4 - Epidemiology and The Evidence Bas - 2019 - Withrow and MacEwen S Small Anima

4
Epidemiology and the Evidence-Based

Medicine Approach
AUDREY RUPLE, BRENDA N. BONNETT, AND RODNEY L. PAGE
Epidemiology is defined as the study of the distribution and deter- and molecular characterization of tumors), therapeutic interven-
minants of disease in populations. Historically epidemiologic tions (e.g., interventional surgery and targeted, small-molecule
methods were used primarily in veterinary populations for the chemotherapy), and the expanding field of genomics in cancer
investigation of outbreaks and/or epidemics of infectious disease, research. This is attributable in part to the presumption that
yet the philosophies, attitudes, methodologies, and application most clients want care for their pets at a level similar to that they
of epidemiology are in fact more broadly applicable to research themselves receive. Therefore many approaches and interventions
and clinical practice, regardless of species, disease, or discipline. In have been adopted from human medicine and applied to animals
fact, epidemiologic principles form the foundation of evidence- despite considerable gaps in evidence as to their efficacy and/or
based medicine (EBM), an approach to the practice of health care effectiveness in the veterinary clinical situation. In addition, even
that is now well accepted in the human and veterinary medicine where a sufficient quantity of studies is present, the quality and
fields. For a clinician, using the EBM approach involves a com- consistency of reporting is frequently inadequate to allow system-
mitment to base all decisions on the best available evidence and atic review or adequate comparison between studies.1 This issue is
to be explicit about the level and quality of evidence on which not unique to oncology and has spawned efforts to improve the
decisions are based. An extensive literature is available on EBM reporting of veterinary studies, with a longer term goal of improv-
and evidence-based practice in the human medicine field (e.g., ing the quality of work.1–7 To approach a level of care in veterinary
The Cochrane Collaboration [http://www.cochrane.org/]) and oncology truly similar to that in humans, there will need to be an
in the veterinary field (e.g., Evidence-Based Veterinary Medicine increased focus on EBM. Further information and articles perti-
Association [https://ebvma.org/]). The EBM approach can and nent to challenges of applying EBM in practice can be found on
should be applied to all interventions, including diagnosis and the website of the Evidence-Based Veterinary Medicine Associa-
prognosis, choice of preventive and clinical therapies applied to tion (https://ebvma.org) and the Centre for Evidence-Based Vet-
individuals, and decisions about health policy or control programs erinary Medicine (http://www.nottingham.ac.uk/cevm/).
for populations. In this chapter, we focus on quantifying the occurrence of
Pathophysiology forms the basis of our understanding of cancer (incidence, prevalence) and risk factors for cancer (causal
health and disease, but this knowledge, even combined with clin- reasoning, associations). An evidence-based approach to diagno-
ical acumen and experience, is not sufficient grounds for decision sis, prognosis, and selection of therapeutic interventions will be
making across the spectrum of activities of health professionals. proposed, although other authors in this text will present specific
To have confidence that our interventions will be beneficial, we details of diagnosis, prognosis, and therapy for specific cancers.
need to understand that personal and expert opinion are only Rather than presenting an exhaustive or systematic review of the
anecdotal evidence, unless they are based on a valid appraisal of literature in this chapter, we will highlight the relevant literature.
available evidence from the literature. In addition to embrac- Our aim is to provide a guide for the application of epidemiologic
ing the philosophy of EBM, all clinicians must develop the principles to oncology, in general and for clinical practice.
knowledge and skills, such as information management, critical
appraisal, and causal reasoning, that are needed to assess evidence Measures of Disease Occurrence
to determine that their chosen interventions are both efficacious
and effective (see glossary of terms in Table 4.1). Unfortunately, Complete and accurate cancer surveillance data are the founda-
especially in veterinary medicine, there are many gaps in our evi- tion needed to make appropriate conclusions about the burden
dence base, in terms of both validity and relevance of published of disease, to make recommendations for cancer prevention and
studies. control, and for the design of analytic studies to identify causal
In veterinary medicine, in general and in certain special- associations between exposures and cancer risk. Here we cover the
ties including oncology, the trend has been toward a heightened measures used to quantify cancer occurrence such as incidence,
sophistication of practice, including the use of advanced technolo- prevalence, and proportional measures and the types of data used
gies in diagnostic testing (e.g., state-of-the-art imaging techniques to calculate them.
81
82 PA RT I The Biology and Pathogenesis of Cancer
TABLE 4.1 Glossary of Terms

Term Definition Comments
Efficacy How well a treatment works in those who receive it (e.g., May be proved in laboratory studies or clinical
correct formulation, dose). trials.
