABA.19-23 Individuals homozygous for Arg-16 have been shown
to have a greater and more rapid response to albuterol treatment than individuals who are
homozygous for Gly-16.19,20 Other studies have shown that patients homozygous for Arg-16
who receive a regularly scheduled SABA long-term experience decreases in peak expira-
tory flow readings (PEFR) and forced expiratory volume (FEV1). This patient population also
exhibited a worsening of daily symptoms, increases in the amount of rescue medication
required, and an increased number of exacerbations.21-23 In clinical practice, patients who
require SABAs consistently throughout the day would generally have additions, intensifi-
cations, or modifications to their maintenance asthma therapy medications. These studies
suggest that regular SABA therapy may necessitate more timely maintenance medication
adjustment in patients who are homozygotes of Arg-16.
Table 11-2
Significant B2AR SNPs
Frequency of
SNP (Homozygotes)14,16,17,130,131 Response to b2–Agonists12,14,26
Arg16Gly Overall: 54.8–60.4 Fewer exacerbations and reduced need for rescue
White: 60.7 medication for patients receiving regular SABA
Asian: 46.0 therapy
Black: 45.0
Gln27Glu Overall: 24.7–47.2 Reduced risk of developing tachyphylaxis
White: 36.1
Asian: 13.6
Black: 18.0
Thr164Ile Overall: 5.0 Decreased binding of drug to receptor
→ possible decrease in drug activity
SABA = short-acting beta-2 agonist; SNPs, single nucleotide polymorphisms.
Due to the relationship between genotype and response to SABAs, as well as the
controversy surrounding the use of LABAs, several studies have been conducted to deter-
mine the relationship between LABAs and the Gly-16 SNP. Unfortunately, the results of
these studies are conflicting. Bleeker and colleagues conducted a pharmacogenetic analy-
sis of two trials that used either salmeterol or formoterol, which showed no difference
between genotypes in exacerbations, spirometric measures (PEFR and FEV1), or adverse

effects.24 All of the subjects in this analysis received ICS in addition to a LABA. Conversely,
a similar analysis conducted by Weschler and colleagues of two large studies utilizing
salmeterol showed that those subjects who were homozygous for the wild-type receptor
had lower spirometric measures as well as increased symptoms and albuterol use regard-
less of concurrent ICS use.25 Those homozygous for Gly-16 showed an improvement in
spirometric and symptomatic measures during the study periods. A retrospective study in
546 children and young asthmatics showed a two-fold increase in exacerbations in those
with Arg-16. In patients receiving salmeterol, the risk increased to over three times that of
those with Gly-16.
CLINICAL PEARL
An association has been established between the Arg16Gly polymorphism
and response to β2-agonists. However, the effects of this polymorphism
depend on whether LABAs or SABAs are used, and whether SABAs are
used long-term on a regular or an as-needed basis.
GLN27GLU (318C>G). Some studies have shown an association between the Glu-27 poly-
morphism and a protective effect against bronchial hyperresponsiveness and severity of
asthma.17 Glu-27 may also afford protection against tachyphylaxis to β2-agonist therapy.
A study involving 80 Spanish asthmatic patients showed that patients with Gln-27 were
more likely to experience tachyphylaxis than patients with the Glu-27 polymorphism.26
THR164ILE (730C>T). The Ile-164 is a rare polymorphism compared to both Arg16Gly and
Gln27Glu. In vitro studies have suggested that the activity of the receptors displaying this
polymorphism is about half that of wild-type receptors.14 The binding affinities for albuter-
ol, terbutaline, formoterol, and salmeterol were all found to be lower (K[i]s being 1.2–3.0
fold higher) for Thr164Ile when studied in transfected fibroblasts.27 Furthermore, studies in
transgenic mice showed a decreased cardiac response to isoproterenol when compared to
mice with the wild-type receptor, leading to the hypothesis that those with this polymor-
phism could also have a lower baseline bronchodilator response to inhaled β2-agonists.
OTHER VARIANTS AFFECTING RESPONSE TO β2-AGONISTS. Candidate gene studies and

GWAS have resulted in identification of polymorphisms on other genes within B2AR signal-
ing pathways that influence response to β2-agonists. Polymorphisms within the CRHR2,
ADCY9, ARG1, ARG2, NOS3, THRB, SLC24A4, SLC22A15, and SPATS2L genes have been
associated with alterations in the therapeutic response to β2-agonists.28 In a study involv-

ing asthmatic children, it was found that the Ile772Met polymorphism on the ADCY9 gene
contributed to an enhanced bronchodilator response to albuterol in children that were also
receiving budesonide.29 Polymorphisms within ARG1, ARG2, and NOS3 can influence the
production of the endogenous bronchodilator nitric oxide and thereby influence response
to β2-agonists. GWAS have identified a polymorphism in the promoter region of SPATS2L.

This study also highlights the potential of GWAS in identifying novel genes that may be
involved in regulating the response to β2-agonists.30 Polymorphisms found in genes coding
for the solute carrier group of ion transporters have also been found to influence bron-
chodilator response.31 These studies highlight the polygenic nature of asthma in that a
variety of different polymorphisms found in different genes can influence the response to
β2-agonists. Additional studies will be required to better define the role and importance of
each of these genes as well as the combined effect of polymorphisms in these genes on
bronchodilator response.
Inhaled Corticosteroids
ICS are the most potent and effective medications for long-term control of asthma
symptoms. As such, the majority of patients with asthma receive an ICS as part of their
treatment regimen.1 The ICS exert their pharmacologic activity by binding to intracellular
glucocorticoid (GC) receptors. Activation of GC receptors alters gene transcription and may
also affect post-translational events resulting in down-regulation of pro-inflammatory
mediators and up-regulation of anti-inflammatory mediators.32 As with β2-agonists, ICS
are administered via inhalation to minimize their systemic effects.33 Available ICS agents
include beclomethasone dipropionate, budesonide, ciclesonide, flunisolide, fluticasone
propionate, and mometasone furoate. Concerns regarding adverse effects of ICS, includ-
ing reduced linear growth in children, have been substantiated. However, these risks are
balanced by the efficacy of ICS in the control of asthma, provided the lowest effective
doses are used.1
Interpatient variability in response to corticosteroids has been demonstrated in a number
of disease states including inflammatory bowel disease (IBD), psoriasis, nephritic syndrome,
and various cancers. The term glucocorticoid resistance has been coined to define this vari-
ability. In asthma, it has been shown that 5-10% of patients will have a reduced response
to ICS.13 This number increases to 35% in those with severe disease and to nearly 40% in
black patients with asthma. A 12-week study showed that 22% of subjects experienced
a decrease in their FEV1 of 5% or more with inhaled beclomethasone therapy, while 10%
improved by more than 40%.34 Another study showed 38% of subjects respond minimally
to beclomethasone or fluticasone over 24 weeks of therapy.35 In both of these studies, the
average response was approximately a 10% increase in FEV1.
There is less evidence regarding variability in the frequency and severity of adverse
effects associated with ICS, most likely because systemic adverse effects are relatively
rare. However, one study showed a significant relationship between doses of inhaled triam-
cinolone, a highly bioavailable ICS, and decreases in bone density. Analysis of individual
response showed a large amount of variability in the degree of bone loss experienced
between patients who were independent of the number of puffs per day. This finding
suggests that factors other than dose were contributing to this adverse effect.36

Due to the broad effects of corticosteroids, it is more difficult to pinpoint appropriate
genetic targets for focus in pharmacogenomic studies involving ICS. Studies involving
genes that code for several key receptors, however, have shown promise in identifying
hypo- and hyper-responders to ICS.
CORTICOTROPIN-RELEASING HORMONE RECEPTOR-1. The gene that encodes for

corticotropin-releasing hormone receptor-1 (CRHR1) was chosen as a potential target for
research because it is the primary receptor mediating the release of adrenocorticotropic
hormone (ACTH), a major regulator of glucocorticoid and catecholamine synthesis.37,38
This gene, located on chromosome 17q21-22, has been implicated in the pathogenesis of
inflammatory diseases and is located in a region linked to asthma in some genome-wide

screens. It is hypothesized that alterations in the expression or function of this receptor as
a result of genetic variation can lead to decreased ACTH release. This, in turn, will decrease
cortisol release in response to inflammation and possibly upregulate glucocorticoid recep-
tors. When patients with this variant are given exogenous corticosteroids, they will likely
have a more pronounced therapeutic response.38
Variations in CRHR1 have been associated with a 2–4 fold increase in response to ICS
when compared to those without the variations.38 One SNP in particular, rs242941, has been
linked to a significant increase in ICS response. This SNP, however, is intronic so it does
not have any effect on the CRHR1 unless it is present in a specific haplotype, designated
GAT. This haplotype is present in 27% of whites. A retrospective pharmacogenetic analysis
showed that, while all patients on ICS had improved FEV1 after 8 weeks of therapy, GAT
homozygotes experienced a two- to three-fold increase in that improvement. Heterozygotes’
response fell in between that of homozygotes and those without the GAT haplotype.
T-BOX EXPRESSED IN T CELLS (TRANSCRIPTION FACTOR T-BET). Transcription factor

T-bet plays an important role in the inflammatory process. It influences the development
of naïve T-lymphocytes, induces the production of interferon-gamma, and represses the
production of interleukins 4 and 5.39 The gene that encodes for T-bet, TBX21, is also locat-
ed on chromosome 17q21.27 It has been shown that asthmatic patients have a decreased
number of T-cells that express TBX21, and that deletion of this gene in mice results in
airway hyperresponsiveness. Furthermore, several mutations of TBX21 have been associ-
ated with both asthma susceptibility and severity in humans.40 The SNP rs2240017, which
results in the substitution of glutamine for histidine at the 33rd amino acid (Gln-33), has
been linked to a significant increase in response to ICS.26 A retrospective pharmacogenetic
analysis in 701 children showed that those with the SNP had about a three-fold improve-
ment in their PC20 (provocative concentration of histamine causing a 20% fall in FEV1; a
measure of airway responsiveness) when compared to those with the wild-type gene. In
fact, the mean PC20 for those subjects on ICS with the SNP was 27.7 mg/mL. Anything
>25 mg/mL indicates “normal” airway responsiveness. FEV1 increased to a similar extent in
all subjects receiving ICS, regardless of the presence of the SNP. The subjects included in
this analysis, about 4.5% of the study population, were all heterozygotes for the SNP. The
frequency of homozygosity for this beneficial mutation varies greatly based on race: black
0.4% to 2%; white 2.7% to 3.0%; Hispanic 7.1%; and Korean 11.8%.40
GLUCOCORTICOID RECEPTOR GENE (GR/NR3C1). Another logical genetic target would be

the gene that encodes for the glucocorticoid receptor (GR) itself. This receptor has two
naturally occurring isoforms: GRα, which is functional and involved in regulating proinflam-
matory mediators, and GRβ, which has no hormone-binding activity and is actually consid-
ered an endogenous inhibitor of actions mediated by the GR.13 An imbalance of either of
these isoforms due to a genetic anomaly can lead to glucocorticoid resistance. It is known
that 95% of patients with glucocorticoid resistance have type I resistance, which is associ-
ated with an increased expression of GRβ.41,42 The patient will often present with severe
systemic side effects but will have minimal therapeutic effects. Conversely, type II gluco-
corticoid resistance is associated with a decrease in GRα, which results in a generalized
primary cortisol resistance. These patients typically do not experience either therapeutic
or adverse effects when administered ICS. However, the specific mutation(s) that result(s)
in this imbalance have not yet been identified.
OTHER VARIANTS AFFECTING RESPONSE TO ICS. Due to the broad effects of cortico-
steroids and their interactions with other pathways, additional polymorphisms have also
been identified through either candidate gene or GWAS. The heat shock-organizing protein
plays an important role in GR hetero-complex association. SNPs on STIP1, the gene encod-
ing the heat shock-organizing protein, have been shown to correlate with improvement
in lung function following treatment with corticosteroids.43 CYP3A4*22 allele has been
shown to be associated with improved asthma control in children receiving fluticasone
propionate (FP).44 This study suggests that decreased CYP3A4 activity may increase expo-
sure to FP and might help in predicting response to FP. GWAS have identified associations
between polymorphisms in novel genes such as the T gene and the GLCCI1 gene and lung
function response to corticosteroid treatment.28
Leukotriene Modifiers
The term LT modifier represents two classes of drugs: LT receptor antagonists (LTRAs),
which include montelukast and zafirlukast, and the 5-lipoxygenase (ALOX5) inhibitor zileu-
ton. LTRAs directly antagonize the leukotriene receptor CysLT1 preventing endogenous LT
binding, while zileuton inhibits the 5-lipoxygenase enzyme responsible for synthesizing
LTs from arachidonic acid. In both cases, LT-mediated signaling is diminished resulting in
reduced mucus secretion, airway edema, and bronchospasm.45
LT modifiers are clinically effective for the treatment of asthma; however, their efficacy
is considered to be inferior to that of ICS and LABAs (in combination with an ICS). Thus,
they are recommended as non-preferred alternative therapy (see Table 11-1). However, a
distinguishing, potentially attractive, characteristic of these medications compared with
those previously discussed is that LT modifiers are administered via the oral, not inhala-
tional, route.1
Although not as widely studied as β2-agonists and ICS, one study showed the distribution
of responses with montelukast to be similar to that seen with inhaled beclomethasone.
Of those receiving montelukast, 42% had an increase in FEV1 of more than 10%, while 34%
of patients had no improvement or a worsening in FEV1.34 Because LT appears linked to
asthma susceptibility and severity, it is theorized that variations in therapeutic response to
this class of drugs are at least partially mediated by the concentration of LT.16 This suggests
that asthma may be mediated by factors other than LT and that LT modifiers will be less
effective in those patients with lower LT concentrations. Conversely, higher concentrations
of these mediators could indicate a better response to LT modifiers.
Serious adverse effects with the LTRAs are relatively rare, but zileuton has been linked
to hepatotoxicity and even rare causes of hepatic failure. One safety surveillance study
showed that 4.4% of subjects receiving zileuton (600 mg 4 times daily) had elevations
in alanine aminotransferase to greater than 3 times the upper limit of normal; 1.3% had
elevations greater than 8 times the upper limit.46 Women were more likely than men to
experience these significant elevations, as were the elderly. Although no genetic studies
have been performed, it is possible that the genetic mutations discussed below, or others,
could contribute to this adverse effect.
Studies have been conducted to determine if the variability in response to LT modifier

therapeutics, as described above, is a result of polymorphisms in genes involved in the
production of proteins that affect the pharmacodynamics or pharmacokinetics of LT modi-
fiers. Polymorphisms in genes encoding proteins that bind to LTs (CYSLTR1 and CYSLTR2),
regulate their production (ALOX5, LTC4S, and LTA4H), or influence disposition of LT modi-
fier drugs (MRP1 and SLCO2B1) have been associated with altered response to LT modifiers.
Studies involving each of these classes of proteins are summarized below.
ARACHIDONATE 5-LIPOXYGENASE GENE (ALOX5). ALOX5 is involved very early in the

LT synthesis process, converting arachidonic acid to 5-hydroperoxyeicosatetraenoic acid
(5-HPETE) and then converting 5-HPETE to LTA4.47 The promoter gene for this enzyme,
located on chromosome 10q11, contains a regulatory region with five tandem repeats of
the binding motif.48,49 Table 11-3 shows the frequencies of the variants of this and other
enzyme promoter genes in the LT synthesis pathway and highlights the racial differences
that exist.50
It has been theorized that certain variant alleles result in a decrease in LT production;
therefore, it is likely the airway obstruction experienced in patients with these altered
promoter genes is caused by mechanisms other than LT. Patients with these variants will
be less responsive to LT modifiers.16 This theory is supported by two studies that showed
significant increases in FEV1 (9.1-18.8%) in subjects homozygous for the wild-type gene,
compared to a small decrease (1.1-2.3%) in those with mutations of the gene.16,51 Another
Table 11-3
Frequency and Effects of Variant Promoter Genes in the LT Synthesis
Pathway
Variant Promoter Homozygotes Heterozygotes Effect on Effect on
Gene (%)50 (%)50 FEV 16,50,51,54 Exacerbations50
1
ALOX5 Overall: <1 Overall: 36 +/– –

White: 3.6 White: 33
Black: 24.6 Black: 57
LTC4 synthase Overall: 11 Overall: 38 + –
LTA4 hydrolase Overall: 8 Overall: 44 NE +

ALOX5, 5-lipoxygenase; FEV1, forced expiratory volume in 1 second; LT, leukotriene; LTA4, leukotriene A4;
LTC4, leukotriene C4.
+, increased.
–, decreased.
+/–, conflicting evidence. NE, not evaluated.
study showed that those homozygous for the polymorphism had no change in FEV1,
β2-agonist use, or number of asthma exacerbations after 12 weeks of montelukast treat-
ment, while those homozygous and heterozygous for the wild-type gene had significant
improvements in each of these parameters.52 Other research, however, has shown oppo-
site outcomes. One study that compared percent change in FEV1 and exacerbations with
montelukast showed that homozygotes for the variant gene had a fifteen-fold greater
increase in FEV1 when compared to homozygotes for the wild-type gene.50 In addition,
those with at least one mutant allele had a 73% reduction in exacerbations (at least one
during the 6-month study period). Obviously, further studies are necessary to clarify this
apparent discrepancy.
LT C4 SYNTHASE PROMOTER GENE (LTC4S). LTC4 synthase is a late-stage enzyme in the

LT production process that is responsible for converting LTA4 to LTC4.53 One specific variant
in the promoter gene for this enzyme, SNP A(−444)C (rs730012) located on chromosome
5q35, has been identified in 32% of non-asthmatic patients, 56% of those with severe
asthma, and 76% of those with aspirin intolerant asthma (AIA).54 This variant enhances the
expression of LTC4 synthase, thus increasing LTC4 production. It is hypothesized that since
LTC4 and its products contribute so significantly to asthma physiology, those with this SNP
would be good responders to LT modifiers. This theory is supported by a small study that
involved administering zafirlukast for 2 weeks to 23 subjects with severe asthma and 31
non-asthmatic subjects. Those subjects with at least one variant allele for the promoter
gene had a 9% increase in FEV1, compared to a 12% decrease in those with the wild-type
gene. Furthermore, subjects homozygous or heterozygous for the mutant gene had an 80%
reduction in exacerbations while taking montelukast.50 Interestingly, those with at least
one variant allele receiving placebo also had a significant decrease in exacerbations (69%)
when compared to those homozygous for the wild-type gene.
LEUKOTRIENE A4 HYDROLASE PROMOTER GENE (LTA4H). The LTA4H enzyme is respon-

sible for converting LTA4 to LTB4. The presence of one SNP (rs2660845) in its promoter
gene, located on chromosome 12q22, has been linked with up to a four-fold increase in the
risk of patients having at least one exacerbation when taking montelukast.50 This differ-
ence was not observed between genotypes in those receiving placebo. The effect on LTA4
hydrolase activity as a result of this SNP is unknown. It is possible that the increase in exac-
erbations results from a decrease in the activity of LTA4H, which leads to more LTA4 being
converted to LTC4 and increased asthma symptoms. Conversely, the variant could increase
the activity of LTA4H, thus decreasing the production of the cys-LT. This would result in a
decreased therapeutic response to LT modifiers.
CYSTEINYL LT RECEPTOR GENES (CYSLTR1 AND CYSLTR2). Cysteinly LTs bind to the cyste-
ine LT receptors and mediate the effects correlated with the pathophysiology of asthma.
Montelukast, pranlukast, and zafirlukast are LTRAs that bind to the CYSLTR1 receptor
and inhibit the effects of cysteinly LTs mediated through this receptor. A study of 89 AIA
patients found that a −634 C>T promoter polymorphism can help predict LTRA require-
ments in the management of AIA.55 A study analyzing the association between response
to montelukast and polymorphisms in 10 genes found that polymorphisms in CYSLTR2
resulted in a significantly higher improvement in morning peak expiratory flow. These
polymorphisms likely predispose individuals to high cysteinyl LT concentrations, thereby
causing an enhanced response to LTRA therapy.56
SOLUTE CARRIER ORGANIC ANION TRANSPORTER FAMILY MEMBER 2B1 GENE (SLCO2B1).
The ability of polymorphisms, in genes encoding proteins that affect LTRA uptake and
concentrations, have been investigated. A coding, non-synonymous polymorphism in
SLCO2B1 (c. 935G>A) was shown to result in significantly lower morning concentrations of
montelukast in patients receiving a dose the previous evening (n = 80 subjects).57 Another
study in a relatively smaller population of 16 volunteers showed conflicting results and

found no effect of the same polymorphism on montelukast plasma concentrations.58
A third study involving 24 subjects also found no association between the c.935G>A
polymorphism and plasma levels of montelukast.59 The conflicting results suggest that
SLCO2B1 polymorphisms may only have a minor role in affecting montelukast plasma
levels. Additional studies with larger sample sizes would be needed to clearly define the
role of SLCO2B1 polymorphisms in influencing plasma concentrations and response to
montelukast.
Theophylline
Theophylline is a methylxanthine that exerts its pharmacologic activity by non-selectively
inhibiting phosphodiesterase, thereby increasing intracellular cAMP concentrations in vari-
ous tissues. In respiratory tissue, this causes bronchodilation. In addition, theophylline may
also have anti-inflammatory properties.60 Clinical use of theophylline, however, is limited
by its extra-respiratory effects including cardiac, gastrointestinal, and central nervous
system stimulation. Theophylline has a narrow therapeutic index necessitating routine
monitoring of serum drug concentrations and dose adjustments when necessary to maxi-
mize efficacy and minimize risk of toxicity.1
Due to the narrow therapeutic index of theophylline and potential for toxicity, pharma-
cogenomic studies have focused on genes that ultimately affect serum drug levels.
CYTOCHROME P450 1A2 GENE (CYP1A2). Theophylline is metabolized primarily via

demethylation and hydroxylation by hepatic cytochromes P450 (CYP) 1A2, CYP3A3, and
CYP2E1. A polymorphism in the 5’-flanking region of the CYP1A2 gene at position −2964
(G>A) has been associated with reduced theophylline clearance.61 This discovery followed
initial findings that the same polymorphism was associated with differences in metabolism
of a similar methylxanthine, caffeine.62 Based on these data, it has been postulated that
patients with the −2964G>A polymorphism in the CYP1A2 gene may require lower doses of
theophylline to reduce risk of toxicity.61
Clinical Implications and Testing

The data for each of the genetic variants discussed above demonstrates that response to
asthma medication is affected by the presence or absence of certain polymorphisms. Knowl-
edge of specific genetic variants could help healthcare providers individualize therapy and
make adjustments before a patient has an exacerbation or adverse response to therapy.
There are tests currently available for identifying several of the polymorphisms discussed
above, but they are not commonly used in practice. The debate continues regarding the
clinical, legal, and ethical implications surrounding widespread genetic testing. The impact
of pharmacogenomic testing on clinical decision making, patient privacy, and costs of
healthcare is discussed throughout this book. For example, there are proven racial differ-
ences in the frequencies of many of the mutations covered here. African Americans are
less likely than whites to have the beneficial variants in β2AR and CRHR1, but they are more
likely to have the potentially positive mutation in ALOX5. Could this finding lead to genetic
testing prompted by racial background, prior to initiating or altering therapy, or in obtaining
healthcare insurance?
There is some overlap in genetic polymorphisms with regards to the pathophysiol-
ogy and susceptibility of asthma with that of both chronic obstructive pulmonary disease
(COPD) and allergic rhinitis/atopic dermatitis.63-66 This genetic overlap is important because
of the many similarities in the treatment of these conditions. As per the 2007 Global Initia-
tive for Chronic Obstructive Lung Disease guidelines, β2-agonists and corticosteroids are
indicated for a large proportion of patients with COPD. Similarly, the 2008 update of the
World Health Organization’s Allergic Rhinitis and its Impact on Asthma guidelines state
intranasal corticosteroids are the most effective treatment for allergic rhinitis, and the LT
modifier montelukast should be considered in all patients over the age of 6.67,68 Although
no definitive studies have been done, given these similarities in etiology and treatment, it
is likely that the genetic variants identified as having a potential effect on asthma therapy
will have similar effects in COPD and allergic rhinitis.
The pharmacogenomic data related to β2-agonists, ICS, LT modifiers, and theophyl-
line make it clear that genotyping could help to guide asthma therapy. However, the lack
of long-term outcomes studies, the small sample size in many of the studies, and limited
replication of the studies, make it hard to advocate for universal genetic screening of
asthmatic patients.
CYSTIC FIBROSIS
Introduction
CF is a hereditary disorder that influences multiple organs and organ systems. CF affects
approximately 1 in 3,500 newborns and is the most common autosomal recessive genetic
disorder in the United States. In the 1970s, the median survival age of a patient with CF
in the United States was in the mid-teens.69 However, due to significant advances in the
diagnosis and treatment of CF, in 2013 the median predicted survival age has increased to
>40 years.70
CF is caused by mutations in the CFTR gene on chromosome 7 responsible for encod-
ing the CFTR protein. The CFTR protein performs various physiologic functions, including
regulation of transmembrane chloride ion transport. On a cellular level, this mutation and
the resulting absence of CFTR protein activity is associated with exocrine gland dysfunc-
tion causing viscous secretions in organs including the pancreas, intestine, and lungs.71 This
widespread pathology causes a diverse clinical presentation. Pulmonary manifestations may
include cough, shortness of breath, and wheezing and are sometimes associated with recur-
rent respiratory tract infections. Gastrointestinal and pancreatic manifestations may include
steatorrhea and malnutrition. Neonates classically present with meconium ileus. CF may
also affect the reproductive system causing obstructive azoospermia in males and reduced
fertility in females.69,72 The diagnosis of CF is based on clinical symptoms or a sibling history

of CF that is confirmed through documentation of CFTR dysfunction (Table 11-4).
When CF was initially described, survival past childhood was uncommon. Advances in
the diagnosis of CF and the treatment of complications related to CF have dramatically
increased the lifespan of patients with the disorder; however, effective methods to correct
Table 11-4
Diagnosis of CF69
a
One or more clinical features characteristic of CF, or
History of CF in a sibling, or
Positive newborn screening test (elevated immunoreactive trypsinogen)

AND
Laboratory evidence of an abnormality in the CFTR gene or protein (positive sweat chloride or
nasal potential tests), or
Identification of CF causing CFTR mutations on both copies of the CFTR gene
a
Chronic sinopulmonary disease manifested by colonization/infection with typical CF pathogens, chronic cough
and sputum production, persistent abnormalities on chest radiograph such as atelectasis, bronchiectasis,
infiltrates, and hyperinflation, airway obstruction with wheezing and air trapping, nasal polyps in conjunction
with radiographic or tomographic abnormalities of the paranasal sinuses; gastric and nutritional abnormalities
including meconium ileus, distal intestinal obstruction syndrome, rectal prolapse, pancreatic insufficiency,
recurrent acute pancreatitis, chronic pancreatitis, prolonged neonatal jaundice, chronic hepatic disease manifested
by clinical or histological evidence of focal biliary cirrhosis or multilobar cirrhosis obstructive, and nutritional
deficiencies resulting in failure to thrive, protein malnutrition with edema or complications secondary to fat-soluble
vitamin deficiency; salt-wasting syndromes such as acute salt depletion or chronic metabolic alkalosis, or genetic
abnormalities causing obstructive azoospermia.
CF, cystic fibrosis; CFTR, cystic fibrosis transmembrane conductance regulator.
the genetic defect, and thus affect a cure, have until recently remained elusive. Despite the
fact that CF has been commonly referred to as a pediatric disorder, more and more prac-
titioners who usually deal with adult populations are now caring for CF patients. In 1970
the median predicted survival for a CF patient was 16 years; however, in 2006, the median
predicted survival had increased to 36.9 years.73 Adult CF patients are more common today;
in 2006, 44.6% of the patients in the Cystic Fibrosis Foundation Registry were 18 years of
age or older.73
In 2006, 362 CF deaths were recorded in the CF database sponsored by the Cystic
Fibrosis Foundation.74 Over 90% of all CF deaths are related to pulmonary complications
of the disease.75
Pathophysiology
CF is caused by a genetic defect in the production of the CFTR protein, which is a member
of the adenosine triphosphate (ATP)-binding cassette transporter ATPases. CFTR is a 1480
amino acid glycoprotein that functions as a cAMP-regulated chloride channel. The protein
consists of two transmembrane domains, which contribute to the ion pore with two cyto-
plasmic nucleotide-binding domains (NBDs) linked by a cytosolic regulatory domain. Chlo-
ride transport is controlled by cAMP-dependent protein kinase A (PKA) phosphorylation of
the R-domain with ATP binding and hydrolysis at the NBDs.76,77
CFTR is expressed on the apical plasma membrane of the epithelial cell, where it is
part of a multiprotein assembly in close proximity to a number of other ion channels and
membrane receptors.76 In addition to acting as a gated-chloride channel, CFTR appears to

play a role in the regulation of other apical ion transport processes, including the epithe-
lial sodium channel (ENaC) whose activity is inhibited by CFTR in the normal airway.78,79
In addition to CFTR, luminal chloride secretion in airway epithelial cells can also occur
through alternative chloride channels such as those activated by P2Y2 receptors or intra-
cellular calcium. These fluid and ion secretion and absorption processes are responsible
for maintaining appropriate airway surface liquid (ASL) hydration.
In the airways, the most important factor in mucus clearance is hydration of the airway
surface. The primary physiologic defect in the airways in CF is a thick, tenacious mucus that
is poorly cleared due to dehydration of the ASL. The ASL consists of fluid in the mucus
layer and the periciliary liquid (PCL). The mucus layer of the airway lies over the PCL,
which is a low viscosity polyanionic fluid or gel layer that facilitates ciliary movement.75
Thus, the PCL serves as a lubricant between the mucus layer and the airway allowing cili-
ary movement of the mucus layer. ASL volume regulation is not completely understood;
however, in the normal lung it appears that chloride secretion by the CFTR and alterna-
tive chloride channels, and sodium absorption via ENaC, work in conjunction to maintain
ASL. The balance between sodium absorption and chloride secretion maintains the proper
hydration and tonicity of secretions. In CF, the CFTR is either missing or nonfunctional; thus,
this balance is disrupted. In addition to the loss of CFTR chloride secretion, ENaC-related
sodium absorption appears to be increased leading to dehydration of the ASL, collapse
of the PCL, and the inability of airway cilia to clear mucus.75,79 Epithelial cells in CF retain
their ability to secrete chloride via non-CFTR chloride channels; however, this function is
not sufficient to maintain ASL homeostasis.
The dehydration of the mucus and the loss of the PCL results in adhesion of mucus
to the airway surface and the formation of mucus plaques. Adhesive secretions obstruct
submucosal glands and the distal airways.77 Mucus stasis and airway obstruction occur. As
a result of mucus stasis, inhaled bacteria are not efficiently cleared, and bacterial coloniza-
tion develops resulting in a neutrophilic inflammation. Mucus stasis may prevent neutrophil
migration and the diffusion of antimicrobial substances produced in the lungs, decreasing
the efficiency of the immune reaction. In addition, neutrophil elastase and other related
proteases damage structural proteins in the lungs leading to bronchiectasis and decreased
opsonophagocytosis, which perpetuates chronic infection.80 Furthermore, deoxyribonucleic
acid (DNA) and other breakdown products released upon neutrophil death contribute to
decreased mucus viscoelasticity. Thick, viscous mucus may also create a hypoxic environ-
ment that is favorable for bacterial growth and biofilm development. In addition, evidence
exists demonstrating impaired production or trafficking of CFTR may cause activation
of nuclear factor-kappa B resulting in increased interleukin-8, the principal neutrophil
chemoattractant in the lung; therefore, airway inflammation may also occur independent
of infection in the CF lung.80,81
The genetic defect in CF results in episodic exacerbations of acute viral and bacte-
rial pulmonary infections, leading to further airway structural damage, bronchiectasis, air
trapping, and hyperinflation of the lungs. Increased airway obstruction and a progressive
decline in lung function in turn lead to chronic hypoxia.
Genetic Defects
The expression and activation of CFTR is complex. Briefly, messenger ribonucleic acid
(mRNA) transcripted from DNA in the nucleus migrates to cytoplasmic ribosomes where
amino acid translation occurs. Protein synthesis then occurs at the membrane of the endo-
plasmic reticulum followed by glycosylation and folding of the protein in the Golgi body.
In normal cells, appropriately folded CFTR is then trafficked to the cell surface. Truncated,
unstable, or misfolded protein is degraded by the endoplasmic reticulum.82
Mutations of the CFTR can be classified based on how they disrupt CFTR function (Table
11-5). Class I mutations are splicing or nonsense mutations and are caused by a premature
stop codon, which results in an unstable mRNA or a shortened or unstable protein that
is degraded in the cell. Class I mutations result in the CFTR not being expressed at the
cell membrane. Molecules that suppress premature stop codons and allow translation
to continue may be used to increase apical membrane protein expression in this class of
mutations. This type of mutation is associated with 5-10% of CFTR mutations and results in
a severe phenotype exhibiting both pulmonary disease and pancreatic insufficiency.83 The
W1282X mutation is the most common mutation among CF patients of Ashkenazi Jewish
descent where it is seen in up to 60% of alleles; however, it is rare in other populations.84
Class II mutations are mutations caused by impaired processing of the CFTR protein,
which results in degradation of CFTR by the endoplasmic reticulum degradation processes,
and a lack of functional CFTR expression at the cell membrane. Class II mutations are
similar to Class I mutations; however, they result in a more severe form of CF that includes
respiratory disease and pancreatic insufficiency. The most common Class II mutation,
∆F508, is caused by the deletion of a phenylalanine at the 508 amino acid position. This
mutation is seen in 70% of defective alleles and 90% of CF patients in the United States.77
Because of its frequency, most research has been directed at understanding and correcting
this defect. It is thought that calnexin and the heat shock protein HSC70 are involved in
the trafficking of the protein as it matures. They may associate with the protein and assist
it in the folding process before releasing it to the Golgi body for glycosylation. Because the
∆F508 mutation impedes proper folding, a prolonged association with calnexin and HSC70
results in ubiquitination and degradation in the endoplasmic reticulum.82,85 It is estimated
that greater than 99% of ∆F508 protein is degraded by this mechanism.82 When expressed
at the membrane, the ∆F508 mutation retains chloride-channel activity, but it does not
Table 11-5
Classification of CFTR Dysfunction
Class I Class II Class III Class IV Class V Class VI
Functional Premature Amino acid Amino acid Amino acid Promoter or Truncated at
defect stop codon deletion substitu- substitu- splicing C-terminus
tion tion errors
Results on Shortened Misfolding Dysregula- Altered Decreased Unstable at
CFTR dysfunc- of protein tion of channel CFTR plasma
expression tional that is protein at archi- expres- membrane
and/or protein degraded cellular tecture sion at
function that is in ER membrane leading to the apical
degraded decreased membrane
in ER conduc- with
tance or differing
channel levels of
gating activity
Representa- G542X, ∆F508, G551D, R117H, D565G, Q1412X
tive geno- R553X, N1303K, G1349D R334W, G576A
types W1282X G85E R234P,
D1152H
CFTR, cystic fibrosis transmembrane conductance regulator; ER, endoplasmic reticulum.
function as well as the wild-type channel.77 This decrease in relative function may be
related to a decreased open time for the channel.86
Class III mutations are caused by full-length proteins that are properly processed
and trafficked to the membrane but have significantly decreased chloride ion transport
capabilities. The lack of channel activity appears to be caused by the protein’s resistance
to phosphorylation, ATP binding, or hydrolysis. The most common Class III mutation is a
glycine to aspartic acid interchange at amino acid position 551 (G551D), which is present
among 3.1% of CF chromosomes.82,87 Patients with Class III mutations usually present with
severe disease that includes both pancreatic insufficiency and respiratory disease.
Class IV mutations, like Class III mutations, are properly processed and expressed at
the apical membrane. The protein is appropriately regulated by PKA and cAMP; however,
amino acid substitutions result in changes in channel architecture that alter chloride
conductance or channel gating.88 These mutations are relatively rare. When expressed on
the apical membrane, the ∆F508 mutant CFTR retains some chloride conductance and may
exhibit Class IV mutant properties. Because the CFTR protein is appropriately expressed
on the apical membrane and maintains some function, the mutation is associated with
pancreatic sufficiency and milder disease.89
Class V mutations result from promoter or splicing errors, leading to decreased expres-
sion of the functional protein at the apical membrane. This may be due to decreased
numbers of the CFTR protein or adequate numbers with decreased function. Approxi-
mately 13% of CFTR mutations result in pre-mRNA splicing errors.90 Presentation of CF in
patients with Class V mutations is quite variable, and disease severity has been shown to
be inversely related to the number of correctly spliced transcripts.88 Mutations that gener-
ate both correctly and aberrantly spliced transcripts (3849 + 10kb C→T) confer a milder
phenotype with more variable disease, whereas mutations that completely abolish exon
recognition (621 + 1 G→T) result in an absence of correctly spliced transcripts and a rela-
tively severe phenotype.90 Additional classification categories have been used to describe
mutations that result in increased protein turnover at the membrane or proteins presenting
with altered regulatory properties.77
Class VI mutations result from expression of a truncated CFTR protein that lacks
70-98 residues at the C-terminus of CFTR. An example of this is Q1412X, which lacks 70
amino acids at the C-terminus and has reduced stability at the apical membrane surface.91
CLINICAL PEARL
Although the most common mutation in Caucasians in the United
States is the ∆F508 mutation, the prevalence of mutation in other races
and ethnicities is not as predictable. Population mixing may alter the
prevalence of mutation in different racial and ethnic groups over time.
Treatment
Current treatment options available to help manage the symptoms of CF are outlined
in Table 11-6. Management of CF symptoms consists of replacing pancreatic enzymes,
correcting nutritional deficiencies, and preventing pulmonary deterioration through avert-
ing and treating infectious exacerbations. These measures are supportive/palliative and
Table 11-6
Pharmacotherapy of CF72,132,133
Therapy Comments
Microencapsulated pancreatic Pancreatic insufficiency leads to malnutrition and fat-

enzymes soluble vitamin deficiency due to decreased digestion
and absorption of dietary fat and protein and fat-soluble
vitamins.
Multivitamin It is used for supplementation of fat-soluble vitamins.
Percussion and postural drainage It increases clearance of
airway mucus.
Recombinant human DNAse It decreases viscosity of airway mucus, increases clearance
of airway mucus, improves pulmonary function, and
decreases frequency of pulmonary exacerbations.
Hypertonic saline (7%) It may improve hydration of airway surface liquid,
improves lung function and quality of life, and decreases
exacerbations.
β-agonists/theophylline It improves pulmonary function in patients with reactive
airway disease. Responsiveness to bronchodilators should
be identified prior to initiating chronic therapy.
Corticosteroids Oral corticosteroids (1–2 mg/kg) decrease inflammation
associated with disease but have undesirable effects. The
efficacy of inhaled corticosteroids is not proven.
Ibuprofen High-dose ibuprofen decreases CF-associated inflammation;
however, there is an increased risk of gastrointestinal
adverse effects.
Aerosolized tobramycin It improves lung function and quality of life. It decreases
pulmonary exacerbations and intravenous antibiotic use.
Azithromycin It is recommended for use in patients 6 years of age or older
with Pseudomonas aeruginosa. Treatment improves lung
function.
CF, cystic fibrosis.
help in managing the clinical manifestations of CF. There has been a strong interest in
the development of drugs that can restore CFTR activity and significantly alter the course
of CF.83 Because CF is caused by a variety of different mutations, different drugs will be
necessary for treating it based on the underlying genetic defect. The two main classes of
drugs that are being developed for modulating CFTR activity are potentiators and correc-
tors. Molecules aimed at improving the function of the CFTR protein by increasing chloride
conductance are termed “potentiators.” Agents aimed at improving the expression of CFTR
on the apical membrane through enhanced gene transcription, protein processing, or traf-
ficking of the protein are termed “correctors.”