Effectiveness How well a treatment works in those to whom it is offered. Studies must occur in the environment and under
conditions and with patients typical of those to
whom it will be offered in practice.
Compliance How closely a treatment protocol is followed. Influenced by clinician, client, patient, formulation,
duration, and so forth.
Coherence How well findings reflect our understanding of biologic Limited by our current understanding.
relationships/pathophysiology.
Consistency The extent to which new findings agree with previously Limited by the current literature, traditional
published findings. approaches, funding, and so forth.
Experimental studies Traditional research approach done in a laboratory or highly Potential for high validity, generally lower relevance
controlled environment. to the clinical situation.
External validity The extent to which a study’s findings can be extrapolated to A function of the study population, methods, data
a wider population. Similar terms include relevance and collection, treatments, and so forth.
generalizability.
Incidence rate The rate at which new events occur in a population: (Number of Cancer incidence rates are available from popu-
new events in a specified period) ÷ (Number of individuals at lation-based data (e.g., cancer registry data) or
risk during this period) × 10n prospective (cohort or longitudinal) studies.
Internal validity The extent to which a study’s findings are likely correct for that Likelihood that systematic bias is responsible for
study population. the study findings reduces its validity (e.g.,
because of bias in selecting study participants,
measuring the exposure, and confounding).
Observational studies Epidemiologic studies that use existing comparisons in the Examples: (1) Case-control study: Researcher
species of interest in its “natural” environment (often observes/describes exposures in individuals
client-owned animals, perhaps in veterinary practice set- selected based on presence/absence of the
tings). outcome; (2) Cohort study: Individuals with dif-
ferent exposures are followed and incidence of
outcome(s) is observed.
Randomized controlled trial (RCT) Randomized refers to the random allocation of exposure. Researcher exerts control over which individuals
Controlled refers to appropriate comparison groups (e.g., receive which treatments or exposures and
placebo or standard treatment). observes outcomes.
Trial is generally conducted in a clinical setting.
Prevalence The number of events in a given population at a designated Taking the number of canine cancers that are
time: Number of events at a designated time ÷ Number of observed in a clinic or several clinics during a
individuals at risk at the designated time. designated period of time and dividing by the
total number of patients seen during the same
period is a proportional measure, not preva-
lence.
Proportional morbidity or mortality The number of events (e.g., disease, death) in a limited Proportional measures are used when the underly-
population (e.g., animals presenting to the clinic, total ing population at risk is not known.
deaths) at a designated time.

Incidence because of the scarcity of animal cancer registries and lack of infor-
Incidence, or the number of newly diagnosed cancer cases divided mation about the total animal population (census data) at risk.
by the total population at risk over a specified period of time, is the Cancer incidence data has been provided from several popula-
most useful disease occurrence statistic for comparison between tion-based cancer registries (Table 4.2). Estimates of canine can-
populations over time. Incidence data are especially valid when cer incidence range from 99.3 to 804 per 100,000 dog-years.8–17
they are generated from a large population-based cancer registry Variation in estimates may be due in part to differences in actual
with histologically confirmed cases and complete ascertainment cancer risks and/or variation in the base population. These regis-
of the population at risk within a defined geographic area or theo- tries included information from all cancer cases identified within
retically from large prospective, longitudinal, or cohort studies. a specified geographic region from a well-defined and enumerated
True incidence data are rarely obtainable in veterinary populations population. One of the earliest, well-known cancer registries for
CHAPTER 4 Epidemiology and the Evidence-Based Medicine Approach 83
TABLE 4.2 Characteristics of Population-Based Companion Animal Cancer Registries and Cancer Incidence
Registry Period Cases/Population at Risk Incidence/Prevalence
California Animal Neoplasm Registry (CANR)8,9 1963–1966 1624/80,006 dogs 381.2/100,000 dogs over 3-year
448/54,786 cats period.
155.8/100,000 cats over the 3-year
period.
Tulsa Registry of Canine and Feline Neo- 1972–1973 899 cases/63,504 dogs; 59 cases/11,909 cats 1126 cases per 100,000 dogs; 470
plasms10 (Tulsa, Oklahoma) cases per 100,000 cats
Norwegian Canine Cancer Registry11,12 (Oslo, 1990–1998 14,401 tumors/census of dogs in Norway in Boxers: 28 and 14/1000 dogs per
Norway) 1992–9313 year for total and malignant
tumors, respectively.
Bernese mountain dogs: 10 and
4/1000 dogs/year for total and
malignant tumors, respectively.
Genoa Registry of Animal Tumors13 (Genoa, 1985–2002 3303/107,981; 1,943,725 dog-years Males: 99.3/100,000 dog-years
Italy) Females: 272.1/100,000 dog-years,
for total tumors (malignant and
benign).
Animal Tumor Registry14 (Venice, Italy) 2005 2509 dogs; 494 cats/296,318 dogs; 214,683 282, 143 and 140/100,000 dogs
cats for total, malignant and benign
tumors, respectively; 77, 63,
14/100,000 cats for total,
malignant and benign tumors,
respectively.
Swiss Cancer Registry15 1955–2008 Registered Swiss dog population 13/100,000 dogs in 1955 to
695/100,000 dogs in 2008.
Danish Veterinary Cancer Registry (DVCR)16 2005–2008 1523 dogs/dogs registered in the Danish Dog Breeds with standardized morbidity
Registry as of August 2006 ratios ≥2: Boxer, Bernese mountain
dog, and West Highland white
terrier. Measures for all dogs were
not provided.
Piedmont Canine Cancer Registry17 2001–2008 1175 tumors/dogs recorded in registration 804/100,000 dogs for malignant
system in Piedmont, Italy tumors; 897/100,000 dogs for
benign tumors

companion animals was the California Animal Neoplasm Regis- residents of two provinces in northern Italy over a 3-year period start-
try.8,17 This comprehensive effort began in 1963 with the goal of ing in 2005. Earlier prevalence data for feline cancers have ranged
identifying all neoplasms diagnosed over a 3-year period among from 51.9/100,000 cat-years from the California Animal Neoplasm
animals living in the San Francisco Bay Area Counties of Alameda Registry8,18 to 470.2/100,000 cat-years from the Tulsa Registry.10
and Contra Costa. The denominator was estimated by conduct- In addition to population-based cancer registries, cancer occur-
ing a survey in a probability sample of households in Alameda rence data are abundantly available from veterinary teaching hos-
County to derive the age, sex, and breed distribution of pets and pital databases and insurance databases. A caution to be noted
to determine whether the household had used veterinary services. when interpreting cancer occurrence information from hospital-
Additional information on former and existing cancer registries based registries is that data may be inconsistently recorded or
for companion animals has been comprehensively reviewed.18,19 inaccurate and the size and characteristics of the underlying popu-
lation at risk are not known20,21; thus neither true incidence nor
prevalence measures can be calculated. Instead, the proportional
Prevalence morbidity ratio (PMR) is used to quantify cancer occurrence. For
Cancer prevalence information from population-based registries example, the PMR for a particular tumor type among a single
is also useful for surveillance and comparison between popula- breed is calculated as follows:
tions. Prevalence is the number of total cancer cases divided by the (Number of tumor type in breed ÷
number of dogs in the population at risk at one point in time. For number of total tumors in breed) ÷
example, the prevalence of canine cancer in April 2005 was 143 (Number of tumor type in all other breeds ÷
per 100,000 dogs in an Italian population (Table 4.2).8,9 number of total tumors in all other breeds).
Feline cancer prevalence has been reported from a population-
based registry in Italy as 63 per 100,000 cats14 (see Table 4.2). These Proportional measures are not to be interpreted as prevalence or
data were based on a telephone survey conducted among 214,683 incidence of cancer occurrence. As an example, Craig et al presents
proportional statistics from a necropsy database and concludes that would die before 10 years of age, whereas the values for golden
golden retrievers have an increased “risk” of tumors similar to that retrievers are somewhat different (30%, 50%). Of course, there may
for Boxers.22 However, only the proportion of dead dogs that had be true differences between the two study populations, and/or the
cancer are available in that study, and these data cannot be used to differences may be influenced by referral bias and the high propor-
estimate risk. Although proportional measures, such as those pre- tion of unclassifiable deaths in the VMDB study.
sented in an article by Fleming et al,23 have some usefulness for To further illustrate this example, using just the breeds in Fig.