Two disease-modifying drugs that correct the underlying defect in CF have been
recently approved for the treatment of CF. The approval of these drugs heralds a new era
in the treatment of CF and exemplify how new drugs are being developed to target specific
mutations.
Ivacaftor (Kalydeco)
Ivacaftor (VX-770) is an orally administered, small molecule drug that acts as a potentia-
tor and enhances the gating activity of CFTR channel. It was identified via high-throughput
screening from a library consisting of over 230,000 compounds.92 Ivacaftor was originally
evaluated for treatment in patients who carried a G551D mutation in at least one allele.
G551D mutations result in a glycine to aspartic acid change that allows localization of CFTR
to the cell surface but diminishes their gating ability. In two 48-week, placebo-controlled,
phase III trials in patients aged 12 and older (STRIVE) or 6-11 years (ENVISION), 150 mg
ivacaftor every 12 hours resulted in a significant improvement in lung function as measured
by FEV1.93,94 Safety and efficacy of ivacaftor were monitored in patients who completed the
STRIVE and ENVISION trials for an additional 96 weeks in a phase III, open-label extension
study (PERSIST).95 The improvements in FEV1 were maintained, and the drug was found to
be generally well tolerated at nearly three years of treatment. Ivacaftor received approval
from the FDA in 2012 for the treatment of patients who tested positive for a G551D-CFTR
variant in at least one allele. In 2014, the FDA extended the approval to eight additional
Class III mutations (G1244E, G1349D, G178R, G551S, S1251N, S1255P, S549N, S549R) and for
a Class IV mutation (R117H).96
Because ivacaftor acts as a potentiator, it is only effective when malfunctioning CFTR
is expressed on the cell surface. Thus, ivacaftor alone is unlikely to be effective for treat-
ing patients suffering from Class I and Class II mutations such as ∆F508 mutation and
G542X, which affect the expression or localization of CFTR to the cell surface. Based on
the mechanism of action and available clinical data, ivacaftor is not currently indicated for
patients with CF due to mutations other than the selected Class III and Class IV mutations
listed above.
Lumacaftor/Ivacaftor Combination (Orkambi)
The approval of ivacaftor marked a significant milestone in the treatment of CF, as it was
the first drug to be approved that treated the underlying defect in CF. However, ivacaftor
alone is not effective for treating individuals who are homozygous for the ∆F508 mutation.
∆F508 is the most common cause of CFTR mutation, and approximately 45% of the CF
patients in North America are homozygous for the ∆F508 mutation.97 A combination of a
“potentiator” such as ivacaftor with any other “corrector” molecule is likely going to be the
most effective treatment for patients homozygous for the ∆F508 mutation. The “correc-
tor” can improve CFTR folding and trafficking, while the “potentiator” can help increase
the probability of channel opening.97,98 Lumacaftor (VX-809) is a corrector that has been
demonstrated to improve folding and trafficking of ∆F508 CFTR to the cell surface. In addi-
tion, it has been shown to help improve stabilization of the partially rescued ∆F508 CFTR
at the cell surface by binding to the protein.99
A phase 2 study in patients homozygous for the ∆F508 mutation showed that a
combination of lumacaftor and ivacaftor resulted in improvements in FEV1 and also modest
improvement in sweat chloride concentration.100 Two phase 3 randomized, double-blind,
placebo-controlled trials (TRAFFIC and TRANSPORT) were conducted to test the efficacy
and safety of ivacaftor with two different doses of lumacaftor. Patients were randomly
assigned to one of the following three groups: 600 mg lumacaftor once daily with 250
mg ivacaftor every 12 hours, 400 mg lumacaftor and 250 mg ivacaftor every 12 hours, or
lumacaftor and ivacaftor matched placebos every 12 hours. A total of 1,122 patients were
randomized (559 in TRAFFIC and 563 in the TRANSPORT study), and the regimen was for
24 weeks. A significant improvement in FEV1 and reductions in rate of pulmonary exacer-

bations at both lumacaftor doses, for patients receiving the lumacaftor-ivacaftor combi-
nation, was seen by day 15 and continued through the 24-week regimen. Adverse effects
reported by the treatment and placebo groups were comparable and suggested that the
drug combination has an acceptable side-effect profile.97 Orkambi (lumacaftor/ivacaftor)
was approved on July 2, 2015 for the treatment of CF patients 12 years of age and older
who have two copies of the ∆F508 mutation in the CFTR gene.101
CLINICAL PEARL
Kalydeco (lumacaftor) is indicated for CF caused by the G551D, G1244E,
G1349D, G178R, G551S, S1251N, S1255P, S549N, S549R or the R117H
mutations. Orkambi (lumacaftor/ivacaftor) is indicated for patients with two
copies of ∆F508 mutation in their CFTR gene.
Other Protein Modulators in Development

Other compounds are being investigated for their ability to correct the underlying defect in
CF by targeting the processing, trafficking, and activity of CFTR.102
GENTAMICIN. Gentamicin increases CFTR expression in CFTR nonsense mutation trans-

fected cells and in a bronchial epithelial cell line expressing a nonsense mutation.103,104
Gentamicin nasal drops (given 3 times a day for 14 days) caused a significant decrease in
nasal potential difference (NPD), a measurement of chloride secretion and sodium absorp-
tion across the epithelia, in response to a chloride-free isoproterenol solution in patients
homozygous for the W1282X stop mutation (n = 4) or heterozygous for the W1282X and
G542X (n = 3), ∆F508 (n = 1), or 3849 + 10kb C→T (n = 1) mutation.105 No differences were
seen in basal NPD.
In a followup double-blind, placebo-controlled crossover trial, nasally administered
gentamicin was evaluated in 19 CF patients homozygous for the W1282X mutation (n = 11) or
heterozygous for the W1282X and ∆F508 mutations (n = 8) and five patients homozygous for
the ∆F508 mutation.106 In patients carrying the W1282X mutation (homozygous or hetero-
zygous), a significant decrease in basal and isoproterenol-treated NPD was associated with
gentamicin administration; however, the response was not seen in all patients. When homo-
zygous and heterozygous W1282X patients were evaluated separately, NPD was significantly
decreased in the homozygous population but not the heterozygous population. No changes
in basal NPD or response to chloride-free isoproterenol solution were seen in patients
homozygous for ∆F508. Ex vivo analysis of full-length CFTR protein in nasal epithelial cells
before and after gentamicin treatment in two heterozygous patients who had a response to
therapy demonstrated an increased cellular membrane localization of CFTR.
Another study evaluating intranasal gentamicin and tobramycin in CF patients hetero-
zygous for a premature stop mutation (n = 11) or CF patients without a stop mutation
(n = 18) found no differences in basal or isoproterenol stimulated NPD over 28 days.107

Ex vivo evaluation of nasal epithelial cells likewise demonstrated no increases in membrane
localization of CFTR. The effect of systemically administered gentamicin was evaluated in
five CF subjects heterozygous for one premature stop mutation and five CF subjects with-
out premature stop mutations.108 Subjects were administered gentamicin for 7 days with
doses adjusted to achieve peak serum concentrations of 8–10 mg/L and trough concentra-
tions <2 mg/L. Chloride sweat test, basal NPD, and response to chloride-free isoproterenol
administration were measured at baseline, during (days 3, 4, 5, and 6) and after gentamicin
administration, and again 1 to 4 weeks after treatment. No difference between the two
groups was seen in the chloride sweat test or basal NPD; however, patients with the stop
mutation had a significant increase in the number of NPD readings, indicating increased
chloride transport. As with the other studies, not all patients responded.
Differences in these trials may be related to population differences because it has
been demonstrated that response to gentamicin may be related to initial levels of CFTR
expression, enhanced nonsense-mediated mRNA decay, or to the type of stop muta-
tion.109,110 Furthermore, early studies demonstrating success with gentamicin have been
conducted in populations of patients having at least one mutation containing W1282X. It
has been demonstrated that W1282X CFTR retains partial chloride channel function that is
enhanced after suppression of the stop codon.111 Finally, although none of the trials report-
ed significant adverse effects, long-term systemic administration of gentamicin would raise
safety concerns because of its well-documented toxicities.
ATALUREN. Because of the early success with gentamicin and the recognized toxicity and
administration issues, considerable interest exists in the development of compounds that
can safely suppress premature stop codons. Ataluren is an orally available oxadiazole iden-
tified via high throughput screening of low molecular weight compounds that may have the
potential to treat disease caused by nonsense mutations. At low concentrations, ataluren
has been shown to promote dose-dependent read through in nonsense mutations and is a
more potent nonsense suppressor than gentamicin.112
Phase I trials of ataluren indicated the drug was safe for further clinical study.113 A
phase II trial in adult CF patients with at least one nonsense mutation in the CFTR gene
evaluated the effectiveness of ataluren given in two 28-day cycles.114 The first cycle (n =
23) consisted of 16 mg/kg per day of ataluren given in 3 divided doses for 14 days followed
by 14 days without drug. In cycle two, 21 of the cycle one patients received ataluren 40 mg/
kg/ day in 3 divided doses for 14 days followed by 14 drug-free days. Significant increases
were seen in total chloride transport in both treatment cycles, and total chloride transport
entered the normal range for 57% of patients in cycle 1 and 43% of the patients in cycle
2. Adverse effects were mild. Efficacy and safety were also studied for patients receiv-
ing ataluren 3 times daily for 12 weeks.115 A positive improvement in pulmonary function
was observed over time in this study. A randomized, placebo-controlled phase III trial to
assess safety and efficacy of ataluren showed that the side effects were similar between
the placebo and treatment groups.116 However, no significant improvement in lung function
was observed in the general nonsense mutation CF population. Only patients not receiving
chronic inhaled tobramycin seemed to show an improvement in lung function with atal-
uren. Based on these studies, ataluren seems to be well tolerated and may be useful for
nonsense mutation CF patients who are not receiving chronic inhaled tobramycin.
SODIUM BUTYRATE. Class V mutations are caused by aberrantly spliced transcripts. The
severity of the disease depends on the number of correctly spliced transcripts. One method
for increasing the numbers of correctly spliced transcripts, and thus the expression of CFTR
at the apical membrane, is to overexpress splicing factors which increases the amount of
correctly spliced RNA.117 Sodium butyrate has been previously shown to modify the alterna-
tive splicing pattern of exon 7 in the survival motor neuron-2 (SMN2) gene, thus increasing
the number of full-length SMN mRNA transcripts and exon 7 containing SMN protein in
spinal muscular atrophy lymphoid cells.118 Based on these results, the effect of sodium
butyrate function and expression of CFTR in an epithelial cell line from a nasal polyp of a
patient with the 3849 + 10kb C→T mutation was evaluated.90 Sodium butyrate significantly
decreased the amount of aberrantly spliced transcripts and activated CFTR. The utility of
these findings is unknown as sodium butyrate is not well suited for clinical use; however,
other agents such as valproic acid have also demonstrated the ability to increase SMN2
transcripts and proteins and may have application in this area.
SODIUM 4-PHENYLBUTYRATE. Sodium 4-phenylbutyrate (4-PBA) is a derivative of sodium

butyrate that has been shown to increase trafficking of ∆F508.119 It is thought that 4-PBA
reduces the expression of heat shock protein HSC-70, resulting in decreased HSC-70 asso-
ciation with ∆F508 leading to decreased ∆F508 degradation in the endoplasmic reticulum.83
The effect of 4-PBA on NPD and sweat chloride response was evaluated in CF patients
homozygous for the ∆F508 mutation in a randomized, double-blind, placebo-controlled
trial.120 Subjects were given placebo or 19 grams of 4-PBA in divided doses 3 times a day.
Patients in the pilot trial treated with 4-PBA (n = 9) showed significant improvement in
NPD after infusion of a chloride-free isoproterenol solution when compared to placebo-
treated patients. No differences were seen in amiloride sensitive NPD or sweat chloride
concentrations.
A phase I/II trial of 4-PBA in 19 adult CF patients homozygous for ∆F508−CFTR
demonstrated a significant increase in isoproterenol-stimulated chloride transport in nasal
epithelia. The maximal response was seen with a daily dose of 20 grams for 1 week.121 Peak
response occurred between days 3 and 4 with a decrease in response seen at day 7. It is
possible that this decrease in response could be related to increased ENaC activity because
a recent in vitro study showed that 4-PBA induces a time-dependent increase in ENaC
protein in the apical membrane in conjunction with increased apical amiloride-sensitive
sodium current.122 Therefore, increased CFTR expression may be offset by increased ENaC-
related sodium absorption.
GENISTEIN. Genistein acts as a potentiator and has been investigated for its ability to
increase CFTR channel opening and treat the deficient channel function caused by Class
III mutations. Genistein, a soybean-derived isoflavone, has been shown to increase CFTR
chloride transport in both wild-type and mutant CFTR. In patients with at least one G551D
mutation, perfusion of the nasal mucosa with genistein was shown to hyperpolarize NPD
in both healthy and CF patients indicating it stimulates chloride conductance in the nasal
epithelia.123 It appears that genistein enhances CFTR activity by decreasing the closing rate
of the channel at low concentrations, but it may inhibit channel activity at high doses by
binding to a low affinity site that decreases the opening rate of the channel.124,125 Genis-
tein’s clinical usefulness may be limited by its low potency and rapid metabolism.
PDE5 INHIBITORS. PDE5 inhibitors have been shown to correct ∆F508-CFTR trafficking
and CFTR chloride conductance.126 Intraperitoneal injections of sildenafil, vardenafil, and
tadalafil have been shown to increase CFTR-mediated chloride transport in ∆F508del mice.
Vardenafil was also found to stimulate chloride conductance through the normal CFTR
protein.126 These promising early results suggest that PDE5 inhibitors could be useful for
rescuing ∆F508-CFTR trafficking and restoring deficient chloride conductance caused by
Type II mutations.
Additional disease-modifying agents that are in development include Riociguat,

QBW251, N91115 and QR-010. The progress of these agents through the discovery and
regulatory process can be monitored at the CF Foundation website (http://www.cff.org/
research/drugdevelopmentpipeline/).127
Gene Therapy
Because defective genes are responsible for CF, introduction of a normal copy of the CFTR
gene can help to correct the underlying defect in CF. Adeno-associated viruses (AAV) have
been tested for their ability to deliver the complete human CFTR cDNA to the lungs, sinus,
and nose of individuals with CF.128 Aerosolized AAV vectors encoding CFTR were found
to be safe in a study involving 102 subjects. However, no significant improvement in lung
function was observed, most likely owing to the challenges associated with delivering
genes to the lungs. More recently, a randomized, double-blind, placebo-controlled, phase
2b trial conducted in the UK found a modest but significant improvement in lung function
following delivery of plasmid DNA encoding the CFTR gene using liposomes.129 The results
reported in this study are very encouraging, but several challenges will need to be over-
come before gene therapy finds a place in the treatment of CF.
SUMMARY
Pharmacogenetic testing can help guide drug therapy for both asthma and CF. Genetic
variants have been identified that influence the response to treatment with β2-agonists,
inhaled corticosteroids, LT modifiers, and theophylline in asthmatic patients. The applica-
tion of this research can help practitioners to optimize therapy based on genotype and
reduce negative outcomes. Mutations in the CFTR gene are responsible for CF. The approv-
al of ivacaftor and lumacaftor/ivacaftor for specific mutations has ushered in a new era in
the treatment of CF. Although each of these drugs treats the underlying defect in CF and
improves CFTR function, they are indicated for different mutations and demonstrate how
personalized medicine is revolutionizing the way diseases are treated.
REFERENCES
1. National Asthma Education and Prevention Program. Expert Panel Report 3: Guidelines for the diag-
nosis and management of asthma.: U.S. Department of Health & Human Services, Public Health
Services, National Institutes of Health, National Heart, Lung, and Blood Institute; 2007.
2. Noguchi E, Arinami T. Candidate genes for atopic asthma: Current results from genome screens.
Am J Pharmacogenomics. 2001;1:251-261.
3. Howard T, Meyers D, Bleecker E. Mapping susceptibility genes for asthma and allergy. J Allergy Clin
Immunol. 2000;105(suppl):S477-481.
4. Cho SH. Pharmacogenomic approaches to asthma treatment. Allergy Asthma Immunol Res.
2010;2(3):177-182.
5. Busse WW, Lemanske RF Jr. Asthma. N Engl J Med. 2001;344:350-362.
6. Barnes PJ. Beta-adrenergic receptors and their regulation. Am J Respir Crit Care Med. 1995;152(3):838-
860.
7. Castle W, Fuller R, Hall J, et al. Serevent nationwide surveillance study: comparison of salmeterol
with salbutamol in asthmatic patients who require regular bronchodilator treatment. BMJ. 1993 Apr
17;306(6884):1034-1037.
8. Mann RD, Kubota K, Pearce G, et al. Salmeterol: a study by prescription-event monitoring in a UK
cohort of 15,407 patients. J Clin Epidemiol. 1996 Feb;49(2):247-250.
9. Lanes SF, Lanza LL, Wentworth CE III. Risk of emergency care, hospitalization, and ICU stays for acute
asthma among recipients of salmeterol. Am J Respir Crit Care Med. 1998 Sep;158(3):857-861.
10. Nelson HS, Weiss ST, Bleecker ER. The Salmeterol Multicenter Asthma Research Trial: a compari-
son of usual pharmacotherapy for asthma or usual pharmacotherapy plus salmeterol. Chest. 2006
Jan;129(1):15-26.
11. Salpeter SR, Buckley NS, Ormiston TM, et al. Meta-analysis: effect of long-acting beta-agonists on
severe asthma exacerbations and asthma-related deaths. Ann Intern Med. 2006 Jun 20;144(12):904-
912.
12. Ortega VE, Hawkins GA, Peters SP, et al. Pharmacogenetics of the [beta]2-adrenergic receptor gene.
Immunol Allergy Clin North Am. 2007;27:665-684.
13. Morrow T. Implications of pharmacogenomics in the current and future treatment of asthma. J Manag
Care Pharm. 2007;13:497-505.
14. Liggett SB. The pharmacogenetics of B2-adrenergic receptors: Relevance to asthma. J Allergy Clin
Immunol. 2000;105(suppl):S487-492.
15. Portelli M, Sayers I. Genetic basis for personalized medicine in asthma. Expert Rev Respir Med.
2012;6(2):223-236.
16. Drazen JM, Silverman EK, Lee TH. Heterogeneity of therapeutic responses in asthma. Br Med Bull.
2000;56:1054-1070.
17. Contopoulos-Ioannidis DG, Manoli EN, Ioannidis JP. Meta-analysis of the association of beta2-
adrenergic receptor polymorphisms with asthma phenotypes. J Allergy Clin Immunol. 2005;115:963-
972.
18. Litonjua AA, Gong L, Duan QL, et al. Very important pharmacogene summary ADRB2. Pharmacogenet
Genomics. 2010;20(1):64-69.
19. Martinez FD, Graves PE, Baldini M, et al. Association between genetic polymorphisms of the beta2-
adrenoceptor and response to albuterol in children with and without a history of wheezing. J Clin
Invest. 1997;100(12):3184-3188.
20. Lima JJ, Thomason DB, Mohamed MH, et al. Impact of genetic polymorphisms of the beta2-
adrenergic receptor on albuterol bronchodilator pharmacodynamics. Clin Pharmacol Ther.
1999;65(5):519-525.
21. Israel E, Chinchilli VM, Ford JG, et al. Use of regularly scheduled albuterol treatment in asthma:
genotype-stratified, randomised, placebo-controlled cross-over trial. Lancet. 2004;364:1505-1512.
22. Israel E, Drazen JM, Liggett SB, et al. The effect of polymorphisms of the beta(2)-adrenergic receptor
on the response to regular use of albuterol in asthma. Am J Respir Crit Care Med. 2000;162:75-80.
23. Taylor DR, Drazen JM, Herbison GP, et al. Asthma exacerbations during long term beta agonist use:
influence of beta 2 adrenoceptor polymorphism. Thorax. 2000;55:762-767.
24. Bleecker ER, Postma DS, Lawrance RM, et al. Effect of ADRB2 polymorphisms on response to
long-acting beta2-agonist therapy: a pharmacogenetic analysis of two randomised studies. Lancet.
2007;370:2118-2125.
25. Wechsler ME, Lehman E, Lazarus SC, et al. beta-Adrenergic receptor polymorphisms and response to
salmeterol. Am J Respir Crit Care Med. 2006;173:519-526.
26. Telleria JJ, Blanco-Quirós A, Muntion S, et al. Tachyphylaxis to beta2-agonists in Spanish
asthmatic patients could be modulated by beta2-adrenoceptor gene polymorphisms. Respir Med.
2006;100(6):1072-1078.
27. Green SA, Rathz DA, Schuster AJ, et al. The Ile164 beta(2)-adrenoceptor polymorphism alters salme-
terol exosite binding and conventional agonist coupling to G(s). Eur J Pharmacol. 2001;421(3):141-147.
28. Ortega VE, Meyers DA, Bleecker ER. Asthma pharmacogenetics and the development of genetic
profiles for personalized medicine. Pharmgenomics Pers Med. 2015;8:9-22.
29. Tantisira KG, Small KM, Litonjua AA, et al. Molecular properties and pharmacogenetics of a polymor-
phism of adenylyl cyclase type 9 in asthma: interaction between beta-agonist and corticosteroid
pathways. Hum Mol Genet. 2005;14(12):1671-1677.
30. Himes BE, Jiang X, Hu R, et al. Genome-wide association analysis in asthma subjects identifies
SPATS2L as a novel bronchodilator response gene. PLoS Genet. 2012;8(7):e1002824.

31. Drake KA, Torgerson DG, Gignoux CR, et al. A genome-wide association study of bronchodilator
response in Latinos implicates rare variants. J Allergy Clin Immunol. 2014;133(2):370-378.
32. Umland SP, Schleimer RP, Johnston SL. Review of the molecular and cellular mechanisms of action
of glucocorticoids for use in asthma. Pulm Pharmacol Ther. 2002;15(1):35-50.
33. Kelly HW. Comparison of inhaled corticosteroids: an update. Ann Pharmacother. 2009;43(3):519-527.
34. Malmstrom K, Rodriguez-Gomez G, Guerra J, et al. Oral montelukast, inhaled beclomethasone, and
placebo for chronic asthma: A randomized, controlled trial. Ann Intern Med. 1999;130:487-495.
35. Szefler SJ, Martin RJ, King TS, et al. Significant variability in response to inhaled corticosteroids for
persistent asthma. J Allergy Clin Immunol. 2002;109:410-418.
36. Israel E. Genetics and the variability of treatment response in asthma. J Allergy Clin Immunol.
2005;115(suppl):S532-S538.
37. Weiss S, Litonjua A, Lange C, et al. Overview of the pharmacogenetics of asthma treatment.
Pharmacogenomics J. 2006;6:311-326.
38. Tantisira K, Lake S, Silverman E, et al. Corticosteroid pharmacogenetics: Association of sequence
variants in CRHR1 with improved lung function in asthmatics treated with inhaled corticosteroids.
Hum Molec Gen. 2004;13:1353-1359.
39. Tantisira KG, Hwang ES, Raby BA, et al. TBX21: A functional variant predicts improvement in asthma
with the use of inhaled corticosteroids. Proc Nat Acad Sci USA. 2004;101:18099-18104.
40. Raby BA, Hwang E-S, Steen KV, et al. T-Bet Polymorphisms are associated with asthma and airway
hyperresponsiveness. Am J Respir Crit Care Med. 2006;173:64-70.
41. Leung D, Spahn J, Szefler S. Steroid-unresponsive asthma. Semin Respir Crit Care Med. 2002;23:387-
398.
42. Bray P, Cotton R. Variations of the human glucocorticoid receptor gene (NR3C1): pathological and in
vitro mutations and polymorphisms. Hum Mutat. 2003;21:557-568.
43. Hawkins GA, Lazarus R, Smith RS. The glucocorticoid receptor heterocomplex gene STIP1 is associated
with improved lung function in asthmatic subjects treated with inhaled corticosteroids. J Allergy Clin
Immunol. 2009;123(6):1376-1383.
44. Stockmann C, Fassl B, Gaedigk R. Fluticasone propionate pharmacogenetics: CYP3A4*22 polymor-
phism and pediatric asthma control. J Pediatr. 2013;162(6):1222-7, 1227.e1-2.
45. Busse WW. Does leukotriene modulation of eosinophil function explain the therapeutic effectiveness
of receptor antagonists in some patients with asthma? Clin Exp Allergy. 2001;31(6):806-807.
46. Watkins PB, Dube LM, Walton-Bowen K, et al. Clinical pattern of zileuton-associated liver injury:
results of a 12-month study in patients with chronic asthma. Drug Saf. 2007;30:805-815.
47. Weschler M, Israel E. Pharmacogenetics of treatment with leukotriene modifiers. Curr Opin Allerg Clin
Immunol. 2002;2:395-401.
48. Hoshiko S, RÃdmark O, Samuelsson B. Characterization of the human 5-lipoxygenase gene
promoter. Proc Nat Acad Sci USA. 1990;87:9073-9077.
49. Drazen JM, Yandava CN, Dube L, et al. Pharmacogenetic association between ALOX5 promoter
genotype and the response to anti-asthma treatment. Nat Genet. 1999;22:168-170.
50. Lima JJ, Zhang S, Grant A, et al. Influence of leukotriene pathway polymorphisms on response to
montelukast in asthma. Am J Respir Crit Care Med. 2006;173:379-385.
51. Anderson W, Kalberg C, Edwards L, et al. Effects of polymorphisms in the promoter region of
5-lipoxygenase and LTC4 synthase on the clinical response to zafirlukast and fluticasone. Eur Respir J.
2000;16(suppl):183S.
52. Telleria J, Blanco-Quiros A, Varillas D, et al. ALOX5 promoter genotype and response to montelukast
in moderate persistent asthma. Resp Med. 2008;102:857-861.
53. Wechsler M, Israel E. Pharmacogenetics of treatment with leukotriene modifiers. Curr Opin Allerg Clin
Immunol. 2002;2:395-401.
54. Sampson AP, Siddiqui S, Buchanan D, et al. Variant LTC4 synthase allele modifies cysteinyl leukotriene
synthesis in eosinophils and predicts clinical response to zafirlukast. Thorax. 2000;55(suppl):S28-31.
55. Kim SH, Ye YM, Hur GY, et al. CysLTR1 promoter polymorphism and requirement for leukotriene recep-
tor antagonist in aspirin-intolerant asthma patients. Pharmacogenomics. 2007;8(9):1143-1150.
56. Klotsman M, York TP, Pillai SG, et al. Pharmacogenetics of the 5-lipoxygenase biosynthetic pathway
and variable clinical response to montelukast. Pharmacogenet Genomics. 2007;17(3):189-196.
57. Mougey EB, Feng H, Castro M, et al. Absorption of montelukast is transporter mediated: a common
variant of OATP2B1 is associated with reduced plasma concentrations and poor response. Pharmaco-
genet Genomics. 2009;19(2):129-138.
58. Tapaninen T, Karonen T, Backman JT, et al. SLCO2B1 c.935G>A single nucleotide polymorphism
has no effect on the pharmacokinetics of montelukast and aliskiren. Pharmacogenet Genomics.
2013;23(1):19-24.
59. Kim KA, Lee HM, Joo HJ, et al. Effects of polymorphisms of the SLCO2B1 transporter gene on the
pharmacokinetics of montelukast in humans. J Clin Pharmacol. 2013;53(11):1186-1193.
60. Weinberger M, Hendeles L. Theophylline in asthma. N Engl J Med. 1996 May 23;334(21):1380-1388.
61. Obase Y, Shimoda T, Kawano T, et al. Polymorphisms in the CYP1A2 gene and theophylline metabo-
lism in patients with asthma. Clin Pharmacol Ther. 2003;73(5):468-474.
62. Nakajima M, Yokoi T, Mizutani M. Genetic polymorphism in the 5’-flanking region of human CYP1A2
gene: effect on the CYP1A2 inducibility in humans. J Biochem. 1999;125(4):803-808.

63. James AJ, Reinius LE, Verhoek M, et al. Increased YKL-40 and chitotriosidase in asthma and chronic
obstructive pulmonary disease. Am J Respir Crit Care Med. 2016;193(2):131-142.
64. Chien Y-H, Hwu W-L, Chiang B-L. The genetics of atopic dermatitis. Clin Rev Allergy Immunol.
2007;33:178-190.
65. Molfino N. Current thinking on genetics of chronic obstructive pulmonary disease. Curr Opin Pulm
Med. 2007;13:107-113.
66. Wood A, Stockley R. The genetics of chronic obstructive pulmonary disease. Resp Res. 2006;7:130-143.
67. Global Strategy for the Diagnosis, Management and Prevention of COPD, Global Initiative for Chronic
Obstructive Lung Disease (GOLD). www.goldcopd.org. Accessed November 15, 2015.
68. Allergic rhinitis and its impact on asthma. World Health Organization. http://www.whiar.org/. Accessed
November 15, 2015.
69. Farrell PM, Rosenstein BJ, White TB, et al. Guidelines for diagnosis of cystic fibrosis in newborns
through older adults: Cystic Fibrosis Foundation consensus report. J Pediatr. 2008;153(suppl):S4-S14.
70. Cystic Fibrosis Foundation. Patient Registry Annual Data Report to the Directors 2013. Bethesda, MD:
Cystic Fibrosis Foundation; 2013.
71. Gibson RL, Burns JL, Ramsey BW. Pathophysiology and management of pulmonary infections in cystic
fibrosis. Am J Respir Crit Care Med. 2003;168(8):918-951.
72. Wright CC, Vera YY. Chapter 18. Cystic fibrosis. In: DiPiro JT, Talbert RL, Yee GC, et al., eds. Pharmaco-
therapy: A Pathophysiologic Approach, 9th ed. New York, NY: McGraw-Hill; 2014. Available at: http://
accesspharmacy.mhmedical.com.ccl.idm.oclc.org/content.aspx?bookid=689&Sectionid=48811466.
73. Cystic Fibrosis Foundation. Patient Registry 2006 Annual Data Report. Bethesda, MD: Cystic Fibrosis
Foundation; 2008.
74. Cystic Fibrosis Foundation. Available at: http://www.cff.org/. Accessed November 15, 2015.
75. Boucher RC. Airway surface dehydration in cystic fibrosis: pathogenesis and therapy. Ann Rev Med.
2007;58:157-170.
76. Gadsby DC, Vergani P, Csanády L. The ABC protein turned chloride channel whose failure causes
cystic fibrosis. Nature. 2006;440:477-483.
77. Rowe SM, Miller S, Sorscher EJ. Cystic fibrosis. N Engl J Med. 2005;352:192-201.
78. Mehta A. CFTR: more than just a chloride channel. Pediatric Pulmonol. 2005;39:292-298.
79. Donaldson SH, Boucher RC. Sodium channels and cystic fibrosis. Chest. 2007;132:1631-1636.
80. Elizur A, Cannon CL, Ferkol TW. Airway inflammation in cystic fibrosis. Chest. 2008;133:489-495.
81. Weber AJ, Soong G, Bryan R, et al. Activation of NF-κB in airway epithelial cells is dependent on CFTR
trafficking and Cl-channel function. Am J Physiol Lung Cell Mol Physiol. 2001;281:L71-L78.
82. Kerem E. Pharmacological induction of CFTR function in patients with cystic fibrosis: mutation specific
therapy. Pediatric Pulmonol. 2005;40:183-196.
83. Kerem E. Mutation specific therapy in CF. Peadiatr Resp Rev. 2006;7(suppl):S166-S169.
84. Shoshani T, Augartan A, Gazit E, et al. Association of a nonsense mutation (W1128X), the most
common mutation in the Ashkenazi Jewish cystic fibrosis patients in Israel, with presentation of
severe disease. Am J Hum Genet. 1992;50:222-228.
85. Roomans GM. Pharmacological approaches to correcting the ion transport defect in cystic fibrosis. Am
J Respir Med. 2003;2:413-431.
86. Zeitlin P. Novel pharmacologic therapies for cystic fibrosis. J Clin Investig. 1999;103:447-452.
87. Becq F. On the discovery and development of CFTR chloride channel activators. Curr Pharm Design.
2006;12:471-484.
88. MacDonald KD, McKenzie KR, Zeitlin PL. Cystic fibrosis transmembrane regulator protein mutations.
‘Class’ opportunity for novel drug innovation. Pediatr Drugs. 2007;9:1-10.
89. Rubenstein RC. Targeted therapy for cystic fibrosis. Cystic fibrosis transmembrane conductance regula-
tor mutation-specific pharmacologic strategies. Mol Drug Ther. 2006;10:293-301.
90. Nissim-Rafinia M, Aviram M, Randell SH, et al. Restoration of the cystic fibrosis transmembrane
conductance regulator function by splicing modulation. EMBO Rep. 2004;5:1071-1077.
91. Rowntree RK1, Harris A. The phenotypic consequences of CFTR mutations. Ann Hum Genet.
2003;67(Pt 5):471-485.
92. Brodlie M, Haq IJ, Roberts K, et al. Targeted therapies to improve CFTR function in cystic fibrosis.
Genome Med. 2015;7(1):101.
93. Sheridan C. First cystic fibrosis drug advances towards approval. Nat Biotechnol. 2011;29(6):465-466.
94. Deeks ED. Ivacaftor: a review of its use in patients with cystic fibrosis. Drugs. 2013;73(14):1595-1604.
95. McKone EF, Borowitz D, Drevinek P, et al. Long-term safety and efficacy of ivacaftor in patients
with cystic fibrosis who have the Gly551Asp-CFTR mutation: a phase 3, open-label extension study
(PERSIST). Lancet Respir Med. 2014;2(11):902-1110.
96. Lenherr N, Lura M, Trachsel D. Ivacaftor in a young boy with the rare gating mutation S549R—use of
lung clearance index to track progress: a case report. BMC Pulm Med. 2015;15:123.
97. Wainwright CE, Elborn JS, Ramsey BW, et al. Lumacaftor-ivacaftor in patients with cystic fibrosis
homozygous for Phe508del CFTR. N Engl J Med. 2015;373(3):220-231.
98. Matthes E, Goepp J, Carlile GW, et al. Low free drug concentration prevents inhibition of F508del
CFTR functional expression by the potentiator VX-770 (ivacaftor). Br J Pharmacol. 2015;Oct 22.
99. Eckford PD, Ramjeesingh M, Molinski S. VX-809 and related corrector compounds exhibit second-
ary activity stabilizing active F508del-CFTR after its partial rescue to the cell surface. Chem Biol.
2014;21(5):666-678.
100. Boyle MP, Bell SC, Konstan MW. A CFTR corrector (lumacaftor) and a CFTR potentiator (ivacaftor)
for treatment of patients with cystic fibrosis who have a phe508del CFTR mutation: a phase 2
randomised controlled trial. Lancet Respir Med. 2014;2(7):527-538.
101. Gohil K. Pharmaceutical approval update. P & T. 2015;40(9):567-568.
102. Pettit RS, Fellner C. CFTR modulators for the treatment of cystic fibrosis. P & T. 2014;39(7):500-511.
103. Howard M, Frizzell RA, Bedwell DM. Aminoglycoside antibiotics restore CFTR function by overcoming
premature stop mutations. Nat Med. 1996;2:467-469.
104. Bedwell DM, Kaenjk A, Benos DJ, et al. Suppression of a CFTR premature stop mutation in a bronchial
epithelial cell line. Nat Med. 1997;3:1280-1284.
105. Wilschanski M, Famini C, Blau H, et al. A pilot study of the effect of gentamicin on nasal potential
difference measurements in cystic fibrosis patients carrying stop mutations. Am J Respir Crit Care
Med. 2000;161:860-865.
106. Wilschanski M, Yahav Y, Yaacov Y, et al. Gentamicin-induced correction of CFTR function in patients
with cystic fibrosis and CFTR stop mutations. N Engl J Med. 2003;349:1433-1441.
107. Clancy JP, Rowe SM, Bebok Z, et al. No detectable improvements in cystic fibrosis transmembrane
conductance regulator by nasal aminoglycosides in patients with cystic fibrosis with stop mutations.
Am J Resp Cell Mol Biol. 2007;37:57-66.
108. Clancy JP, Bebok Z, Ruiz F, et al. Evidence that systemic gentamicin suppresses premature stop muta-
tions in patients with cystic fibrosis. Am J Respir Crit Care Med. 2001;163:1683-1692.
109. Linde L, Boelz S, Nissim-Rafinia M, et al. Nonsense-mediated mRNA decay affects nonsense transcript
levels and governs response of cystic fibrosis patients to gentamicin. J Clin Investig. 2007;117:683-692.
110. Sermet-Gaudelus I, Renouil M, Fajac A, et al. In vitro prediction of stop-codon suppression by intrave-
nous gentamicin in patients with cystic fibrosis: a pilot study. BMC Medicine. 2007;5:5.
111. Rowe SM, Varga K, Rab A, et al. Restoration of W1282X CFTR activity by enhanced expression. Am J
Resp Cell Mol Biol. 2007;37:347-356.
112. Welch EM, Barton ER, Zhuo J, et al. PTC124 targets genetic disorders caused by nonsense mutations.
Nature. 2007;447:87-91.
113. Hirawat S, Welch EM, Elfring GL, et al. Safety, tolerability, and pharmacokinetics of PTC124, a nonamino-
glycoside nonsense mutation suppressor, following single- and multiple-dose administration to
healthy male and female adult volunteers. J Clin Pharmacol. 2007;47:430-444.

114. Kerem E, Hirawat S, Armoni S, et al. Effectiveness of PTC124 treatment of cystic fibrosis caused by
nonsense mutations: a prospective phase II trial. Lancet. 2008;372:719-727.
115. Wilschanski M, Miller LL, Shoseyov D, et al. Chronic ataluren (PTC124) treatment of nonsense muta-
tion cystic fibrosis. Eur Respir J. 2011;38(1):59-69.
116. Kerem E, Konstan MW, De Boeck K, et al. Ataluren for the treatment of nonsense-mutation
cystic fibrosis: a randomised, double-blind, placebo-controlled phase 3 trial. Lancet Respir Med.
2014;2(7):539-547.
117. Nissim-Rafinia M, Kerem B. Splicing modulation as a modifier of the CFTR function. Prog Mol Subcell
Biol. 2006;44:233-254.
118. Chang JG, Hsieh-Li HM, Jong YJ, et al. Treatment of spinal muscular atrophy by sodium butyrate. Proc
Natl Acad Sci USA. 2001;98(17):9808-9813.

119. Rubenstein RC, Egan ME, Zeitlin PL. In vitro pharmacologic restoration of CFTR-mediated chloride
transport with sodium 4-phenylbutyrate in cystic fibrosis epithelial cells containing ∆F508-CFTR.
J Clin Investig. 1997;100:2457-2465.
120. Rubenstein RC, Zeitlin PL. A pilot clinical trial of oral sodium 4-phenylbutyrate (Buphenyl) in
∆F508-homozygous cystic fibrosis patients: partial restoration of nasal epithelial CFTR function.
Am J Resp Crit Care Med. 1998;157:484-490.
121. Zeitlin PL, Diener-West M, Rubenstein RC, et al. Evidence of CFTR function in cystic fibrosis after
systemic administration of 4-phenylbutyrate. Mol Ther. 2002;6:119-126.
122. Pruliere-Escabasse V, Planes C, Escudier E, et al. Modulation of epithelial sodium channel traf-
ficking and function by sodium 4-phenylbutyrate in human nasal epithelial cells. J Biol Chem.
2007;282:34048-34057.
123. Illek B, Zhang L, Lewis NC, et al. Defective function of the cystic fibrosis-causing missense mutation
G551D is recovered by genistein. Am J Physiol. 1999;277(4 Pt 1):C833-C839.
124. Wang F, Zeltwanger S, Yang IC-H, et al. Actions of genistein on cystic fibrosis transmembrane conduc-
tance regulator channel gating. Evidence for two binding sites with opposite effects. J Gen Physiol.
1998;111:477-490.
125. Moran O, Zegarra-Moran O. A quantitative description of the activation and inhibition of CFTR by
potentiators: genistein. FEBS Letters. 2005;579:3979-3983.
126. Noel S, Dhooghe B, Leal T. PDE5 inhibitors as potential tools in the treatment of cystic fibrosis. Front
Pharmacol. 2012;3:167.
127. Cystic Fibrosis Foundation. Drug Development Pipeline. Available at: https://tools.cff.org/research/
drugdevelopmentpipeline/. Accessed November 15, 2015.
128. Moss RB, Milla C, Colombo J, et al. Repeated aerosolized AAV-CFTR for treatment of cystic fibrosis: a
randomized placebo-controlled phase 2B trial. Hum Gene Ther. 2007;18(8):726-732.
129. Alton EW, Armstrong DK, Ashby D, et al. Repeated nebulisation of non-viral CFTR gene therapy in
patients with cystic fibrosis: a randomised, double-blind, placebo-controlled, phase 2b trial. Lancet
Respir Med. 2015;3(9):684-691.
130. Reihsaus E, Innis M, MacIntyre N, et al. Mutations in the gene encoding for the beta 2-adrenergic
receptor in normal and asthmatic subjects. Am J Respir Cell Mol Biol. 1993;8:334-339.
131. Sabato MF, Irani A-M, Bukaveckas BL, et al. A simple and rapid genotyping assay for simultaneous
detection of two ADRB2 allelic variants using fluorescence resonance energy transfer probes and
melting curve analysis. J Mol Diagn. 2008;10:258-264.
132. Flume PA, O’Sullivan BP, Robinson KA, et al. Cystic fibrosis pulmonary guidelines: chronic medications
for maintenance of lung health. Am J Respir Crit Care Med. 2007;176:957-969.
133. Dalcin P deTR, Abreu e Silva FA de. Cystic fibrosis in adults: diagnostic and therapeutic aspects. J Bras
Pneumol. 2008;34:107-117.
CHAPTER
12
Toxicogenomics
Helen E. Smith, MS, PhD, RPh

After completing this chapter, the reader BIOMARKER—a biological characteristic
should be able to that can be measured to indicate the
presence of a biologic or pathogenic
Define key terminology used in the field
process, predict a pharmacological
of toxicogenomics.
response to a therapeutic intervention,
Discuss the tools used in toxicogenomics indicate an exposure, or predict the risk
and toxicogenetics. of disease or adverse effect due to an
Discuss how toxicogenomics is utilized in exposure.
drug development. DEOXYRIBONUCLEIC ACID (DNA)—a self-
Discuss the potential and current clinical replicating material present in nearly all
use of toxicogenomics and toxicogenetics living organisms as the main constituent
to personalize the use of medicines. of chromosomes.
MESSENGER RIBONUCLEIC ACID (mRNA)—the
ribonucleic acid polymer derived from
DNA through the process of transcription
providing the code for protein production
through translation.
METABOLOMICS—the study of the relative
production of metabolites in animals,
cells, or tissues. Metabolomics may be
used in toxicological studies to evaluate
the effects of an exposure on protein
expression and activity.
MONOGENETIC—a phenotypic trait that is
influenced by one gene and its variants.
POLYGENETIC—a phenotypic trait that is
influenced by more than one gene and
its variants.
PROTEOMICS—the study of the relative
levels of protein expression and activity

in animals, cells, or tissues. Proteomics
may be used in toxicological studies to
evaluate the effects of an exposure on
gene and mRNA expression.
357
TOXICOGENETICS—the study of how an individual’s genetic makeup affects gene expression,

protein expression and activity, and metabolism in response to exposures to potentially
toxic compounds.
TOXICOGENOMICS—the study of how the genome as a whole responds to exposures to
potentially toxic compounds.