describing patterns within a breed, they are very risky to use for 4.1 and data from the Swedish insurance database, if one ranked
comparison across breeds in which population-based measures are the breeds based on actual numbers of dogs that died because of
unavailable and the degree of referral bias is unknown. In addition, tumors (e.g., perhaps how an oncology clinician would perceive
in those data, 40% of deaths could not be classified pathologically the “risk” based on dogs that present to a specialty clinic), golden
and the unclassified proportion showed extreme variation across retrievers would be number one because they are among the more
breed (e.g., 16%–60%). Fig. 4.1 shows a comparison between pro- numerous breeds in this population. Likewise, if one ranked the
portional mortality ratios and true mortality rate using a subset of breeds by the proportion of dead dogs that had tumors (e.g., simi-
data from Bonnett et al, a study with information on the population lar to what would be reported in analysis of postmortem data),
at risk and data from the Veterinary Medical Database (VMDB) the top three would be BMDs, Boxers, and golden retrievers. So,
study.23,24 For golden retrievers, 30% of deaths (before 10 years of in these examples, as has been frequently reported in the United
age in the Swedish insurance population) were a result of cancer.24 States, based on proportional statistics, golden retrievers would be
For Leonbergers and Boxers, the proportional mortality was 28% labeled as one of the highest risk breeds. However, in looking at
and 37%, respectively.24 Proportional values for these three breeds the true incidence based on these Swedish data, they do not have
may be similar, but, in fact, Leonbergers and Boxers have a risk for an increased risk (before 10 years of age) compared with all breeds.
death resulting from cancer (before 10 years of age) that is almost There is likely considerable misunderstanding of the occurrence
four times as high as that for golden retrievers (approximately 200 of cancer in dogs in the United States because of the lack of
deaths per 10,000 years-at-risk versus 55 [p > 0.05]).24 Irish wolf- accurate incidence data and confusion about the interpretation
hounds and Bernese mountain dogs (BMDs) have an equal risk of proportional statistics. Of course, where a breed is very com-
(approximately 300 deaths resulting from tumors per 10,000 dog- mon, such as the golden retriever, and given that a considerable
years at risk [DYAR]), but tumors account for more than 40% of proportion of them die of cancer, that will represent an important
deaths in BMDs and only 22% in Irish wolfhounds.24 Note that population burden of disease, even if they are not truly the “high-
these are deaths before 10 years of age. Comparing the proportional est risk” breed. This is why golden retrievers were selected for the
mortality values between the two studies, values for BMDs are very largest prospective cohort study conducted in a dog population,
similar (42%, 45%), perhaps because almost all dogs of this breed the Golden Retriever Lifetime Study (GRLS), to investigate the
Yearly cancer mortality rate (deaths due to cancer per 10,000 dog-years-at-risk in Swedish dogs ≤ 10 years of age).*
Proportional cancer mortality (%) (deaths due to cancer ÷ all deaths in Swedish dogs ≤ 10 years of age).*
Relative Frequency (%) (deaths due to cancer ÷ eligible deaths in dogs from VMDB).† 306
296
197 203
168
119
105
50 55 50
45 41 45
34 30 37 44 32
21 27 23 23 26 28 22
18 14 20 13 na
1 1 1 3 3,4 4,5 5,6 ,6 ,7 7

iever reeds iever dland ne an er xer5
un d6 Dog
etr etr Da m rg Bo tain
R All B nR wfoun at ber nbe lfho un
rador lde Ne Gre Do Leo h Wo Mo
Lab Go Iris nes
e
r
Be
• Fig. 4.1 Comparison of true mortality rate (blue column) and proportional mortality (red column) for
selected dog breeds. True mortality is reported as the total number of deaths resulting from cancer per
10,000 dog-years-at-risk (e.g., the cancer mortality in Labrador retrievers was 45 deaths per 10,000
dogs per year). The proportional mortality is reported as the percentage of deaths resulting from cancer
compared with all deaths reported in that breed (e.g., 21% of deaths in Labrador retrievers were caused
by cancer). The 95% confidence intervals for mortality rates overlap for breeds with the same number
(e.g., mortality risk in Labrador retrievers and golden retrievers was not different from that for all breeds
combined, but was different from the other breeds listed in the table).24,25
risk factors for occurrence of cancer.25 In addition, the Swedish within the Swedish data is quite informative given that the limi-
data only include dogs up to 10 years of age; it is unknown how tations occur equally across breeds. Crudely comparing overall
the statistics would look if dogs of all ages were included. As the mortality rates for cancer, BMDs were approximately six times
authors of this study discuss, for cancer (or any cause of death) more likely to die of cancer compared with all dogs combined
that occurs at older ages, a dog must live long enough to experi- (306 versus 50 per 10,000 DYAR, respectively).24 Where it was
ence it (i.e., not die at a younger age because of any other cause), possible to do more sophisticated analyses, BMDs were shown to
and deaths before 10 years of age are relevant to focus on for can- be 17 times more likely to die of cancer, compared with baseline,
cer prevention.24 and adjusting for age, gender, and breed.33 Even if specifics of the
population may not be the same as other populations, data such as
these are important for identifying high-risk breeds. Comparison
Sources of Information on Cancer Occurrence across populations and over time is needed, with due consider-
One of the largest clinic-based databases is the VMDB.26 This ation of data issues.
database was started in 1964 by the National Cancer Institute, Studies on Swedish insurance data have also presented statistics
includes patient data from 26 university teaching hospitals in on morbidity and mortality in cats.38,39 As with dogs, the diagno-
the United States and Canada, and contains more than 7 million ses are made by veterinarians, but further details are unavailable.
records from all species covering the full range of diagnoses, includ- The overall age-standardized mortality rate for death resulting
ing cancer. The VMDB is a widely used source of cancer surveil- from cancer in insured Swedish cats (generally <12 years of age)
lance information for companion animals; however, as discussed was 37 per 10,000 cat-years at risk. The most common types of
previously (and presented in Fig. 4.1), there is no information on cancer in the Swedish data were mammary, stomach/intestinal,
the base population in these studies and only proportional mea- and lymphoma. Siamese breeds were at increased risk of death
sures can be calculated. Given that the data sources are teaching caused by neoplasia; mammary cancer was the most common
hospitals, the patients and disease diagnoses represented are likely type,40 in agreement with an earlier study.41 Differences between
to be influenced by referral bias, resulting in estimates of disease populations and data are no doubt affected by differences in vari-
frequency that may not be typical of those seen in the general dog ous factors (e.g., spay/neuter rates and age structure of the popula-
and cat population. In an analysis using VMDB medical records it tions). Further study of neoplasia in cats is needed.
was concluded that substantial referral bias may indeed exist in the Notwithstanding the previous discussion on the relative pau-
data set, and the authors suggested the accuracy of prevalence esti- city of population-based, histologically confirmed data on the
mates measured from the VMDB could be improved by statistical incidence of cancer, there is no doubt that certain breeds are at
adjustment on the basis of geographic proximity of the patient’s high risk for cancer (e.g., the BMD, flat-coated retrievers, Boxers,
residence to the nearest university teaching hospital.27 and Scottish terriers). In the section that follows, we will pres-
The use of primary care data for investigations regarding the ent an overview of known and suspected risk factors for specific
occurrence of cancer in companion animals is less affected by cancers, including breed-specific risks, which is not strictly lim-
referral bias, but may be influenced by misclassification bias.28 ited by the level of evidence or quality of studies or data but that
To date, primary care data have been an underused resource for reflects the current state of knowledge. An important concern for
determining cancer incidence, but advances in large-scale data col- the future, however, is that without true incidence data we are
lection and management suggest primary care data may become limited in our ability to track changes in occurrence over time,
increasingly accessible for this purpose. For instance, the Vet- as proportional measures are influenced by changes both in the
erinary Companion Animal Surveillance System (VetCompass), numerator and denominator. In other words, an increase in the
which began collecting clinical data from primary practices in the popularity of a breed may lead to its apparent overrepresentation
United Kingdom in 2009, now holds data on nearly 6 million in proportional cancer measures; a change in the distribution of
animals collected from more than 500 veterinary practices across breeds may affect cancer prevalence, without any actual change in
the United Kingdom (as of August 2017).29 VetCompass began breed risk. Without incidence measures, it would be impossible to
collecting clinical data in Australia in 201630 and pilot projects are accurately evaluate the effectiveness of programs aimed at prevent-
underway in Spain, Germany, and New Zealand.31 Projects related ing or controlling disease.
to cancer outcomes in companion animals are already underway,29
and these data will likely become an increasingly important source Factors Associated with Cancer Risk
of information for research concerning cancer occurrence.
Two well-established insurance databases are from the United Observational studies are the tools of epidemiology used to iden-
Kingdom32 and from Sweden.24,32,33 A notable limitation of these tify and characterize the determinants of cancer risk. Informa-
databases is that not all cases are histologically confirmed. The tion from descriptive studies such as case series may help generate
benefits and limitations of these data have been discussed exten- hypotheses but is not adequate as a basis for evidence-based cancer
sively in the literature.34 From the UK database, using data from prevention strategies. Results from case series are also no longer
1997 through 1998, cancer incidence among 130,684 dogs at risk accepted for publication in at least one major veterinary medi-
was 747.9 per 100,000 dog-years.35 From the Swedish data, the cal journal.6 Analytic observational studies such as case-control
overall mortality rate for cancer was 50 per 10,000 dog-years-at- and cohort designs, on the other hand, are used to test research
risk (which equates to 500 per 100,000).24 Osteosarcoma (OSA) hypotheses, and when well designed, can provide valuable infor-
incidence rates were 6.1 and 5.0 dogs per 10,000 dog-years for mation for cancer prevention strategies.