TRANSCRIPTOMICS—the evaluation of mRNA expression levels in cells or tissues.
Transcriptomics are also referred to as expression profiles. Transcriptomics may be used
in toxicological studies to evaluate the effects of an exposure on mRNA expression.
INTRODUCTION
Classic toxicology often utilizes tools such as clinical chemistry, histology, and electron
microscopy to measure the effects of a toxin on a tissue. Newer toxicologic methods
include evaluating the genome’s response to toxin exposure and how the genome of an
organism in turn affects its response to the exposure. Gene transcription, protein expres-
sion and activity, and metabolic activity may all be evaluated to measure the genome
response to toxic exposures. Toxicogenetics is the study of how an individual organism’s
genetic makeup affects its response to environmental stressors or toxins. The terms toxi-
cogenomics and toxicogenetics (as well as pharmacogenomics and pharmacogenetics) are
often used synonymously, but there is a distinction. Toxicogenomics focuses on the genome
as a whole, while toxicogenetics focuses on variants of individual genes. Toxicogenetics
investigations usually evaluate the influence of the variants of one or just a handful of
genes, rather than the whole genome. The distinction between these two terms is appar-
ent in their applications.
The discipline of toxicogenomics grew out of pharmacogenomics in the 1990s as
pharmacogenomic tools began to be applied to toxicologic questions. The development of
microarray technologies at that time enabled high throughput investigations of the effects
of exposures on gene expression. This allowed the field of toxicogenomics to move forward
rapidly.1 Toxicogenetics has been a part of clinical medicine for some time, although it was
not until recently that it was given that name. One of the first examples of the applica-
tion of toxicogenetic concepts to medicine was the discovery during the Korean War that
soldiers of certain ethnic backgrounds developed severe hemolysis when given the anti-
malarial primaquine.2 Deficiencies in the activity of the glucose 6-phosphate dehydro-
genase (G6PD) enzyme caused this susceptibility to hemolysis. Originally, G6PD activity
in patient red blood cells was measured to predict the amount of risk for this adverse
response to primaquine, whereas now genotyping can be done to predict this risk.
Toxicogenomics is used with the goal of understanding environmental exposures’
deleterious effects as well as the underlying causes of medications’ adverse effects. Toxico-
genomics is, therefore, very pertinent to the field of pharmacy. It is emerging as a tool in
both drug development and medicine. In drug development, toxicogenomic methods are
used to investigate the mechanisms and to predict the toxicities of medications. Toxico-
genetics is applied in clinical medicine to identify patients at risk for developing adverse
drug reactions based on their genotypes.
The following sections of this chapter will discuss the tools utilized in toxico-
genomics and toxicogenetics and the use of toxicogenomics in drug development.
Chapter 12 • Toxicogenomics 359
Examples will be provided of current and potential applications of toxicogenetics in

clinical pharmacy practice.
CASE STUDY—TOXICOGENOMICS OF ABACAVIR

A.B. is a 45-year-old Caucasian male who is positive for human immunodeficiency virus
(HIV+). Clinicians wish to prescribe abacavir for A.B. for a second time. A.B. has no other
underlying disease states. The patient tolerated abacavir in the past, showing no signs
of hypersensitivity. Although this medication is one that caregivers feel is best to use
in treating this patient, A.B. is quite anxious about taking the medication again. He read
that it is possible to develop a severe hypersensitivity reaction to this drug, even though
it was previously well tolerated. As a pharmacist involved in this patient’s care, what
recommendations would you make in an effort to help A.B. decide whether or not to take
abacavir again?
Questions
1. What genetic variant should A.B. be genotyped for in an effort to determine his risk for devel-
oping abacavir-induced hypersensitivity? You discuss with A.B. what is currently known about
the association between a patient’s genotype and his risk for hypersensitivity to abacavir.
2. What is the rationale for screening patients for this polymorphism before the administration
of abacavir?
3. If the patient is homozygous for the wild-type allele, may it be assumed with certainty he will
not develop hypersensitivity to abacavir?
Answers
1. HLA-B*57:01. It is known that patients who carry the HLA-B*57:01 variant have an increased
risk for developing a life-threatening hypersensitivity to abacavir. All patients initiating abaca-
vir therapy or those who have been on it before, if their HLA-B genotype is unknown, should
be evaluated for their HLA-B status. Abacavir is not recommended for patients with the HLA-
B*57:01 genotype.3,4
2. Screening patients for this polymorphism will identify patients very likely to develop abacavir
hypersensitivity. If a patient carries an HLA-B*57:01 allele, clinicians should be extremely cau-
tious if they choose to administer this drug to such patients, as the presence of that allele
puts the patients at significant risk of suffering a serious adverse drug effect.3,4
3. No. Patients who carry the HLA-B*57:01 variant have a significantly greater risk of develop-
ing abacavir hypersensitivity compared to patients who do not carry HLA-B*57:01. However,
patients who are homozygous wild type should be monitored for signs of abacavir hypersen-
sitivity even though they do not have a known genetic risk factor predisposing them to de-
veloping abacavir hypersensitivity.3,4 There is a small incidence of hypersensitivity occurring in
patients who do not have the HLA-B*57:01 genotype. These patients have as yet undiscovered
risk factors for abacavir hypersensitivity.
TOOLS OF TOXICOGENOMICS AND TOXICOGENETICS

The development and implementation of tools used in toxicogenomics and toxicogenetics

investigations have changed rapidly over time, and they will probably continue to do so as
technologies evolve. Their use has sometimes been challenging as scientists struggle with
choosing the best technological ways to approach the questions they wish to address and
how best to analyze the results produced when using these tools. In general, these tools
provide data to evaluate changes in gene expression, protein production, or metabolite

production as a result of an exposure to a toxic substance. On the flip side, the data may
also be used to evaluate how different genotypes affect responses to exposures. Two
general approaches are taken. The first involves comparison of simultaneous changes in
gene expression, protein production, or metabolite production as a result of an exposure.
The second involves evaluation of sequential changes in gene expression, protein produc-
tion, or metabolite formation as a result of an exposure and allows for an investigation
into biological pathways of response. Both approaches provide a means of describing the
genomic response and investigating the mechanism of a toxicological response.5,6 The
data produced may be used to classify unknown compounds and predict the toxicity of
new compounds, as well as to investigate mechanisms of toxicity, and predict the risk of
patients developing adverse effects from exposures. Toxicogenomics is currently being
utilized in the environmental and occupational health fields, in various chemical industries,
in the pharmaceutical and biopharmaceutical industries, and in medicine.
One of the most significant tools in toxicogenomics to date is the microarray. There
are several kinds of array technologies, with deoxyribonucleic acid (DNA) arrays being the
first to become available. DNA arrays include gene expression arrays, genotyping arrays,
and identifying arrays. These arrays each have different applications in toxicogenomics.
Gene expression arrays measure the relative changes in gene expression as a result of an
exposure in cultured cells or the tissues of an exposed organism compared with the gene
expression patterns of cells or tissues prior to exposure. These changes in gene expression,
when measured as changes in messenger ribonucleic acid (mRNA) expression, are called
transcript profiles. Studies using transcript profiles may be designed to evaluate gene
expression changes across different doses, across exposures at different times in growth or
development, for different lengths of time, or across different species. With the develop-
ment of arrays that can measure the response of thousands of genes and, in some cases,
the entire genome of various organisms, the potential to investigate toxic responses of a
huge number of genes at the genome level and across a variety of animals has evolved.
Genotyping arrays are another DNA array technology commonly used in toxicogenom-
ics. Genotyping by detecting single nucleotide polymorphisms (SNPs), alternative splice
variants, or chromosome copy number may be done using SNP arrays, alternative splice
arrays, or comparative genomic hybridization arrays, respectively. Assays used to identify
contaminating organisms in foods, feeds, or water represent a third type of DNA array
commonly used in toxicogenomics. Many contaminants may be identified by their unique
genetic fingerprint.
Toxicogenomic data from DNA microarray studies may be complemented by proteomics
and metabolomics studies. Proteomics is the study of protein expression and activity,
whereas metabolomics is the study of metabolite production. Proteomics and metabo-
lomics may also be used in toxicity studies to evaluate the effects of an exposure. For
a detailed discussion of proteomics and metabolomics methodologies, see reviews by
Boguski and McIntosh7 and Kaddurah-Daouk et al.8
Other techniques used in toxicogenetics can characterize an individual’s genetic make-

up. Indirect determinations of a patient’s genotype involve using a surrogate marker of the
genotype, such as determining the kinetics of an exogenously administered probe drug in a
patient or measuring enzyme activity in accessible tissues such as blood. These techniques
identify a patient’s phenotype. Some phenotypes are known to correlate to a particular
genotype, thus identifying the patient’s genotype indirectly.
Direct genotyping of patient DNA may be done using a number of techniques such as
restriction fragment length polymorphism assays, Sanger biochemistry in semi-automated
capillary-based assays, or the high-throughput hybridization genotyping microarray
presented above. The use of microarrays for genotyping allows for a large number of
alleles to be identified at one time for one patient or for many patients to be genotyped
at the same time. However, next-generation DNA sequencing methods may replace micro-
arrays for high-throughput genotyping. These next-generation DNA sequencing techniques
such as cyclic-array sequencing methods may prove to be less costly and perhaps more
accurate compared to microarray genotyping methods. A comprehensive discussion of
these techniques is beyond the scope of this chapter, but information about them may be
found in recent reviews.9-11
APPLICATION OF TOXICOGENOMICS TO DRUG DEVELOPMENT

Toxicogenomics is opening the door to advancing our knowledge of toxicity mechanisms
and may provide tools to identify potentially hazardous compounds more quickly. This is
important for compounds used in industry, those that are environmental contaminants, and
for the development of new medications. In drug development, traditional toxicology stud-
ies are conducted on potential new medications as part of the drug development process.
Figure 12-1 illustrates where toxicogenomics could be incorporated into drug development.
Traditionally, a variety of toxicity assessments are conducted on these new drug enti-
ties depending on their anticipated toxicity. These assessments include the use of in vitro
assays as well as testing compounds for toxicity in animals (animal bioassays). These stud-
ies provide data for predicting toxicities and early identification of general mechanisms of
toxicity. These studies are typically done early in the drug development process. Animal
bioassays are also used late in the drug development process to thoroughly investigate
the mechanisms of the toxicities observed in the earlier assays and to predict long-term
effects, such as the potential for carcinogenicity.
The process of identifying and understanding the potential toxicities of a new drug
takes time, is very expensive, may require relatively large amounts of the new compound,
and often requires large numbers of animals. For these reasons, alternative methods for
evaluating the toxicities of a new drug are being proposed. By providing data that better
identify successful candidates initially, the tools of toxicogenomics may help reduce costs
of toxicity testing in the drug development process (i.e., by reducing the number of possible
new drugs run through the entire development process).
Gene expression data are being used to predict the toxicity of drugs and to assist in the
identification of mechanisms of toxicity. When using gene expression microarrays to predict
drug toxicity, an assumption is made that structurally similar drugs or drugs with the same
mechanisms of toxicity will have similar effects on gene expression profiles. The transcript
profiles of novel compounds can be compared to the transcript profiles generated for
known compounds. The databases of transcript profiles of known compounds must be
well characterized with regard to the toxicological and pathological responses they cause
in biological systems for this comparison to be valid and useful.

This approach to the classification of unknown compounds, the prediction of
compound toxicity, and the investigation of mechanisms of toxicity have been validated.
Waring et al. showed that compounds with similar toxicity mechanisms produced similar
gene expression profiles in transcription microarrays in rat hepatocytes treated for 24 hours
Drug target selection and validation

Application of omics tools and informatics to determine the precise molecular mechanism of a
disease. Select and validate the most important drug target(s).
Computer-aided drug design

Select the best drug candidates for preclinical study. Cheminformatics (toxicogenomics) will
Number of candidates and cost involvement

help to determine the possible toxicities of drug candidates based on chemical structure.
Preclinical study (In vitro)
Drug safety
Use of omics tools and bioinformatics (toxicogenomics) to evaluate the toxicity profiles of
those best candidates (found in the computer-aided drug design phase). Select the best of the
best candidates based on in vitro risk:benefit ratio.
Preclinical study (In vivo)

candidates (found in vitro preclinical studies). Select the best candidates based on in vivo
risk:benefit ratio. Determine the PK, PD, initial human dose and ADME for selected best
candidates.
Clinical study: phase I

best candidates (found in vivo preclinical study). Select a few best candidates for new
clinical trials (Phase II and III).
FIGURE 12-1 Toxicogenomics in drug discovery. ADME = absorption, distribution,

metabolism, and elimination; PD = pharmacodynamic; PK = pharmacokinetic.
Source: Reprinted from Khan SM, Baghdasarian, A, Fahlman, RP, et al. Current status and
future prospects of toxicogenomics in drug discovery. Drug Discovery Today. 2014;19(5):562-
578. © 2014 with permission from Elsevier.
with known hepatotoxins.12 Waring et al. also demonstrated that the gene expression in
rats’ livers treated with known hepatotoxins correlated with the histopathology and clini-
cal chemistry parameters measured in those treated rats.13 This second study by Waring,
which is quite important, showed that changes in gene expression as a result of exposures
and measured by microarrays are valid when compared to traditional measures of toxicity
such as histopathology and clinical chemistry. Hamadeh et al. showed that animals treated
with agents in the same pharmacological class (e.g., all animals were treated with peroxi-
some proliferators) had very similar gene expression profiles. They also demonstrated that
animals treated with drugs of differing chemical classes (e.g., peroxisome proliferators
versus enzyme inducers) have very different gene expression profiles.14,15 Table 12-1 lists the
compounds tested in the studies by Waring and Hamadeh mentioned above.

For microarrays to be useful in predicting toxicity mechanisms or classification of
unknown compounds into toxic categories, the transcript profiles of the test compounds
need to be compared to the transcript profiles of model compounds that have been well
characterized. Hamadeh et al. illustrated that unidentified compounds could be identified
Table 12-1
Compounds Used in Microarray Validity Testing
Compounds Tested References
Known Hepatotoxins 12, 13

Allyl alcohol
Amiodarone
Aroclor 1254
Arsenic
Carbamazepine
Carbon tetrachloride
Diethylnitrosamine
Dimethyl-formamide
Diquat
Etoposide
Indomethacin
Methapyrilene
Methotrexate
3-Methylcholanthrene
Monocrotaline
Peroxisome Proliferators 14, 15
Clofibrate
Wyeth 14,643 Gemfibrozil
Enzyme Inducer 14, 15
Phenobarbital
using comparisons of cDNA microarrays.15 In their study, the investigators created a database
of transcription profiles generated by exposing rats to compounds that are either enzyme
inducers or peroxisome proliferators. The expression profiles from these known compounds
were then compared to the expression profiles generated from the rats’ livers exposed to
unidentified compounds. The investigators were blinded to the identity of the unknown
compounds the rats were exposed to as they compared the gene expression profiles. The
unknown compounds were correctly identified using the transcription database of the
known compounds.15 The compounds investigated in this study are also listed in Table 12-1.
Several public and commercial databases of transcript profiles of model compounds
have been developed.16 Primary databases are those that publish in-house expression
data; secondary databases are those that publish in-house and collaborator data. Tertiary
databases are those that house data from unrelated parties meeting submission criteria.17
Table 12-2 lists the currently existing public transcript profile databases. Investigators may
submit their expression array data to these public databases for the benefit of the scientific
community. Data submitted to these databases must meet the quality criteria listed in the
Minimal Information About Microarray Experiments (MI-AME) standards.16,19
Microarrays will likely improve the ability to predict drug toxicity by streamlining the
drug development process. Data from expression arrays showing the effects of a drug
exposure on genomic expression may identify drugs early in the drug developmental
process that are likely to fail due to toxicity. Such early identification of potential failures
would preclude the need to continue with more toxicity studies. Microarray expression
data may also be very useful in contributing to the early understanding of mechanisms of
toxicity and may help direct further, more traditional toxicity testing to investigate mecha-
Table 12-2
Public Transcript Profile Databases
Database Name Publisher Website References
Gene Expression National Center for http://www.ncbi.nlm.nih.gov/geo 16, 17

Omnibus Biotechnology
(NCBI GEO) Information (NCBI)
ArrayExpress European Molecular http://www.ebi.ac.uk/arrayexpress 16, 18

Biology Laboratory—
European
Bioinformatics
Institute (EMBL-EBI)
nisms of toxicity or may reduce the need for these studies. In a review of the usefulness of
toxicogenomics in the pharmaceutical industry, Lühe et al. suggest that toxicogenomics is
already able to predict the toxicities of many compounds, especially those that are hepa-
totoxins or nephrotoxins.16
Another application of pharmacogenomics in drug development is the potential to limit
drug toxicity in volunteers during clinical trials. Proven biomarkers of toxicity generated from
transcript profiles can be used to identify subjects at risk for known adverse reactions and,
thus, prevent these subjects from participating in studies that might be dangerous to them.
During drug development, the pharmaceutical industry is already realizing some of these
anticipated advantages of microarrays. These advantages, however, are offset by several
disadvantages. The cost effectiveness of toxicogenomics in predicting toxic events may be
limited to those toxicities that are usually seen after long-term exposures, such as the devel-
opment of cancer. Furthermore, the current difficulties in interpreting the large amounts of
data generated from microarray studies limits our ability to understand the results of multi-
ple exposure expression studies or compare toxic effects across various species.16
Proteomics and metabolomics techniques will also be useful in the drug development
process by providing data that complements data generated from microarray and tradi-
tional toxicology studies. These techniques will contribute to investigations of mechanisms
of drug toxicity and drug toxicity prediction, but at this time their use is limited by the large
amount of data generated from these methodologies. Techniques for analyzing these data
are still being developed. If this challenge of data analysis is overcome, the integration
of genomics, proteomics, and metabolomics is likely to lead to a better understanding of
adverse drug responses.
CURRENT AND POTENTIAL APPLICATIONS OF

TOXICOGENETICS TO CLINICAL PRACTICE
Pharmacogenomics is proving useful in drug development and toxicity screening, as

discussed. Pharmacogenetics evaluates the influence of individual gene variants on the
response to a drug and is also useful in the individualization of medicines. Several exam-
ples illustrate advances in the knowledge of genetic variants that have identified relation-
ships between these variants and the risk for drug toxicity. Genetic polymorphisms may
increase the adverse effects of drugs either by increasing the potential for side effects of
a drug used alone or by increasing the likelihood of drug–drug interactions when drugs are
used simultaneously. Genes that influence drug toxicity have been generally grouped into
one of three categories: those that code for drug-metabolizing enzymes, those that code
for transporters, and those that code for human leukocyte antigens (HLAs).20 The variants
of drug-metabolizing enzymes and transporters may affect the pharmacokinetics of vari-

ous drugs. When polymorphisms decrease the activity of metabolizing enzymes, biologi-
cally active parent drugs or metabolites may accumulate and thus reach toxic levels. Two
such examples are the polymorphisms found in cytochrome P450s (CYPs) and in gluta-
thione S-transferase (GST). Several of the known polymorphisms in these genes are well
characterized regarding their influence of drug pharmacokinetics.
Attempts to discover correlations between genotypes and the risk of developing drug
toxicities are pursued with the hope that the information found may be clinically useful.
Understanding the relationship between genotype and risk for toxicity may aid in identify-
ing patient populations whose polymorphisms put them at relatively higher risk for drug
toxicities compared to populations without those polymorphisms. This information may
help at-risk patients avoid using drugs that could prove harmful to them. This information
may also be useful in determining appropriate doses for patients with varying drug toxicity
risks if their genotype allows for limited use of the drug.
However, not all correlations between genotype and risk of drug toxicity identified so
far are clinically utilized at this time. Other variables may complicate and confound the
clinical usefulness of known correlations between genotypes and drug response. Different
populations may have different genetic variants associated with the same adverse drug
response. A monogenetic risk for drug toxicity in the first population, and an unidentified
polygenetic or other multifactorial risk of drug toxicity in the second population may cause
this discrepancy across populations. Genetic biomarkers that have potential as predictive
risk factors of adverse drug effects need extensive validation before they may be clinically
useful. Currently, attempts have been made to develop guidelines for drug dosing that
incorporate polygenic traits.
There are some medications for which the known relationship between genotype and
risk of toxicity is being utilized clinically in an attempt to reduce the incidence of severe
adverse effects. Documents recently published focus on pharmacogenomics and toxi-
cogenomics, which are useful to clinicians wishing to understand the known and clinically
relevant associations between phenotype and genotype. One such document is the Table
of Pharmacogenomic Biomarkers in Drug Labeling from the U.S. Food and Drug Administra-
tion (FDA).21 This document lists all the drugs on the U.S. market with labeling that includes
information on genomic biomarkers. Of course, patients for whom a drug may be prescribed
are the usual subject of the genotyping, especially when testing is performed to avoid drug
toxicities. In some instances, however, genotyping may be done on an infectious agent such
as the HIV virus or cells of a tumor to determine the virus’ or the tumor cells’ suscepti-
bility to a medication. This table provides information that will allow clinicians to decide
whether genotyping is warranted—to ensure efficacy or avoid adverse drug responses. The
table lists the drugs that have genotyping in their labeling but does not specify whether
genotyping is mandated.
Another resource is a series of peer-reviewed, evidence-based medicine guidelines for
using pharmacogenomics in clinical practice from the Clinical Pharmacogenetics Imple-
mentation Consortium (CPIC) of the Pharmacogenomics Research Network. These guid-

ance documents are available for several medications, some of which can cause significant
toxicities in patients and for which the understanding of the influence of a genotype on
the phenotype is well understood. These guidance documents provide information to
clinicians on whether or how these medications should be used in patients depending on
their genotype. An explanation of how the evidence is weighted can be found in an article
by Relling and Klein.22 A list of the documents published to date and in-depth information
on pharmacogenomics can be found on the website PharmGKB (www.PharmGKB.org).23 A
variety of medications included in this series of documents have toxicities associated with
genetic variants including anticoagulants, chemotherapeutics, antimicrobials, and anti-
convulsants. Below is a discussion of several medications where using a patient’s genotype
may decrease the incidence of adverse effects.
Warfarin
Warfarin is a narrow therapeutic index drug used as an anticoagulant, and the most
common adverse effect is an increased risk for bleeding events. With warfarin, it is often
difficult to determine the starting dose in some patients and then to maintain their anti-
coagulant levels appropriately. At this time, it is known that variants of the warfarin-metab-
olizing enzymes CYP2C9 and vitamin K epoxide reductase complex 1 (VKORC1) contribute
to the variability in required starting and maintenance doses for patients. Several studies
have found that patients with the lower activity variants of the metabolizing enzymes
CYP2C9, CYP2C9*2, and CYP2C9*3, are at higher risk for over-anticoagulation or bleeding
complications on this medication compared to patients with normal activity CYP2C9.24-26
The gene for VKORC1 also has several polymorphic alleles, and the 1173C/T variant has
been found to be significantly associated with dose variability and to contribute to the risk
of patients experiencing over-anticoagulation or bleeding events too.27 It is now known
that not all the variability of warfarin dosing needs are accounted for by the *2 and *3 vari-
ants of the CYP2C9 gene and the 1173C/T variant of the VKORC1 gene. A current discovery
shows that more recently identified genetic variants of CYP2C9 influence dosing needs in
some populations.28 The CPIC guidance document on warfarin dosing based on the *2 and
*3 variants of CYP2C9 and the 1173C/T variant of the VKORC1 gene is available to help
clinicians initiate warfarin therapy, reducing the risk of adverse bleeding events.29
CLINICAL PEARL
Most adverse drug effects will be polygenetic rather than monogenetic in
nature, complicating our understanding and use of pharmacogenomic
information.
Primaquine
A deficiency in G6PD puts patients at risk for developing severe, life-threatening oxidative
hemolytic anemia when exposed to certain medications. A patient may have a G6PD defi-
ciency because the gene coding for that enzyme has variants that decrease or eliminate
the enzyme’s activity. The most significant variants have the highest prevalence in people
of African, Mediterranean, or Asian ancestry. Patients with these backgrounds should be
evaluated for G6PD deficiency before they are administered certain drugs. During the Kore-
an War, the G6PD deficiency was found to be responsible for the severe hemolytic anemia
suffered by some soldiers with the use of primaquine. Other medications that can cause
hemolytic anemia in the presence of G6PD deficiency include the antimicrobial dapsone
used to treat leprosy and rasburicase used to treat hyperuricemia in cancer patients. The
FDA currently recommends that patients be evaluated for their G6PD status before they
are given rasburicase and dapsone.30,31
Carbamazepine
Carbamazepine is a medication used primarily as an anticonvulsant. Very serious derma-
tological reactions that include Stevens-Johnson Syndrome (SJS) and toxic epidermal
necrolysis (TEN) can occur in some patients taking this medication. An association has been
found between the risk of developing SJS and TEN and the presence of a genetic variant
of HLA-B, HLA-B*15:02.32-34 This association appears to be most prevalent in patients of
Asian ancestry.35 The FDA has included information in carbamazepine labeling about the
significance of this association to prompt clinicians to genotype patients for their HLA-B
genotype and help to prevent the life-threatening adverse effects this drug can cause.
Patients who have been identified as having at least one allele for HLA-B*15:02 should
not be treated with carbamazepine unless the benefits clearly outweigh the risks.36,37
The FDA cautions, however, that the usefulness of genotyping patients based on ethnic-
ity before the administration of carbamazepine is limited due to the variability in the
rates of HLA-B*15:02 prevalence in ethnic groups.36 Those patients who do not carry the
HLA-B*15:02 allele may still develop carbamazepine-associated severe cutaneous adverse
effects.38 Therefore, HLA-B*15:02 is not a universal genetic biomarker for this occurrence of
these adverse effects. Efforts to find other genetic biomarkers linked to the development
of carbamazepine-induced SJS and TEN are continuing.
CLINICAL PEARL
Clinicians cannot assume that a polymorphism associated with an
adverse drug effect in one population will also be associated with that
adverse drug effect in another population.
Abacavir
Another variant of the HLA-B gene, HLA-B*57:01, has been found to be associated with
a severe hypersensitivity reaction to abacavir. This hypersensitivity reaction is not the
SJS or TEN seen with carbamazepine that is linked to the HLA-B*15:02 variant, but it is
a clinical syndrome involving multiple organs. Abacavir is a reverse transcriptase inhibi-
tor used to treat HIV. Mallal et al. found that prospectively screening HIV patients for
HLA-B*57:01 reduced the incidence of hypersensitivity reactions to abacavir.39 Patients
should be screened for their HLA-B*57:01 status before using abacavir for the first time, and
in patients who have taken and tolerated the medication before but whose HLA-B*57:01
status is not known. Abacavir should not be administered to any patient that is HLA-B*57:01
positive.3,4 A negative finding for the presence of HLA-B*57:01 does not mean that a patient
will not develop sensitivity to the drug, but the chances are much smaller compared to
patients who carry a HLA-B*57:01 allele.
CLINICAL PEARL
Clinicians should not assume the absence of an allele known to be
associated with an adverse drug effect in a patient guarantees that the
adverse effect will not occur.
6-Mercaptopurine
6-Mercaptopurine (6-MP) is an antimetabolite antineoplastic agent used mostly to treat
various leukemias, especially acute lymphoblastic leukemia. It is an active metabolite of the
prodrug azathioprine, an antimetabolite used primarily to prevent transplant rejection, rheu-
matoid arthritis, and inflammatory gastrointestinal diseases. 6-MP is metabolized to cyto-
toxic metabolites that cause significant adverse effects. 6-MP is inactivated in part by thiol
methylation catalyzed by the polymorphic thiopurine S-methyltransferase (TPMT) enzyme.
Low-activity variants of the TPMT enzyme have been identified. Patients who are genetic
heterozygotes for the low-activity variant of TPMT have intermediate levels of TPMT activity,
while those who are homozygotes for the low-activity allele have low or no TPMT activity.
Decreased TPMT activity puts patients at risk for developing significant myelotoxicity
with azathioprine and 6-MP administration, because high levels of 6-MP are then avail-
able for metabolism to the cytotoxic metabolites that accumulate. TPMT-deficient homo-
zygotes have a serious risk of developing life-threatening myelotoxicity on conventional
doses of azathioprine and 6-MP.40 TPMT genotyping has been found to improve the predic-
tion of hematologic side effects of these drugs.41,42 Therefore, a patient’s TPMT genotype
or phenotype status should be determined before administering azathioprine and 6-MP to
prevent this life-threatening side effect.43-45
Irinotecan
Irinotecan is an antineoplastic agent with a mechanism of action due to its inhibition of topoi-
somerase I. Irinotecan is metabolized to an active metabolite SN-38 primarily via hepatic
carboxylesterase activity. Both the parent drug and SN-38 bind to topoisomerase I after it
complexes with DNA, preventing repair of the single-strand breaks caused by topoisomerase
I. This complex of irinotecan or SN-38 with the topoisomerase and DNA is cytotoxic, which is
both its therapeutic mechanism of action and the mechanism for its toxicity. SN-38 is gluc-
uronidated by the polymorphic enzyme uridine 5’-diphospho (UDP)-glucuronosyltransferase
1A1 (UGT1A1), allowing it to be eliminated eventually from the body. A low-activity variant of
UGT1A1, UGT1A1*28, has been identified and is associated with an increased risk of cancer
patients developing severe neutropenia during irinotecan treatment due to increasing levels
of SN-38.46,47 The FDA recommends genotyping patients for their UGT1A1 status and reducing
the irinotecan starting dose for patients homozygous for the low-activity alleles.48
The aforementioned examples show current clinical applications of toxicogenetics to
prevent adverse drug responses. Identifying patients at risk for developing severe adverse
effects to some medications based on their genotype is very useful when the association
between the genotype and adverse effect has been well characterized. There are many
associations between genotype and drug response that are not yet fully characterized but
may be useful in the future. For example, genetic variants of CYP2C9 may significantly alter
the pharmacokinetics of medications used to treat cardiovascular disease, nonsteroidal anti-
inflammatory drugs, and hypoglycemic agents as well as warfarin (discussed previously).
The utility of genotyping patients receiving these medications for their CYP2C9 status to
prevent adverse effects has not been unequivocally proven. Perhaps with more characteriza-
tion of the associations between these genotypes and the adverse effects caused by these
drugs, genotyping will prove useful in preventing adverse effects of these particular drugs.49
Besides the significance of polymorphisms in drug-metabolizing enzymes, polymor-
phisms in genes coding for drug transporter enzymes such as the multidrug resistance
transporter 1 (MDR1) may also prove to be clinically significant in the development of
adverse drug reactions. Again, genotyping patients for known MDR1 variants is not yet vali-
dated to the extent to be clinically useful, but it may be in the near future.50
Another very interesting yet potential use of pharmacogenomics in the clinical setting
is evaluating biomarkers for identifying idiosyncratic adverse reactions. The International
Serious Adverse Events Consortium (SAEC) led by the FDA was formed in 2007 to further
this application of pharmacogenomics. Efforts are being made to identify genetic variants
that may be predictive of these rare, non-dose related but potentially serious adverse reac-
tions to medications. The SAEC is currently working to identify variants that may have a
role in the idiopathic development of serious hepatic and dermal adverse drug responses.51
SUMMARY
Regulatory agencies, academic institutions, and industry groups have joined forces in the
form of various consortiums, centers, or funding agencies to promote the use, validation,
and effectiveness of toxicogenomic and toxicogenetic technologies and their applica-
tions. One agency working on it is the National Institute of Environmental Health Sciences
(NIEHS), which created the National Center for Toxicogenomics (NCT) in September 2000.
The mission of the center was “to promote the evolution and coordinated use of gene
expression technologies and to apply them to the assessment of toxicological effects in
humans.”52 Its primary goal was to create a reference system of human gene expression
data and develop a database of chemical effects in biological systems. Its second goal was
to gain more understanding of mechanisms of toxicity.52 NCT is no longer in existence in its
original form but has been redistributed throughout NIEHS.
While the NCT was in its original form, to meet its stated goals, the NCT formed the
Toxicogenomics Research Consortium (TRC) in 2000. This consortium was formed and fund-
ed by the NIEHS Division of Extramural Research and Training (DERT) with the solicitation of
applications for extramural researcher participation. Selected participants coordinated their
research efforts in toxicogenomics investigations and had the support of the NIEHS extra-
mural staff and NIEHS NCT as well as access to NCT-supported resource contractors. The
majority of the researchers involved in the consortium focused their toxicogenomics efforts
in the field of environmental health. The goals for the TRC included the “enhancement of
research in the broad area of environmental stress responses using microarray gene expres-
sion profiling; development of standards and practices that will allow analysis of gene
expression data across platforms and provide an understanding of intra- and interlaboratory
variation; contribute to the development of a robust relational database that combines toxi-
cological endpoints with changes in gene expression profiles; improve public health through
better risk detection, and earlier intervention in disease processes.”53
Parallel to the efforts of the TRC is a consortium of pharmaceutical companies under
the coordinating guidance of the Health and Environmental Sciences Institute of the
International Life Sciences Institute. Members of this group are evaluating the world-wide
harmonization efforts of gene expression data and analysis.54
The NIEHS has developed databases as tools for storing, accessing, and comparing
data generated by pharmacogenomics studies. NIEHS has developed the toxicogenomics
data repository—Chemical Effects in Biological Systems (CEBS)—that was the original
mission of the NCT. This repository is publicly accessible and includes information on study
design, clinical chemistry, and histopathology data associated with the study subjects as
well as microarray and proteomics data generated in the submitted studies.55 Access to
CEBS can be gained by going to http://cebs.niehs.nih.gov. A second data repository has
also been developed by the NIEHS in the Mount Desert Island Biological Laboratory, a
Marine and Freshwater Biomedical Science Center of the NIEHS. This data repository, the
Comparative Toxicogenomics Database, allows scientists to investigate the interactions
among chemicals, genes, proteins, and diseases thought to be influenced by environmental
exposures. Although these databases focus on the adverse effects of environmental expo-
sures, they are still of interest to healthcare providers because the patient environment
will influence the etiology of many diseases as well as the effectiveness or risk of their
pharmacological treatment.
Another key government agency that is coordinating efforts to develop toxicogenomics
and toxicogenetics is the FDA, which created the National Center for Toxicological Research
(NCTR) to conduct research and provide technical advice and training to FDA scientists.
These services are provided to assist the FDA in making science-based regulatory decisions
with the goal of improving the health of the American public. NCTR research focuses on
understanding mechanisms of toxicity and the further development of new technologies
for assessing human exposure, susceptibility to disease, and risk.
Besides providing leadership in the development of toxicogenomics and toxicogenet-
ics, the FDA is encouraging the pharmaceutical and biopharmaceutical industry to submit
pharmacogenomics data with investigational new drug applications, new drug applica-
tions, and biologic license applications. Their guidance documents, Guidance for Industry:
Pharmacogenomic Data Submissions, delineates when pharmacogenomics and pharma-
cogenetics data must be submitted to the agency and when it may be voluntarily submit-
ted.56-58 At this time, there is concern in the pharmaceutical and biopharmaceutical industry
regarding how toxicogenomics data will be interpreted and used in the drug development
process. The FDA and companies developing drugs and biologicals are making efforts to
understand the use of toxicogenomics in the drug development process.58
The importance of toxicogenomics is exemplified by the many efforts of scientists,
clinicians, and government agencies to discover, validate, and use toxicogenomics and toxi-
cogenetics in drug discovery, toxicity prediction and understanding, and the personalization
of medicine. Although limitations still exist in our ability to use the information generated
by toxicogenomics investigations, there is promise in its eventual (more complete) utility.
However, as we continue learning about the influence of genetic variants on drug toxicity
and response, the application of this information will likely become more difficult because
most adverse responses to medications, when influenced by genomics, will rarely be mono-
genetic. These types of examples will likely be limited when compared to the numbers of
adverse drug reactions that will prove to be polygenic. The adverse effects of medications
will also likely be found to influence the effects of the medications themselves on multi-
ple genes, as well as by other environmental factors influencing the expression of those
same genes. This moves toxicogenomics into the realm of another new discipline called
systems biology. This science is attempting to look at how gene–gene interactions and
gene–environment interactions influence an organism’s responses to exposures to either
environmental factors or medications.59
It is important that pharmacists keep current with the evolution of toxicogenomics and
toxicogenetics because professionals in the field of pharmacy are involved in drug develop-
ment, investigations of adverse effects, and the safe use of medications in clinical practice.
Familiarity and frequent perusal of medical literature focusing on toxicogenomics and
toxicogenetics investigations and new clinical applications is necessary for the practitioner
who wishes to keep abreast of this rapidly evolving field.
REFERENCES
1. Weber WW. Toxicogenomics: history and current applications. ASM News. 2004;70:364-370.
2. Cappellini MD, Fiorelli G. Glucose-6-phosphate dehydrogenase deficiency. Lancet. 2008;371:64-74.
3. Ziagen® (package insert). Research Triangle Park, NC: GlaxoSmithKline; March 2015.
4. Martin MA, Klein TE, Dong BJ, et al. Clinical Pharmacogenetics Implementation Consortium Guidelines
for HLA-B genotypes and abacavir dosing. Clin Pharmacol Ther. 2012;91(4):734-738.
5. Schmidt CW. Toxicogenomics. Environ Health Perspect. 2002;110:A750-A755.
6. Hayes KR, Bradfield CA. Advances in toxicogenomics. Chem Res Toxicol. 2005;18:403-414.
7. Boguski MS, McIntosh MW. Biomedical informatics for proteomics. Nature. 2003;422:233-237.
8. Kaddurah-Daouk R, Kristal BS, Weinshilboum RM. Metabolomics: a global biochemical approach to
drug response and disease. Ann Rev Pharmacol Toxicol. 2008;48:653-683.
9. Shendure J, Mitra RD, Varma C, et al. Advanced sequencing technologies: methods and goals. Nature
Reviews Genetics. 2004;5:335-344.
10. Shendure J, Hanlee J. Next-generation DNA sequencing. Nat Biotechnol. 2008;26:1135-1145.
11. Metzker ML. Sequencing technologies—the next generation. Nat Rev Genet. 2010;11:31-46.
12. Waring JF, Ciurlionis R, Jolly RA, et al. Microarray analysis of hepatotoxins in vitro reveals a correlation
between gene expression profiles and mechanisms of toxicity. Toxicol Lett. 2001;120:359-368.
13. Waring JF, Jolly RA, Ciurlionis R, et al. Clustering of hepatotoxins based on mechanism of toxicity
using gene expression profiles. Toxicol Appl Pharmacol. 2001;175:28-42.
14. Hamadeh HK, Bushel PR, Jayadev S, et al. Gene expression analysis reveals chemical-specific profiles.
Toxico Sci. 2002;67:219-231.
15. Hamadeh HK, Bushel PR, Jayadev S, et al. Prediction of compound signature using high density gene
expression profiling. Toxico Sci. 2002;67:232-240.
16. Lühe A, Suter L, Ruepp S, et al. Toxicogenomics in the pharmaceutical industry: Hollow promises or
real benefit? Mutat Res. 2005;75:02-115.
17. Edgar R, Domrachev M, Lash AE. Gene expression omnibus: NCBI gene expression and hybridization
array data repository. Nucleic Acids Res. 2002;30:207-210.
18. Brazma A, Parkinson H, Sarkans U, et al. ArrayExpress—a public repository for microarray gene expres-
sion data at the EBI. Nucleic Acids Res. 2003;31:68-71.
19. Brazma A, Hingamp P, Quackenbush J, et al. Minimum information about a microarray experiment
(MIAME)-toward standards for microarray data. Nature Genet. 2001;29:365-371.
20. Wilke RA, Lin DW, Roden DM, et al. Identifying genetic risk factors for serious adverse drug reactions:
current progress and challenges. Nat Rev Drug Discovery. 2007;6:904-916.
21. U.S. Food and Drug Administration. Table of Pharmacogenomic Biomarkers in Drug Labeling. Avail-
able at: http://www.fda.gov/drugs/scienceresearch/researchareas/pharmacogenetics/ucm083378htm.
Accessed August 26, 2015.

22. Relling MV, Klein TE. CPIC: Clinical Pharmacogenetics Implementation Consortium of the Pharma-
cogenomics Research Network. Clin Pharmacol Ther. 2011;89(3):464-467.
23. PharmGKB. The Pharmacogenomics Knowledgebase. https://www.pharmgkb.org. Accessed August 26,
2015.
24. Coumadin® (package insert). Princeton, NJ: Bristol-Myers Squibb Company; October 2011.
25. Momary KM, Shapiro NL, Viana MA, et al. Factors influencing warfarin dose requirements in African
Americans. Pharmacogenomics. 2007;8:1535-1544.
26. Lima MV, Ribeiro GS, Mesquita ET, et al. CYP2C9 genotypes and the quality of anticoagulation control
with warfarin therapy among Brazilian patients. Eur J Clin Pharmacol. 2008;64:9-15.
27. Limdi NA, Veenstra DL. Warfarin pharmacogenetics. Pharmacotherapy. 2008;28:1084-1097.
28. Drozda K, Wong S, Patel SR, et al. Poor warfarin dose prediction with pharmacogenetic algo-
rithms that exclude genotypes important for African Americans. Pharmacogenet Genomics. 2015
Feb;25(2):73-81.
29. Johnson JA, Gong L, Whirl-Carrillo M, et al. Clinical pharmacogenetics implementation consortium
guidelines for CYP2C9 and VKORC1 genotypes and warfarin dosing. Clin Pharmacol Ther. 2011; Oct
90(4):625-629.
30. Elitek® (package insert). New York, NY: Sanofi-Synthelabo Inc.; May 2015.
31. Aczone® (package insert). Fort Collins, CO: QLT USA; July 2015.
32. Chung WH, Hung SI, Hong HS, et al. Medical genetics: a marker for Stevens-Johnson syndrome.
Nature. 2004;428-486.
33. Hung SI, Chung WH, Jee SH, et al. Genetic susceptibility to carbamazepine-induced cutaneous
adverse drug reactions. Pharmacogenet Genomics. 2006;16:297-306.
34. Man CB, Kwan P, Baum L, et al. Association between HLA-B*1502 allele and antiepileptic drug-induced
cutaneous reactions in Han Chinese. Epilepsia. 2007;48:1015-1018.
35. Lonjou C, Thomas L, Borot N, et al. A marker for Stevens-Johnson syndrome: ethnicity matters.
Pharmacogenomics J. 2006;6:265-268.
36. Tegretol® (package insert). East Hanover, NJ: Novartis Pharmaceuticals Corporation; December 2007.
37. Leckband SG, Kelsoe1 JR, Dunnenberger HM, et al. Clinical pharmacogenetics implementation consor-
tium guidelines for HLA-B genotype and carbamazepine dosing. Clin Pharmacol Ther. 2013; Sep
94(3):324-328.
38. Kaniwa N, Saito Y, Aihara M, et al. HLA-B locus in Japanese patients with anti-epileptics and allo-
purinol-related Stevens-Johnson syndrome and toxic epidermal necrolysis. Pharmacogenomics.
2008;9:1617-1622.
39. Mallal S, Phillips E, Carosi G, et al. HLA-B*5701 screening for hypersensitivity to abacavir. N Engl J Med.
2008;358:569-579.
40. Eichelbaum M, Ingelman-Sundberg M, Evans WE. Pharmacogenomics and individualized drug therapy.
Ann Rev Med. 2006;57:119-137.
41. Relling MV, Hancock ML, Rivera GK, et al. Mercaptopurine therapy intolerance and heterozygosity at
the thiopurine S-methyltransferase gene locus. J Natl Cancer Inst. 1999;91:2001-2008.
42. Heckmann JM, Lambson EM, Little F, et al. Thiopurine methyltransferase (TPMT) heterozygosity and
enzyme activity as predictive tests for the development of azathioprine-related adverse events. J
Neurol Sci. 2005;231:71-80.
43. Imuran® (package insert). San Diego, CA: Prometheus Laboratories Inc; May 2008.
44. Purinethol® (package insert). Sellersville, PA: DSM Pharmaceuticals Inc; August 2003.
45. Relling MV, Gardner EE, Sandborn WJ, et al. Clinical pharmacogenetics implementation consortium
guidelines for thiopurine methyltransferase genotype and thiopurine dosing. Clin Pharmacol Ther.
2011; Mar 89(3):387-391.
46. Innocenti F, Undevia SD, Iyer L, et al. Genetic variants in the UDP-glucuronosyltransferase 1A1 gene
predict the risk of severe neutropenia of irinotecan. J Clin Oncol. 2004;22:1382-1388.
47. Marcuello E, Altés A, Menoyo A, et al. UGT1A1 gene variations and irinotecan treatment in patients
with metastatic colorectal cancer. Br J Cancer. 2004;91:678-682.
48. Camptosar® (package insert). New York, NY: Pfizer; December 2014.
49. Kirchheiner J, Seeringer A. Clinical implications of pharmacogenetics of cytochrome P450 drug
metabolizing enzymes. Biochimica et Biophyicas Acta. 2007;1770:489-494.
50. Zhou S, Di YM, Chan E, et al. Clinical pharmacogenetics and potential application in personalized
medicine. Curr Drug Metab. 2008;9:738-784.
51. Holden A. The innovative use of large-scale industry biomedical consortium to research the genetic
basis of drug induced serious adverse events. Drug Discover Today: Technologies. 2007;4:5-87.
52. Tennant RW. The National Center for Toxicogenomics: using new technologies to inform mechanistic
toxicology. Environ Health Perspect. 2002;110:A8-A10.
53. Medlin J. Toxicogenomics research consortium sails into uncharted waters. Environ Health Perspect.
2002;110:A744-A746.
54. Pennie W, Pettit SD, Lord PG. Toxicogenomics in risk assessment: an overview of an HESI collaborative
research program. Environ Health Perspect. 2004;112:417-419.
55. Waters M, Stasiewicz S, Merrick BA, et al. CEBS—chemical effects in biological systems: a public data
repository integrating study design and toxicity data with microarray and proteomics data. Nucleic
Acids Res. 2008;36:D892-D900.