males and females, respectively,36 and among females, breast can- The case-control study design is the most commonly used
cer incidence was 111 dogs per 10,000 dog-years.37 The limita- observational study design in veterinary epidemiology research
tions of the Swedish data are that deaths are mainly in dogs 10 and in cancer epidemiology research in general. This is the most
years of age or younger and it is unknown whether the diagnosis efficient study design, in terms of cost and time, when evaluat-
has been validated by histology,20 but comparison across breeds ing associations with relatively rare outcomes, such as specific
cancers. Unfortunately, as data collection is often retrospective, TABLE 4.3 Guidelines for Interpreting Clinical
many potential sources of bias must be considered. The features Relevance from Odds Ratios or Relative
of an ideally conducted study (e.g., with the least opportunity for Risk Measures
systematic bias) include the complete ascertainment of all newly
diagnosed cases with histopathologic confirmation of primary Inverse Association ≈ Positive Association ≈
tumors and a random (or matched) selection of controls from the Decreased Risk Clinical Relevance Increased Risk
same base population as the cases. In a population-based case- 1.0 Not evident 1.0
control study design,42,43 we can assume that if a control subject
had been diagnosed with the tumor of interest, that control would 0.7 to <1.0 Weak >1.0–1.5
have been a case in the study (i.e., the controls are from the same 0.5 to <0.7 Moderate >1.5–2.0
base population as the cases). The goal of the control group is to
represent the exposure experience of the base population. For this 0.3 to <0.5 Strong >2.0–3.5
reason, we are not interested in selecting the “healthiest” subjects <0.3 Very strong >3.5
as our comparison group.
In a hospital-based case-control study, both cases and con-
trols are selected from the same hospital(s). The limitation with
this design is that we cannot generalize the study results to a
clearly defined base population. However, this design is valid The RR and OR are similarly interpreted. A value greater than
and can still provide meaningful results. When using hospital- 1.0 indicates that the exposure is positively associated with dis-
based case-control study design, it is preferable to randomly ease (increases risk), whereas a value less than 1.0 indicates that
or systematically sample from the noncase population and the exposure is inversely associated with disease (decreases risk).
to not include animals that have been diagnosed with other A value of 1.0 indicates there is no association between expo-
cancers.44,45 sure and disease. The 95% confidence interval (CI) indicates the
In a prospective cohort study, a group of animals is defined precision of the RR or OR, and if the 95% CI includes 1.0, we
on the basis of exposure and followed over time to compare interpret the RR or OR to be statistically nonsignificant. It must
the incidence of disease (or other specified outcome) among be remembered, however, that statistical significance does not
the exposed and unexposed groups. The results obtained from a necessarily equate with clinical importance. For the latter, the
prospective cohort study are advantageous compared to results magnitude of the effect is also important to consider. Table 4.3
obtained from a case-control study for many reasons. One pri- shows suggested guidelines for interpretation of risk estimates.
mary advantage is that we can assume temporality, or that the When considering whether to implement preventive measures
exposure came before the disease, when associations are observed or health interventions at the population level, the following,
from prospectively collected data. Systematic errors resulting in addition to the risk estimate, are also relevant: prevalence of
from selection bias (e.g., referral bias) and differential recall bias the factor (i.e., likelihood of exposure) and the prevalence of the
(e.g., misclassification of exposure by disease status) are also not disease. These values are used to estimate the attributable risk or
major concerns when interpreting results from a well-performed risk-reduction measures.