56. U.S. Department of Health and Human Services; Food and Drug Administration; Center for Drug Evalu-
ation and Research (CDER); Center for Biologics Evaluation and Research (CBER); and Center for
Devices and Radiological Health (CDRH). Guidance for Industry: Pharmacogenomic Data Submissions.
Rockville, MD: Food and Drug Administration; March 2005.
57. Khan SM, Baghdasarian A, Fahlman RP, et al. Current status and future prospects of toxicogenomics in
drug discovery. Drug Discov Today. 2014;19(5):562-578.
58. Freeman K. Toxicogenomics data: the road to acceptance. Environ Health Perspect. 2004;112:A678-685.
59. Olden K. Toxicogenomics—a new systems toxicology approach to understanding of gene-environment
interactions. Ann NY Acad Sci. 2006;1076:703-706.
CHAPTER
13
The Pharmacogenetics of
Addiction
Patti W. Adams, PhD and Erika E. Tillery, PharmD, BCPP, CGP

After completing this chapter, the reader ADDICTION— formerly referred to as
should be able to substance dependence in the Diagnostic
Explain the role of reward pathways and and Statistical Manual of Mental
aversion centers in the neurophysiology Disorders, fourth edition (DSM-IV-TR),
a chronic, relapsing brain disease with
of drug dependence.
specific neurophysiology, course, and
Describe how neuroadaptations set of treatment. Dopamine is a key
contribute to development of drug neurotransmitter in addiction.
dependence and withdrawal symptoms.
DEPENDENCE—physiological dependence
Discuss the pharmacodynamic and occurring in response to repeated
pharmacokinetic consequences of administration of many types of drugs,
genetic polymorphisms and understand including drugs of abuse. An individual
how their effects contribute to the risk of may function normally only in the
drug dependence and addiction. presence of the substance, and absence
Describe the significance of of the substance leads to physical
pharmacogenomics in the treatment of disturbance and withdrawal.
addiction related to specific substance NEUROADAPTATION—molecular change that
use disorders including alcohol use occurs in neural pathways in response to
disorder, opioid use disorder, and tobacco the presence of a drug.
use disorder.
TOLERANCE—a need for markedly increased
Evaluate a patient case scenario amounts of the substance to achieve
and identify interventions based on intoxication or desired effect; a markedly
pharmacogenetic variables associated diminished effect is observed with
with medications used to treat a continued use of the same amount of
substance use disorder. substance.
WITHDRAWAL—a common reaction in long-
term drug users after the drug is no
longer administered. Characteristics of
the withdrawal syndrome vary by the
substance abused but generally include
marked physiological and psychological

disturbances; substance is taken to
relieve or avoid withdrawal symptoms.
375
INTRODUCTION
Substance abuse is a major public health problem in the United States and many parts of
the world. According to the 2013 National Survey on Drug Use and Health, an estimated
24.6 million people aged 12 or older in the United States report current (i.e., past month)
use of illicit drugs (e.g., marijuana/hashish, cocaine/crack, heroin, hallucinogens, inhalants,

or prescription medications used nonmedically).1 In 2013, more than half (52.2%) of Ameri-
cans aged 12 or older reported being current drinkers of alcohol, which translated to an
estimated 136.9 million individuals. An estimated 66.9 million Americans aged 12 or older
were current users of a tobacco product, which represented 25.5% of the population.1 Of
the nicotine products used, cigarettes were smoked most often followed by cigars and then
smokeless tobacco and pipes.
Although not all users of drugs of abuse will develop dependence, those who do
may have a lifelong struggle with the deleterious effects on their livelihood and quality
of life. In addition, many individuals with addiction may receive pharmacologic treatment
with minimal results. Emerging data on the pharmacogenomic aspects of both drugs of
abuse and medications used to treat substance use disorders indicate that polymorphisms
in genes may influence the pharmacodynamic and pharmacokinetic properties of drugs of
abuse as well as the medications used in the treatment of addiction.2 Such data empha-
size the importance of considering an individual’s genotype when treating substance use
disorders.
CASE STUDY−TOBACCO USE DISORDER

M.K. is a 32-year-old female who wants to quit smoking. She has a 15-pack/year history
and has smoked approximately 1 pack per day since the age of 17. M.K. attempted to quit
smoking several times in the past 5 years with the nicotine patch, gum, and lozenge,
but she has been unsuccessful each time. She would like to know about alternative
treatments and wishes to stop smoking for life.
Questions
1. What region of the brain plays an important role in the addicting and rewarding pathways of
nicotine?
2. What genetic polymorphisms exist that may alter treatment responses to nicotine replace-
ment therapy such as the nicotine patch, gum, and lozenge?
3. From which U.S. Food and Drug Administration (FDA)-approved medications for nicotine de-
pendence/tobacco use disorder would M.K. benefit?
THE NEUROPHYSIOLOGY OF DRUG DEPENDENCE

Determinants of Dependence Potential
The potential for a drug to cause dependence in an individual is affected by three factors:
1. The drug’s ability to activate reward pathways.

2. The drug’s ability to produce neuroadaptations with repeated administration.
3. The drug’s propensity for producing physical and emotional withdrawal symptoms when the
agent is discontinued.
Drugs of abuse exert their activities by binding to receptors and inducing an effect.
Typically the effect is mediated by activation or inhibition of a target receptor, followed
Chapter 13 • The Pharmacogenetics of Addiction 377
by an alteration of membrane permeability, enzymatic activity, or signal transduction

cascades. A common downstream effect of drugs of abuse is manipulation of central
reward and aversion pathways. Because this is dependent on the drug’s ability to access
the central nervous system (CNS), the rate of accumulation in and clearance from the CNS
compartment impact the intensity of the reward signal and withdrawal symptoms, respec-
tively. Therefore, genetic polymorphisms that affect any of these properties of drugs can
also impact an individual’s susceptibility to becoming dependent on a substance as well
as the likelihood of relapse during periods of abstinence.
Reward Pathways
Although drugs of abuse may act on a diverse array of pharmacological targets in multiple
brain regions, activation of dopaminergic pathways in the mesolimbic system is central to
the development of drug dependence and addiction. The mesolimbic system is formed
by multiple brain regions located at distinct locations in the brain. This system is largely
considered to originate in the ventral tegmental area (VTA), a small structure located in
the midbrain region of the brainstem (Figure 13-1). Neural projections extending from the
VTA form connections to other brain regions that are associated with cognition, learn-
ing, emotional memory, and reward-seeking behavior such as the nucleus accumbens,
prefrontal cortex, amygdala, and hippocampus.3,4 In general, activation of reward centers
in the mesolimbic system reinforces critical behaviors such as those related to reproduc-
tion, feeding, and exploration while activation of aversion centers suppresses adverse
behaviors.5,6 Reward and aversion centers are located throughout the mesolimbic system,
but reward centers tend to be highly concentrated in the nucleus accumbens, where they
produce a range of feelings from mild pleasure to euphoria. In contrast, aversion centers
tend to be more concentrated in the amygdala, where their activation is associated with
feelings of fear, sadness, or loss. Drugs of abuse promote addiction by essentially hijacking
the mesolimbic reward system. Withdrawal syndromes resulting from abrupt drug discon-
tinuation in a dependent patient are associated with depression of reward center activity,
activation of aversion centers, or a combination of these effects.
Neurons from the VTA project to and influence the activity of many other brain regions.
The VTA neurons are composed predominantly of dopaminergic neurons (approximately
60-65%), with most of the remaining neurons being GABAergic (approximately 30-35%),
and a small population of glutamatergic neurons. Activation of mesolimbic dopaminergic
neural projections extending from the VTA to the nucleus accumbens is associated with
activation of reward pathways in the brain and is a common event triggered by drugs of
abuse.7 However, more recent studies have shown that the populations of dopaminergic
neurons extending from the VTA are quite heterogeneous and are distinguished by differ-
ent gene expression patterns, electrophysiological properties, and co-transmitters secreted
(i.e., glutamate and GABA).6,8 These newer studies suggest that through distinct popula-
tions of dopaminergic neurons, the VTA may contribute regulatory effects on both reward
centers and aversion centers.
In addition to regulating multiple brain regions, VTA neurons also receive input from
neuronal projections originating from multiple brain regions (Figure 13-1). Some of these
include glutamatergic input from the lateral habenula, glutamatergic input from the medial
prefrontal cortex, and GABAergic inputs from the nucleus accumbens.9-11 Through such
complex interactions, midbrain dopamine signaling is increased by stimuli that produce
rewards and are inhibited by aversive stimuli. For example, glutamatergic signaling from
Nucleus Lateral Habenula

Accumbens Prefrontal Cortex
GABAergic Glutamatergic
Ventral
Tegmental Area
Dopaminergic
GABAergic
Glutamatergic
Nucleus Accumbens Cognition

Prefrontal Cortex Reward-Seeking
Amygdala Behavior
Hippocampus Emotional Memory
Influence
FIGURE 13-1 Overview of important brain regions involved in reward pathways. Multiple
brain regions form the mesolimbic system. Key reward signals involve dopamine signaling
that originates from neurons in the ventral tegmental area (VTA) and project to the nucleus
accumbens. Neurons originating in the VTA are composed predominantly of dopaminergic
neurons, but also include smaller populations of GABAergic and glutamatergic neurons,
which inhibit (GABA) or promote (glutamate) neurotransmission by downstream neurons. In
addition to dopaminergic signaling to the nucleus accumbens, VTA neurons also project to
other brain regions of the mesolimbic system including the prefrontal cortex, amygdala, and
hippocampus, which have important roles in cognition, learning, emotional memory, and
reward-seeking behavior. The VTA also receives regulatory input from other brain regions
including the nucleus accumbens, prefrontal cortex, and lateral habenula.
the prefrontal cortex may activate reward pathways in response to desirable stimuli. In
contrast, glutamatergic inputs from the lateral habenula to the rostromedial tegmental
nucleus (also known as the tail of the VTA) activate GABAergic neurons that actually
inhibit subpopulations of dopaminergic neurons extending to the nucleus accumbens.12
Therefore, signals from brain regions outside of the mesolimbic system increase midbrain
dopaminergic neuronal activity in the presence of rewarding stimuli and reduce midbrain
dopamine signaling in the presence of aversive stimuli. Addictive substances activate the
mesolimbic dopamine system by interacting with specific molecular targets, either within
or outside of the mesolimbic system.
Neuroadaptation, Dependence, and Withdrawal

Neuroadaptations are molecular changes in neural pathways that occur in response to
the presence of a drug. These neuroadaptations are associated with the development
of tolerance because an increase in drug concentration is required to produce a desired
effect that was previously produced with a lower dose. For example, if the presence of
high levels of a drug leads to hyperactivation of a receptor, a portion of the population of
that receptor type may become desensitized to compensate for the excessive stimulation.
Upregulation of the particular drug receptor may then occur to offset widespread recep-
tor desensitization, and many of these newly produced receptors may also eventually
undergo desensitization. These changes are appropriate for muting the effect of the drug
on the receptor, as long as the drug is present, and are considered neuroadaptive changes.
Such neuroadaptations lead to higher concentrations of drugs being required to produce a
desired response and, thus, form the foundation of tolerance to a drug. Another common
example of a neuroadaptive change includes the reduction of receptor density (i.e., through
reduced expression, reduced insertion in the neural membrane, increased internalization
and degradation of the receptor) in response to hyperactivation of a receptor.
Withdrawal symptoms generally occur during periods of abstinence in the dependent
individual when the plasma and CNS concentrations of the drug fall dramatically. In the
absence of drug, the neuroadaptations that had previously developed are inappropriate.
Withdrawal symptoms can vary widely depending on the specific agent of dependence and
the individual’s degree of dependence on that agent. However, development of withdrawal
symptoms is commonly associated with reduced dopamine signaling in reward pathways
and increased dopamine signaling in aversion pathways. Alterations in neurotransmission
to other brain regions also occur and may lead to psychological, somatic, and behavioral
changes during withdrawal.
POLYMORPHISMS AFFECTING RISK OF DEPENDENCE FOR

MULTIPLE AGENTS
Mesolimbic Dopamine Pathways
Because the mesolimbic dopamine pathways form the common pathway for develop-
ment of reward and addiction, variations that occur in the genes encoding key mediators
of this pathway have the potential to affect an individual’s risk of dependence to multiple
substances (Table 13-1). Polymorphisms that may affect the function of the mesolimbic
dopamine reward pathways occur in genes encoding dopamine receptors, dopamine trans-
porters, and key enzymes involved in dopamine metabolism.
A key molecule for transducing the dopamine signal is the dopamine receptor.
Dopamine receptors are G protein-coupled receptors and are classified into the D1-like
and D2-like families. The D1-like family includes D1 and D5 dopamine receptors. These
receptors are Gs-coupled receptors that lead to increased production of cyclic adenosine
monophosphate (cAMP) when activated. Members of the D2-like family are Gi-coupled,
with activation leading to decreased cAMP production. The D2 family includes the D2, D3,
and D4 dopamine receptors. In the nucleus accumbens of the mesolimbic pathway, the D2
family appears to be the most prevalent, particularly the D2 dopamine receptor, which is
Table 13-1
Select Polymorphisms Affecting Dopamine Signaling
Alterations in Components That Transduce the Dopamine Signal
Allele (Gene) Nucleotide or Structural Change Comments References
rs1076560 Intronic variant; minor T allele Association with increased 13-15

(DRD2) promotes formation of D2L over risk of cocaine and opioid
D2S during splicing dependence
Taq1A variant Polymorphism located in 3′-UTR; Association with increased 16–21
(DRD2) produces reduced D2 receptor risk of alcohol, opioid,
density methamphetamine, and
nicotine dependence and
gambling addiction
VNTR variants Multiple variants; variable number May affect ligand-dependent 22, 23
(DRD4) (2–10) of tandem repeats (of 16 receptor activation
codons) in exon 3 or expression level;
presence of 7-repeat allele
associated with increased
nicotine use
Alterations in Components Responsible for Clearance and/or Metabolism of Dopamine
rs4680 SNP replacing valine with Reduced level and activity of 26

(COMT) methionine in codon 158 of the COMT protein; conflicting
long form of COMT and codon data on risk of addiction
108 of the short form of COMT
VNTR variants Variable number (9–10) of tandem 9-repeat allele associated 24, 25
(SLC6A3) repeats (of 40 bp sequence in with lower protein levels
3′-UTR) and greater response
to smoking cues than
10-repeat allele
COMT, catechol-O-methyltransferase; VNTR, variable number tandem repeat.
encoded by the DRD2 gene. The DRD2 gene produces two splice variants of the dopamine
receptor, D2L and D2S. The D2S variant is shorter due to the removal of 29 codons from
exon 6 during splicing. This variant functions as a presynaptic autoreceptor, which reduces
dopaminergic neurotransmission on neurons where it is expressed. The D2L variant is full
length and functions mostly postsynaptically, where it transduces dopamine signaling in
mesolimbic pathways.
Given the role of D2 dopamine receptors in reward pathways, polymorphisms for these
receptors have been studied for their possible effects on the development of substance
abuse disorders (Table 13-1). Polymorphisms in DRD2 that produce nonsynonymous muta-
tions (i.e., those that result in an amino acid change in the protein) are uncommon. However,
DRD2 variants that produce changes in gene expression and splicing events have been
described and are associated with alterations in susceptibility for substance abuse. One
such polymorphism is the rs1076560 D2 receptor variant.13-15 The minor (T) allele of this
variant produces an intronic change that results in increased production of the D2L form
during splicing. As a result, more copies of the postsynaptic D2 dopamine receptors are
produced. The physiological effect is an increase in sensitivity to dopamine in tissues that
express the D2 receptor, including in the mesolimbic pathways. This DRD2 variant has been
studied specifically to assess its effect on substance abuse. Data show an association of
the minor allele of this variant with increased risk of opioid and cocaine dependence.13-15
Another example of a DRD2 polymorphism that alters the risk of development of
substance abuse disorders is the A1 allele of the Taq-1A variant. This polymorphism occurs
in the 3′-untranslated region (UTR) of the DRD2 gene and results in reduced D2 receptor
density in the brain.16 Specific investigations into the occurrence of this allele in patients
with substance abuse disorders reveal that the A1 allele is associated with nicotine, alco-
hol, opioid, and methamphetamine dependence.17-20 Interestingly, this allele has also been
associated with increased risk for addictive gambling.21
Polymorphisms in the D4 dopamine receptor, which is encoded by the DRD4 gene,
have also been implicated in increasing the risk for substance abuse and addiction. A
number of the DRD4 polymorphisms occur as variable number of tandem repeat (VNTR)
variants. These VNTR variants are commonly characterized by 2-10 tandem repeats of
nucleotides in exon 3 that compose 16 codons each. Among the VNTR variants of DRD4,
the most common are those containing 4, 7, or 2 tandem repeats. The number of repeating
tandem sequences in the VNTR variants has been linked to changes in receptor expres-
sion level and is believed to occur either through changes in gene expression or through
changes in messenger ribonucleic acid (mRNA) stability and translation efficiency.22 These
variants may also influence susceptibility to chemical dependence. For example, the DRD4
7-tandem repeat variant, which is associated with reduced D4 dopamine receptor levels,22
has been linked to increased smoking quantity23 although the mechanism by which this
relationship occurs is not clear.
In addition to polymorphisms in the receptors that transduce the dopamine signal,
genetic variability affecting dopamine metabolism and clearance mechanisms have also
been implicated in altered susceptibility to chemical dependence (Table 13-1). Once
secreted into extracellular space between neurons during neurotransmission, dopamine
is rapidly cleared from the synaptic region by dopamine transporters (DATs) located in
neuronal membranes. The rate of clearance from the cleft regulates the duration that the
dopamine signal persists. Genetic changes that reduce the expression or functionality of
the DATs would prolong the synaptic presence of dopamine and, thus, the dopamine signal.
The gene that encodes the DAT is SLC6A3. Polymorphisms in SLC6A3 have been described
that produce variable numbers (9–10) of tandem repeats consisting of a 40-nucleo-
tide sequence in the 3′-UTR of the gene. The 9-repeat allele has been associated with
increased response to smoking cues compared to the 10-repeat allele.24,25 This has been
hypothesized to be due to reduced DAT protein levels, which would prolong the dopamine
signal in the synaptic space.
Termination of the dopamine signal is regulated by dopamine metabolism, which is
mediated by a variety of enzymes. Polymorphisms that reduce the activity of dopamine-
metabolizing enzymes would potentially prolong the dopamine signal, while those that
increase the enzymatic activity would potentially shorten the duration of the dopamine
signal. One relevant polymorphism is the rs4680 variant of the COMT gene. The COMT
gene encodes the dopamine-metabolizing enzyme catechol-O-methyltransferase (COMT).

This enzyme catalyzes the last reaction in the two-reaction pathway that converts dopa-
mine to its urinary metabolite homovanillic acid. The rs4680 variant is associated with
reduced COMT enzyme levels and activity.26 However, the implications of this change on
addiction susceptibility are unclear.
Opioid Pathways
Opioids are agents that produce their pharmacologic effects via activation of opioid recep-
tors. Therapeutically, opioids are used in a variety of applications including analgesic, anes-
thetic, antidiarrheal, and antitussive uses. In addition to these therapeutic effects, opioids
can also produce reward, tolerance, and withdrawal symptoms associated with high abuse
potential. Opioids are produced endogenously in humans (i.e., enkephalins, endorphins,

dynorphins) where they influence mood, anxiety level, perception of reward, and goal-
directed behavior. Because many drugs of abuse increase secretion of endogenous opioids,
genetic polymorphisms that affect the functionality of opioid signaling pathways can influ-
ence susceptibility to dependence.
Activation of Reward Pathways
The mu opioid receptor, which is the primary target of most opioid agents, and the kappa
opioid receptor, which is the target of the endogenous dynorphins, both interact with
the reward and aversion pathways of the brain. Opioid receptors are predominantly Gi/
G0-linked G-protein coupled receptors. Receptor activation produces decreased adenylyl
cyclase activity, decreased cAMP production, reduced opening of presynaptic voltage-
gated Ca2+ (calcium) channels, and increased post-synaptic K+ (potassium) conductance.
Thus, opioid receptor activation reduces the release of pain-mediating substances presyn-
aptically and produces a hyperpolarizing effect postsynaptically, both of which reduce the
conduction of pain impulses.
In the mesolimbic reward pathways, mu opioid receptors are located on GABAergic
presynaptic nerve terminals where receptor activation results in reduced neurotransmis-
sion of GABA (Figure 13-2). In the mesolimbic system, such GABAergic neurons are usually
responsible for providing tonic inhibition of the VTA dopaminergic neurons that extend to
the nucleus accumbens and to other brain regions. Therefore, in the presence of opioid,
tonic inhibition is removed (i.e., disinhibition), and dopaminergic signaling to the nucleus
accumbens reward centers increases. In the early stages of opioid abuse, this positive rein-
forcement drives continued drug use.
The kappa opioid receptor is also expressed in the mesolimbic system. Activation of
kappa opioid receptors has been associated with anhedonia, dysphoria, and aversion as
well as increased risk of developing depression in response to stress.27,28 Receptor activa-
tion has also been shown to reduce dopamine release in the nucleus accumbens.29-31 Thus,
kappa opioid receptor activation is believed to have an opposing effect on mesolimbic
dopamine secretion to that of mu opioid receptor activation and may be important in
modulating the dopaminergic response to drugs of abuse.
Neuroadaptation, Tolerance, and Withdrawal
Tolerance to opioids develops as chronic exposure to opioids induces compensatory
changes in the mu opioid receptor signaling cascade. Although some degree of opioid
receptor desensitization may occur with chronic exposure to opioid agonists, data suggest
the major neuroadaptation that leads to opioid tolerance involves changes downstream
of the opioid receptor. One of the best characterized changes involves upregulation of
adenylyl cyclase activity. As long as opioids are present, adenylyl cyclase is maintained in
an inactivated state, disinhibition of VTA GABAergic neurons is maintained, and dopamine
signaling to the nucleus accumbens is high. Over time, prolonged low activity levels of
adenylyl cyclase (in the VTA GABAergic neuron) triggers increased expression of adenylyl
A Mu Opioid Receptors
GABAergic
neurons Dopaminergic
(Tonic Inhibition) neuron *
* *
*
B
Endorphins
Enkephalins
Exogenous Opioids Mu Opioid Receptors:
“Disinhibition”
GABAergic
neurons Dopaminergic
(Tonic Inhibition) neuron *
** * **
****** **
GABA INCREASED
* Dopamine REWARD
FIGURE 13-2 Activation of mu opioid receptors produces increased dopamine release

in reward pathways. In A: GABAergic neurons normally produce tonic inhibition of
dopaminergic neurons that extend to the ventral tegmental area (VTA). This tonic inhibition
prevents high levels of reward center activity in the absence of an appropriate stimulus. Mu
opioid receptors are present at presynaptic sites on these GABAergic neurons. In B: When
activated by endogenous opioids (i.e., endorphins, enkephalins) or by exogenous opioids
(i.e., morphine, heroin), mu opioid receptors mediate inhibition of GABAergic neurons.
This mu opioid receptor-induced reduction in GABA signaling allows the downstream
dopaminergic neuron to increase its neurotransmission to the VTA. Therefore, mu opioid
receptor activation in the mesolimbic pathway results in increased dopamine signaling to the
VTA and increased perception of reward.
cyclase. More drug may be needed at this time to convert and maintain the newly produced
adenylyl cyclase molecules in an inactive state (i.e., tolerance). After these neuroadapta-
tions have occurred, the expanded pool of adenylyl cyclase begins to reactivate if sufficient
levels of opioid are not available to maintain it in an inactive state. The result is a rebound
hyperactivation of adenylyl cyclase activity followed by production of very high levels of
cAMP, increased Ca2+ entry through voltage-gated channels, and decreased K+ efflux. The
net effect of these responses is hyperactivation of the GABAergic neuron on which the
mu opioid receptors are located. The increased GABA signaling in the VTA leads to abrupt
suppression of dopaminergic neurotransmission to the nucleus accumbens, which contrib-

utes to withdrawal symptoms.
Activation of kappa opioid receptors may also contribute to the undesirable symp-
toms associated with withdrawal. In a drug-dependent state, chronic drug use is associated
with increased expression of dynorphins (the endogenous kappa opioid receptor agonists),
which can reduce dopamine signaling in the mesolimbic and mesocortical systems. This
coincides with the negative symptoms associated with chronic drug use, such as anhedonia
and depressed mood. A number of animal studies suggest that, during drug withdrawal
in a dependent state, pharmacologic blockade of the kappa opioid receptor attenuates
withdrawal symptoms.32-35 Kappa opioid receptor activation may also be involved in stress-
induced responses such as worsening of withdrawal symptoms and increased relapse risk.35
A role for kappa opioid receptors in mediating the aversive symptoms associated with drug
use is important, because avoidance of these undesirable symptoms drives continued drug
use in the later stages of drug dependence.
Polymorphisms Affecting Opioid Pharmacodynamics
The response to opioids is mediated via the opioid receptor family, which includes three
receptor subtypes: mu, kappa, and delta opioid receptors. Opioid receptors are postulated
to have roles in mediating reward and aversion responses to multiple agents with abuse
potential.36 The most widely studied is the mu opioid receptor, which appears to have an
important role in transducing reward signals when activated in mesolimbic pathways.37,38
In contrast, kappa opioid receptors have been implicated in attenuating reward signals
and enhancing aversion signals, suggesting a role in withdrawal symptoms and in trig-
gering relapse.39 Kappa opioid receptors may also promote drug use during periods of
stress because under stressful conditions kappa receptors exhibit increased activity, an
effect that involves increased corticotropin-releasing factor signaling.40,41 The role of delta
opioid receptors has remained largely enigmatic until recently, when evidence has revealed
that activation of this receptor subtype is associated with reduced anxiety and enhanced
mood.42 However, conflicting data have precluded determination of whether or how this
receptor may play a role in development and maintenance of addiction. Therefore, the
remainder of this section will focus primarily on the mu opioid and kappa opioid receptor-
mediated pathways.
MU OPIOID RECEPTOR. The OPRM1 gene (chromosome 6q25.2) that encodes the mu
opioid receptor exhibits a high degree of genetic variation, including numerous single
nucleotide polymorphisms (SNPs) and multiple splice variants.43,44 For most of these SNPs,
studies demonstrating functional impacts on opioid dependence are either lacking or have
produced conflicting results. Therefore, the implications of mu opioid receptor polymor-
phisms in alteration of abuse potential remain largely unclear. However, among the OPRM1
SNPs, the A118G nucleotide substitution in exon 1 (rs1799971), which produces an N40D
amino acid change in the N-terminal region of the receptor, has been extensively charac-
terized.45 The minor 118G allele (which encodes aspartic acid at position 40) of this variant
exhibits three-fold greater affinity for binding to the endogenous opioid β-endorphin than
the 118A allele.46 This allele has also been shown in three different studies to be associated
with increased risk for heroin addiction.47-49
In addition to influencing the risk for opioid dependence, the rs1799971 polymorphism
(A118G) may influence the risk for alcohol dependence. Homozygous expression of the
118G allele is associated with increased dopamine signaling in mesolimbic pathways and
with increased reward (i.e., euphoria, intoxication) following alcohol use.50,51 In addition, the
118G allele was found to be more common than the 118A allele in families where alcohol-
ism is common. However, one cannot conclude from these findings that similar results will
be observed for all addictive agents that converge on the opioid pathways. For example,
at least a portion of nicotine-induced reward is known to be mediated via elevations in
endogenous opioids and activation of mu opioid receptors in reward centers.52-54 Despite

this relationship, the 118G allele of rs1799971 does not appear to increase the risk of nico-
tine dependence as it does for heroin and alcohol.55 In addition to the many SNPs that
exist, the OPRM1 gene has been predicted to produce a large number of splice variants.
Further investigation of these putative variants is also needed.
KAPPA OPIOID RECEPTOR. The OPRK1 gene (chromosome 8q11.2) that encodes the kappa
opioid receptor also exhibits a high degree of genetic variability.45,56 Many of these vari-
ants have been implicated in addiction susceptibility. Numerous intronic variants have
been described, and at least five have been linked to alcohol or cocaine dependence or
to increased risk of drug use and relapse in heroin and methadone users.56-58 Some of
these influence susceptibility to dependence for multiple drugs of abuse. For example, the
intronic OPRK1 variant rs6473797 and the synonymous variant rs1051660 are each associat-
ed with both alcohol and opioid dependence.56,58,59 Others may exert influence only as part
of a haplotype. For example, no association for the synonymous variant rs702764 (encodes
A281A in exon 4) was shown for alcohol dependence when the variant was analyzed
alone.56 However, a haplotype of four SNPs that includes this variant was demonstrated to
have significant association with the amount of alcohol used in a heroin-addicted popula-
tion undergoing methadone maintenance treatment.57 Such findings indicate that the influ-
ence of specific polymorphisms on alcohol use and dependence may be quite complex.
Finally, it is important to remember that findings from genetic association studies
are commonly inconsistent when analyzed across different populations (e.g., comparing
populations of European, Asian, and African descent). This suggests that the significance
of specific polymorphisms or even haplotypes may vary based on differences in the over-
all genetic background of populations from specific regions.58,60 Further investigations
are needed to establish clear relationships between OPRK1 variants and drug abuse and
dependence.
Another important gene that impacts signaling through the kappa opioid receptor is
the PDYN (prodynorphin) gene. This gene is composed of four exons, with exon 4 encod-
ing the dynorphin proteins. Dynorphins are upregulated in chronic drug use and attenuate
drug-induced dopamine signaling in the mesolimbic system. The PDYN gene produces a
number of variants that are candidates for influencing the susceptibility to dependence
on opioids or other drugs of abuse. For example, at least nine PDYN variants have been
associated with alcohol dependence (Table 13-2).56 Among these, only one occurred
in exon 4 (rs6045819), and this variant encodes a synonymous polymorphism. Others
occurred in regions upstream (rs10854244) or downstream (rs60045784) of the gene in the
3′-untranslated region (rs2235749, rs910080, rs10485703), in intronic regions (rs6035222
and rs6045868), or in the promoter (rs1997794). A number of these variants are thought to
increase dependence susceptibility by reducing PDYN gene expression levels.61
Data suggest that at least some of the PDYN variants may increase susceptibility
to dependence on multiple substances. SNP rs1997794 is associated with both alcohol
and opioid dependence.56,62 SNP rs2235749 (3′-UTR) has been implicated in alcohol
Table 13-2
Select Polymorphisms in the Opioid Pathway Associated with
Addiction to Nonopioid Substances
Variant Type of Polymorphism Increased Dependence Risk

OPRM1 Variants
rs1799971 Nonsynonymous (N40D) Alcohol, heroin, opioid
PDYN Variants
rs6045819 Synonymous (exon 4) Alcohol

rs2235749 3′-UTR Alcohol, cocaine
(haplotype-dependent)
rs910080 3′-UTR Alcohol
rs10485703 3′-UTR Alcohol
rs6035222 Intronic Alcohol
rs6045868 Intronic Alcohol
rs1997794 (promoter) Promoter Opioid and alcohol
rs10854244 Upstream of gene Alcohol
rs60045784 Downstream of gene Alcohol
OPRK1 Variants
rs6473797 Intronic Opioid, alcohol

rs1051660 Synonymous Opioid, alcohol
OPRD1 Variants
rs1042114 Nonsynonymous (F27C) Ethanol, cocaine, opioid

rs2234918 Synonymous Ethanol, cocaine, opioid
dependence when assessed in the rs2235749–rs10485703 haplotype and has been impli-
cated in cocaine dependence when assessed in the rs2235749–rs910080–rs910079 haplo-
type.56,63 Interestingly, the haplotype containing the 3′-UTR SNPs rs1022563, rs2235749,
and rs910080 was significantly associated with opioid dependence. However, when SNP
rs1022563 (3′-UTR polymorphism) was assessed alone, a significant association with opioid
dependence was observed only when the sample population was analyzed by gender (i.e.,
significant association observed only in females).62,64 Such findings emphasize the need for
additional studies to refine our understanding of the contextual impacts of SNPs expressed
within haplotypes.
DELTA OPIOID RECEPTOR. The OPRD1 gene encodes the delta opioid receptor. The delta
opioid receptor binds primarily to the endogenously produced enkephalins and is thought
to have a role in mood regulation. The OPRD1 gene is located on chromosome 1p36 and
contains three exons. Although the mechanisms by which the delta opioid receptor may
influence drug dependence are unclear, several polymorphisms of the OPRD1 gene have
been investigated for associations with substance abuse and dependence. Two polymor-
phisms, including the nonsynonymous Phe27Cys variant (rs1042114) and the synonymous
Gly307Gly variant (rs2234918), have both been associated with ethanol, cocaine, and
opioid dependence.65,66 Additionally, a number of intronic variants (rs2236857, rs581111,
rs2236861, rs3766951, rs678849) have been associated with increased risk of heroin
addiction.58,67-69 However, ethnic background and gender appear to influence the reported
associations. The promoter variant rs569356 possesses an A>G substitution approximately
2kb upstream of the OPRD1 transcription start site. The rs569356 minor G allele has been
shown to increase promoter activity, suggesting that this polymorphism produces increased
expression levels.70 Furthermore, the G allele has been significantly associated with opioid
dependence in a European American population,65 suggesting that increased expression of
OPRD1 may increase susceptibility to dependence.
INTERACTION OF SPECIFIC AGENTS WITH REWARD

PATHWAYS: NICOTINE
Nicotine is commonly ingested by inhalation as a component of cigarette smoke or vapor
or by absorption across the oral mucosa when administered through other tobacco prod-
ucts. Once ingested, nicotine is readily absorbed into the bloodstream, where it is rapidly
delivered to the cerebral vasculature. In the cerebral circulation, nicotine readily penetrates
the blood–brain barrier to access its target sites in the CNS. Whether the administration
route is by inhalation or absorption across oral mucosa, nicotine accesses the brain within
seconds and can rapidly accumulate to reach pharmacologically active concentrations.
The rapid accumulation of nicotine in the brain produces the “rush” perceived by nicotine
users, and this “rush” is considered one of the rewarding effects that positively reinforces
nicotine use.71-76
Activation of Reward Pathways

Nicotine activation of the reward pathways in the brain occurs via an indirect mechanism
(Figure 13-3). Nicotine exerts its effects in the CNS by binding to and activating nicotinic
acetylcholine receptors (nAChRs), which are ligand-gated cation channels that produce
depolarizing (excitatory) effects in neurons when activated. Activation of nAChRs on
glutamatergic and GABAergic neurons promotes neurotransmission from these neurons,
thereby producing excitatory or inhibitory influences, respectively, on the VTA dopaminer-
gic neurons.73,74,76 The initial effect of nicotine on dopaminergic neurotransmission from the
VTA is minimal, as both the inhibitory (GABA) and excitatory (glutamate) signals increase.
However, GABA signaling rapidly decreases as nAChRs on GABAergic neurons undergo
desensitization. This results in a net increase in dopaminergic secretion due to unopposed
glutamate-induced stimulation.77,78 The increased dopamine signaling in the nucleus
accumbens of the mesolimbic system is associated with the rewarding (positive reinforce-
ment) effects of nicotine use.
Neuroadaptation, Dependence, and Withdrawal

The driving factor for re-administration of nicotine, through smoking or the use of other
nicotine-containing products, is to repeat the pleasurable feelings that result from reward
pathway activation.73,76 Initially, this repeated activation of the reward pathways is achieved
by maintaining nicotine levels at concentrations in the brain that allow for saturation
A B
GABAergic GABAergic
neuron (-) neuron (-)
Dopaminergic Dopaminergic
neuron * neuron ** ***
** * * * **
+ +
+++++ ** + +
+++++ * *** **
+ + * **
Glutamatergic Glutamatergic
neuron (+) neuron (+)
Nicotine
GABA
+ Glutamate
* Dopamine
FIGURE 13-3 Schematic representation of nicotine effects on dopaminergic

neurotransmission in the mesolimbic pathway. Nicotine indirectly regulates dopaminergic
signaling in mesolimbic pathways. Nicotine activates nAChRs located on GABAergic and
glutamatergic neurons that synapse on dopaminergic neurons in the ventral tegmental area
(VTA). GABA exerts hyperpolarizing effects (through GABA receptor activation), whereas
glutamate exerts depolarizing effects (through NMDA receptor activation) on dopaminergic
neurons. In A: Initially, increased secretion of both GABA and glutamate occur in the VTA
in response to nicotine. In B: However, nAChRs located on GABAergic neurons are thought
to rapidly desensitize, resulting in increased dopaminergic neurotransmission due to
unopposed glutamate signaling in the VTA. Differential desensitization rates of nAChRs on
glutamatergic and GABAergic neurons may be due to differences in subunit composition of
the receptors at the two locations.
Source: Reprinted with permission from Zdanowicz MM, Adams PW. The pharmacogenetics
of nicotine dependence and smoking cessation therapies. J Pharmacogenomics
Pharmacoproteomics. 2014;5(4):Article 138. All rights reserved.
of specific populations of nAChRs.79 Over time, nAChRs maintained in a saturated state

undergo desensitization, with some subtypes of nAChRs becoming desensitized faster than
others.79,80 As increasing numbers of nAChRs convert to a desensitized state, increased
expression of these subtypes of nAChRs occurs.80-82 Such upregulation is likely a homeo-
static response to receptor desensitization.
As a result of receptor upregulation, more nicotine is now needed to achieve receptor
occupancy sufficient to maintain desensitization of some nAChR populations while achiev-
ing activation of other nAChR populations (i.e., those that are capable of ligand-induced
activation). This balance is likely required to re-experience activation of the reward pathway.
These molecular changes are thought to underlie the early stages of dependence in the
patient. After such neuroadaptations occur, the dramatic drops in nicotine concentrations
that occur between nicotine dosing result in large populations of unoccupied receptors.
Withdrawal symptoms such as anxiety, anhedonia, irritability, and difficulty concentrat-
ing are thought to result from the combination of reduced dopamine signaling to nucleus
accumbens and increased dopamine signaling to the prefrontal cortex.83
In summary, susceptibility to development of nicotine dependence can be impacted

by both pharmacodynamic and pharmacokinetic influences. Genetic changes that alter
the transduction of the nicotine signal or the capacity of reward and aversion pathways to
become activated can produce altered intensity of both positive and negative responses
to nicotine, thereby influencing the likelihood of habitual use of nicotine. Likewise, genetic
changes that alter the rate of nicotine metabolism and clearance can change the frequency
and intensity of withdrawal symptoms, which affect smoking behaviors and susceptibility
to dependence.
Pharmacogenomics Affecting Pharmacodynamics of Nicotine

Genetic variation in any molecular component that contributes to transduction of the
nicotine signal can affect overall nicotine response and has potential to influence the risk
of developing dependence. Among these, polymorphisms in genes encoding the nAChR
subunits are among the most widely studied. Nicotinic AChRs are pentameric transmem-
brane receptors, which function as ligand-gated cation channels. Neuronal nAChRs occur
in numerous subtypes, each of which is composed of different combinations of alpha and
beta subunits. The number of nAChR subtypes is quite large because each subunit type
(i.e., α or β subunit type) exists in multiple isoforms (at least 12 total isoforms identified α2
– α10 and β2 – β4).73,74,76,84 Distinct pharmacologic properties such as ligand binding affinity,
ion permeability, current amplitude, and desensitization rate are determined by the specific
subunit composition of each nAChR subtype.85-87 All receptors contain alpha subunits,
which are critical for formation of the ligand binding site, and many receptors also contain β
subunits, which influence the receptor’s pharmacologic properties. Therefore, receptors are
either homomeric and formed by an assembly of only α subunits, or they are heteromeric
and formed by a combination of α and β subunits (most commonly with stoichiometry of
[αx]2[βy]3).84
The nicotinic acetylcholine receptor subunits have numerous polymorphisms, some
of which have been associated with altered response to nicotine or altered susceptibility
to nicotine dependence (Table 13-3). For example, the rs16969968 variant of the CHRNA5
gene contains a nonsynonymous mutation that produces a D398N change in the amino
acid sequence of the nAChR α5 subunit. This change results in a reduction in nicotine-
induced receptor activation.88 The rs16969968 variant also has been associated with
increased smoking quantity and increased risk of nicotine dependence.89-91 The CHRNA3
polymorphisms rs1051730 and rs6495308 have been associated with increased smok-
ing intensity (as defined by depth and frequency of inhalation) and increased number of
cigarettes smoked per day as well as increased risk of nicotine dependence.89,92-94 The
specific mechanisms by which these variants are associated with increased susceptibility
to nicotine dependence are unclear. However, the rs1051730 variant contains a synonymous
mutation in exon 3 of the gene, and the rs6495308 variant contains an intronic change.
Thus, the effects of these variants are due to mechanisms other than amino acid changes
in the coding sequence of the receptor subunit.
Other nAChR polymorphisms have been associated with reduced risk of nicotine
dependence. For example, nonsynonymous, gain-of-function mutations (rs61737499 and
rs12914008) have been identified in the CHRNB4 gene that encodes the β4 subunit of the
nAChR. The nAChR β subunit has been linked to regulation of nAChR activity, and increased
nAChR signaling in certain brain pathways increases aversion to nicotine. Indeed, expres-
sion of these gain-of-function variants in the medial habenula is associated with increased
Table 13-3
Select Polymorphisms in Nicotinic Acetylcholine Receptor Subunits
Affecting Risk of Nicotine Dependence
Polymorphisms Associated with Increased Risk of Nicotine Dependence
rs1051730 Synonymous SNP in exon 5 (no Associated with greater 89, 92, 93
(CHRNA3) amino acid change) smoking intensity and
quantity
rs6495308 Intronic variant TT allele associated with 94
(CHRNA3) increased cigarettes
smoked per day
rs16969968 Missense mutation produces Reduced ligand response 88–91
(CHRNA5) D398N amino acid change of α5 subunit; associated
with increased smoking
rs13277254 Nucleotide change upstream of Molecular effects unclear 136
(CHRNA6/B3) gene cluster
Polymorphisms Associated with Decreased Risk of Nicotine Dependence
rs578776 Nucleotide change in 3’-UTR A allele has protective 88, 91