prospective cohort study. There is a need for more longitudinal Findings from all observational studies are influenced by sys-
(as opposed to retrospective), preferably population-based stud- tematic error to some degree because there is inherent bias in the
ies to strengthen the quality of evidence in the field of veterinary methods used to select the study population, measure exposures,
oncology. and identify the outcome. The opportunity for any one study to
Regardless of the observational study design used, nondiffer- report an association that is due in part to chance is a real concern,
ential exposure misclassification will be a major concern, and the even with the use of valid study design methods and statistical
possibility and extent of misclassification should be considered analyses. Confidence in the evidence for a particular association
when interpreting observational study results. Methods by which is strengthened when it is observed repeatedly in multiple popula-
exposure misclassification can be reduced include using a precise tions and with the use of more and more rigorous study design
and accurate questionnaire that has been properly validated or methods. Meta-analysis is a technique whereby results from mul-
incorporating the use of biomarkers of exposure that can be quan- tiple, similar studies can be combined to increase the power of
tified into the study design. findings. Several examples are available from the human litera-
To estimate the magnitude of an association between an expo- ture relating to nutritional risk factors associated with pancreatic,
sure and a cancer type, the relative risk, or risk ratio (RR), and breast, and colon cancer.46–48 Unfortunately, in small animal
odds ratio (OR) are calculated from data collected from cohort oncology, studies have neither been performed nor reported con-
studies or cross-sectional and case-control studies, respectively. sistently enough nor are there an adequate number of studies
The RR is calculated as follows: conducted to support meta-analyses being conducted regularly at
this time. Notwithstanding these limitations, Table 4.4 presents
RR = Incidence among exposed subjects ÷ risk factors, including breed risks, for some of the more common
Incidence among unexposed subjects cancers in dogs and cats for which there are at least reasonable
where incidence is the number of events divided by total ani- estimations of association.
mal-time of follow-up. The OR can be used to estimate the RR The identification of modifiable risk factors for canine can-
when incidence data are not available. The OR is calculated as cers is the first step in eventually reducing incidence. Table 4.5
follows: presents analytic studies used to test hypotheses that selected
factors were either associated with an increased or decreased
OR = (Number of exposed cases ÷ number of unexposed cases) ÷ risk of canine and feline cancers. Characteristics of the study
(Number of exposed controls ÷ number of unexposed controls) design and analytic methods are highlighted as strengths
TABLE 4.4 Commonly Diagnosed Cancers and Suspected Risk Factors

Cancers Suspected Risk Factors
Common Canine Cancers
Mammary carcinoma Obesity, increasing age, high dietary fat intake, late age at spay, and some breeds (e.g., English Springer spaniel, pointer,
poodle, Boston terrier, Dachshund, German shepherd, Chihuahua)
Osteosarcoma High weight, high height, increasing age, early castration/spay, some breeds (e.g., Irish wolfhound, Saint Bernard, Great
Dane, Rottweiler, Irish setter, Doberman Pinscher, golden retriever, Labrador retriever, Leonberger)
Transitional cell carcinoma of Being neutered, exposure to phenoxy-acid containing herbicides, frequent flea dipping, increasing age, some breeds (e.g.,
the urinary bladder Scottish terrier, Beagle, Shetland sheepdog, Wirehaired fox terrier, West Highland white terrier)
Mast cell tumors Some breeds (e.g., Boxer, Rhodesian ridgeback, Vizsla, Boston terrier, Weimaraner, Chinese Shar-Pei, Bullmastiff, Dutch
pug, Labrador retriever, American Staffordshire terrier, golden retriever, English setter, English pointer), increasing age
Lymphoma ETS, exposure to chemicals containing 2,4-dichlorophenoxyacetic acid, increasing age, some breeds (e.g., Bullmastiff,
Boxer, Scottish terrier, Gordon setter, Irish wolfhound, Basset hound, golden retriever)
Common Feline Cancers
Lymphoma FeLV, FIV, ETS increasing age
Sarcoma Vaccine injection
Cutaneous squamous cell Solar irradiation

carcinoma
ETS, Environmental tobacco smoke; FeLV, feline leukemia virus; FIV, feline immunodeficiency virus.

TABLE 4.5 Selected Observational Studies of Canine and Feline Cancers by Type of Exposure
Exposure Main Findings Strengths/Limitations
ETS
Reif, 199854 Positive trend for number of packs smoked by owner and increased risk of Strengths: Evaluation of nose size as an effect modifier with
canine nasal cancer among long-nosed (dolichocephalic) dogs. biologic plausibility; collected information on potential
confounders.
Limitations: Use of controls with cancer.
Reif, 199255 Statistically nonsignificant positive association for living with ≥1 versus no Strengths: High participation rates among cases and
smokers and canine lung cancer risk. Association was stronger among controls.
short-nosed dogs (brachycephalic or mesocephalic). Limitations: Use of controls with cancer; limited statistical
power.
Marconato, Any ETS exposure was positively associated with canine lymphoma, Strengths: Population-based study design.
200943 compared with no exposure. Limitations: Use of controls with cancer; limited ETS expo-
sure information was collected.
Bertone, 200256 Strong, statistically significant association for any household ETS exposure Strengths: Statistical power to evaluate trends; cases con-
and malignant lymphoma in cats. Statistically significant trend reported firmed by biopsy; respectable response rate among the
for a stronger association with increasing years of ETS exposure. cases and controls (>65%); use of a detailed question-
naire to assess ETS and other environmental exposures.