(CHRNA3) effects; molecular effects
not clear
rs61737499 Missense mutation produces Gain-of-function mutation; 95, 96
(CHRNB4) T375I amino acid change increased aversion
rs12914008 Missense mutation produces T91I Gain-of-function mutation; 95, 96
(CHRNB4) amino acid change increased aversion
SNP, single nucleotide polymorphism.
nicotine aversion in mice.95 Likewise, these variants have been associated with reduced
susceptibility to nicotine dependence in humans.96
The nAChR polymorphisms discussed here include only representatives of those that
have been linked to functional changes in nicotine responsiveness. Many more polymor-
phisms in genes that encode the nAChR subunits have been described, although not all
have been assessed for functional implications with respect to risk of developing nicotine
dependence. Other polymorphisms probably have functional impacts on development of
nicotine use disorder, but more research is needed in this area.
Pharmacogenomics Affecting Pharmacokinetics of Nicotine

Due to the rapid onset of action and the short plasma half-life of nicotine, smokers can
titrate plasma nicotine levels (i.e., by changing depth and frequency of inhalation) to
achieve the desired effects.76 Because metabolism of nicotine is associated with termina-
tion of nicotine effects, any genetic change that changes the rate of nicotine metabolism
can affect an individual’s smoking behaviors and risk of developing nicotine dependence.
Individuals who metabolize nicotine more slowly (i.e., slow metabolizers) have prolonged
systemic nicotine exposure following ingestion and, thus, are likely to display smoking
habits such as reduced intensity or frequency of inhalation. Also, the gradual clearance of
nicotine that would be associated with slow metabolizers means nicotine abstinence is less
likely to induce severe withdrawal symptoms in these individuals. Therefore, slower rates
of nicotine metabolism are associated with reduced smoking and reduced incidence of
nicotine dependence.97-101 In contrast, individuals who are rapid metabolizers of nicotine are
more likely to experience severe withdrawal symptoms because nicotine is rapidly cleared
from CNS receptors; they exhibit more aggressive smoking behaviors and increased smok-
ing frequency in an attempt to avoid nicotine withdrawal.102-104
Nicotine is metabolized by hepatic enzymes through a variety of pathways (Figure
13-4) with excretion of nicotine and its metabolites almost entirely via the kidneys. Only
8-10% of nicotine is excreted unchanged; therefore, metabolic pathways play important
roles in its elimination. The primary pathway for metabolism of nicotine is mediated by
cytochrome P450 2A6 (CYP2A6). CYP2A6 converts nicotine to the iminium ion. This reac-
tion is followed by the aldehyde dehydrogenase-catalyzed conversion of this intermediate
to cotinine, which composes approximately 75% of the urinary metabolites of nicotine
detected in urine. CYP2A6 is highly polymorphic (nearly 100 variants described; see www.
cypalleles.ki.se), with many variants affecting enzymatic activity levels (Table 13-4). At least
two variants have been identified that increase enzymatic activity. These variants are asso-
ciated with either increased gene copy number (i.e., CYP2A6*1x2, through gene duplication
(S)-nicotine-N-β -glucuronide
T (3-5%)
UG
Aldehyde
CYP2A6/2B6 oxidase
Nicotine Imminium ion Cotinine
(~75%)
FM
O3
Nicotine-N '-oxide
(4-7%)
FIGURE 13-4 Important pathways for nicotine metabolism in humans. The majority of
nicotine is converted to cotinine, which composes approximately 75% of the total urinary
metabolites of nicotine. Cotinine is produced by a two-step process catalyzed by a CYP450
enzyme and cytosolic aldehyde oxidase. CYP2A6 is the major CYP450 enzyme catalyzing
the first reaction, with about 80% of total cotinine formation dependent on the activity of
this enzyme. CYP2B6 has a minor role in catalyzing the first reaction in cotinine formation,
but the enzyme can assume a larger role in individuals with deficiencies in CYP2A6 activity.
Other pathways contributing to nicotine metabolism include the N-oxidation of nicotine
by flavin-containing monooxygenase 3 (FMO3) and glucuronidation of nicotine by UDP-
glucuronosyltransferases (UGT). Other pathways of nicotine metabolism have minor roles,
each contributing to the formation of less than 2% of the total urinary metabolites of nicotine.
About 8-10% of nicotine is excreted in the urine unchanged.
Source: Reprinted with permission from Ann Jose, Editorial Assistant, Journal
Pharmacogenomics and Pharmacoproteomics. The pharmacogenetics of nicotine
dependence and smoking cessation therapies. J Pharmacogenomics Pharmacoproteomics.
2014;5:Article 138. All rights reserved.
Table 13-4
CYP2A6 Variants with Altered Enzymatic Activity
Polymorphisms Associated with Increased Activity
Alleles
CYP2A6*1x2, CYP2A6*1B
Polymorphisms Associated with Absent or Reduced Activity
CYP2A6*1H, CYP2A6*2, CYP2A6*3, CYP2A6*4, CYP2A6*5, CYP2A6*6, CYP2A6*7, CYP2A6*9,

CYP2A6*10, CYP2A6*11, CYP2A6*12, CYP2A6*17, CYP2A6*18, CYP2A6*19, CYP2A6*20,
CYP2A6*21, CYP2A6*22, CYP2A6*23, CYP2A6*24, CYP2A6*25, CYP2A6*26, CYP2A6*27,
CYP2A6*28, CYP2A6*35
events) or increased mRNA stability (i.e., CYP2A6*1B).103,105-107 However, most polymorphisms

result in reduced CYP2A6 activity through reduced gene expression (i.e., reduced promoter
activity), absent gene expression (i.e., gene deletion), or reduced functionality of the protein
product (i.e., nonsynonymous SNPs). Many variants have been demonstrated to impact
susceptibility to nicotine dependence.104,108
Polymorphisms in enzymes catalyzing minor reactions in the metabolism of nicotine
have been found to have no significant impact on nicotine metabolism (i.e., CYP2B6 vari-
ants) or have not been evaluated for effects on nicotine metabolism (i.e., UGT variants).109,110
However, nonfunctional CYP2A6 alleles (i.e., CYP2A6*4) have been associated with a shift
in metabolic pathways such that CYP2B6-, FMO-, and UGT-catalyzed reactions assume
a more prominent role in the metabolism of nicotine.111 These findings suggest that poly-
morphisms in other nicotine-metabolizing enzymes could impact nicotine inactivation and
clearance in the absence of fully functional CYP2A6.
PHARMACOGENOMICS AFFECTING MANAGEMENT OF

ADDICTION
The Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5), describes
addiction under substance-related and addictive disorders.112 The specific substance-
related disorders include the following drugs of abuse: alcohol, caffeine, cannabis, halluci-
nogens, inhalants, opioids, sedatives/hypnotics/anxiolytics, stimulants, and tobacco. Because
gambling behaviors activate reward systems similar to those activated by drugs of abuse,
gambling disorder is classified in the same section of the DSM-5 under nonsubstance-
related disorders. For the purpose of this chapter, the most common substances of abuse
(e.g., tobacco/nicotine, alcohol, opioids) are the focus because of the propensity for treat-
ment with therapeutic agents.
To receive a diagnosis of a substance use disorder, one must exhibit a problematic
pattern of substance use leading to clinically significant impairment or distress, displayed

within a 12-month period by at least two of the following:
1. Substance often taken in larger amounts or over a longer period than was intended.
2. Persistent desire or unsuccessful efforts to cut down or control use.
3. A great deal of time spent in activities required to obtain the substance, use the substance,
or recover from its effects.
4. Craving or a strong desire or urge to use the substance.

5. Recurrent use of the substance results in a failure to fulfill major role obligations at work,
school, or home (e.g., repeated absences from work or poor work performance related to
substance, substance-related absences, suspensions, or expulsions from school, neglect of
children or household).
6. Continued use despite substance-related social or interpersonal problems (e.g., arguments

with a spouse about consequences of intoxication, physical fights).
7. Important social, occupational, or recreational activities are given up or reduced because of
substance use.
8. Recurrent use in spite of hazardous situations (e.g., driving an automobile/operating a ma-
chine when impaired or smoking in bed).
9. Continued use despite knowledge of having a persistent or recurrent physical or psychologi-
cal problem likely caused or exacerbated by the substance.
10. Tolerance.
11. Characteristic withdrawal syndrome of the substance occurs.112
In addition to pharmacotherapy for substance use disorders, most treatment strategies
involve complementary nonpharmacological interventions such as regularly attending
12-step support groups (i.e., Alcoholics Anonymous, Narcotics Anonymous) or participat-
ing in cognitive behavioral therapy. Treatment approaches utilizing medications should be
tailored to the patient’s needs and carefully considered based on the potential pharma-
cogenetic variations. Several therapeutic agents are available to treat individuals suffering
from substance use disorders.
CLINICAL PEARL
Individuals who struggle with addiction may receive a diagnosis of a
substance use disorder corresponding to the particular substance abused
(i.e., alcohol use disorder, opioid use disorder, tobacco use disorder).
Pharmacotherapy for Alcohol Use Disorder

Three medications are FDA approved for use in the pharmacologic management of alcohol
use disorder: disulfiram, naltrexone, and acamprosate. Disulfiram is an irreversible inhibitor
of aldehyde dehydrogenase (ALDH) that works by blocking the second stage of alcohol
metabolism where acetaldehyde is metabolized to acetic acid. The result is a buildup of
the toxic acetaldehyde that produces hypotension, flushing, nausea, and vomiting when
alcohol is consumed. In persons of Asian descent who possess an ALDH2*2 genetic vari-
ant of the ALDH enzyme, acetaldehyde is metabolized slowly and accumulation ensues.113
Disulfiram’s effect at deterring individuals from alcohol consumption is greatly reduced in
those with genetically compromised ALDH function.114
Because alcohol stimulates the release of β-endorphins from neurons, naltrexone (an
opioid antagonist) is another treatment option that has been used for alcohol dependence
since the 1990s. Naltrexone prevents the uptake of endorphins released during alcohol
consumption and decreases dopamine release at the nucleus accumbens, thus altering the
positive reinforcement and reward associated with alcohol use.115 Pharmacogenomic stud-
ies have identified a polymorphism, Asn40Asp, on the mu opioid receptor gene (OPRM1)
that impacts receptor binding and stimulation by a factor of three and so impacts an
alcoholic’s response to naltrexone. Individuals possessing this genetic variant treated with
naltrexone exhibited prolonged time to relapses compared to those lacking the trait.116
Acamprosate is indicated for the maintenance of abstinence from alcohol and has
multiple actions, including antagonism at the N-methyl-D-aspartate receptor and activa-
tion of GABA receptors, thus promoting the GABA/glutamate balance. Acamprosate does
not undergo hepatic metabolism, and the drug is primarily excreted through the renal
pathway.117 Although acamprosate appears to be the preferred treatment for achieving
abstinence from alcohol when compared to oral naltrexone, evidence is limited regarding
the pharmacogenetic variations and response to treatment with acamprosate so further
studies are warranted. Selecting a medication for an alcohol use disorder should be patient
specific.
CLINICAL PEARL
Naltrexone and acamprosate have demonstrated small-to-moderate effect
sizes superior to placebo in the treatment of alcohol use disorders.
Pharmacotherapy for Opioid Use Disorder

Individuals with an opioid use disorder may be offered methadone, buprenorphine, or
buprenorphine with naloxone. Methadone is a synthetic opioid receptor agonist that is
FDA approved for the detoxification associated with opioid withdrawal and for substitution
in the maintenance phase of opioid dependence. Due to its pharmacokinetic parameters,
methadone has a high interindividual variability, resulting in wide-ranging responses to
treatment with a given dose.2 The major CYP isoenzymes involved in methadone metabo-
lism are CYP3A4 and CYP2B6. CYP2B6 is one of the most polymorphic CYP genes with
more than 100 described SNPs. Individuals whose livers show high levels of CYP2B6
compared to CYP3A4 have a higher stereoselective metabolism (S>R) of methadone.118
Because methadone may also act as a p-glycoprotein substrate, ABCB1 polymorphisms are
suggested to contribute to the interindividual variability of methadone kinetics and affect
dosage requirements.119 Methadone doses have ranged from 55 mg/day to 921 mg/day in a
70-kg patient to produce plasma methadone levels of 250 ng/mL.2
Buprenorphine tablets and sublingual film are FDA approved for opioid use disorder
treatment, including induction and maintenance. Buprenorphine is a partial opioid agonist
with very high affinity for the mu receptors and exhibits slow dissociation. Buprenorphine
is also available as a combination product with naloxone at a 4:1 (buprenorphine to nalox-
one) ratio. Naloxone is FDA approved as monotherapy for reversing the effects of opioids
in known and/or suspected overdose situations. Naloxone alone is administered parentally
and is metabolized via glucuronidation. Studies indicate that the presence of the 118G
allele on the OPRM1 receptor produces a strong cortisol response to naloxone.120
Buprenorphine is metabolized to norbuprenorphine primarily via CYP3A4 and to a
lesser extent via CYP2C8. This may account for the drug–drug interactions and increased
risk of CNS depression when potent CNS depressants are simultaneously administered.121
Buprenorphine and norbuprenorphine undergo further phase II metabolism by UGT1A1,
UGT2B7, and UGT1A3. However, data suggest that polymorphisms in genes encoding these
enzymes do not impact treatment response.122
CLINICAL PEARL
Pharmacotherapy options for opioid use and alcohol use disorders
include the opioid antagonists naloxone and naltrexone. Pharmacogenetic
variations in the mu opioid receptor gene may alter clinical response.
Pharmacotherapy for Tobacco Use Disorder

Nicotine is the primary psychoactive component of tobacco. Due to the highly addictive
nature of nicotine, individuals with tobacco use disorder may find it difficult to remain
abstinent. To reduce cravings for tobacco products and blunt withdrawal symptoms, the
first-line therapy for smoking cessation is nicotine replacement therapy (NRT).123 NRT is
available as a gum, transdermal patch, lozenge, and inhaler and nasal spray; selection is
based on patient preference.
In spite of the wide variety and accessibility of FDA-approved NRT products, effects on
smoking cessation have been minimal with long-term abstinence rates less than 25%.124
One’s genetic profile may contribute to treatment response, and studies have identified
various pharmacogenetic variants affecting patient outcomes with NRT that illustrate
the possible need for pharmacogenetic testing to guide therapy selection. Individu-
als who express the CYP2A6*1B high activity allele are less likely to achieve abstinence
from smoking when treated with NRT.125 On the other hand, smokers with reduced activity
CYP2A6 variants are twice as likely to quit smoking compared to individuals who do not
carry the allele. The COMT rs4680 polymorphism reduces COMT activity, and individuals
with this genotype are more likely to successfully abstain from smoking with the use of
the NRT patch.126 Other polymorphisms affecting treatment response with NRT include
the CHRNA5−CHRNA3−CHRNB4 variants in the nAChR subunits, the DRD2-141C insertion/
deletion of the dopamine D2 receptor, the DRD4 dopamine D4 receptors, and the Asp40
variant in the OPRM1 gene.127-129 The outcomes are described in Table 13-5.
Bupropion sustained release under the brand Zyban® is FDA approved as an aid for
smoking cessation treatment.130 The mechanism of action of bupropion in smoking cessa-
tion is theorized to enhance the dopaminergic signaling in the mesolimbic pathways,
which is important for the perception of nicotine reward.123 Bupropion may also serve as a
weak agonist at certain nAChR subunits. The primary enzyme responsible for metabolizing
bupropion is CYP2B6. Smokers identified as slower metabolizers with a lower activity of
CYP2B6 had higher relapse rates and more cravings for cigarettes, yet bupropion was able
to mitigate the effects of this genotype in female smokers.131 Individuals with the CYP2B*6
genotype benefit significantly from bupropion treatment when compared to smokers with
the CYP2B*1 genotype. Other pharmacogenetic variations associated with bupropion are
summarized in Table 13-5.
CLINICAL PEARL
Although many genes involved in nicotine dependence and withdrawal

overlap, the genes contributing to successful smoking abstinence with
bupropion are not necessarily the same as those affecting success with
NRT.
Table 13-5
Pharmacogenetic Factors Affecting Response to Treatment of
Substance Use Disorders2,11,114,115,117,123,124,126-129,131,133,135
Substance Use Polymorphisms Affecting Treatment Response

Disorder Pharmacotherapy and/or Relapse
Alcohol Naltrexone Asn40Asp on the OPRM1 gene results in longer time to

relapse
Treatment response greater in presence of the 118G allele
(90%) versus the 118A allele
Disulfiram ALDH genetic changes reduce response
Opioid Methadone Stereoselective (S>R) correlated with high levels of
CYP2B6
ABCB1 polymorphisms impact methadone kinetics and
may influence dose requirements
Buprenorphine CYP3A4 and CYP2C8 impact metabolism and
+/− naloxone elimination of buprenorphine
No impact of UGT1A1 or UGT2B7 on buprenorphine
kinetics reported
Presence of 118G allele on OPRM1 yields greater response
in buprenorphine + naloxone
Tobacco NRT CYP2A6*1B increases nicotine metabolism and reduces
treatment response
COMT rs4680 increases response to transdermal NRT
CHRNA5−CHRNA3−CHRNB4 high-risk variants are more
likely to respond to NRT or bupropion
Del C allele of the DRD2-141C I/D polymorphism produces
higher quit rates versus homozygous Ins C allele
Lower abstinence rates with VNTR polymorphism in
DRD4 gene
Smokers with the OPRM1 Asp40 variant may require
extended high-dose (21 mg) patch treatment as
maintenance therapy to achieve abstinence
Bupropion CYP2B6 slow metabolizers yield poor response with
higher rates of relapse
CYP2B*6 (bupropion > placebo)
CYP2B*1 (bupropion = placebo)
DRD2-141C I/D: homozygous Ins C allele response > Del C
allele response
DRD4 exon III VNTR increased bupropion efficacy
SLC6A3/SLC6A4 altered bupropion efficacy
Varenicline CHRNA5−CHRNA3−CHRNB4 polymorphisms associated
with varied varenicline response
CHRNB2 minor allele carriers have a lower incidence of
nausea
I/D, insertion/deletion; NRT, nicotine replacement therapy; VNTR, variable number tandem repeat.
Varenicline, brand name Chantix®, is another non-nicotine prescription medication indi-

cated for smoking cessation. It is a partial agonist/antagonist at the nicotinic α4β2 recep-
tor that blunts cravings associated with nicotine due to the release of dopamine and also
reduces the pleasurable sensations associated with smoking.132 Because long-term smok-
ing abstinence rates with varenicline are low (approximately 20-30%), a large interindivid-
ual variability exists. Polymorphisms in the CHRNA5–CHRNA3–CHRNB4 are associated with

altered response to varenicline treatment.133 Nausea is the most common adverse effect
associated with varenicline, and it is one of the most common reasons cited for discontinu-
ation.134 Smokers with minor variants in the CHRNB2 allele experience fewer episodes of
nausea associated with varenicline when compared to those without the minor allele.135
SUMMARY
A considerable body of evidence highlights the need for genetic testing in patients with
substance use disorders. Pharmacogenomic considerations should include how genetic
polymorphisms impact the patient’s response to therapy during treatment of the addic-
tion/substance use disorder as well as the likelihood of relapse. For substance use disor-
ders, genetic variability can influence success and relapse rates in complex ways because
it can cause alterations in (1) the pharmacodynamics of the drug of abuse, (2) the phar-
macodynamics of the therapeutic agent, (3) the metabolism of the drug of abuse, and (4)
the metabolism of the therapeutic agent. Therefore, pharmacotherapy selection should
be specific and based on the genetic makeup of the individual. The treatment plan should
be tailored to include strategies for success aimed at both achieving and maintaining
abstinence.
Research continues to explore genetic variations present in patients with substance
use disorders. As personalized medicine grows, the need is greater for determining the
pharmacogenetic differences affecting one’s response to the substance of abuse and the
medications prescribed for treating addiction. Larger studies are needed to evaluate the
costs associated with improving medication selection by way of pharmacogenomic testing.
REFERENCES
1. Substance Abuse and Mental Health Services Administration (SAMHSA), Results from the
2013 National Survey on Drug Use and Health: Summary of National Findings, NSDUH Series
H-48, HHS Publication No. (SMA) 14-4863. Rockville, MD: Substance Abuse and Mental Health
Services Administration, 2014. Available at: http://www.samhsa.gov/data/sites/default/files/
NSDUHresultsPDFWHTML2013/Web/NSDUHresults2013.pdf. Accessed November 4, 2015.
2. Meyer MR, Maurer HH. Absorption, distribution, metabolism and excretion pharmacogenomics of drugs
of abuse. Pharmacogenomics. 2011;12(2):215-33. DOI:10.2217/PGS.10.171. PMID: 21332315.
3. Bjorklund A, Dunnett SB. Dopamine neuron systems in the brain: an update. Trends Neurosci.
2007;30(5):194-202.
4. Polter AM, Kauer JA. Stress and VTA synapses: implications for addiction and depression.
Eur J Neurosci. 2014;39(7):1179-1188.
5. Kelley AE, Berridge KC. The neuroscience of natural rewards: relevance for addictive drugs. J Neurosci.
2002;22(9):3306-3311.
6. Lammel S, Lim BK, Malenka RC. Reward and aversion in a heterogeneous midbrain dopamine system.
Neuropharmacology. 2014;76 Pt.B:351-359.
7. Nestler EJ. Is there a common molecular pathway for addiction? Nat Neurosci. 2005;8(11):1445-1449.
8. Volman SF, Lammel S, Margolis EB, et al. New insights into the specificity and plasticity of reward and
aversion encoding the mesolimbic system. J Neurosci. 2013;33(45):17569-17576.
9. Margolis EB, Toy B, Himmels P, et al. Identification of rat ventral tegmental area GABAergic neurons.
PLoS One. 2012;7(7):e42365.
10. Hikosaka O. The habenula: from stress evasion to value-based decision-making. Nat Rev Neurosci.
2010;11(7):503-513.
11. Lecca S, Meye FJ, Mameli M. The lateral habenula in addiction and depression: an anatomical, synaptic,
and behavioral overview. Eur J Neurosci. 2014;39(7):1170-1178.
12. Lammel S, Lim BK, Ran C, et al. Input-specific control of reward and aversion in the ventral tegmental
area. Nature. 2012;491(7423):212-217.
13. Zhang Y, Bertolino A, Fazio L, et al. Polymorphisms in human dopamine D2 receptor gene affect
gene expression, splicing, and neuronal activity during working memory. Proc Natl Acad Sci USA.
2007;104(51):20552-20557.
14. Moyer RA, Wang D, Papp AC, et al. Intronic polymorphisms affecting alternative splicing of human
dopamine D2 receptor are associated with cocaine abuse. Neuropsychopharmacology. 2011;36(4):

753-762.
15. Clarke TK, Weiss AR, Ferarro TN, et al. The dopamine D2 (DRD2) SNP rs1076560 is associated with
opioid addiction. Ann Hum Genet. 2014;78(1):33-39.
16. Pohjalainen T, Rinne JO, Nagren K, et al. The A1 allele of the human D2 dopamine receptor gene
predicts low D2 receptor availability in healthy volunteers. Mol Psychiatry. 1998;3(3):256-260.
17. Smith L, Watson M, Gates S, et al. Meta-analysis of the association of the Taq1A polymorphism
with the risk of alcohol dependency: a HuGE gene-disease association review. Am J Epidemiol.
2008;167(2):125-138.
18. Lawford BR, Young RM, Noble EP, et al. The D(2) dopamine receptor A(1) allele and opioid depen-
dence: association with heroin use and response to methadone treatment. Am J Med Genet.
2000;96(5):592-598.
19. Chen CK, Hu X, Lin SK, et al. Association analysis of dopamine D2-like receptor genes and meth-
amphetamine abuse. Psychiatr Genet. 2004;14(4):223-226.
20. Munafo M, Clark T, Johnstone E, et al. The genetic basis for smoking behavior: a systematic review and
meta-analysis. Nicotine Tob Res. 2004;6(4):583-597.
21. Comings DE, Rosenthal RJ, Lesieur HR, et al. A study of the dopamine D2 receptor gene in pathologi-
cal gambling. Pharmacogenetics. 1996;6(3):223-234.
22. Shoots O, Van Tol HH. The human dopamine D4 receptor repeat sequences modulate expression.
Pharmacogenomics J. 2003;3(6):343-348.
23. Perkins KA, Lerman C, Grottenthaler A, et al. Dopamine and opioid gene variants are associated with
increased smoking reward and reinforcement owing to negative mood. Behav Pharmacol. 2008;19(5-
6):641-649.
24. Franklin TR, Lohoff FW, Wang Z, et al. DAT genotype modulates brain and behavioral responses
elicited by cigarette cues. Neuropsychopharmacology. 2009;34(3):717-728.
25. Franklin TR, Wang Z, Li Y, et al. Dopamine transporter genotype modulation of neural responses to
smoking cues: confirmation in a new cohort. Addict Biol. 2011;16(2):308-322.
26. Chen J, Lipska BK, Halim N, et al. Functional analysis of genetic variation in catechol-O-
methyltransferase (COMT): effects on mRNA, protein, and enzyme activity in postmortem human
brain. Am J Hum Genet. 2004;75(5):807-821.
27. Pfeiffer A, Brantl V, Herz A, et al. Psychomimesis mediated by kappa opiate receptors. Science.
1986;233(4765):774-776.
28. Carlezon WA Jr, Beguin C, DiNieri JA, et al. Depressive-like effects of the kappa-opioid receptor
agonist salvinorin A on behavior and neurochemistry in rats. J Pharmacol Exp Ther. 2006;316(1):
440-447.
29. Zhang Y, Butelman ER, Schlussman SD, et al. Effect of the kappa opioid agonist R-84760 on cocaine-
induced increases in striatal dopamine levels and cocaine-induced place preference in C57BL/6J mice.
Psychopharmacology (Berl). 2004;173(1-2):146-152.
30. Zhang Y, Butelman ER, Schlussman SD, et al. Effect of the endogenous kappa opioid agonist
dynorphin A (1-17) on cocaine-evoked increases in striatal dopamine levels and cocaine-induced place
preference in C57BL/6J mice. Psychopharmacology (Berl). 2004;172(4):422-429.
31. Chefer VI, Czyzyk T, Bolan EA, et al. Endogenous kappa-opioid receptor systems regulate mesoaccum-
bal dopamine dynamics and vulnerability to cocaine. J Neurosci. 2005;25(20):5029-5037.
32. Chartoff E, Sawyer A, Rachlin A, et al. Blockade of kappa opioid receptors attenuates the devel-
opment of depressive-like behaviors induced by cocaine withdrawal in rats. Neuropharmacology.
2012;62(1):167-176.
33. Beardsley PM, Howard JL, Shelton KL, et al. Differential effects of the novel kappa opioid receptor
antagonist, JDTic, on reinstatement of cocaine-seeking induced by footshock stressors vs cocaine
primes and its antidepressant-like effects in rats. Psychopharmacology (Berl). 2005;183(1):118-126.
34. Carey AN, Borozny K, Aldrich JV, et al. Reinstatement of cocaine place-conditioning prevented by the
peptide kappa-opioid receptor antagonist arodyn. Eur J Pharmacol. 2007;569(1-2):84-89.
35. Land BB, Bruchas MR, Lemos JC, et al. The dysphoric component of stress is encoded by activation of
the dynorphin kappa-opioid system. J Neurosci. 2008;28(2):407-414.
36. Wassum KM, Ostlund SB, Maidment NT, et al. Distinct opioid circuits determine the palatability and
the desirability of rewarding events. Proc Natl Acad Sci USA. 2009;106(30):12512-12517.
37. Contet C, Kieffer BL, Befort K. Mu opioid receptor: a gateway to drug addiction. Curr Opin Neurobiol.
2004;14(3):370-378.
38. LeMerrer J, Becker JA, Befort K, et al. Reward processing by the opioid system in the brain. Physiol Rev.
2009;89(4):1379-1412.
39. Wee S, Koob GF. The role of the dynorphin-kappa opioid system in the reinforcing effects of drugs of
abuse. Psychopharmacology (Berl). 2010:210(2):121-135.
40. Bruchas MR, Land BB, Chavkin C. The dynorphin/kappa opioid system as a modulator of stress-
induced and pro-addictive behaviors. Brain Res. 2010;1314:44-55.
41. Van’t Veer A, Yano JM, Carroll FI, et al. Corticotropin-releasing factor (CRF)-induced disruption of
attention in rats is blocked by the kappa-opioid receptor antagonist JDTic. Neuropsychopharmacology.
2012;37(13):2809-2816.
42. Pradhan AA, Befort K, Nazaki C, et al. The delta opioid receptor: an evolving target for the treatment
of brain disorders. Trends Pharmacol Sci. 2011;32(10):581-590.
43. Knapman A, Connor M. Cellular signaling of non-synonymous single nucleotide polymorphisms of the
human mu-opioid receptor (OPRM1). Br J Pharmacol. 2015;172(2):349-363.
44. Crist RC, Berrettini WH. Pharmacogenetics of OPRM1. Pharmacol Biochem Behav. 2014;123:25-33.
45. Bauer IE, Soares JC, Nielsen DA. The role of opioidergic genes in the treatment outcome of drug
addiction pharmacotherapy: a systematic review. Am J Addict. 2015;24(1):15-23.
46. Bond C, LaForge KS, Tian M. Single-nucleotide polymorphism in the human mu opioid receptor gene
alters beta-endorphin binding and activity: possible implications for opiate addiction. Proc Natl Acad
Sci USA. 1998;95(16):9608-9613.
47. Kumar D, Chakraborty J, Das S. Epistatic effects between variants of kappa-opioid receptor gene and
A118G of mu-opioid gene increase susceptibility to addiction in Indian population. Prog Neuropsycho-
pharmacol Biol Psychiatry. 2012;36(2):225-230.
48. Deb I, Chakraborty J, Gangopadhyay PK, et al. Single-nucleotide polymorphism (A118G) in exon 1 of
OPRM1 gene causes alteration in downstream signaling by mu-opioid receptor and may contribute to
the genetic risk for addiction. J Neurochem. 2010;112(2):486-496.
49. Kapur S, Sharad S, Singh RA, et al. A118g polymorphism in mu opioid receptor gene (oprm1): associa-
tion with opiate addiction in subjects of Indian origin. J Integr Neurosci. 2007;6(4):511-522.
50. Ray LA, Hutchison KE. A polymorphism of the mu-opioid receptor gene (OPRM1) and sensitivity to the
effects of alcohol in humans. Alcohol Clin Exp Res. 2004;28(12):1789-1795.
51. Arias AJ, Armeli S, Gelernter J, et al. Effects of opioid receptor gene variation on targeted nalmefene
treatment in heavy drinkers. Alcohol Clin Exp Res. 2008;32(7):1159-1166.
52. Davenport KE, Houdi AA, Van Loon GR. Nicotine protects against mu-opioid receptor antagonism by
beta-funaltrexamine: evidence for nicotine-induced release of endogenous opioids in brain. Neurosci
Lett. 1990;113(1):40-46.
53. Pomerleau OF. Endogenous opioids and smoking: a review of progress and problems. Psychoneuro-
endocrinology. 1998;23(2):115-130.
54. Walters CL, Cleck JN, Kuo YC, et al. Mu-opioid receptor and CREB activation are required for nicotine
reward. Neuron. 2005;46(6):933-943.
55. Zhang L, Kendler KS, Chen X. The mu-opioid receptor gene and smoking initiation and nicotine depen-
dence. Behav Brain Funct. 2006;2:28.
56. Xuei X, Dick D, Flury-Wetherill L, et al. Association of the kappa-opioid system with alcohol depen-
dence. Mol Psychiatry. 2006;11(11):1016-1024.

57. Wang SC, Tsou HH, Chung RH, et al. The association of genetic polymorphisms in the kappa-opioid
receptor 1 gene with body weight, alcohol use, and withdrawal symptoms in patients with methadone
maintenance. J Clin Psychopharmacol. 2014;34(2):205-211.
58. Levran O, Londono D, O’Hara K, et al. Genetic susceptibility to heroin addiction: a candidate gene
association study. Genes Brain Behav. 2008;7(7):720-729.
59. Yuferov V, Fussell D, LaForge KS, et al. Redefinition of the human kappa opioid receptor gene (OPRK1)
structure and association of haplotypes with opiate addiction. Pharmacogenetics. 2004;14(12):
793-804.
60. Levran O, Londono D, O’Hara K, et al. Heroin addiction in African Americans: a hypothesis-driven
association study. Genes Brain Behav. 2009;8(5):531-540.
61. Levran O, Yuferov V, Kreek MJ. The genetics of the opioid system and specific drug addictions. Hum
Genet. 2012;131(6):823-842.
62. Clarke TK, Krause K, Li T, et al. An association of prodynorphin polymorphisms and opioid dependence
in females in a Chinese population. Addict Biol. 2009;14(3):366-370.
63. Yuferov V, Ji F, Nielsen DA, et al. A functional haplotype implicated in vulnerability to develop cocaine
dependence is associated with reduced PDYN expression in human brain. Neuropsychopharmacology.
2009;34(5):1185-1197.
64. Wei SG, Zhu YS, Lai JH, et al. Association between heroin dependence and prodynorphin gene poly-
morphisms. Brain Res Bull. 2011;85(3-4):238-242.
65. Zhang H, Kranzler HR, Yang BZ, et al. The OPRD1 and OPRK1 loci in alcohol or drug dependence:
OPRD1 variation modulates substance dependence risk. Mol Psychiatry. 2008;13(5):531-543.
66. Mayer P, Rochlitz H, Rauch E, et al. Association between a delta opioid receptor gene polymorphisms
and heroin dependence in man. Neuroreport. 1997;8(11):2547-2550.
67. Nelson EC, Lynskey MT, Heath AC, et al. Association of OPRD1 polymorphisms with heroin depen-
dence in a large case-control series. Addict Biol. 2014;19(1):111-121.
68. Crist RC, Clarke TK, Ang A, et al. An intronic variant in OPRD1 predicts treatment outcome for opioid
dependence in African-Americans. Neuropsychopharmacology. 2013;38(10):2003-2010.
69. Clarke TK, Crist RC, Ang A, et al. Genetic variation in OPRD1 and the response to treatment for opioid
dependence with buprenorphine in European-American females. Pharmacogenomics J. 2014;14(3):
303-308.
70. Zhang H, Gelernter J, Gruen JR, et al. Functional impact of a single-nucleotide polymorphism in the
OPRD1 promoter region. J Hum Genet. 2010;55(5):278-284.
71. Balfour DJ. Neuroplasticity within the mesoaccumbens dopamine system and its role in tobacco
dependence. Curr Drug Targets CNS Neurol Disord. 2002;1(4):413-421.
72. Benowitz NL. Pharmacology of nicotine: addiction, smoking-induced disease, and therapeutics. Annu
Rev Pharmacol Toxicol. 2009;49:57-71.
73. Dani JA, Jenson D, Broussard JI, et al. Neurophysiology of nicotine addiction. J Addict Res Ther.
2011;S1(1).
74. D’Souza MS, Markou A. Neuronal mechanisms underlying development of nicotine dependence: impli-
cations for novel smoking-cessation treatments. Addict Sci Clin Pract. 2011;6(1):4-16.
75. Dani JA, DeBiasi M. Mesolimbic dopamine and habenulo-interpeduncular pathways in nicotine with-
drawal. Cold Spring Harb Perspect Med. 2013;3(6).
76. Benowitz NL. Nicotine addiction. N Engl J Med. 2010;362(24):2295-2303.
77. Mansvelder HD, Keath JR, McGehee DS. Synaptic mechanisms underlie nicotine-induced excitability of
brain reward areas. Neuron. 2002;33(6):905-919.
78. Wooltorton JR, Pidoplichko VI, Broide RS, et al. Differential desensitization and distribution of nicotinic
acetylcholine receptor subtypes in midbrain dopamine areas. J Neurosci. 2003;23(8):3176-3185.
79. Hukkanen JP, Jacob P III, Benowitz NL. Metabolism and disposition kinetics of nicotine. Pharmacol Rev.
2005;57(1):79-115.
80. Marks MJ, Burch JB, Collins AC. Effects of chronic nicotine infusion on tolerance development and
nicotinic receptors. J Pharmacol Exp Ther. 1983;226(3):817-825.
81. Schwartz RD, Kellar KJ. Nicotinic cholinergic receptor binding sites in the brain: regulation in vivo.
Science. 1983;220(4593):214-216.
82. Buisson B, Bertrand D. Chronic exposure to nicotine upregulates the human (alpha)4(beta)2 nicotinic
acetylcholine receptor function. J Neurosci. 2001;21(6):1819-1829.

83. Carboni E, Bortone L, Giua C, et al. Dissociation of physical abstinence signs from changes in extra-
cellular dopamine in the nucleus accumbens and in the prefrontal cortex of nicotine dependent rats.
Drug Alcohol Depend. 2000;58(1-2):93-102.
84. Kalamida D, Poulas K, Avramopoulou V, et al. Muscle and neuronal nicotinic acetylcholine receptors:
structure, function, and pathogenicity. FEBS J. 2007;274(15):3799-3845.
85. Xiao Y, Kellar KJ. The comparative pharmacology and up-regulation of rat neuronal nicotinic recep-
tor subtype binding sites stably expressed in transfected mammalian cells. J Pharmacol Exp Ther.
2004;310(1):98-107.
86. Wu J, Liu Q, Yu K, et al. Roles of nicotinic acetylcholine receptor beta subunits in function of human
alpha4-containing nicotinic receptors. J Physiol. 2006;576(Pt 1):103-118.
87. Kuryatov A, Onksen J, Lindstrom J. Roles of accessory subunits in alpha4beta2(*) nicotinic receptors.
Mol Pharmacol. 2008;74(1):132-143.

88. Bierut LJ, Stitzel JA, Wang JC, et al. Variants in nicotinic receptors and risk for nicotine dependence.
Am J Psychiatry. 2008;165(9):1163-1171.
89. Munafo MR, Timofeeva MN, Morris RW, et al. Association between genetic variants on chromosome
15q25 locus and objective measures of tobacco exposure. J Natl Cancer Inst. 2012;104(10):740-748.
90. Saccone SF, Hinrichs AL, Saccone NL, et al. Cholinergic nicotinic receptor genes implicated in a nico-
tine dependence association study targeting 348 candidate genes with 3713 SNPs. Hum Mol Genet.
2007;16(1):36-49.
91. Saccone NL, Saccone SF, Hinrichs AL, et al. Multiple distinct risk loci for nicotine dependence identi-
fied by dense coverage of the complete family of nicotinic receptor subunit (CHRN) genes. Am J Med
Genet B Neuropsychiatr Genet. 2009;150B(4):453-466.
92. Le Marchand L, Derby KS, Murphy SE, et al. Smokers with the CHRNA lung cancer-associated variants
are exposed to higher levels of nicotine equivalents and a carcinogenic tobacco-specific nitrosamine.
Cancer Res. 2008;68(22):9137-9140.
93. Thorgeirsson TE, Geller F, Sulem P, et al. A variant associated with nicotine dependence, lung cancer
and peripheral arterial disease. Nature. 2008;452(7187):638-642.
94. Bousman CA, Rivard C, Haese JD, et al. Alpha-5 and -3 nicotinic receptor gene variants predict nico-
tine dependence but not cessation: findings from the COMMIT cohort. Am J Med Genet B Neuro-
psychiatr Genet. 2012;159B(2):227-235.
95. Slimak MA, Ables JL, Frahm S, et al. Habenular expression of rare missense variants of the beta4
nicotinic receptor subunit alters nicotine consumption. Front Hum Neurosci. 2014;8:12.
96. Haller G, Druley T, Vallania FL, et al. Rare missense variants in CHRNB4 are associated with reduced
risk of nicotine dependence. Hum Mol Genet. 2012;21(3):647-655.
97. Pianezza ML, Sellers EM, Tyndale RF. Nicotine metabolism defect reduces smoking. Nature.
1998;393(6687):750.
98. Schoedel KA, Hoffman EB, Rao Y, et al. Ethnic variation in CYP2A6 and association of genetically slow
nicotine metabolism and smoking in adult Caucasians. Pharmacogenetics. 2004;14(9):615-626.
99. Ho MK, Mwenifumbo JC, Zhao B, et al. A novel CYP2A6 allele, CYP2A6*23, impairs enzyme function
in vitro and in vivo and decreases smoking in a population of Black-African descent. Pharmacogenet
Genomics. 2008;18(1):67-75.
100. Liu T, David SP, Tyndale RF, et al. Associations of CYP2A6 genotype with smoking behaviors in
southern China. Addiction. 2011;106(5):985-994.
101. Zhu AZ, Binnington MJ, Renner CC, et al. Alaska native smokers and smokeless tobacco users with
slower CYP2A6 activity have lower tobacco consumption, lower tobacco-specific nitrosamine expo-
sure, and lower tobacco-specific nitrosamine bioactivation. Carcinogenesis. 2012;34(1):93-101.
102. Kubota T, Nakajima-Taniguchi C, Fukuda T, et al. CYP2A6 polymorphisms are associated with nico-
tine dependence and influence withdrawal symptoms in smoking cessation. Pharmacogenomics J.
2006;6(2):115-119.
103. Rao Y, Hoffman E, Zia M, et al. Duplications and defects in the CYP2A6 gene: identification,
genotyping, and in vivo effects on smoking. Mol Pharmacol. 2000;58(4):747-755.
104. Raunio H, Rahnasto-Rilla M. CYP2A6: genetics, structure, regulation, and function. Drug Metab Drug
Interact. 2012;27(2):73-88.
105. Fukami T, Nakajima M, Yamanaka H, et al. A novel duplication type of CYP2A6 gene in African-
American population. Drug Metab Dispos. 2007;35(4):515-520.

106. Wang J, Pitarque M, Ingelman-Sundberg M. 3′-UTR polymorphism in the human CYP2A6 gene affects
mRNA stability and enzyme expression. Biochem Biophys Res Commun. 2006;340(2):481-497.
107. Mwenifumbo JC, Lessov-Schlaggar CN, Zhou Q, et al. Identification of novel CYP2A6*1B variants: the
CYP2A6*1B allele is associated with faster in vivo metabolism. Clin Pharmacol Ther. 2008;83(1):115-121.
108. Mwenifumbo JC, Tyndale RF. Genetic variability in CYP2A6 and the pharmacokinetics of nicotine.
109. Al Koudsi N, Tyndale RF. Hepatic CYP2B6 is altered by genetic, physiologic, and environmental factors
but plays little role in nicotine metabolism. Xenobiotica. 2010;40(6):381-392.
110. Binnington MJ, Zhu AZ, Renner CC, et al. CYP2A6 and CYP2B6 genetic variation and its association
with nicotine metabolism in South Western Alaska Native people. Pharmacogenet Genomics.
2012;22(6):429-440.
111. Yamanaka H, Nakajima M, Nishimura K, et al. Metabolic profile of nicotine in subjects whose CYP2A6
gene is deleted. Eur J Pharm Sci. 2004;22(5):419-425.

112. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed.
Arlington, VA: American Psychiatric Association; 2013.
113. Thomasson HR, Crabb DW, Edenberg HJ, et al. Alcohol and aldehyde dehydrogenase polymorphisms
and alcoholism. Behav Genet. 1993;23:131-136.
114. Agarwal DP. Genetic polymorphisms of alcohol metabolizing enzymes. Pathol Biol. 2001; 49(9):
703-709.
115. Kenna GA, McGeary JE, Swift RM. Pharmacotherapy, pharmacogenomics, and the future of alcohol
dependence treatment, part 1. Am J Health-Syst Pharm. 2004;61:2272-2279.
116. Oslin DW, Berrettini W, Kranzler HR, et al. A functional polymorphism of the μ-opioid receptor gene
is associated with naltrexone response in alcohol-dependent patients. Neuropsychopharmacology.
2003;28:1546-1552.
117. Anton RF, O’Malley SS, Ciraulo DA, et al. Combined pharmacotherapies and behavioral interven-
tions for alcohol dependence: the COMBINE study: a randomized controlled trial. JAMA. 2006;
295(17):2003-2017.
118. Ray LA, MacKillop J, Hutchinson KE. Pharmacogenetics of the mu-opioid receptor and the treatment
of addictions. Personalized Medicine. 2007;4(2):217-220.
119. Crettol S, Déglon JJ, Besson J, et al. ABCB1 and cytochrome P450 genotypes and phenotypes:
influence on methadone plasma levels and response to treatment. Clin Pharmacol Ther.
2006;80(6):668-681.
120. Levran O, Yeferov V, Kreek MJ. The genetics of the opioid system and specific drug addictions.
Hum Genet. 2012;131:823-842.
121. Cone EJ, Gorodetzky CW, Yousefnejad D, et al. The metabolism and excretion of buprenorphine in
humans. Drug Metab Dispos. 1984;12(5):577-581.
122. Chang Y, Moody DE. Glucuronidation of buprenorphine and norbuprenorphine by human liver
microsomes and UDP-glucuronosyltransferases. Drug Metab Lett. 2009;3(2):101-107.
123. Zdanowicz MM, Adams PW. The pharmacogenetics of nicotine dependence and smoking cessation
therapies. J Pharmacogenomics Pharmacoproteomics. 2014;5(4):1-20. doi:10.4172/2153-0645.1000138.
124. Munafo MR, Shields AE, Berrettini WH, et al. Pharmacogenetics and nicotine addiction treatment.
125. Malaiyandi V, Lerman C, Benowitz NL, et al. Impact of CYP2A6 genotype on pretreatment smoking
behavior and nicotine levels from and usage of nicotine replacement therapy. Mol Psychiatry.
2006;11(4):400-409.
126. Johnstone EC, Elliot KM, David SP, et al. Association of COMT Val108/158Met genotype with smoking
cessation in a nicotine replacement therapy randomized controlled trial. Cancer Epidemiol Biomarkers
Prev. 2007;16(6):1065-1069.
127. Chen LS, Baker TB, Piper ME, et al. Interplay of genetic risk factors (CHRNA5-CHRNA3-CHRNB4) and
cessation treatments in smoking cessation success. Am J Psychiatry. 2012;169(7):735-742.
128. Lerman C, Jepson EP, Wileyto PE, et al. Role of functional genetic variation in the dopamine D2
receptor (DRD2) in response to bupropion and nicotine replacement therapy for tobacco dependence:
results of two randomized clinical trials. Neuropsychopharmacol. 2006;31(1):231-242.
129. Lerman C, Wileyto EP, Patterson F, et al. The functional μ-opioid receptor (OPRM1) Asn40Asp variant
predicts short-term response to nicotine replacement therapy in a clinical trial. Pharmacogenomics J.
2004;4(3):184-192.
130. Zyban [package insert]. Research Triangle Park, NC: GlaxoSmithKline; January 2015.
131. Lerman C, Shields PG, Wileyto EP, et al. Pharmacogenetic investigation of smoking cessation treat-
ment. Pharmacogenetics. 2002;12(8):627-634.
132. Chantix (package insert). New York, NY: Pfizer Labs; October 2014.
133. King DP, Paciga S, Pickering E, et al. Smoking cessation pharmacogenetics: analysis of varenicline and
bupropion in placebo-controlled clinical trials. Neuropsychopharmacology. 2012;37:641-650.
134. Garrison GD, Dugan SE. Varenicline: a first-line treatment option for smoking cessation. Clin Ther.
2009;31:463-491.
135. Swan GE, Javitz HS, Jack LM, et al. Varenicline for smoking cessation: nausea severity and variation in
nicotinic receptor genes. Pharmacogenomics J. 2012;12:349-358.
136. Saccone NL, Schwantes-An TH, Wang JC, et al. Multiple cholinergic nicotinic receptor genes affect
nicotine dependence risk in African and European Americans. Genes Brain Behav. 2010; 9(7):741-750.
CHAPTER
14
Pharmacogenomics and
Diabetes Mellitus
Launa M. J. Lynch, PhD and Kimberly L. Barefield, PharmD, BCPS, CDE

After completing this chapter, the reader DEOXYRIBONUCLEIC ACID (DNA)—a self-
should be able to replicating material present in nearly all
Identify genetic variations that are
of chromosomes.
associated with impacting the clinical
response of oral drugs used in the FASTING BLOOD GLUCOSE—a blood glucose
treatment of diabetes mellitus. test done when the person has been
fasting; also referred to as fasting plasma
Differentiate between the genetic
glucose (FPG). Fasting is defined as no
variations with clinical utility and those
caloric intake for at least 8 hours.
that still need further investigation.
GAIN-OF-FUNCTION ALLELE—a genetic
Identify the polymorphisms associated
mutation resulting in enhanced or new
with an increased risk of developing
protein function.
diabetes mellitus.
GLYCATED HEMOGLOBIN (HbA1c)—a form of
Design a medication therapy plan based
hemoglobin generated when hemoglobin
on the genetic information for metformin
binds with glucose. The glycated
and sulfonylureas.
hemoglobin persists in the body for
8-12 weeks. It is used to determine the
average plasma glucose levels and
provides information on long-term blood
glucose control.
LOSS-OF-FUNCTION ALLELE—a genetic
mutation, which results in a reduced or
absent protein function.
ORAL GLUCOSE TOLERANCE TEST (OGTT)—
measures glucose levels after a person
drinks an equivalent of 75 g glucose load;
a positive diabetic test would measure
≥200 mg/dL at 2 hours post-glucose load.
SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—
the most common genetic variation in

which there is a variation in a single
nucleotide, a DNA building block.
405
TYPE 1 DIABETES MELLITUS (T1DM)—a disease that results from β-cell destruction, usually
leading to absolute insulin deficiency and hyperglycemia.
TYPE 2 DIABETES MELLITUS (T2DM)—a progressive disease characterized by insulin
resistance and a progressive insulin secretory defect. As pancreatic β-cell dysfunction
progresses, there is worsening of blood glucose level control.
INTRODUCTION
Diabetes mellitus currently affects 29.1 million Americans—9.3% of the population—and is
the primary cause of death for 69,071 patients each year.1 Approximately 90-95% of diabet-
ic patients are diagnosed with type 2 diabetes mellitus (T2DM). This disease increases the
risk for microvascular complications leading to renal failure, blindness, and amputations
as well as cardiovascular complications. The estimated direct costs for patients diagnosed
with diabetes mellitus is $176 billion per year and $69 billion in indirect costs (i.e., disability,
work loss, premature mortality).
Currently, patients with T2DM are most commonly treated with metformin, if not
contraindicated and if tolerated by the patient, or another oral agent.2 However, a patient-
centered approach is recommended based on patient preferences, costs, adverse effects,
hypoglycemia risk, and effect on weight.2,3 After 3 months of initial therapy, if the glycated
hemoglobin (HbA1c) level is not at goal of <7%, the practitioner should consider a combina-
tion of metformin with one of the following options discussed later in the chapter: sulfo-
nylureas, thiazolidinediones, DPP-4 inhibitors, SGLT2 inhibitors, GLP-1 receptor agonists,
or basal insulin.2 The meglitinides may be used in place of the sulfonylureas in patients
with irregular meal patterns or in the case of postprandial hypoglycemia on sulfonylurea
therapy. Other drugs (i.e., α-glucosidase inhibitors, pramlintide, bromocriptine) are used
less frequently due to modest efficacy, adverse effects, and frequency of administration.
Combination dual therapy should be considered in patients with an HbA1c level ≥9% for
timelier lowering of the HbA1c level to goal range.
There is considerable interpatient variability in the effect of oral antidiabetic drugs
(OADs) that is both nonbiological and biological. The variability in drug disposition and
tolerability may be related to nonbiological factors such as compliance, physician prescrib-
ing practices, or patient access to healthcare. Biological factors are related to the phar-
macokinetic and pharmacodynamics characteristics of the drug.4 These biologic variations
may be due to genetic factors that influence the efficacy and safety of OADs because they
may alter drug absorption, hepatic uptake and metabolism, renal excretion, and the extent
of systemic drug distribution.
CLINICAL CASE STUDY−PHARMACOGENOMIC

CONSIDERATIONS FOR ORAL ANTIDIABETIC
THERAPY
After being diagnosed with T2DM, a 68-year-old man presents to the pharmacotherapy
clinic. His laboratory results show a serum creatinine of 1.1 mg/dL and a HbA1c of 9.2%.
The patient is started on dual therapy with metformin 500 mg bid and glipizide 5 mg
bid, along with counseling on diet and exercise. The patient has a past medical history
Chapter 14 • Pharmacogenomics and Diabetes Mellitus 407
of myocardial infarction and is currently taking ASA 162 mg daily, metoprolol tartrate
50 mg bid, atorvastatin 40 mg daily, and lisinopril 20 mg daily. Three months later, the
patient returns for followup to evaluate this current therapy. His laboratory results show
HbA1c is 6.8%, and he is complaining of frequent hypoglycemia in the morning as well
as gastrointestinal adverse effects.
Questions
1. What genetic polymorphisms may explain the patient’s response to the prescribed antidia-
betic regimen?
2. What other therapeutic options might be considered for treatment?
INSULIN SECRETAGOGUES
Sulfonylureas
Sulfonylureas are divided into first- and second-generation agents. Six sulfonylureas are
currently available in the United States. The available first-generation sulfonylureas are
tolbutamide, tolazamide, and chlorpropamide. Glyburide, glipizide, and glimepiride are the
second-generation sulfonylureas available by prescription. Sulfonylureas inhibit the ATP-
sensitive potassium (KATP) channels on the pancreatic β-cells. The KATP channels control
the resting membrane potential of pancreatic β-cells. Closure of the KATP channel causes
membrane depolarization of the pancreatic β-cell and stimulates the release of insu-
lin.5,6 The KATP channel is composed of four sulfonylurea receptor 1 (SUR1) subunits and
four inward-rectifier potassium ion channel (Kir6.2) subunits. The SUR1 is the regulatory
subunit, while the Kir6.2 forms the pore of the KATP channel.7,8 The SUR1 regulatory subunit
is encoded by the ATP-binding cassette, sub-family C, member 8 (ABCC8) gene. The Kir6.2
pore subunit is encoded by the potassium inwardly-rectifying channel, subfamily J, member
11 (KCNJ11) gene.6 In these genes, gain-of-function mutations can cause the KATP channel to
be in an open state that hyperpolarizes the membrane and impairs insulin release. Muta-
tions that cause a loss of function promote closure of the KATP channel and cause the
membrane to depolarize, which in turn leads to insulin hypersecretion.6,9
The binding sites for sulfonylurea on the KATP channel have been identified.10 The
A-site binds to the sulfonylurea moiety of the drug and is exclusively found on the SUR1
subunit. The B-site binds to the carboxamido moiety of the drug and is found in the SUR1
subunit and the Kir6.2 subunit.11 The majority of first-generation sulfonylureas (gliclazide,
chlorpropamide, tolbutamide, and tolazamide) bind to the A-site. The majority of the
second-generation sulfonylureas (glibenclamide, glipizide, glyburide, and glimepiride4)
binds to both sites—designated as the AB-site. Examples of an A-site binding drug and an
AB-site binding drug are shown in Figure 14-1.
ABCC8 Gene Variants’ Effect on Sulfonylurea-Induced Insulin Secretion
The polymorphisms in the ABCC8 gene have been investigated by multiple studies. The −3c
→ t (rs1799854) polymorphism and the synonymous Thr759Thr (rs1801261) polymorphism

have been associated with a 40% decrease in insulin secretion after a tolbutamide injec-
tion.12 Population studies done with Caucasian, Danish, and French Caucasian individuals
have found an association between carriers of the −3c → t and the Thr759Thr polymor-
phisms and an increased risk of developing T2DM.12-14 One study found an association
between the −3c → t polymorphism and development of T2DM in Dutch Caucasians.15 In
Chlorpropamide
CI
Sulfonylurea moiety
Glibenclamide
Sulfonylurea moiety
CI
Carboxamido moiety
FIGURE 14-1 Structures and binding sites for sulfonylurea drugs. There are two binding
sites for a sulfonylurea on the KATP channel, an A-site and a B-site. The A-site binds to the
sulfonylurea moiety. The B-site binds to the carboxyamido moiety. The majority of the first-
generation sulfonylureas bind to the A-site. The second-generation sulfonylureas have higher
binding affinity because they have both the sulfonylurea and carboxyamido moieties and
bind to the AB-site.
studies done with populations from north and south India, researchers did not find an asso-
ciation between the −3c → t polymorphism and development of T2DM.16-18 Data from two
studies suggest the −3c → t polymorphism does not influence sulfonylurea response.19,20
Based on the limited study design for both studies, further investigation is merited.
A common polymorphism of the ABCC8 gene is the Ser1369Ala (S1369A, rs757110)
variant, which has been consistently associated with differences in T2DM patient response
to sulfonylurea treatment. Zhang and colleagues investigated the effects of the Ser1369A-
la variant in patients with T2DM treated with 40 mg of gliclazide for 8 weeks.21 At week 8,
the A allele carriers had a 1.6% reduction in HbA1c levels and the S/S homozygote patients
had a 0.76% reduction in HbA1c levels (P = 0.044) when compared to baseline. The A allele
seemed to confer an increased sensitivity to sulfonylurea treatment that resulted in a

greater reduction in HbA1c levels. In these two patient groups, no change was observed in
fasting plasma glucose (FPG) levels. The discrepancy between the HbA1c levels and the
FPG levels is unclear, but the HbA1c levels capture the postprandial glucose levels that are
not reflected in FPG levels and could explain the differences.21
Feng and colleagues evaluated the influences of gene variants in the ABCC8 gene in
patient response to a gliclazide 40 mg twice daily treatment for 8 weeks.22 The patients
(n = 1,268) were divided into two cohorts based on the geographic location of the treat-
ment hospital. In a pooled analysis of all participants, there was a 2.8% decrease in FPG
in patients with the S/A genotype and a 7.7% decrease in FPG in A/A genotype carriers
when compared to patients with the S/S genotype. Patients received an oral glucose toler-
ance test (OGTT) and had their glucose levels tested 2 hours later. At the 2-hour mark,
serum glucose levels in patients with the S/A genotype showed −10.8% and with the A/A
genotype showed −11.9% when compared to patients with the S/S genotype. Reductions in
the HbA1c levels from baseline to week 8 were 1.5% in the S/A genotype, 1.7% in the A/A
genotype, and 1.4% in the S/S genotype. These results were not significant between the
groups. The odds of responding to gliclazide therapy were higher with the S/A genotype
(OR: 1.4, 95% CI: 1.0−2.1) and the A/A genotype (OR: 2.2; 95% CI: 1.4−3.6) when compared to
S/S homozygotes.22
Two studies have investigated the association of the Ser1369Ala variant of the ABCC8
gene and the risk of severe hypoglycemia due to sulfonylurea treatment. Thirty-two
Japanese patients with T2DM and 111 German patients with T2DM treated with a sulfo-
nylurea who were admitted to the emergency department with severe hypoglycemia were
genotyped for the Ser1369Ala polymorphism. In these patients, no association was found
between the Ser1369Ala polymorphism and increased risk of hypoglycemia.23,24
KCNJ11 Gene Variant Effect on Sulfonylurea-Stimulated Insulin Secretion
The Glu23Lys (E23K, rs5219) is the most widely studied polymorphism of the KCNJ11 gene
and is referred to as the variant K23 allele. The variant K23 allele has been identified as
a risk allele for T2DM. In some clinical studies, the variant K allele was associated with
decreased insulin secretion in humans as the result of a gain-of-function mutation.7,25,26
Patients with the K23 variant allele tend to have lower levels of insulin secretion when
treated with a sulfonylurea. As a result of this decreased insulin secretion, these patients
experience fewer hypoglycemic adverse side events when compared to those with the
wild-type allele. Holstein and colleagues observed that 46% of patients with T2DM with
the K23 variant allele were associated with a decreased risk of sulfonylurea-induced hypo-
glycemia when compared to the 31% who experienced hypoglycemia events (OR: 0.54; 95%
CI: 0.30−0.98; P = 0.04). Those patients with the K23 variant allele had higher HbA1c levels
compared to patients who had the wild-type E allele.29
In the United Kingdom Prospective Diabetes Study (UKPDS), no significant relation-
ship was found between being a carrier of the variant K23 allele and sulfonylurea failure
in 364 newly diagnosed patients with T2DM. In this study, the patients were treated with
chlorpropamide.4,27 Sesti and colleagues conducted a study comprised of 525 Caucasian

patients with T2DM treated with glibenclamide. In their study, 66.8% of patients who
failed sulfonylurea treatment were found to have the variant K23 allele as compared to
the 58% of patients who did not fail sulfonylurea therapy (OR: 1.45; 95% CI: 1.01−2.09).28
Several reasons may explain the discrepancies between the UKPDS and Sesti and
colleagues’ findings:
1. The definitions of secondary sulfonylurea treatment failure were different; the UKPDS au-
thors defined sulfonylurea failure as patients who needed additional therapy to control
hyperglycemia. Sesti and colleagues defined sulfonylurea failure as progression to insulin
therapy.
2. The duration of therapy before failure was different for both studies. The UKPDS duration
was 1 year, which may not have allowed for those patients with the K23 variant allele to ex-
perience sulfonylurea failure. The study conducted by Sesti and colleagues had a duration
of 12 years.
3. The two studies used different sulfonylureas. The UKPDS used chlorpropamide as the treat-
ment drug (which binds to the A-site of the KATP channel), while Sesti and colleagues used
the sulfonylurea glibenclamide (which binds to the AB-site of the KATP channel).4
4. The UKPDS recruited newly diagnosed patients who were treatment naïve, while Sesti and
colleagues recruited patients who had been treated previously for T2DM. Patients who were
newly diagnosed would be expected to have better β-cell function compared with those pa-
tients who had the disease for a longer period of time.26-28,30
The KCNJ11 and ABCC8 genes are next to each other on chromosome 11, and they are
separated by only 5 kb. A strong linkage disequilibrium occurs due to the close proximity
of the two genes. Most carriers of a KCNJ11 K23 allele also usually carry an ABCC8 Ala1369
risk allele.31,32 Lang and colleagues compared the structure-activity relationships in a KATP
channel containing the E23/S1369, a no-risk haplotype, to a KATP channel containing the
risk K23/A13639 haplotype. The KATP channels containing the risk K23/A13639 haplotype
were significantly less sensitive to inhibition by chlorpropamide, glimepiride, and tolbuta-
mide.33
Neonatal Diabetes Mellitus Linked to Mutations in KCNJ11 and ABCC8 Genes
Neonatal diabetes mellitus (NDM) occurs when insulin therapy is required during the first 6
months of life. The incidence of the disease is estimated to be 1 in 200,000 to 1 in 400,000
live births.34-38 Transient NDM comprises 50-60% of NDM cases and resolves within 18
months of birth.39 Permanent NDM encompasses 40-60% of NDM cases and requires
insulin treatment throughout life.39 Current evidence indicates that NDM is a monogenic
disorder. Mutations in the insulin gene (INS),40-42 glucokinase gene (GCK),43,44 and the
genes encoding the subunits of the KATP channels in the pancreatic β-cells45-49 (KCNJ11,
ABCC8) are associated with the occurrence of NDM. Heterozygous mutations in the KCNJ11
and ABCC8 genes are associated with the most common forms of NDM.11,30 The gain-of-
function mutations in the KCNJ11 and ABCC8 genes impair insulin release and result in
hyperglycemia in patients with NDM.11
Before the genetic basis for the disease was discovered, the treatment for NDM was
daily insulin therapy. Patients with NDM with either the KCNJ11 or ABCC8 genes have been
successfully treated with sulfonylureas that correct the defect in the KATP channels and
allow these patients to move from insulin injections to long-term sulfonylurea thera-
py.36,50-52 Determination of the sulfonylurea dosage depends on which mutation in the KATP
channel the patient has acquired. Patients with the ABCC8 mutation exhibit a milder case of
NDM. Sulfonylurea treatment should be effective in these patients because there have been
no reports of the mutations in the ABCC8 gene altering sulfonylurea treatment.11 Patients
with the KCNJ11 gene mutation have varying responses to sulfonylurea therapy.53,54 In the
pancreatic β-cell, when levels of ATP increase and bind to the KATP channel, the result is
closure of the channel and ultimately the release of insulin. Patients with the KCNJ11 gene
mutation that reduces the binding affinity of inhibitory ATP have been successfully treated
with a sulfonylurea.55 Mutations that increase the intrinsic probability of the ion chan-
nel being open will require a higher dose of a sulfonylurea to achieve the desired clinical
outcome. For example, a dose for a NDM patient with this type of mutation would be 2.5 mg/
kg/day of glyburide, but a common dose of glyburide for a pediatric patient with T2DM is 0.2
mg/kg/day. In some cases, patients with these types of mutations required a combination of
insulin and a sulfonylurea to achieve the desired therapeutic goals.11
CLINICAL PEARL
NDM is a monogenic disorder. The most common cause of NDM is
mutations in the ABCC8 or KCNJ11 genes of the KATP channel found in
pancreatic β-cells. Identification of these mutations has allowed for the
majority of these patients to be switched from insulin therapy to oral
sulfonylurea therapy. Patients with the ABCC8 gene mutation have milder
cases of NDM and respond well to sulfonylurea therapy. In some cases,
patients with the KCNJ11 gene mutation need a combination of insulin and
a sulfonylurea for treatment of the disease.
TCF7L2 Gene Variants and Sulfonylurea Treatment Failure

The Wnt signaling pathway has roles in regulatory key events during embryonic develop-
ment, including cell fate determination and organogenesis. The extracellular Wnt signal
stimulates a β-catenin dependent and a β-catenin independent pathway. The Wnt/β-
catenin dependent signaling pathway is activated under hyperglycemic conditions. The
transcription factor 7-like 2 (TCF7L2) is a component of the Wnt/β-catenin signaling
pathway. The TCF7L2 gene encodes for the transcription factor (Tcf-4), which is a key
regulator of pancreatic β-cell proliferation and function, glucose homeostasis, production
of glucagon-like peptide 1, and lipid metabolism.56 Polymorphisms in the TCF7L2 gene can
alter the levels of glucagon-like peptide-1 (GLP-1). The two gene polymorphisms of TCF7L2
that are in linkage disequilibrium are rs7903146 (C>T) and rs12255372 (G>T). Both poly-
morphisms are consistently strong predictors of an increased risk of developing T2DM.57-60
Pearson and colleagues used the Genetics of Diabetes Audit and Research Tayside
(GoDARTS) database to determine the impact of TCF7L2 polymorphisms on 901 sulfo-
nylurea-treated Scottish patients with T2DM.61 Failure to attain an HbA1c level of ≤7%
within 1 year of treatment initiation was observed in the rs12255372 T/T homozygotes
more often than with the G/T or G/G genotypes. Additionally, 57% of the rs12255372 T/T
homozygotes failed to reach the HbA1c level target (<7%) when compared to 41% of the
rs12255372 G/T genotypes and 40% of the rs12255372 G/G homozygotes (P = 0.006). The
odds of sulfonylurea treatment failure were higher in patients with the rs12255372 T/T
genotype versus the rs12255372 G/G genotype (OR: 2.16; 95% CI: 1.21−3.86; P = 0.009). The
odds of sulfonylurea treatment failure were higher in patients with the rs7903146 T/T
genotype versus the rs7903146 C/C genotype (OR: 1.90; CI: 1.90−3.33; P = 0.024).61 Schroner
and colleagues found similar results with the rs7903146 polymorphism in 87 patients with
T2DM treated with a combination of sulfonylurea and metformin.62 During the 6 months
of treatment, patients with the rs7903146 T/T or C/T genotypes had a smaller reduction in
HbA1c levels and FPG when compared to the C/C genotype.63
Insulin Receptor Substrate-1 Gene Variants and Sulfonylurea Treatment Failure

The insulin receptor substrate-1 (IRS-1) is an endogenous signal-transduction protein
involved in the insulin signaling pathway. The IRS-1 protein mediates the metabolic effects
of insulin.7,63 Two studies have investigated the effects of the Gly972Arg polymorphism
of the IRS-1 gene on sulfonylurea therapy. Sesti and colleagues investigated the associa-
tion between an increased risk of secondary sulfonylurea failure and the presence of the
Gly972Arg polymorphism (rs1801278) on the IRS-1 gene.64 There were 477 patients with
T2DM enrolled in the study. Secondary sulfonylurea failure was defined as the requirement
of insulin as a result of uncontrolled hyperglycemia (FPG >300 mg/dL). Patients who were
carriers of the Arg972 allele were twice as likely to be at risk of experiencing secondary
sulfonylurea failure as noncarriers. This finding was independent of age, sex, body mass
index, and duration of diabetes mellitus.64
In the second study, Seeringer and colleagues examined the association between
secondary sulfonylurea failure and carriers of the IRS-1 gene Glyc972Arg polymorphism in
213 patients with T2DM treated with insulinotropic drugs (e.g., sulfonylureas, meglitinides,
and/or insulin).65 In the sulfonylureas study, 47 patients had T2DM. Secondary failure was
defined as the point in which the patient was switched from oral drug therapy to insulin
therapy. Patients taking a sulfonylurea who were carriers of the Arg972 allele had aver-
age HbA1c levels of 8.7 ± 1.3% compared to 7.6 ± 1.1% for noncarriers (P = 0.005). For all
213 patients in the study, those who were carriers of the Arg972 allele had average HbA1c
levels of 8.3 ± 1.8% compared to 7.6 ± 1.3% for noncarriers (P = 0.005).65
Nitric Oxide Synthase 1 Adaptor Protein Gene Variants and Decreased Insulin
Release
Nitric oxide synthase 1 adaptor protein (NOS1AP) regulates neuronal nitric oxide synthase
activity. In the pancreatic β-cell, neuronal nitric oxide synthase has been shown to affect
voltage-gated calcium channels and reduce intracellular calcium levels, which in turn may
result in decreased insulin release.66 Becker and colleagues assessed patients for whether
sulfonylurea treatment was affected by NOS1AP gene polymorphisms in 183 patients with
T2DM. There was no difference in dose adjustment needed for tolbutamide, gliclazide, or
glimepiride in patients with the rs10494366 T>G NOS1AP genotype and those without this
polymorphism. However, dose adjustments were needed when patients were prescribed
glibenclamide. The prescribed doses of glibenclamide were higher in patients with the
rs10494366 T/G genotype when compared to patients with the wild-type T/T genotype.
The risk of mortality was also assessed, and the study found that patients with the variant
G allele were at a higher risk of mortality if using glibenclamide when compared to patients
with the wild-type T/T genotype (HR: 2.8; 95% CI: 1.09−7.22). The risk of mortality was lower
in patients prescribed either tolbutamide or glimepiride who were variant G allele carriers
compared to wild-type carriers (HR: 0.30 for tolbutamide, HR: 0.18 for glimepiride).67
Hepatocyte Nuclear Factor Gene Variants and Early Development of Diabetes
Mellitus
The hepatocyte nuclear factor 1 alpha (HNF1α) gene is involved in pancreatic β-cell differ-
entiation and function. Mutations in HNF1α, hepatocyte nuclear factor 1 beta (HNF1β), and
hepatocyte nuclear factor 4 alpha (HNF4α) are associated with the development of matu-
rity-onset diabetes of the young that starts in adolescence or early adulthood.68 Carriers
of this mutation will present with hyperglycemia after 10 years of age and develop T2DM
by the age of 55 years as a result of progressive pancreatic β-cell dysfunction.69,70 The
dysfunction in the pancreatic β-cell is associated with a decrease in insulin secretion due
to the reduced ability of glucose to close the KATP channels.71,72 Sulfonylureas can improve
glycemic control and should be considered first-line therapy in these patients.8,73
Pearson and colleagues compared gliclazide and metformin treatment in patients with
the HNF1α mutation (rs2259816).74 Patients prescribed gliclazide had a 5-fold decrease
in fasting glucose when compared to those prescribed metformin. Patients using glicla-
zide with the HNF1α mutation were also shown to have a 4-fold greater response than
patients without the HNF1α mutation.74 Shepherd and colleagues later confirmed patients
with T2DM with the HNF1α mutation who were switched from insulin therapy to gliclazide
therapy were able to continue on the sulfonylurea therapy alone for 39 months and main-
tain a HbA1c level of 6.9%. Eventually, these patients required insulin therapy due to the
progressive nature of the dysfunction in pancreatic β-cells associated with the disease.75
Mutations in the HNF4α gene causes a slower progressing pancreatic β-cell dysfunction.
These patients share clinical characteristics similar to patients with the HNF1α gene muta-
tion and are sensitive to sulfonylurea treatment.73 Morever, the HNF1β gene is involved in
the development of the pancreas, the renal tract, and the Mullerian tract. Patients with
mutations in the HNF1β gene characteristically have T2DM, progressive renal disease, and
female genital tract abnormalities.76 Carriers of mutated HNF1β gene (rs11868513) have
insulin resistance and do not respond well to sulfonylureas.77
Effect of CYP2C9 and CYP2C19 Gene Variants on Metabolism of Sulfonylureas
Sulfonylureas are metabolized in the liver by the cytochrome P-450 enzyme system
isoenzyme CYP2C9. The most common CYP2C9 allele has been designated as CYP2C9*1
and is considered the wild-type allele.78 The two common nonsynonymous variants are
Arg133Cys (CYP2C9*2, rs1799853) and Ile359Leu (CYP2C9*3, rs1057910).26 These CYP2C9
variants are loss-of-function mutations. Kirchheiner and colleagues studied the effects of
the CYP2C9 variants in healthy volunteers being treated with sulfonylureas.79,80 After the
subjects took the sulfonylureas orally, the CYP2C9 variant carriers were associated with
increased plasma drug concentration and decreased drug clearance when compared to
wild-type carriers.
Zhou and colleagues examined the therapeutic response in carriers of the CYP2C9
variant alleles in 1,073 patients being treated with sulfonylureas.81 The data were collected
from GoDARTS; of the 1,073 patients in the study, 6% were carriers of CYP2C9 (*2/*2, or
*2/*3 or *3/*3) variant alleles. Carriers of the CYP2C9 variants had a 0.5% greater reduction
in HbA1c levels compared to those patients with the wild-type CYP2C9 homozygotes (P =
0.003). Patients with the CYP2C9 variants treated with sulfonylureas were 3.4 times more
likely to achieve the clinical desired HbA1c levels of <7% when compared to patients with
the wild-type CYP2C9 alleles (P = 0.009).81
Patients with the poor metabolizer (PM) phenotypes (*2/*3 and *3/*3) were at greater
risk of a severe hypoglycemic event during sulfonylurea therapy (OR: 5.2).82 The frequency
of the CYP2C9 variant alleles in Caucasians was 11% (CYP2C9*2) and 7% (CYP2C9*3). Only
4% of Asians tend to have the CYP2C9 PM phenotype.30 In the Asian population, the poly-
morphic CYP2C19 gene appeared to play a more important role in the pharmacokinetics of
sulfonylureas. The two common variants are CYP2C19*2 (G>A, rs4244285) and CYP2C9*3
(G>A, rs4986893). The CYP2C19*2 gene variant is observed in 10 to 15% of Caucasians and
40 to 50% of Asians. The CYP2C19*3 gene variant is rare in Caucasians (<1%), whereas the
variant is more common in Asians (8%).83 Zhang and colleagues studied the effects of
gliclazide metabolism when taken orally by healthy Chinese males who were carriers of
the CYP2C19 PM phenotype and the CYP2C9*1 wild-type allele. The plasma concentration
of gliclazide was 4.3-fold higher in CYP2C19 PMs when compared to the wild-type carriers,
and the half-life of the drug was extended from 15.1 to 44.5 hours.84
Meglitinides
Meglitinides inhibit KATP channels on the pancreatic β-cells to stimulate the release of
insulin. The mechanism of action is similar to the sulfonylureas. When compared to sulfo-
nylureas, meglitinides have a shorter duration of action, better control of postprandial
hyperglycemia, and lower risk of severe hypoglycemia. The pancreatic β-cell KATP channel
is composed of four SUR1 subunits encoded by the ABCC8 gene and four Kir6.2 subunits
encoded by the KCNJ11 gene.6 Nateglinide and repaglinide are the two meglitinides avail-
able. Binding of repaglinide involves the SUR1 subunit of the KATP channel and a separate
distinct binding site on the pancreatic β-cell. Nateglinide binds competitively to the SUR
subunits of the KATP channel. Repaglinide has a higher binding affinity for the SUR1 subunit
over the SUR2 subunit when compared to repaglinide.85
Effect of ABCC8 and KCNJ11 Gene Variants on Meglitinide-Generated Insulin
Secretion
Genetic mutations in the ABCC8 or KCNJ11 genes encoding the KATP channel have been
studied in patients who were treated with repaglinide. The therapeutic effects of repa-
glinide in 100 Chinese patients with T2DM who were carriers of genetic mutations in either
the ABCC8 or KCNJ11 genes was investigated.86 For the KCNJ11 gene, the K/K homozygote
is considered the wild type. Baseline HbA1c levels were found to be higher in patients with
the E/K and K/K genotypes when compared to patients who were E/E homozygotes (P =
0.0103). After 24 weeks of treatment with repaglinide, the change in HbA1c levels from the
baseline were lower in patients with the E/K and K/K genotypes when compared to those
with E/E homozygotes (P = 0.0221). There was also a significant difference in the failure
of repaglinide treatment between the three groups. Nearly 43% of E/E homozygotes,
19% of E/K heterozygotes, and 18% of K/K homozygotes failed repaglinide treatment (P
= 0.0364). The other genetic variant of the KATP channel investigated was exon16-3T/C
ABCC8 (rs1799854). For this variant, the wild type is the C/C homozygote. There were no
significant changes in HbA1c or fasting glucose levels when T/T, T/C, and C/C genotypes of
the exon16-3 of the ABCC8 gene were compared. The C/C homozygotes were more insulin
sensitive after repaglinide treatment when compared to the T/T and T/C genotypes.86
KCNQ1 Gene Variants and Efficacy of Repaglinide
The voltage-gated potassium channel is encoded by the KCNQ1 gene. This voltage-gated
potassium channel is located in several tissues, including the pancreas. The efficacy of
repaglinide treatment in Chinese patients with T2DM and the association between and the
KCNQ1 mutations rs2237892 (C>T) and rs2237895 (C>A) was determined.87 The FPG levels
were higher in carriers of the rs2237892 C allele compared to A allele carriers. Patients
with the rs2237895 T/T homozygote had lower levels of glucose following repaglinide
treatment when compared to patients with the rs2237895 C allele.87

Effect of SLCO1B1 Gene Variants on the Pharmacokinetic Properties of
Repaglinide
Organic anion-transporting polypeptides (OATPs) are responsible for the membrane
transport of numerous compounds. The OATP family consists of 11 members, and the one
involved in the pharmacokinetics of repaglinide is OATP1B1. OATP1B1 is located mainly in

the sinusoidal membranes of hepatocytes and mediates the transport of OADs from the
portal venous blood to into the hepatocytes.6 The OATP1B1 is encoded by the solute carrier
organic anion-transporter family, member 1B1 (SLCO1B1) gene. The most common variants
of the SLCO1B1 gene are Asn130Asp (rs2306283) and Val174Ala (rs4149056).
Kalliokoski and colleagues performed several studies in healthy Caucasian volunteers

to determine the effect of SLCO1B1 gene polymorphisms on the pharmacokinetic param-
eters of meglitinides.88-90 One study demonstrated that the SLCO1B1 Val174Ala polymor-
phism affected the pharmacokinetics of repaglinide at currently available clinical doses.
The area under the curve (AUC) plasma concentration was higher in carriers with the C/C
genotype when compared to the T/T genotype (P ≤0.001). The maximum concentration
of repaglinide was higher in subjects with the C/C genotype when compared to the T/T
homozygotes.89 Although there was no significant effect between the SLCO1B1 gene poly-
morphisms and the pharmacokinetics of nateglinide,88 Zhang and colleagues found the
AUC of nateglinide to be higher in carriers with the T/C genotype (P = 0.016) and the C/C
genotype (P = 0.001) when compared to the T/T homozygotes.91 However, this was a small
study done with healthy Chinese volunteers so the difference in the results between the
two studies may be due to the small sample size and different ethnicity of the volunteers.
Effect of TCF7L2 Gene Variants on Repaglinide Efficacy
During embryonic development, the Wnt/β-catenin signaling pathway regulates cell fate
determination and organogenesis. The transcription factor Tcf-4 is part of the Wnt/β-
catenin dependent pathway. TCF7L2 is the gene that encodes for Tcf-4 transcription factor.
The levels of GLP-1 are altered by polymorphisms in the TCF7L2 gene. Production of GLP-1,
glucose homeostasis, pancreatic β-cell function, and lipid metabolism are also regulated
by Tcf-4.56 The TCF7L2 rs290487C>T polymorphism was studied in Chinese patients with
T2DM.92 Carriers of the T/T genotype showed better repaglinide efficiency due to lower
levels of fasting insulin, decreased triglycerides, and decreased low-density lipoprotein
cholesterol when compared to carriers of the C/C or C/T genotype (P <0.05).92
NAMPT Gene Variants’ Decrease of Postprandial Insulin Levels
Nicotinamide phosphoribosyltransferase (NAMPT) is the rate-limiting enzyme involved
in the biosynthesis of nicotinamide adenine dinucleotide (NAD). The enzyme NAMPT
converts nicotinamide into nicotinamide mononucleotide. Defects in the NAMPT enzyme
inhibit NAD production. NAD biosynthesis is important for pancreatic β-cell function and
is involved in glucose homeostasis.93 Sheng and colleagues examined the effects of the
NAMPT −3186C>T polymorphism (rs11977021) in 35 Chinese subjects with the mutation
treated with rapaglinide.94 In carriers with the C/T genotype, the polymorphism was associ-
ated with decreases in the postprandial serum insulin levels when compared to C/C or T/T
homozygotes (P <0.05).94
Carriers of SLC30A8 Gene Variants and Increased Risk for Developing T2DM
Variants of the SLC30A8 gene that encodes the zinc transporter 8 (ZnT8) have been
reported to contribute to an increased risk of developing T2DM. The ZnT8 is mainly

expressed in pancreatic β-cells and is responsible for the uptake of zinc into intracellular
vesicles where zinc plays a role in storage and secretion of insulin. The nonsynonymous
SLC30A8 polymorphism Arg325Trp (rs13266634) is associated with a higher risk of devel-
oping T2DM in Asian and European populations but not in the African population. Individu-
als with a C/C genotype had a 33% increased risk of developing T2DM when compared to
those who were carriers of the T/T and C/T genotypes.95
A study done by Huang and colleagues with Chinese subjects (n = 48) examined the
effects of the SLC30A8 polymorphisms on repaglinide response.96 Subjects who were
carriers of the C/T and T/T genotypes had a better response to repaglinide treatment
compared to those who were C/C genotype carriers. The C/T and T/T genotypes had lower
fasting insulin serum levels (P <0.05) and lower postprandial insulin serum levels (P <0.01)
when compared to C/C homozygotes. The effects of the other common polymorphism of
SLC30A8 (Arg325Gln, rs16889462) on repaglinide treatment was also studied. Patients
with the rs16889462 G/A genotype exhibited lower FPG levels, lower postprandial plasma
glucose levels, and lower HbA1c levels when compared to G/G homozygotes.96
Effects of MDR1 Gene Variants on Repaglinide Pharmacokinetics
The multidrug resistant 1 (MDR1) gene encodes for the P-glycoprotein (P-gp) transporter,
an efflux transporter that has an important role in drug disposition and distribution. The
P-gp is found in high concentrations in the luminal membrane of barrier organs such as
the intestine and blood-brain barrier.97 The function of this transporter is to export xenobi-
otics from cells into extracellular spaces.98 The MDR1 G2677T polymorphism (Ala893Thr,
rs2032582) was studied for its effects on repaglinide pharmacokinetics. In 24 Chinese
subjects being treated with repaglinide, the plasma concentration of the drug was higher
in carriers of the G/T and T/T genotypes than those with the G/G carriers (P = 0.007).99
Effects of CYP2C9 Gene Variants on Nateglinide Metabolism
Nateglinide is metabolized mainly in the liver by the cytochrome P-450 enzyme system
isoenzyme CYP2C9 and to a lesser extent by CYP3A4 and CYP2D6. The most common
CYP2C9 allele has been designated as CYP2C9*1, and it is considered the wild-type
allele.78 The two common nonsynonymous variants are Arg133Cys (CYP2C9*2, rs1799853)
and Ile359Leu (CYP2C9*3, rs1057910); they are considered loss-of-function variants.26
Carriers of the CYP2C9*3 genotype were found to have reduced clearance of nateglinide
when compared to carriers of the wild-type allele (P <0.01). Even with the reduced clear-
ance of nateglinide, no significant difference was discovered in plasma glucose, insulin, or
glucagon levels in those individuals. Subjects taking nateglinide who were carriers of the
CYP2C9*2 allele were found to have similar wild-type pharmacokinetic parameters.100
Effects of CYP2C8 Gene Variants on Repaglinide Metabolism
Repaglinide is mainly metabolized by the CYP2C8 and CYP3A4 isoenzymes. The CYP2C8
is a polymorphic enzyme, and the CYP2C8*3 allele has two linked amino acid substitutions:
R139K and K399R. Caucasians are 14% likely to be a carrier of the CYP2C8*3 variant allele.
The data are inconsistent about the impact of the CYP2C8*3 allele on repaglinide pharma-
cokinetics. Niemi and colleagues showed there was a 45-48% decrease in plasma concen-
tration levels of repaglinide in heterozygous carriers of the CYP2C8*3 polymorphism.101
However, Bidstrup and colleagues as well as Tomalik-Scharte and colleagues found no
statistically significant differences in the repaglinide pharmacokinetics when patients were
treated with repaglinide.102,103
SENSITIZERS
Biguanides
In addition to lifestyle modifications, the biguanide metformin is the recommended initial
therapy, if tolerated, in patients with T2DM.2 The primary physiological effect of metfor-
min is to reduce hepatic glucose output through suppression of gluconeogenesis and

glycogenolysis.104 Other proposed mechanisms include increased tissue uptake of glucose
and utilization, improvement of insulin sensitivity, and decreased gastrointestinal glucose
absorption. The gastrointestinal action leads to adverse effects such as nausea, bloat-
ing, abdominal pain, and diarrhea, which results in intolerance in 5-10% of patients.105 The
glucose-lowering effect is variable, with about 15% of patients not showing a response to
therapy. The effect of therapy on cardiovascular disease (CVD) outcomes is another impor-
tant consideration in choosing a regimen. In the UKPDS study, metformin-treated patients
showed a decrease in myocardial infarction, coronary deaths, and all-cause mortality by
39, 50, and 36%, respectively, in newly diagnosed patients with T2DM with low CVD risk.106
Effect of Drug Transporters on Metformin Response
Metformin is positively charged and hydrophilic at physiological pH.107 It is not metabolized
and undergoes rapid renal elimination. Metformin is dependent on drug transporters for
absorption, distribution, and elimination. The plasma membrane monoamine transport
(PMAT), encoded by SLC29A4 gene, is expressed on the luminal side of enterocytes and
mediates the intestinal absorption of metformin.108 Hepatic uptake of metformin occurs
by organic cation transporter 1 (OCT1), encoded by the SLC22A1 gene, expressed in the
basolateral membrane of hepatocytes.109 Uptake into the proximal tubule cells in the
kidney occurs by OCT2, encoded by SLC22A2 gene, which is expressed at the basolateral
membrane in the kidney tubular cells.110 The multidrug and toxin extrusion transporter 1
(MATE1), encoded by the SLC47A1 gene, and MATE2, encoded by the SLC47A2 gene, which
is expressed in the apical membrane of the renal proximal tubule cells facilitate metformin
excretion from tubular cells into the urine.111,112
Effects of SLC22A1 Gene Variants on Metformin Pharmacokinetics
The actual therapeutic response to metformin is somewhat variable with fewer than
two thirds of patients achieving a HbA1c level of <7%, implying that genetic components
may affect response to therapy.113 The gene encoding OCT1, SLC22A1, has been the most
studied for polymorphisms because it is necessary for the activation of metformin by
transport into the liver. Shu and colleagues were the first to study four different reduced-
function nonsynonymous polymorphisms: R61C (rs12208357), G410S (rs34130495), 420del
(rs72552763), and G465R (rs34059508) of the SLC22A1 gene. Their study examined the
effect of these polymorphisms on metformin activity in two separate trials of healthy
volunteers.114,115 A reduced effect of metformin following OGTTs was reported in those
healthy individuals expressing the OCT1 polymorphisms. Additionally, patients with the
reduced-function alleles had a higher metformin AUC, higher maximal plasma concentra-
tion, and lower oral volume of distribution. A subsequent study by Tzvetkov and colleagues
found an additive increase in renal clearance of metformin with an increase in the number
of reduced-function alleles.116
A prospective study by Christensen and colleagues investigated the effects of poly-

morphisms in the genes encoding OCT1, OCT2, MATE1, MATE2, and PMAT on the trough
steady-state plasma concentration and changes in HbA1c level in 159 patients with T2DM
taking metformin.117 The average metformin steady-state concentration and change in
HbA1c level at 6 and 24 months were inversely correlated with the number of reduced-
function SLC22A1 alleles. The reduced function of OCT1 variants led to less metformin
transported into hepatocytes that, in turn, caused a decreased volume of distribution,
shortened half-life, and lower trough steady-state concentrations. Choi and colleagues
examined the genetic variations of multidrug and toxin extrusion 2 (MATE2-K [SLC47A2]) by
sequencing 272 deoxyribonucleic acid (DNA) samples from four major U.S. ethnic groups on
metformin therapy.118 Two nonsynonymous variants, c.485C>T and c.1177G>A, were associ-
ated with reduced metformin uptake. The most common variant identified, g.130G>A, was
associated with a poorer response to metformin in homozygous patients as assessed by
the relative change in HbA1c level.
The largest population-based study to date, GoDARTS, assessed metformin response
based on HbA1c level reduction and time to monotherapy failure in 1,531 patients with
T2DM who were incident users of metformin.119 These patients were genotyped with R61C
(rs12208357) and 420del (rs72552763) variants. Metformin response was assessed by
measurement of HbA1c level between 6 and 42 months, and through the use of logistic
regression to model the ability of patients to reach a treatment target HbA1c level of <7%
within 18 months after the start of therapy (given their baseline HbA1c level was >7%).
They found that R61C and 420dev variants did not affect the initial HbA1c level reduction
(P = 0.47 and P = 0.92, respectively), the chance of achieving a treatment target (P = 0.083
and P = 0.36), the average HbA1c level on monotherapy up to 42 months (P = 0.44 and P =
0.75), or the occurrence of monotherapy failure (P = 0.85 and P = 0.56). A population-based
study of patients with T2DM who were incident users of metformin found an association
between lower HbA1c reduction and a polymorphism in the rs622342 A>C allele.120 The
decrease in hemoglobin HbA1c level in patients with the AC genotype was 0.29% less than
that in patients with the AA genotype. For patients with the CC genotype, the decrease in
HbA1c level was 0.58% less than that in patients with the AA genotype. The researchers
concluded that patients with the AC or CC genotype had less OCT1 transporter activity and
a reduced capacity to transport metformin into hepatocytes.
Tarasova and colleagues performed the first study examining the association of genet-
ic variation of OCT1, OCT2, and MATE1 with the gastrointestinal adverse effects of metfor-
min therapy.121 A statistically significant association was found between the A allele of the
rs628031 (P = 0.012) as well as 8 bp insertion (rs36056065) in the gene encoding OCT1
(P = 0.02) and the presence of adverse effects. A proposed mechanism of gastrointestinal
intolerance was due to a local increase in drug concentration in the intestinal tissues. The
authors concluded that there is a strong linkage disequilibrium between the A allele and
the 8 bp insertion in the gene encoding OCT1, which may predispose patients on metformin
therapy to gastrointestinal adverse effects.
Effect of SLC22A2 Gene Variants on Metformin Renal Clearance

Studies of the transporter OCT2 polymorphism A270S, encoded by SLC22A2 (rs316019)
on metformin clearance, have been contradictory. Two of the studies found a reduction
in metformin renal clearance in homozygous variant carriers,122,123 while the third study
showed an increase in metformin renal clearance in individuals heterozygous for A270S
variants.124 The study by Christensen and colleagues failed to find an association between
A270S and metformin levels.117
SLC47A1 Gene Variants and Increased Metformin Efficacy
A preliminary study performed by Becker and colleagues examined the effect of single
nucleotide polymorphisms (SNPs) in the SLC47A1 gene encoding MATE1 on the HbA1c level-
lowering effect of metformin.125 The researchers identified all incident metformin users in
the Rotterdam Study (n = 116), a population-based cohort study analyzing the association
between SNPs in the SLC47A1 gene and change in HbA1c level. The researchers found that
the rs2289669 G>A SNP was significantly associated with increased metformin response.
For each minor A allele, there was a decrease in HbA1c level by 0.3% (P = 0.0005) consis-
tent with a reduction in MATE1 transporter activity. No significant association was found
with the other alleles. The Diabetes Prevention Program study was a multicenter, random-
ized genotyping trial enrolling patients at high risk of developing diabetes.126 Patients were
randomized to placebo, metformin, or lifestyle intervention program conducted by exercise
and dietary professionals. The investigators found an association between rs8065082 SNP
in the SLC47A1 gene and lower diabetes incidence in subjects treated with metformin. The
researchers confirmed the findings of Becker and colleagues because the SNP rs8065082
was in high linkage disequilibrium with rs2289669 G>A.
Effect of Ataxia Telangiectasia Mutated Gene on Metformin Efficacy
A genome-wide association study focused on the role of polymorphisms in metformin
transporters in two cohorts of 1,783 Scottish patients from the GoDARTS study and 113
patients from the UKPDS.127 The study identified a common rs11212617 A>C SNP associated
with treatment success. Additionally, the authors proposed the association of the ataxia
telangiectasia mutated (ATM) gene as the causative gene associated with altered glycemic
response to metformin due to the association with insulin resistance, increased T2DM risk,
and adenosine monophosphate-activated protein kinase (AMPK) activation. AMPK activa-
tion leads to suppression of glucose production via gluconeogenesis and increased periph-
eral glucose uptake. A meta-analysis performed by van Leeuwen and colleagues confirmed
the association between ATM and metformin treatment response in patients with T2DM.128
The analysis examined incident users of metformin from the Diabetes Care System West-
Friesland (DCS), the Rotterdam Study from the Netherlands, and the Collaborative Atorvas-
tatin Diabetes Study (CARDS) from the UK for association between the rs11212617 genotype
and metformin treatment success that was defined as achievement of a targeted HbA1c
level ≤7%. In the DCS cohort, an association was observed (P = 0.028); in the Rotterdam
Study cohort, a numerically similar but nonsignificant trend was observed (P = 0.15); and in
the CARDS cohort, there was no significant association. When these three cohorts were
combined with data from the GoDARTS and UKPDS cohorts, the association between the
rs11212617 genotype and success of metformin therapy (P = 0.0000078) was significant.
Thiazolidinediones
Thiazolidinediones (TZDs) act as agonists for the nuclear receptor peroxisome proliferator-
activated receptor-γ (PPARG).129 Activation of this receptor regulates the transcription of

genes involved in glucose and lipid metabolism. There are three known forms of the PPAR
receptor encoded by distinct genes with different tissue expression: PPAR-α (PPARA),
PPAR-γ (PPARG), and PPAR-δ (PPARD).130 TZDs are selective agonists for PPARγ2, which
is primarily expressed in adipose tissue, with minimal activity at PPARγ1 or PPARγ3.131 The
stimulation of PPARγ2 results in increased adipocyte differentiation and fatty acid uptake.
By reducing circulating fatty acid concentrations and lipid availability in the liver and
muscle, an improvement in sensitivity to insulin and reduced hyperglycemia is observed.
The direct mechanism of action of TZDs is unknown; however, data indicate that TZDs
increase insulin sensitivity with direct and indirect effects on adipose tissue and muscle.
Two TZDs are currently available on the market: pioglitazone and rosiglitazone. Both
agents have been shown to improve glycemic control and possibly slow pancreatic β-cell
failure. Rosiglitazone was placed under a restricted access program in 2011 due to increased
cardiovascular risks as compared to pioglitazone.132,133 In 2014, the restriction in the United
States was lifted based on a U.S. Food and Drug Administration (FDA) review of data from
the Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes
(RECORD) trial.134 The FDA concluded that results from the trial showed the major adverse
cardiovascular events in patients being treated with rosiglitazone were no different when
compared to standard-of-care diabetes drugs. The PROspective trial was designed to
investigate the effect of pioglitazone on macrovascular outcomes in patients with T2DM.135
The study concluded that pioglitazone reduced all-cause mortality, nonfatal myocardial
infarction, and stroke in patients who were at high risk of macrovascular events.
AQP2 Gene Variant and Thiazolidinedione-Associated Edema
TZD therapy has been associated with the development of peripheral edema and conges-
tive heart failure as a result of fluid retention.132 A study by Chang and colleagues suggest-
ed that key risk factors for TZD-associated edema included rs296766 polymorphism of
the AQP2 gene coding aquaporin-2 (vasopressin-regulated water channel) and rs12904216
polymorphism of SLC12A1 (solute carrier protein family 12 A, member one) gene coding the
sodium-potassium-2 chloride.136 Additionally, female gender and older age were contribut-
ing risk factors to the development of edema. Another safety concern with the use of TZDs
is the increased risk of bone fractures, which was seen in patients enrolled in the outcome
progressions trial/ADOPT (A Diabetes Outcome Progression Trial) study.137
PPARG Gene Variants and Reduced Risk for Development of T2DM
Several gene variants have been associated with the clinical efficacy of TZD therapy. PPARγ,
tumor necrosis factor (TNF)-α, and adipokinins (e.g., adiponectin [ADIPOQ], leptin, resistin)
are associated with TZD therapy response, specifically variants in PPARG or ADIPOQ. The
Pro12Ala (rs1801282) polymorphism in the PPARG gene was identified in 1997 and has been
studied extensively in relation to the incidence of T2DM.138 A meta-analysis by Gouda and
colleagues concluded the risk of developing T2DM is 14% lower for each minor allele of
this polymorphism.139 The TRIPOD study was the first to provide evidence of an association
between TZD monotherapy in an at-risk population and reduction in the risk of T2DM.140 A
significant reduction in risk was observed for future T2DM in Hispanic women with a previ-
ous history of gestational diabetes mellitus. The investigators found the same therapeutic
effects in the PIPOD trial with pioglitazone.141 The Pro12Ala (rs1801282) polymorphism in
the PPARG gene has been the most studied in relationship to the efficacy of TZDs.
In a study by Kang and colleagues, rosiglitazone was found to be significantly (P =

0.002) more effective in treating T2DM in patients with Pro12Ala (rs1801282) polymorphism,
as evidenced by a significantly greater decrease in FPG and HbA1c levels as compared to
carriers of the wild-type genotype.142 This finding was confirmed in a study by Hsieh and
colleagues in 250 patients with diabetes taking pioglitazone.143 Rosiglitazone efficacy was
also shown to be associated with Thr394Thr (rs2970847) and Gly482Ser (rs8192673) SNPs
of the peroxisome proliferator-activated receptor-γ coactivator-1α, a transcriptional coacti-
vator of PPARG, in Chinese patients with T2DM.144
Gene Variants Affecting Adipokinins
Adipokinins (e.g., adiponectin, leptin, resistin), PPARγ, and TNF-α have been shown to be
increased in patients receiving TZD treatment and enhance the insulin-sensitizing mecha-
nism of TZDs. Variants of adipokinins have been studied for the association with TZD
efficacy. A pilot study by Makino and colleagues suggested that the G/G genotype of
resistin SNP-420 (rs1862513) may be an independent predictor of the reduction in FPG (P
= 0.020) and homeostasis model assessment of insulin resistance (HOMA-IR) (P = 0.012)
in patients receiving pioglitazone.145 In another study, Sun and colleagues investigated the
impact of adiponectin allele T45G and C-11377G polymorphism on the efficacy of rosigli-
tazone in Chinese patients with T2DM.146 The heterozygote genotype resulted in an attenu-
ated rosiglitazone effect on FPG, postprandial glucose, and HOMA-IR as compared to the
homozygote genotype. The association between leptin G-2548A (rs7799039) and TNF-α
G-308A (rs1800629) polymorphisms and rosiglitazone efficacy was investigated by Liu and
colleagues in 245 Chinese patients with T2DM.147 The investigators found an enhanced
rosiglitazone effect on fasting and postprandial serum insulin in patients with AA genotype
of leptin G-2548A as compared to GG+GA genotype (P <0.05). A reduced effect on fasting
serum insulin was observed in patients with GA+AA genotype of TNF-alpha as compared
with the GG genotype (P <0.05).
Effect of CYP2C8 Gene Variants on Thiazolidinediones’ Metabolism
Variation in genes encoding cytochrome P450 enzymes have also been studied for their
effect on TZD efficacy. Pioglitazone is primarily metabolized by CYP2C8 and CYP3A4,
whereas rosiglitazone is metabolized by CYP2C8 and CYP2C9.148,149 Kirchheiner et al. stud-
ied the impact of CYP2C8*3 allele coding for the Arg139Lys (rs11572080) and Lys399Arg
(rs10509681) amino acid substitutions on the pharmacokinetics of rosiglitazone.148 The
clearance rate of rosiglitazone was greater in CYP2C8*1/*3 and *3/*3 patients as compared
to *1/*1. There was no association between CYP2C8 genotype and insulin-sensitizing
effects. Tornio et al. concluded that plasma pioglitazone concentrations were higher in
CYP2C8*1/*3 and *3/*3 as compared to *1/*1 patients.150
ALPHA-GLUCOSIDASE INHIBITORS
Currently, two drugs within this class are on the market today: miglitol and acarbose.
Alpha-glucosidase inhibitors slow the digestion of complex starches by inhibiting the
enzyme alpha-glucosidase in the brush border of the small intestine. Slowing the absorp-
tion results in reduced postprandial blood glucose. The STOP-NIDDM trial investigated
the association of SNPs of PPARα with the conversion from impaired glucose tolerance
to T2DM; the subjects were randomized to either treatment with acarbose or placebo.151
PPARα agonists lower plasma lipid levels, decrease intrahepatic and skeletal muscle lipid
accumulation and adiposity, and normalize insulin and glucose concentrations, which leads
to a reduction in insulin resistance and the risk of T2DM. The investigators reported that
acarbose was more effective in preventing progression to diabetes in those patients with
impaired glucose tolerance carrying risk alleles of PPARA (rs1800206, rs4253778), PPARG
(rs1801282, rs8192678), PPARG coactivator 1α (PGC-1A), and/or HNF4α (HNF-4A, rs4253776,
rs4810424). The investigators also concluded there was an additive effect on the risk of
diabetes with multiple SNPs.
INCRETIN MIMETICS
Glucagon-Like Peptide-1 Analogs and Dipeptidyl Peptidase-4 Inhibitors

The incretin effect is an increased insulin release when glucose is orally ingested versus
when glucose is administered intravenously. The incretin effect is attributed to gastroin-
testinal proteins and incretin hormones that are released upon ingestion of food.152 GLP-1
and gastric inhibitory peptide (GIP) are the two incretin hormones in humans that mediate
incretin effects. GIP is also known as the glucose-dependent insulinotropic polypeptide.
Both incretin hormones stimulate glucose-dependent insulin secretion though receptors
located on the pancreatic β-cells and suppress postprandial glucagon release, slow gastric
emptying, and reduce appetite.152 GLP-1 and GIP are rapidly degraded by the enzyme dipep-
tidyl peptidase-4 (DPP-4).
Two classes of incretin mimetics therapies are in clinical use: (1) GLP-1 analogues that
act as agonists at GLP-1 receptors and (2) DPP-4 inhibitors that prolong the duration of
action of endogenous GLP-1 and GIP by inhibiting the enzymatic breakdown of these incre-
tin hormones. GLP-1 receptor agonists are administered subcutaneously, whereas DPP-4
inhibitors can be given orally because they are not peptides. Exenatide and liraglutide are
the two GLP-1 receptor agonists that are currently available. The inhibitors of the DPP-4
enzyme—sitagliptin, saxagliptin, and linagliptin—are also available.
CTRB1 and CTRB2 Gene Variants and Increased Risk for T1DM
The SNP rs7202877 (T>G) is located near CTRB1 and CTRB2 and has been associated with
an increased risk for developing type 1 diabetes mellitus (T1DM). The CTRB1 and CTRB2
genes encode the digestive enzyme chymotrypsin, which is involved in regulation of the
incretin pathway. The risk allele for rs7202877 is the G allele and has been found to be
protective for T2DM susceptibility. Carriers of the G allele have a 33% increase in GLP-1
levels following insulin stimulation (P = 0.000019).153 Data from patients involved in the
GoDARTS and Dutch DCS West-Friesland studies were used to determine the impact of
the rs7202877 SNP near the CTRB1 and CTRB2 genes in 527 patients with T2DM treated
with either a GLP-1 receptor agonist or a DPP-4 inhibitor. In both studies, the 354 patients
treated with a DPP-4 inhibitor who were T/G or G/G allele carriers had significantly smaller
decreases in HbA1c levels when compared to T/T homozygotes (P = 0.0015). Of the 173
patients treated with a GLP-1 receptor agonist, there was no significant change in the HbA1c
levels in G allele carriers compared to the T/T homozygotes (P = 0.58).153
KCNQ1 Gene Variants and Decreased Insulin Secretion
The pore-forming alpha subunit of voltage-gated potassium channel is encoded by the
KCNQ1 gene. This voltage-gated potassium channel is located in several tissues, including the
pancreas. The effect of GLP-1 and arginine administration in response to dispensed glucose
was studied in 1,578 German subjects who were at high risk for developing T2DM.154 Subjects
were dispensed an OGTT and an intravenous glucose tolerance test, and the effects of
administered GLP-1 and arginine were studied. Subjects were administered GLP-1 first; then
60 minutes later, they were given an arginine bolus to determine the subject’s maximum
insulin secretion level. Insulin secretion was lower in major homozygous carriers with the
KCNQ1 gene mutations rs151290 (A>C), rs2237892 (C>T), and rs2237897 (C>T). Carriers of the
rs151290 (A>C) SNP had a significant increase in GIP levels following the OGTT (P = 0.019).154
TCF7L2 Gene Variants and Increased Risk for Developing T2DM
In hyperglycemic conditions, the Wnt/β-catenin signaling pathway is activated. The tran-
scription factor, Tcf-4, is a component of the Wnt/β-catenin signaling pathway. Both the
gene TCF7L2 that encodes Tcf-4 and the transcription factor regulate levels of GLP-1
production. Tcf-4 is also involved in glucose homeostasis and proliferation and func-
tion of pancreatic β-cells.56 The two gene polymorphisms of TCF7L2 that are in linkage
disequilibrium are rs7903146 (C>T) and rs12255372 (G>T). Both polymorphisms are consis-
tently strong predictors of an increased risk of developing T2DM.57-60 In a study involving
1,110 healthy German volunteers, a significant reduction in GLP-1 secretion was found in
response to insulin release in carriers of both the rs7903146 (C>T) and rs12255372 (G>T)
polymorphic TCF7L2 genes.155 Carriers of the rs7903146 TCF7L2 T/T or T/C variants were
found to be at a higher risk for developing T2DM. These individuals were given an OGTT
and found to have a 50% lower β-cell response (P = 0.01), while their incretin effect was
reduced by 30% (P = 0.02) when compared to subjects with the wild-type genotype.156
WFS1 Gene Variants and Risk for T2DM Development
Wolfram syndrome is an autosomal recessive neurodegenerative disorder that results
from mutations in the WFS1 gene. The WFS1 gene encodes for a polypeptide expressed
in pancreatic islet cells, neurons, and the endoplasmic reticulum. Patients with Wolfram
syndrome present with diabetes insipidus, young-onset nonimmune insulin-dependent
diabetes mellitus, and deafness.157 Studies done in mice where the WFS1 gene was
nonfunctional reported a loss of pancreatic β-cells and impaired insulin secretion.158 Carri-
ers of the mutated WFS1 are predisposed to developing T2DM. Studies have indicated that
carriers of WFS1 develop T2DM due to the decrease in insulin secretion.159,160 Non-diabetic
carriers of the WFS1 gene variant rs10010131 (A>G) were found to have lower insulin secre-
tion during an OGTT when compared to the wild-type genotype (P = 0.03).157
SODIUM-GLUCOSE CO-TRANSPORTER TYPE 2 INHIBITORS

Inhibitors of sodium-glucose co-transporter type 2 (SGLT2) are a new class of glucose-
lowering agents that block the reabsorption of filtered glucose in the renal tubules, leading
to increased urinary glucose excretion.161 Treatment with SGLT2 inhibitors result in reduced
fasting blood glucose and HbA1c. Currently, three agents are on the market: canagliflozin,
dapagliflozin, and empagliflozin. Up to 50 different mutations have been reported of the
SGLT2 gene (SLC5) associated with familial renal glucosuria.162,163 Enigk and colleagues
investigated the effects of a common genetic variation in SGLT2 on human T2DM in 2,590
nondiabetic subjects in Berlin and a cohort of 1,013 Sorbs.164 The SNP SLC5A2 (rs9934336)
in the SLC5 gene was hypothetically associated with an increased 30-minute glucose
concentration, 120-minute insulin concentration, and 120-minute AUC during an OGTT in
907 of the nondiabetic Sorbs (P <0.05). In 2,590 nondiabetic patients from Berlin, the SNP
SLC5A2 was hypothetically associated with increased 120-minute insulin concentrations

(adjusted P <0.05). This suggested a possible role of the SGLT2 gene variant SLC5A2 in
the regulation of glucose homeostasis. Study data indicate that G allele carriers result
in higher glucose levels when treated with SGLT2 inhibitors, which may lead to increased
treatment failures.
SUMMARY
Although the study of T2DM pharmacogenomics is still in its infancy, initial studies indicate
that pharmacogenomics data have the potential to improve the management of diabetes
mellitus. Identification of polymorphic variants in genes affecting both the pharmacokinetic
and pharmacodynamic characteristics of oral antidiabetic medications have led to a greater

understanding of a patient’s risk for developing diabetes mellitus and the potential vari-
ability in drug response. A summary of the current knowledge presented in this chapter has
been summarized in Table 14-1. The gene variants involved in certain types of monogenic
Table 14-1
Summary of the Gene Polymorphisms Involved in the
Pharmacogenetics of Oral Anti-Diabetes Medications
Polymorphisms
Drug Class Gene Studied Effect
Sulfonylurea HNF-1α rs2259816 Sensitivity to hypoglycemic effect73

HNF-1β rs11868513 Does not respond to sulfonylurea
treatment76
KCNJ11 rs5219 K allele decreased insulin secretion
and risk of hypoglycemia; secondary
failure sulfonylurea treatment28,29
ABCC8 rs757110, Decreased insulin secretion; S/S alleles
rs1799854 smaller reduction in HbA1c12,21
CYP2C9*3 rs1057910 Increased risk of hypoglycemia81
CYP2C9*2 rs1799853, Increased risk of hypoglycemia81
TCF7L2 rs12255372, Smaller reduction in HbA1c and failed
rs7903146 to reach target HbA1c goal61
IRS-1 rs1801278 A allele secondary failure of
sulfonylurea treatment64
NOS1AP rs10494366 TG allele reduced clearance66
Meglitinide CYP2C9*3 rs1057910 Reduced nateglinide clearance100
TCF7L2 rs290487 TT alleles better repaglinide efficacy91
SLCO1B1 rs4149056 Increased AUC and Cmax of nateglinide
in TC and CC genotypes: increased
AUC of repaglinide with CC
genotype90
SLC30A8 rs13266634 Increased response on fasting insulin
rs16889462 and postprandial insulin levels with
CT+TT genotypes (rs13266634),
enhanced repaglinide efficacy on
fasting plasma glucose, postprandial
glucose levels and HbA1c with
rs16889462 GA genotype94,95
MDR1 rs2032582 Increased AUC of repaglinide with GT
and TT alleles98
KCNJ11 rs5219 Lower HbA1c levels with EK or KK
genotype85
KCNQ1 rs2237892 Higher levels of fasting plasma
rs2237895 glucose, postprandial glucose and

HbA1c with GA or AA genotype86
NAMPT rs11977021 Lower postprandial insulin levels in CT
genotype93

Summary of the Gene Polymorphisms Involved in the
Pharmacogenetics of Oral Anti-Diabetes Medications
Polymorphisms
Drug Class Gene Studied Effect
Biguanide SLC22A1 rs34130495, Reduced effect following OGTT, higher

rs34059508, AUC, and maximal concentrations
rs7552763, with reduced volume of distri-
rs622342A>C bution, reduced half-life, lower
trough steady-state concentra-
tions113-116,118,119
SLC22A2 rs145450955, Increased Cmax and AUC, reduced renal
rs316019 clearance116, 121-123
SLC47A1 rs2289669G>A, Greater reduction in HbA1c levels and
rs8065082C>T lower diabetes incidence124,125
SLC47A2 rs12943590G>A Lower HbA1c reduction117
ATM rs11212617A>C Associated with treatment success
(achieving an HbA1c <7%)125,127
TZDs PPARG rs1801282 Decrease in FPG and HbA1c137
rs2970847 G>A No effect141
rs8192673 G>A Lower response rate (rosiglitazone); no
effect (pioglitazone)142
PGC-1α rs2970847, Mutations may affect rosiglitazone
rs8192678 treatment143
CYP2C8*3 rs10509681 Decreased AUC and increased clear-
ance148
ADIPOQ rs2241766, Decrease in FPG; homozygotes
rs266729 decrease in HbA1c145
α-Glucosidase PPARA rs1800206, Increased risk for development of
inhibitor rs4253778 DM150
HNF-4A rs4253776 Female carriers of the C allele at
rs4810424 increased risk of developing DM150
PPARG2 rs1801282 Increased risk for development of
DM150
PPARGC1A rs8192678 Increased risk for development of
DM150
DPP-4 CTRB1/2 rs7202877T>G Smaller decrease in HbA1c in minor G
inhibitor allele with DPP-4 inhibitor and no
GLP-1 receptor significant change in HbA1c with
agonist GLP-1 receptor agonists152
SGLT2 SLC5 rs9934336 Increased 30-minute glucose concen-
inhibitor tration, 120-minute insulin concen-
tration, and 120-minute AUC during
OGTT163
AUC, area under the curve; Cmax, maximum concentration; DM, diabetes mellitus; DPP-4, dipeptidyl peptidase-4; FPG,
fasting plasma glucose; GLP-1, glucagon-like peptide-1; HbA1c, glycated hemoglobin; OGTT, oral glucose tolerance
test; PPG, postprandial plasma glucose; SGLT2, sodium-glucose co-transporter type 2; TZDs, thiazolidinediones.
diabetes mellitus, such as NDM, have notably changed the clinical treatment plan for this
subset of patients. Pharmacogenomic data have also increased the understanding of the
effects of SLC22A1 variants and the variable response in metformin therapy. But the clinical
impact of the other genetic variants has, thus far, been limited in patients with polygenic
T2DM. Robust associations between the genetic mutations and drug response need to be
identified and studied to determine if changes in dosing regimens can improve diabetes
mellitus patient outcomes and will allow for the pharmacogenomic data to be translated
into clinical practice.
REFERENCES
1. National Diabetes Statistics Report, 2014. Available at: http://www.cdc.gov/diabetes/pubs/stats
report14/national-diabetes-report-web.pdf. Accessed September 1, 2015.
2. American Diabetes Association. Standards of medical care in diabetes—2015 abridged for primary care
providers. Clin Diabetes. 2015;33(2):97-111.
3. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes: a
patient-centered approach: position statement of the American Diabetes Association (ADA) and the
European Association for the Study of Diabetes (EASD). Diabetes Care. 2012;35(6):1364-1379.
4. Becker ML, Pearson ER, Tkac I. Pharmacogenetics of oral antidiabetic drugs. Int J Endocrinol.
2013;2013:686315.
5. Pearson ER. Pharmacogenetics and future strategies in treating hyperglycaemia in diabetes. Front
Biosci. 2009;14:4348-4362.
6. Mannino GC, Sesti G. Individualized therapy for type 2 diabetes: clinical implications of pharmaco-
genetic data. Mol Diagn Ther. 2012;16(5):285-302.
7. Aquilante CL. Sulfonylurea pharmacogenomics in Type 2 diabetes: the influence of drug target and
diabetes risk polymorphisms. Expert Rev Cardiovasc Ther. 2010;8(3):359-372.
8. Glamoclija U, Jevric-Causevic A. Genetic polymorphisms in diabetes: influence on therapy with oral
antidiabetics. Acta Pharm. 2010;60(4):387-406.
9. Semiz S, Dujic T, Causevic A. Pharmacogenetics and personalized treatment of type diabetes. Biochem
Med. 2013;23(2):154-171.
10. Winkler M, Stephan D, Bieger S, et al. Testing the bipartite model of the sulfonylurea receptor binding
site: binding of A-, B-, and A+B-site ligands. J Pharmacol Exp Ther. 2007;322(2):701-708.
11. Lang V, Light PE. The molecular mechanisms and pharmacotherapy of ATP-sensitive potassium
channel gene mutations underlying neonatal diabetes. Pharmgenomics Pers Med. 2010;3:145-161.
12. Hansen T, Echwald SM, Hansen L, et al. Decreased tolbutamide-stimulated insulin secretion in
healthy subjects with sequence variants in the high-affinity sulfonylurea receptor gene. Diabetes.
1998;47(4):598-605.
13. Inoue H, Ferrer J, Welling CM, et al. Sequence variants in the sulfonylurea receptor (SUR) gene are
associated with NIDDM in Caucasians. Diabetes. 1996;45(6):825-831.
14. Hani EH, Clement K, Velho G, et al. Genetic studies of the sulfonylurea receptor gene locus in NIDDM
and in morbid obesity among French Caucasians. Diabetes. 1997;46(4):688-694.
15. Hart LM, de Knijff P, Dekker JM, et al. Variants in the sulphonylurea receptor gene: association of the
exon 16-3t variant with Type II diabetes mellitus in Dutch Caucasians. Diabetologia. 1999;42(5):617-620.
16. Venkatesan R, Bodhini D, Narayani N, et al. Association study of the ABCC8 gene variants with type 2
diabetes in south Indians. Indian J Hum Genet. 2014;20(1):37-42.
17. Gupta V, Vinay DG, Rafiq S, et al. Association analysis of 31 common polymorphisms with type 2
diabetes and its related traits in Indian sib pairs. Diabetologia. 2012;55(2):349-357.
18. Chavali S, Mahajan A, Tabassum R, et al. Association of variants in genes involved in pancreatic
beta-cell development and function with type 2 diabetes in North Indians. J Hum Gene. 2011;
56(10):695-700.
19. Meirhaeghe A, Helbecque N, Cottel D, et al. Impact of sulfonylurea receptor 1 genetic variability on
non-insulin-dependent diabetes mellitus prevalence and treatment: a population study. Am J Med
Genet. 2001;101(1):4-8.
20. Zychma MJ, Gumprecht J, Strojek K, et al. Sulfonylurea receptor gene 16-3 polymorphism-association
with sulfonylurea or insulin treatment in type 2 diabetic subjects. Med Sci Monit. 2002;8(7):CR512-515.
21. Zhang H, Liu X, Kuang H, et al. Association of sulfonylurea receptor 1 genotype with therapeutic
response to gliclazide in type 2 diabetes. Diabetes Res Clin Pract. 2007;77(1):58-61.
22. Feng Y, Mao G, Ren X, et al. Ser1369Ala variant in sulfonylurea receptor gene ABCC8 is associated
with antidiabetic efficacy of gliclazide in Chinese type 2 diabetic patients. Diabetes Care. 2008;
31(10):1939-1944.
23. Sato R, Watanabe H, Genma R, et al. ABCC8 polymorphism (Ser1369Ala): influence on severe hypo-
glycemia due to sulfonylureas. Pharmacogenomics. 2010;11(12):1743-1750.
24. Holstein JD, Kovacs P, Patzer O, et al. The Ser1369Ala variant of ABCC8 and the risk for severe
sulfonylurea-induced hypoglycemia in German patients with Type 2 diabetes. Pharmacogenomics.
2012;13(1):5-7; author reply 9-10.
25. van Leeuwen N, Swen JJ, Guchelaar HJ, et al. The role of pharmacogenetics in drug disposition and
response of oral glucose-lowering drugs. Clin Pharmacokinet. 2013;52(10):833-854.
26. Distefano JK, Watanabe RM. Pharmacogenetics of anti-diabetes drugs. Pharmaceuticals.
2010;3(8):2610-2646.
27. Gloyn AL, Hashim Y, Ashcroft SJ, et al. Association studies of variants in promoter and coding regions
of beta-cell ATP-sensitive K-channel genes SUR1 and Kir6.2 with Type 2 diabetes mellitus (UKPDS
53). Diabet Med. 2001;18(3):206-212.
28. Sesti G, Laratta E, Cardellini M, et al. The E23K variant of KCNJ11 encoding the pancreatic beta-cell
adenosine 5’-triphosphate-sensitive potassium channel subunit Kir6.2 is associated with an increased
risk of secondary failure to sulfonylurea in patients with type 2 diabetes. J Clin Endocrinol Metab.
2006;91(6):2334-2339.
29. Holstein A, Hahn M, Stumvoll M, et al. The E23K variant of KCNJ11 and the risk for severe sulfonylurea-
induced hypoglycemia in patients with type 2 diabetes. Horm Metab Res. 2009;41(5):387-390.
30. Topić E. The role of pharmacogenetics in the treatment of diabetes mellitus. J Biochem.
2014;33(1):58-70.
31. Huang C, Florez JC. Pharmacogenetics in type 2 diabetes: potential implications for clinical practice.
Genome Med. 2011;3(11):76.
32. Florez JC, Burtt N, de Bakker PI, et al. Haplotype structure and genotype-phenotype correlations
of the sulfonylurea receptor and the islet ATP-sensitive potassium channel gene region. Diabetes.
2004;53(5):1360-1368.
33. Lang VY, Fatehi M, Light PE. Pharmacogenomic analysis of ATP-sensitive potassium channels coex-
pressing the common type 2 diabetes risk variants E23K and S1369A. Pharmacogenet Genomics.
2012;22(3):206-214.
34. Kanakatti Shankar R, Pihoker C, Dolan LM, et al. Permanent neonatal diabetes mellitus: prevalence
and genetic diagnosis in the SEARCH for Diabetes in Youth Study. Pediatr Diabetes. 2013;14(3):174-180.
35. Polak M, Shield J. Neonatal and very-early-onset diabetes mellitus. Semin Neonatol. 2004;9(1):59-65.
36. Stanik J, Gasperikova D, Paskova M, et al. Prevalence of permanent neonatal diabetes in Slovakia and
successful replacement of insulin with sulfonylurea therapy in KCNJ11 and ABCC8 mutation carriers.
J Clin Endocrinol Metab. 2007;92(4):1276-1282.
37. Wiedemann B, Schober E, Waldhoer T, et al. Incidence of neonatal diabetes in Austria—calculation
based on the Austrian Diabetes Register. Pediatr Diabetes. 2010;11(1):18-23.
38. Slingerland AS, Shields BM, Flanagan SE, et al. Referral rates for diagnostic testing support an inci-
dence of permanent neonatal diabetes in three European countries of at least 1 in 260,000 live births.
Diabetologia. 2009;52(8):1683-1685.
39. Hattersley AT, Pearson ER. Minireview: pharmacogenetics and beyond: the interaction of therapeutic
response, beta-cell physiology, and genetics in diabetes. Endocrinology. 2006;147(6):2657-2663.
40. Stoy J, Edghill EL, Flanagan SE, et al. Insulin gene mutations as a cause of permanent neonatal diabe-
tes. Proc Natl Acad Sci USA. 2007;104(38):15040-15044.

41. Edghill EL, Flanagan SE, Patch AM, et al. Insulin mutation screening in 1,044 patients with diabetes:
mutations in the INS gene are a common cause of neonatal diabetes but a rare cause of diabetes
diagnosed in childhood or adulthood. Diabetes. 2008;57(4):1034-1042.
42. Colombo C, Porzio O, Liu M, et al. Seven mutations in the human insulin gene linked to permanent
neonatal/infancy-onset diabetes mellitus. J Clin Invest. 2008;118(6):2148-2156.
43. Njolstad PR, Sagen JV, Bjorkhaug L, et al. Permanent neonatal diabetes caused by glucokinase defi-
ciency: inborn error of the glucose-insulin signaling pathway. Diabetes. 2003;52(11):2854-2860.
44. Njolstad PR, Sovik O, Cuesta-Munoz A, et al. Neonatal diabetes mellitus due to complete glucokinase
deficiency. New Engl J Med. 2001;344(21):1588-1592.
45. Edghill EL, Gloyn AL, Goriely A, et al. Origin of de novo KCNJ11 mutations and risk of neonatal diabetes
for subsequent siblings. J Clin Endocrinol Metab. 2007;92(5):1773-1777.
46. Gloyn AL, Cummings EA, Edghill EL, et al. Permanent neonatal diabetes due to paternal germline
mosaicism for an activating mutation of the KCNJ11 gene encoding the Kir6.2 subunit of the beta-cell
potassium adenosine triphosphate channel. J Clin Endocrinol Metab. Aug 2004;89(8):3932-3935.
47. Vaxillaire M, Populaire C, Busiah K, et al. Kir6.2 mutations are a common cause of permanent neonatal
diabetes in a large cohort of French patients. Diabetes. 2004;53(10):2719-2722.
48. Vaxillaire M, Dechaume A, Busiah K, et al. New ABCC8 mutations in relapsing neonatal diabetes and
clinical features. Diabetes. 2007;56(6):1737-1741.
49. Babenko AP, Polak M, Cave H, et al. Activating mutations in the ABCC8 gene in neonatal diabetes
mellitus. New Engl J Med. 2006;355(5):456-466.
50. Sagen JV, Raeder H, Hathout E, et al. Permanent neonatal diabetes due to mutations in KCNJ11
encoding Kir6.2: patient characteristics and initial response to sulfonylurea therapy. Diabetes.
2004;53(10):2713-2718.
51. Pearson ER, Flechtner I, Njolstad PR, et al. Switching from insulin to oral sulfonylureas in patients with
diabetes due to Kir6.2 mutations. New Engl J Med. 2006;355(5):467-477.
52. Rafiq M, Flanagan SE, Patch AM, et al. Effective treatment with oral sulfonylureas in patients with
diabetes due to sulfonylurea receptor 1 (SUR1) mutations. Diabetes Care. 2008;31(2):204-209.
53. Ashcroft FM. New uses for old drugs: neonatal diabetes and sulphonylureas. Cell Metab. 2010;
11(3):179-181.
54. Stoy J, Greeley SA, Paz VP, et al. Diagnosis and treatment of neonatal diabetes: a United States
experience. Pediatr Diabetes. 2008;9(5):450-459.
55. Zung A, Glaser B, Nimri R, et al. Glibenclamide treatment in permanent neonatal diabetes mellitus
due to an activating mutation in Kir6.2. J Clin Endocrinol Metab. 2004;89(11):5504-5507.
56. Jin T. The WNT signalling pathway and diabetes mellitus. Diabetologia. 2008;51(10):1771-1780.
57. Florez JC, Jablonski KA, Bayley N, et al. TCF7L2 polymorphisms and progression to diabetes in the
Diabetes Prevention Program. New Engl J Med. 2006;355(3):241-250.
58. Grant SF, Thorleifsson G, Reynisdottir I, et al. Variant of transcription factor 7-like 2 (TCF7L2) gene
confers risk of type 2 diabetes. Nat Genet. 2006;38(3):320-323.
59. Cauchi S, El Achhab Y, Choquet H, et al. TCF7L2 is reproducibly associated with type 2 diabetes in
various ethnic groups: a global meta-analysis. J Mol Med. 2007;85(7):777-782.
60. Tong Y, Lin Y, Zhang Y, et al. Association between TCF7L2 gene polymorphisms and susceptibility to
type 2 diabetes mellitus: a large Human Genome Epidemiology (HuGE) review and meta-analysis.
BMC Med Genet. 2009;10:15.
61. Pearson ER, Donnelly LA, Kimber C, et al. Variation in TCF7L2 influences therapeutic response to
sulfonylureas: a GoDARTs study. Diabetes. 2007;56(8):2178-2182.
62. Schroner Z, Javorsky M, Tkacova R, et al. Effect of sulphonylurea treatment on glycaemic control is
related to TCF7L2 genotype in patients with type 2 diabetes. Diabetes Obes Metab. 2011;13(1):89-91.
63. Manolopoulos VG, Ragia G, Tavridou A. Pharmacogenomics of oral antidiabetic medications: current
data and pharmacoepigenomic perspective. Pharmacogenomics. 2011;12(8):1161-1191.
64. Sesti G, Marini MA, Cardellini M, et al. The Arg972 variant in insulin receptor substrate-1 is associated
with an increased risk of secondary failure to sulfonylurea in patients with type 2 diabetes. Diabetes
Care. 2004;27(6):1394-1398.
65. Seeringer A, Parmar S, Fischer A, et al. Genetic variants of the insulin receptor substrate-1 are
influencing the therapeutic efficacy of oral antidiabetics. Diabetes Obes Metab. 2010;12(12):1106-1112.
66. Lajoix AD, Reggio H, Chardes T, et al. A neuronal isoform of nitric oxide synthase expressed in
pancreatic beta-cells controls insulin secretion. Diabetes. 2001;50(6):1311-1323.
67. Becker ML, Aarnoudse AJ, Newton-Cheh C, et al. Common variation in the NOS1AP gene is associated
with reduced glucose-lowering effect and with increased mortality in users of sulfonylurea. Pharma-
cogenet Genomics. 2008;18(7):591-597.
68. Shields BM, Hicks S, Shepherd MH, et al. Maturity-onset diabetes of the young (MODY): how many
cases are we missing? Diabetologia. 2010;53(12):2504-2508.
69. Shepherd M, Ellis I, Ahmad AM, et al. Predictive genetic testing in maturity-onset diabetes of the
young (MODY). Diabet Med. 2001;18(5):417-421.
70. Frayling TM, Evans JC, Bulman MP, et al. Beta-cell genes and diabetes: molecular and clinical charac-
terization of mutations in transcription factors. Diabetes. 2001;50(Suppl 1):S94-100.
71. Stride A, Ellard S, Clark P, et al. Beta-cell dysfunction, insulin sensitivity, and glycosuria precede
diabetes in hepatocyte nuclear factor-1 alpha mutation carriers. Diabetes Care. 2005;28(7):1751-1756.
72. Dukes ID, Sreenan S, Roe MW, et al. Defective pancreatic beta-cell glycolytic signaling in hepatocyte
nuclear factor-1 alpha-deficient mice. J Biol Chem. 1998;273(38):24457-24464.
73. Juszczak A, Owen K. Identifying subtypes of monogenic diabetes. Diabetes Management. 2014;4(1):49-61.
74. Pearson ER, Starkey BJ, Powell RJ, et al. Genetic cause of hyperglycaemia and response to treatment
in diabetes. Lancet. Oct 18 2003;362(9392):1275-1281.
75. Shepherd M, Shields B, Ellard S, et al. A genetic diagnosis of HNF1A diabetes alters treatment and
improves glycaemic control in the majority of insulin-treated patients. Diabet Med. 2009;
26(4):437-441.
76. Lindner TH, Njolstad PR, Horikawa Y, et al. A novel syndrome of diabetes mellitus, renal dysfunction
and genital malformation associated with a partial deletion of the pseudo-POU domain of hepato-
cyte nuclear factor-1 beta. Hum Mol Genet. 1999;8(11):2001-2008.
77. Pearson ER, Badman MK, Lockwood CR, et al. Contrasting diabetes phenotypes associated with hepa-
tocyte nuclear factor-1 alpha and -1 beta mutations. Diabetes Care. 2004;27(5):1102-1107.
78. Kirchheiner J, Brockmoller J. Clinical consequences of cytochrome P450 2C9 polymorphisms. Clin
Pharmacol Ther. 2005;77(1):1-16.
79. Kirchheiner J, Brockmoller J, Meineke I, et al. Impact of CYP2C9 amino acid polymorphisms on
glyburide kinetics and on the insulin and glucose response in healthy volunteers. Clin Pharmacol
Ther. 2002;71(4):286-296.
80. Kirchheiner J, Bauer S, Meineke I, et al. Impact of CYP2C9 and CYP2C19 polymorphisms on tolbu-
tamide kinetics and the insulin and glucose response in healthy volunteers. Pharmacogenetics.
2002;12(2):101-109.
81. Zhou K, Donnelly L, Burch L, et al. Loss-of-function CYP2C9 variants improve therapeutic response to
sulfonylureas in type 2 diabetes: a Go-DARTS study. Clin Pharmacol Ther. 2010;87(1):52-56.
82. Holstein A, Plaschke A, Ptak M, et al. Association between CYP2C9 slow metabolizer genotypes and
severe hypoglycaemia on medication with sulphonylurea hypoglycaemic agents. Br J Clin Pharmacol.
2005;60(1):103-106.
83. Wijnen PA, Op den Buijsch RA, Drent M, et al. Review article: The prevalence and clinical relevance of
cytochrome P450 polymorphisms. Aliment Pharmacol Ther. 2007;26 Suppl 2:211-219.
84. Zhang Y, Si D, Chen X, et al. Influence of CYP2C9 and CYP2C19 genetic polymorphisms on pharmaco-
kinetics of gliclazide MR in Chinese subjects. Br J Clin Pharmacol. 2007;64(1):67-74.
85. Guardado-Mendoza R, Prioletta A, Jimenez-Ceja LM, et al. The role of nateglinide and repaglinide,
derivatives of meglitinide, in the treatment of type 2 diabetes mellitus. Arch Med Sci. 2013;
9(5):936-943.
86. He YY, Zhang R, Shao XY, et al. Association of KCNJ11 and ABCC8 genetic polymorphisms with
response to repaglinide in Chinese diabetic patients. Acta Pharmacol Sin. 2008;29(8):983-989.
87. Dai XP, Huang Q, Yin JY, et al. KCNQ1 gene polymorphisms are associated with the therapeutic effica-
cy of repaglinide in Chinese type 2 diabetic patients. Clin Exp Pharmacol Physiol. 2012;39(5):462-468.
88. Kalliokoski A, Neuvonen M, Neuvonen PJ, et al. Different effects of SLCO1B1 polymorphism on
the pharmacokinetics and pharmacodynamics of repaglinide and nateglinide. J Clin Pharmacol.
2008;48(3):311-321.
89. Kalliokoski A, Neuvonen M, Neuvonen PJ, et al. The effect of SLCO1B1 polymorphism on repaglinide
pharmacokinetics persists over a wide dose range. Br J Clin Pharmacol. 2008;66(6):818-825.
90. Kalliokoski A, Backman JT, Neuvonen PJ, et al. Effects of the SLCO1B1*1B haplotype on the
pharmacokinetics and pharmacodynamics of repaglinide and nateglinide. Pharmacogenet Genomics.

2008;18(11):937-942.
91. Zhang W, He YJ, Han CT, et al. Effect of SLCO1B1 genetic polymorphism on the pharmacokinetics of
nateglinide. Br J Clin Pharmacol. 2006;62(5):567-572.
92. Yu M, Xu XJ, Yin JY, et al. KCNJ11 Lys23Glu and TCF7L2 rs290487(C/T) polymorphisms affect
therapeutic efficacy of repaglinide in Chinese patients with type 2 diabetes. Pharmacol Ther.
2010;87(3):330-335.
93. Revollo JR, Korner A, Mills KF, et al. Nampt/PBEF/Visfatin regulates insulin secretion in beta cells as a
systemic NAD biosynthetic enzyme. Cell Metab. 2007;6(5):363-375.
94. Sheng FF, Dai XP, Qu J, et al. NAMPT -3186C/T polymorphism affects repaglinide response in Chinese
patients with Type 2 diabetes mellitus. Clin Exp Pharmacol Physiol. 2011;38(8):550-554.
95. Cheng L, Zhang D, Zhou L, et al. Association between SLC30A8 rs13266634 polymorphism and type 2
diabetes risk: a meta-analysis. Med Sci Monit. 2015;21:2178-2189.
96. Huang Q, Yin JY, Dai XP, et al. Association analysis of SLC30A8 rs13266634 and rs16889462 polymor-
phisms with type 2 diabetes mellitus and repaglinide response in Chinese patients. Eur J Clin
Pharmacol. 2010;66(12):1207-1215.
97. Fromm MF. Importance of P-glycoprotein at blood-tissue barriers. Trends Pharmacol Sci.
2004;25(8):423-429.
98. Gottesman MM, Pastan I. The multidrug transporter, a double-edged sword. The J Biol Chem.
1988;263(25):12163-12166.
99. Xiang Q, Cui YM, Zhao X, et al. The influence of MDR1 G2677T/a genetic polymorphisms on the
pharmacokinetics of repaglinide in healthy Chinese volunteers. Pharmacology. 2012;89(1-2):105-110.
100. Kirchheiner J, Meineke I, Muller G, et al. Influence of CYP2C9 and CYP2D6 polymorphisms on
the pharmacokinetics of nateglinide in genotyped healthy volunteers. Clin Pharmacokinet.
2004;43(4):267-278.
101. Niemi M, Leathart JB, Neuvonen M, et al. Polymorphism in CYP2C8 is associated with reduced plasma
concentrations of repaglinide. Clin Pharmacol Ther. 2003;74(4):380-387.
102. Bidstrup TB, Damkier P, Olsen AK, et al. The impact of CYP2C8 polymorphism and grapefruit juice on
the pharmacokinetics of repaglinide. Br J Clin Pharmacol. Jan 2006;61(1):49-57.
103. Tomalik-Scharte D, Fuhr U, Hellmich M, et al. Effect of the CYP2C8 genotype on the pharmacokinetics
and pharmacodynamics of repaglinide. Drug Metab Dispos. 2011;39(5):927-932.
104. Basu R, Shah P, Basu A, et al. Comparison of the effects of pioglitazone and metformin on hepatic
and extra-hepatic insulin action in people with type 2 diabetes. Diabetes. 2008;57(1):24-31.
105. DeFronzo RA. Pharmacologic therapy for type 2 diabetes mellitus. Ann Intern Med. 1999;131(4):281-303.
106. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas
or insulin compared with conventional treatment and risk of complications in patients with type 2
diabetes (UKPDS 33). Lancet. Sep 12 1998;352(9131):837-853.
107. Graham GG, Punt J, Arora M, et al. Clinical pharmacokinetics of metformin. Clin Pharmacokinet.
2011;50(2):81-98.
108. Zhou M, Xia L, Wang J. Metformin transport by a newly cloned proton-stimulated organic cation trans-
porter (plasma membrane monoamine transporter) expressed in human intestine. Drug Metab Dispos.
2007;35(10):1956-1962.
109. Wang DS, Jonker JW, Kato Y, et al. Involvement of organic cation transporter 1 in hepatic and
intestinal distribution of metformin. J Pharmacol Exp Ther. 2002;302(2):510-515.
110. Kimura N, Okuda M, Inui K. Metformin transport by renal basolateral organic cation transporter hOCT2.
Pharm Res. 2005;22(2):255-259.
111. Otsuka M, Matsumoto T, Morimoto R, et al. A human transporter protein that mediates the final
excretion step for toxic organic cations. Proc Natl Acad Sci USA. 2005;102(50):17923-17928.
112. Masuda S, Terada T, Yonezawa A, et al. Identification and functional characterization of a new human
kidney-specific H+/organic cation antiporter, kidney-specific multidrug and toxin extrusion 2. J Am Soc
Nephrol. 2006;17(8):2127-2135.
113. Kahn SE, Haffner SM, Heise MA, et al. Glycemic durability of rosiglitazone, metformin, or glyburide
monotherapy. New Engl J Med. 2006;355(23):2427-2443.
114. Shu Y, Sheardown SA, Brown C, et al. Effect of genetic variation in the organic cation transporter 1
(OCT1) on metformin action. J Clin Invest. 2007;117(5):1422-1431.
115. Shu Y, Brown C, Castro RA, et al. Effect of genetic variation in the organic cation transporter 1, OCT1,
on metformin pharmacokinetics. Clin Pharmacol Ther. 2008;83(2):273-280.

116. Tzvetkov MV, Vormfelde SV, Balen D, et al. The effects of genetic polymorphisms in the organic
cation transporters OCT1, OCT2, and OCT3 on the renal clearance of metformin. Clin Pharmacol Ther.
2009;86(3):299-306.
117. Christensen MM, Brasch-Andersen C, Green H, et al. The pharmacogenetics of metformin and its
impact on plasma metformin steady-state levels and glycosylated hemoglobin A1c. Pharmacogenet
Genomics. 2011;21(12):837-850.
118. Choi JH, Yee SW, Ramirez AH, et al. A common 5’-UTR variant in MATE2-K is associated with poor
response to metformin. Clin Pharmacol Ther. 2011;90(5):674-684.
119. Zhou K, Donnelly LA, Kimber CH, et al. Reduced-function SLC22A1 polymorphisms encoding
organic cation transporter 1 and glycemic response to metformin: a GoDARTS study. Diabetes.
2009;58(6):1434-1439.
120. Becker ML, Visser LE, van Schaik RH, et al. Genetic variation in the organic cation transporter 1
is associated with metformin response in patients with diabetes mellitus. Pharmacogenomics J.

2009;9(4):242-247.
121. Tarasova L, Kalnina I, Geldnere K, et al. Association of genetic variation in the organic cation trans-
porters OCT1, OCT2 and multidrug and toxin extrusion 1 transporter protein genes with the gastro-
intestinal side effects and lower BMI in metformin-treated type 2 diabetes patients. Pharmacogenet
Genomics. 2012;22(9):659-666.
122. Song IS, Shin HJ, Shim EJ, et al. Genetic variants of the organic cation transporter 2 influence the
disposition of metformin. Clin Pharmacol Ther. 2008;84(5):559-562.
123. Wang ZJ, Yin OQ, Tomlinson B, et al. OCT2 polymorphisms and in-vivo renal functional consequence:
studies with metformin and cimetidine. Pharmacogenet Genomics. 2008;18(7):637-645.
124. Chen Y, Li S, Brown C, et al. Effect of genetic variation in the organic cation transporter 2 on the renal
elimination of metformin. Pharmacogenet Genomics. 2009;19(7):497-504.
125. Becker ML, Visser LE, van Schaik RH, et al. Genetic variation in the multidrug and toxin extrusion 1
transporter protein influences the glucose-lowering effect of metformin in patients with diabetes:
a preliminary study. Diabetes. 2009;58(3):745-749.
126. Jablonski KA, McAteer JB, de Bakker PI, et al. Common variants in 40 genes assessed for diabetes
incidence and response to metformin and lifestyle intervention in the diabetes prevention program.
Diabetes. 2010;59(10):2672-2681.
127. GoDarts, Group UDPS, Wellcome Trust Case Control C, et al. Common variants near ATM are associ-
ated with glycemic response to metformin in type 2 diabetes. Nat Genet. 2011;43(2):117-120.
128. van Leeuwen N, Nijpels G, Becker ML, et al. A gene variant near ATM is significantly associated with
metformin treatment response in type 2 diabetes: a replication and meta-analysis of five cohorts.
Diabetologia. 2012;55(7):1971-1977.
129. Otto C, Lehrke M, Goke B. Novel insulin sensitizers: pharmacogenomic aspects. Pharmacogenomics.
2002;3(1):99-116.
130. Braissant O, Foufelle F, Scotto C, et al. Differential expression of peroxisome proliferator-acti-
vated receptors (PPARs): tissue distribution of PPAR-a, -b, and -g in the adult rat. Endocrinology.
1996;137(1):354-366.
131. Spiegelman BM. PPAR-gamma: adipogenic regulator and thiazolidinedione receptor. Diabetes.
1998;47(4):507-514.
132. Karalliedde J, Buckingham RE. Thiazolidinediones and their fluid-related adverse effects: facts, fiction
and putative management strategies. Drug Saf. 2007;30(9):741-753.
133. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardio-
vascular causes. New Engl J Med. 2007;356(24):2457-2471.
134. MacDonald MR, Petrie MC, Home PD, et al. Incidence and prevalence of unrecognized myocardial
infarction in people with diabetes: a substudy of the Rosiglitazone Evaluated for Cardiac Outcomes
and Regulation of Glycemia in Diabetes (RECORD) study. Diabetes Care. 2011;34(6):1394-1396.
135. Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular events in
patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In
macroVascular Events): a randomised controlled trial. Lancet. 2005;366(9493):1279-1289.
136. Chang TJ, Liu PH, Liang YC, et al. Genetic predisposition and nongenetic risk factors of thiazolidinedione-
related edema in patients with type 2 diabetes. Pharmacogenet Genomics. 2011;21(12):829-836.
137. Kahn SE, Zinman B, Lachin JM, et al. Rosiglitazone-associated fractures in type 2 diabetes: an analysis
from A Diabetes Outcome Progression Trial (ADOPT). Diabetes Care. 2008;31(5):845-851.

138. Yen CJ, Beamer BA, Negri C, et al. Molecular scanning of the human peroxisome proliferator activated
receptor gamma (hPPAR gamma) gene in diabetic Caucasians: identification of a Pro12Ala PPAR
gamma 2 missense mutation. Biochem Biophys Res Commun. 1997;241(2):270-274.
139. Gouda HN, Sagoo GS, Harding A, et al. The association between the peroxisome proliferator-activat-
ed receptor-gamma2 (PPARG2) Pro12Ala gene variant and type 2 diabetes mellitus: a HuGE review
and meta-analysis. Am J Epi. 2010;171(6):645-655.
140. Buchanan TA, Xiang AH, Peters RK, et al. Preservation of pancreatic beta-cell function and prevention
of type 2 diabetes by pharmacological treatment of insulin resistance in high-risk hispanic women.
Diabetes. 2002;51(9):2796-2803.
141. Xiang AH, Peters RK, Kjos SL, et al. Effect of pioglitazone on pancreatic beta-cell function and
diabetes risk in Hispanic women with prior gestational diabetes. Diabetes. 2006;55(2):517-522.
142. Kang ES, Park SY, Kim HJ, et al. Effects of Pro12Ala polymorphism of peroxisome proliferator-
activated receptor gamma2 gene on rosiglitazone response in type 2 diabetes. Clin Pharmacol Ther.
2005;78(2):202-208.
143. Hsieh MC, Lin KD, Tien KJ, et al. Common polymorphisms of the peroxisome proliferator-activated
receptor-gamma (Pro12Ala) and peroxisome proliferator-activated receptor-gamma coactivator-1
(Gly482Ser) and the response to pioglitazone in Chinese patients with type 2 diabetes mellitus.
Metabolism. 2010;59(8):1139-1144.
144. Zhang KH, Huang Q, Dai XP, et al. Effects of the peroxisome proliferator activated receptor-gamma
coactivator-1 alpha (PGC-1alpha) Thr394Thr and Gly482Ser polymorphisms on rosiglitazone response
in Chinese patients with type 2 diabetes mellitus. J Clin Pharmacol. 2010;50(9):1022-1030.
145. Makino H, Shimizu I, Murao S, et al. A pilot study suggests that the G/G genotype of resistin single
nucleotide polymorphism at -420 may be an independent predictor of a reduction in fasting plasma
glucose and insulin resistance by pioglitazone in type 2 diabetes. Endocr J. 2009;56(9):1049-1058.
146. Sun H, Gong ZC, Yin JY, et al. The association of adiponectin allele 45T/G and -11377C/G polymor-
phisms with Type 2 diabetes and rosiglitazone response in Chinese patients. Br J Clin Pharmacol.
2008;65(6):917-926.
147. Liu HL, Lin YG, Wu J, et al. Impact of genetic polymorphisms of leptin and TNF-alpha on rosiglitazone
response in Chinese patients with type 2 diabetes. European J Clin Pharmacol. 2008;64(7):663-671.
148. Jaakkola T, Laitila J, Neuvonen PJ, et al. Pioglitazone is metabolised by CYP2C8 and CYP3A4 in vitro:
potential for interactions with CYP2C8 inhibitors. Basic Clin Pharmacol Toxicol. 2006;99(1):44-51.
149. Kirchheiner J, Roots I, Goldammer M, et al. Effect of genetic polymorphisms in cytochrome p450 (CYP)
2C9 and CYP2C8 on the pharmacokinetics of oral antidiabetic drugs: clinical relevance. Clin Pharma-
cokinet. 2005;44(12):1209-1225.
150. Tornio A, Niemi M, Neuvonen PJ, et al. Trimethoprim and the CYP2C8*3 allele have opposite effects on
the pharmacokinetics of pioglitazone. Drug Metab Disposs. 2008;36(1):73-80.
151. Andrulionyte L, Kuulasmaa T, Chiasson JL, et al, Group S-NS. Single nucleotide polymorphisms of
the peroxisome proliferator-activated receptor-alpha gene (PPARA) influence the conversion from
impaired glucose tolerance to type 2 diabetes: the STOP-NIDDM trial. Diabetes. 2007;56(4):1181-1186.
152. Jamaluddin JL, Huri HZ, Vethakkan SR, et al. Pancreatic gene variants potentially associated with
dipeptidyl peptidase-4 inhibitor treatment response in Type 2 diabetes. Pharmacogenomics.
2014;15(2):235-249.
153. Hart LM, Fritsche A, Nijpels G, et al. The CTRB1/2 locus affects diabetes susceptibility and treatment
via the incretin pathway. Diabetes. 2013;62(9):3275-3281.
154. Mussig K, Staiger H, Machicao F, et al. Association of type 2 diabetes candidate polymorphisms in
KCNQ1 with incretin and insulin secretion. Diabetes. 2009;58(7):1715-1720.
155. Schafer SA, Tschritter O, Machicao F, et al. Impaired glucagon-like peptide-1-induced insulin
secretion in carriers of transcription factor 7-like 2 (TCF7L2) gene polymorphisms. Diabetologia.
2007;50(12):2443-2450.
156. Villareal DT, Robertson H, Bell GI, et al. TCF7L2 variant rs7903146 affects the risk of type 2 diabetes
by modulating incretin action. Diabetes. 2010;59(2):479-485.
157. Schafer SA, Mussig K, Staiger H, et al. A common genetic variant in WFS1 determines impaired
glucagon-like peptide-1-induced insulin secretion. Diabetologia. 2009;52(6):1075-1082.
158. Ishihara H, Takeda S, Tamura A, et al. Disruption of the WFS1 gene in mice causes progressive
beta-cell loss and impaired stimulus-secretion coupling in insulin secretion. Hum Mol Genet.
2004;13(11):1159-1170.
159. Florez JC, Jablonski KA, McAteer J, et al. Testing of diabetes-associated WFS1 polymorphisms in the
Diabetes Prevention Program. Diabetologia. 2008;51(3):451-457.
160. Sparso T, Andersen G, Albrechtsen A, et al. Impact of polymorphisms in WFS1 on prediabetic
phenotypes in a population-based sample of middle-aged people with normal and abnormal glucose
regulation. Diabetologia. 2008;51(9):1646-1652.
161. Hasan FM, Alsahli M, Gerich JE. SGLT2 inhibitors in the treatment of type 2 diabetes. Diabetes Res
Clin Pract. 2014;104(3):297-322.
162. Yu L, Lv JC, Zhou XJ, et al. Abnormal expression and dysfunction of novel SGLT2 mutations identified
in familial renal glucosuria patients. Hum Genet. 2011;129(3):335-344.
163. Lee YW. Clinical and genetic analysis in a patient with primary renal glucosuria: Identification of a
novel mutation in the gene. Exp Ther Med. 2013;6(6):1532-1534.
164. Enigk U, Breitfeld J, Schleinitz D, et al. Role of genetic variation in the human sodium-glucose
cotransporter 2 gene (SGLT2) in glucose homeostasis. Pharmacogenomics. 2011;12(8):1119-1126.
Chapter 15 • Pharmacogenomic Testing and Drug Labeling 435
PART III
Important Issues in
Pharmacogenomics
CHAPTER 15: Pharmacogenomic Testing and Drug Labeling................437
CHAPTER 16: Pharmacogenomics in Practice:

The Role of the Pharmacist...................................................................................................453
CHAPTER 17: Ethics and Pharmacogenomics.......................................................467
CHAPTER 18: Future of Pharmacogenomics in Clinical Practice......... 491

435
CHAPTER
15
Pharmacogenomic Testing and
Drug Labeling
Kimberly L. Barefield, PharmD, BCPS, CDE; James W. Fetterman, Jr.,

PharmD; and Andria Fetterman, PharmD

After completing this chapter, the reader CLINICAL VALIDITY—the accuracy with which
should be able to a test identifies or predicts a patient’s
clinical status.
Define the clinical implications of
pharmacogenomics in daily practice. GENETIC VARIANTS—an alteration in a gene
distinct from the normal, wild-type allele.
Describe the difference between
the various genetic markers used for PHARMACOGENOMICS—the branch of
determining a patient’s response to pharmacology that deals with the
therapy. influence of genetic variation on drug
response in patients by correlating
Identify which valid biomarkers are
gene expression or single nucleotide
approved by the FDA for use in the
polymorphisms with a drug’s efficacy or
treatment of patients with various
toxicity.
disease states.
VALID BIOMARKERS—a biomarker that is
Discuss the factors that are involved
in evaluation of cost versus benefit of measured in an analytical test system
pharmacogenomic testing. with well-established performance
characteristics and for which there is
an established scientific framework or
body of evidence. This evidence must
elucidate the physiologic, toxicologic,
pharmacologic, or clinical significance of
the test results.
437
INTRODUCTION
The successful mapping of the human genome has made personalized pharmacotherapy a
viable option based on the ability to determine the specific gene variant(s) that can aid safe
and effective drug therapy for a patient’s characteristics. Many alleles have been identified
affecting various diseases and the susceptibility of these diseases to the medications used
for therapy. In this chapter, our purpose is to assist the practitioner in determining (1) which
clinical tests are available for use in practice and (2) for which disease and/or therapy that
particular test will be most valuable. Such knowledge will enable the practitioner to better
care for the patient and ensure optimum response to the drug or dosage chosen.
Several disease states will be reviewed including cardiovascular, respiratory, endocri-
nology, hematology/oncology, central nervous system (CNS)/psychiatric, and infectious
disease. The practicality of pharmacogenomic testing will be discussed within each area as
it relates to clinical practice.
CASE STUDY—PSYCHIATRIC DISORDERS:

EXTREME ANXIETY
B.J., a 25-year-old female, presents with a recent history of extreme anxiety. She has been
on fluoxetine 20 mg daily for about 2 months. She has a history of type 2 diabetes and is
being treated with glyburide 5 mg daily. She is otherwise healthy. B.J. was doing better
after about 2 weeks, but then she began to exhibit the following signs and symptoms:
the inability to sit still and, recently, a persistent tremor as well as excessive sweating.
She has been using two other medications to treat her anxiety prior to fluoxetine but had
to stop them because of severe side effects. Her labs were as follows: comprehensive
metabolic panel within normal limits, blood glucose 175 mg/dL, and A1c 7%; fluoxetine
levels were 2½ times normal limits. Pharmacogenomic testing for CYP2D6 and CYP2C9
were ordered. The results revealed no variants in CYP2C9, but the CYP2D6 genotype was
CYP2D6*3/*4. The CYP2D6 variants are both associated with no enzyme function.
Questions
1. Given the information above, what additional pharmacogenomic testing would you recom-
mend?
2. Is B.J. an extensive or poor metabolizer of CYP2D6?
3. What additional laboratory values would assist in evaluating B.J. for pharmacogenomic
testing?
CLINICAL IMPLICATIONS
Throughout the pharmacy curriculum, it is taught that each drug may work differently
in different patients or each patient may respond differently to the same drug given to
another patient. Known variabilities of drug response due to genetic variations or genetic
mutations are caused by environmental exposure to various toxins.1 According to Zineh,

while a large amount of research is being conducted on the top 200 most prescribed drugs,
a much smaller amount of research is being conducted in the area of pharmacogenomics-
based prescribing for these drugs.2 As a result, such information is not widely found in the
manufacturer’s product insert (PI); however, several more drugs have been added to the
list of valid biomarkers prepared by the U.S. Food and Drug Administration (FDA). Although
more extensive research on the vast majority of drugs needs to be conducted to clarify the
role of pharmacogenomics, drugs with a narrow therapeutic index (e.g., warfarin) have been
well studied and prove to be candidates for the clinical application of pharmacogenomics.
Given the large number of polymorphisms identified thus far, practitioners have diffi-
culty determining which polymorphism(s) have clinical impact. The FDA has approved
several pharmacogenomics-related clinical tests and is now requiring manufacturers to

include that information within their PIs. Approximately 29% of the drugs approved by
the FDA currently contain pharmacogenomics information in their PIs.3,4 Presently, 134
prescription drugs are listed in the FDA Table of Pharmacogenomic Biomarkers for Drug
Labeling; 23 of them are impacted by more than one biomarker for a total of 157 drugs.
This information is based on valid genomic biomarkers, which can help the practitioner
to identify patients who may be responders or nonresponders to various drugs with the
goal of decreasing adverse drug reactions (ADRs) or identifying patients who will respond
better to certain drugs. According to FDA’s definition, a valid biomarker is “one for which
an established and validated assay exists for which an established body of evidence exists
that supports its pharmacological and/or clinical significance.”5
The genomic biomarkers listed on the labels of certain drugs are classified accord-
ing to their effect on clinical response and the indication of risk to the patient. These
genomic biomarkers assist in determining the dose as well as identifying the susceptibility,
resistance, differential diagnosis, and polymorphic targets. Not all of the FDA-approved
labels make recommendations for required pretesting or dosing guidelines based on phar-
macogenomics. Some are provided for informational purposes only, thereby alerting the
practitioner to the possible effects of the biomarker or any polymorphisms that might be
present for that drug.
The following sections review selected biomarkers and their potential impact on the
treatment of cardiovascular diseases, respiratory diseases, endocrinology, hematology/
oncology, CNS/psychiatric disorders, and infectious diseases. Table 15-1 provides a list of
valid biomarkers associated with the FDA-approved drug labels along with the appropriate
recommendations for genomic testing.3,6
CLINICAL PEARL
Approximately 29% of all prescription medications within the United
States have genetic testing options available.
Cardiovascular Diseases
One of the most commonly used drugs in the treatment of various cardiovascular diseases
is warfarin. As a vitamin K antagonist, warfarin is clinically indicated in the treatment of
patients presenting with a variety of diseases requiring anticoagulation such as deep vein
thrombosis, atrial fibrillation, myocardial infarction, pulmonary embolism, and patients with
certain types of artificial heart valves. Warfarin is primarily metabolized by cytochrome

P-450 isoenzymes 2C9 (CYP2C9), which is responsible for the metabolism of approxi-
mately 60-70% of the S-warfarin isomer.7,8 Variations in this enzyme, specifically the single
nucleotide polymorphisms CYP2C9*2 and CYP2C9*3, are responsible for affecting warfarin
metabolism.4,9,10 Those patients who carry either one or both of these variants tend to
Table 15-1
List of Valid Biomarkers Associated with FDA-Approved Drug Labels
Clinical
Presentation Pharmacogenomics Biomarker Drug
Cardiovascular CYP2C9 variants (intermediate or Warfarin

poor metabolizers)
PROC (protein C deficient)
PROS (protein S deficient)
VKORC1 (VKORC1 A allele carriers)
CYP2D6 (poor metabolizers) Metoprolol
Propranolol
Carvedilol
Propafenone
NAT-2 (slow acetylators) Isosorbide and hydralazine
LDLR (heterozygous and homozygous Pravastatin
familial hypercholesterolemia)
CYP2C19 (intermediate or poor Clopidogrel
metabolizers)
CYP2D6 (poor metabolizers) Quinidine
CYP2C19 (poor metabolizers) Prasugrel
CYP2C9 (variant carriers)
CYP3A5 (variant carriers)
CYP2B6 (variant carriers)
CYP2C19 (poor metabolizers) Ticagrelor
Respiratory UGT1A1 (UGT1A1 poor metabolizers) Arformoterol
CYP2D6 (CYP2D6 intermediate or Arformoterol
poor metabolizers)
UGT1A1 poor metabolizer Indacaterol
CFTR (CFTR G551D, G1244E, Ivacaftor
G1349D, G178R, G551S, S1251N,
S1255P, S549R mutation carriers,
F508del mutation homozygotes)
Endocrinology G6PD (G6PD deficient) Glimepiride
Glyburide
Glipizide
Chlorpropamide
Hematology/ See Cardiovascular Warfarin
Oncology F5 (Factor V Leiden carriers) Eltrombopag
SERPINC1 (antithrombin III
deficient)
Del (5q) (chromosome 5q deletion Lenalidomide
positive)
G6PD (deficient) Succimer
Methylene blue
Dabrafenib
BRAF (V600E/K mutation positive) Dabrafenib
ESR1 (estrogen receptor-positive) Exemestane, everolimus
ERBB2 (HER2 protein overexpres- Lapatinib
sion positive or gene over amplifi-
cation positive) Ado-trastuzumab emtansine
ERBB2 (HER2 protein overexpres- Everolimus
sion negative)
ESR1, PGR (hormone receptor- Letrozole
positive)

Clinical
Hematology/ EGFR (EGFR exon 19 deletion or exon Erlotinib

Oncology 21 substitution [L858R] positive)
(continued) EGFR (EGFR protein expression
positive)
TPMT (intermediate or poor Mercaptopurine
metabolizers)
HLA-DQA Lapatinib
HLA-DRB1 (HLA-DQA1*0201 or –
DRB1*0701 allele carriers)
ESR1, PGR (hormone receptor- Anastrozole
positive)
DPYD poor metabolizer Fluorouracil
BCR-ABL (Philadelphia chromosome Omacetaxine
positive)
UGT1A1 poor metabolizer Nilotinib
MS4A1 (CD20 antigen positive) Obinutuzumab
IL2RA (CD25 antigen positive) Denileukin diftitox

EGFR (EGFR protein expression Panitumumab
positive)
KRAS (KRAS codon 12 and 13
mutation negative)
BCR/ABL1 (Philadelphia chromo- Dasatinib
some positive; T315I mutation-
positive)
UGT1A1 poor metabolizers [TA]7/ Pazopanib
[TA]7 genotype)
ESR1, PGR (hormone receptor- Fulvestrant
positive)
PGR (progesterone receptor-positive) Exemestane
ERBB2 (HER2 protein overexpres- Fulvestrant
sion positive) Pertuzumab
Del (17p) (chromosome 17p deletion Ibrutinib
positive)
BCR-ABL1 (Philadelphia chromo- Ponatinib
some positive; T315I mutation
positive)
ALK (ALK gene rearrangement Crizotinib
positive)
KIT (KIT protein expression positive, Imatinib
c-KIT D816V mutation negative)
BCR-ABL1 (Philadelphia chromo-
some positive)
PDGFRB (PDGFR gene rearrange-
ment positive)
FIP1L1-PDGFRA (FIP1L1-PDGFRα
fusion kinase [or CHIC2 deletion]
positive)
DPYD (DPD deficient) Capecitabine
BCR-ABL1 (Philadelphia chromo- Busulfan

some negative)
BCR/ABL1 (Philadelphia chromo- Bosutinib
some positive)
ESR1, PGR Tamoxifen
F5 (Factor V Leiden carriers)

Clinical
Hematology/ F2 (prothrombin 20210A allele Tamoxifen

Oncology positive)
(continued) EGFR (EDFR exon 19 deletion or exon Afatinib
21 substitution) (L858R) positive
UGT1A1 (UGT1A1*28 allele carriers) Irinotecan
TPMT (TPMT intermediate or poor Thioguanine
metabolizers—deficient)
PML-RARA (PMI-RARα Arsenic trioxide
translocation positive)
MS4A1 (CD20 antigen positive) Tositumomab
BRAF (BRAF V600E/K mutation Trametinib
positive)
ERBB2 (HER2 protein Trastuzumab
overexpression positive)
PML/RARA (PML/RARα Tretinoin
translocation positive)
CNS/Psychiatric CYP2D6 (CYP2D6 poor metabolizers) Imipramine
Disorders Nefazodone
Pimozide
Risperidone
Perphenazine
Iloperidone
Desipramine
Fluvoxamine
Nortriptyline
Doxepin
Clozapine
Clomipramine
Fluoxetine
Citalopram
Atomoxetine
Thioridazine
Protriptyline
Aripiprazole
Modafinil
Trimipramine
Amitriptyline
Venlafaxine
Tetrabenazine
Dextromethorphan and quinidine
Vortioxetine
Galantamine
Paroxetine
CYP2D6 (CYP2D6 normal Doxepin
metabolizers)
CYP2C19 (CYP2C19 poor Citalopram
metabolizers)
CYP2C19 (CYP2C19 poor Diazepam
metabolizers) Clobazam
POLG (POLG mutation positive) Divalproex

Valproic acid
Phenytoin
HLA-B (HLA-B*1502 allele carriers) Carbamazepine
HLA-A (HLA-A*3101 allele carriers) Carbamazepine
NAGS, CPS1, ASS1,OTC, ASL, ABL2
(urea cycle enzyme deficient) Valproic acid

Clinical
Infectious CYP2D6 (CYP2D6 poor metabolizers) Terbinafine, quinine sulfate

Diseases NAT1-2 (slow acetylators) Rifampin, isoniazid, and
pyrazinamide

Ebook PDF

Uploaded by

Copyright:

Available Formats

Ebook PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ebook PDF

Uploaded by

Copyright:

Available Formats

All rights reserved.

Because of ongoing research and improvements in technology, the information and

Editorial Project Manager: Ruth Bloom

Library of Congress Cataloging-in-Publication Data

Names: Zdanowicz, Martin M., editor. | American Society of Health-System Pharmacists.

© 2017, American Society of Health-System Pharmacists, Inc. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any

ASHP is a service mark of the American Society of Health-System Pharmacists, Inc.;

To my children, Alex and Olivia,

Preface .............................................................................................................................................................................................................. viii

PART I—Fundamentals of Pharmacogenomics ....................................................................................................... 1

PART II—Applications of Pharmacogenomics in Therapeutics...............................................................213

PART III—Important Issues in Pharmacogenomics..........................................................................................435

Chapter 18: Future of Pharmacogenomics in Clinical Practice....................................................................................................... 491

INDEX .................................................................................................................................................................................... 499

Patti W. Adams, PhD James W. Fetterman, Jr., PharmD

Keri C. Anderson, PharmD, BCPS Edward Grace, PharmD, BCPS (AQ-ID),

Andria Fetterman, PharmD

Kenric B. Ware, PharmD, MBA, AAHIVP

Helen E. Smith, MS, PhD, RPh Martin M. Zdanowicz, PhD, MEd, MA

Todd A. Thompson, PhD

CHAPTER 1: Pharmacogenomics: Current and Future Promise...................3

CHAPTER 2: The Genetic Basis of Pharmacogenomics..................................19

CHAPTER 3: Methodologies in Pharmacogenomics......................................... 59

CHAPTER 4: The Pharmacogenetics of Drug Metabolism.........................107

CHAPTER 5: Pharmacogenomics of Transport Proteins................................151

CHAPTER 6: Pharmacodynamics and Pharmacogenomics......................183

Martin M. Zdanowicz, PhD, MEd, MA

LEARNING OBJECTIVES KEY DEFINITIONS

more alleles for a gene or DNA sequence

SINGLE NUCLEOTIDE POLYMORPHISM (SNP)—a point mutation occurring in >1% of the

methyltransferase (involved in 6-mercaptopurine methylation) and CYP2C19 were also

developed significant neutropenia following treatment with azathioprine.12

SOURCES OF GENETIC VARIABILITY IN DRUG RESPONSE

GENETIC VARIABILITY AND DRUG PHARMACOKINETICS

CYP2C9*2, *3, *8, *11 Decreased Warfarin, phenytoin

CYP3A4*22 Decreased Tacrolimus, paclitaxel, midazolam

P-glycoprotein Decreased Numerous substrates including

The Human CYP Allele Nomenclature Database (http://www.cypalleles.ki.se/) lists only

influenced by numerous drugs that act as inducers (e.g., carbamazepine, phenobarbital,

GENETIC VARIABILITY AND DRUG PHARMACODYNAMICS

a reduced response to asthma medications due to variants in β2-adrenergic receptors and

Gene Product Affected Drug(s) Clinical Impact

Angiotensin-converting ACE inhibitors Reduced efficacy

Despite significant expansion of pharmacogenomic research in the past decade, the

In addition to their role as healthcare providers, pharmacists currently serve as educa-

metabolizing enzymes as biomarkers to identify patients who might be “slow metabolizers”

IMPACT OF PHARMACOGENOMICS ON DRUG DEVELOPMENT

Pharmacogenomics and pharmacogenetics can provide investigators with a tool for

populations or subsets of patients who might be most responsive to a particular medica-

Pharmacogenomics might also aid new drug development by allowing investigators

FUTURE PROMISE OF PHARMACOGENOMICS

Pharmacogenomics might also provide an additional and highly specific means of

If the use of pharmacogenomics is to become fully incorporated into daily clinical

Optimization of drug choice and dose to Expanding the application of

More rapid drug development Cost effectiveness

in clinical practice: a case study of thiopurine methyltransferase genotyping in acute lymphoblastic

LEARNING OBJECTIVES KEY DEFINITIONS

copy of the genome.

CYP2C92, 3, 8, 11 Decreased Warfarin, phenytoin