Limitations: No clear biologic mechanism for the observed
association.
Bertone, 200357 Clinic-based case-control study had ETS exposure positively associated Strengths: Cases confirmed by biopsy; good response
with feline oral SCC. Overall, results do not support a causal relationship rates; use of a detailed questionnaire (see previous
between ETS exposure and feline SCC. entry).
Limitations: Prevalence of ETS exposure was low; limiting
the statistical power to evaluate more than two levels of
exposure.
Continued
TABLE 4.5 Selected Observational Studies of Canine and Feline Cancers by Type of Exposure—cont’d
Pesticides
Hayes, 199163 Any use of chemicals containing 2,4-dichlorophenoxyacetic acid (2,4-D) Strengths: Complete ascertainment of newly diagnosed
positively associated with canine malignant lymphoma, compared with cases; high participation rates among cases and con-
no use. Lymphoma risk increased with greater number of applications trols; collected extensive information on chemical use on
of 2,4-D–containing chemicals. lawns/yards (self-applied and commercially applied).
Limitations: One control group composed of dogs with other
cancers. note: This and other limitations were addressed
in subsequent analyses.
Glickman, Residence location within one mile of a marsh (where chemicals Strengths: Collected information on numerous sources
198965 were used for mosquito control) positively associated with of chemical exposure, including residential location to
canine TCC of the urinary bladder. Receiving flea dips more than industries, pesticide use, flee/tick treatments.
two times/year versus no use was positively associated with TCC. Limitations: 45% of control dogs had malignant neoplasia;
information was not collected on individual dog exposure
to the marsh or on specific chemicals used around the
house/yard.
Glickman, Access versus no access to phenoxy herbicide–treated lawns/yards posi- Strengths: Collected information on brand name and active
200444 tively associated with TCC of the urinary bladder among ingredients for household, lawn, and garden chemicals;
Scottish terriers. No association was observed for lawns/yards results were specific for phenoxy herbicide exposure.
not treated with phenoxy herbicides. Limitations: Limited statistical power to conduct subgroup
analyses.
Raghavan, Use of topical flea/tick products (e.g., shampoos, dips, powders, sprays, Strengths: Collected detailed information on use of flea/tick
2004111 and collars) not associated with TCC of the urinary bladder among products (e.g., type, brand, pattern of use)
Scottish terriers. Limitation: 24% of control dogs had cancer; numbers for
cases and controls were not presented by exposure
level.
Environmental Pollutants
Bettini, 2010112 Pulmonary anthracosis (high versus none) positively associated with Strengths: Histologic confirmation of primary diagnosis of
canine lung cancer risk. lung cancer; exposure assessment determined by histo-
logic scoring of anthracosis; strong biologic mechanism
supporting the a priori hypothesis.
Limitations: Small number of cases limited the statistical
analyses.
Marconato, Living in geographic areas exposed to toxic waste positively associated Strengths: Population-based study design, histologic
200943 with canine cancer risk (all tumors and lymphoma), compared with confirmation of cases; odds ratios were adjusted for age,
living in an unexposed area. No associations observed for canine sex, and breed.
mast cell tumors, canine mammary cancer, or feline cancers. Limitations: Same eligibility criterion (i.e., living at same
address for 2 years before enrollment) was not applied
to controls.
Gavazza, Living in an industrial neighborhood was positively associated with Strengths: Histopathologic or cytologic confirmation of
200166 canine lymphoma risk, compared with living in any other neighborhood. cases; information was collected on potential confound-
Use or storage of paints and solvents was positively associated ers.
with lymphoma risk, compared with no use of chemicals. Limitations: Very low prevalence of exposed cases and
controls; only univariate analyses were conducted.
Bukowski, Cumulative kerosene or coal heat exposure was positively associated Strengths: High participation rate; covariate information
1998113 with sinonasal cancer risk. was compared between respondents and nonrespon-
dents; histopathologic confirmation of cases.
Limitations: Use of controls with cancer.
Endogenous/Exogenous Sex Hormones
Sonnenschein, Earlier age at spaying was inversely associated with canine mammary Strengths: Cases were limited to mammary carcinoma or
199175 cancer. Trend of decreasing risk was observed for younger age at adenocarcinoma.
spaying. Limitations: Controls may not be representative of the base
population.
Ru, 199876 Neutered dogs, regardless of gender, had a greater risk of Strengths: Histologic or radiologic confirmation; large study
osteosarcoma, compared with intact dogs. size; collected information on potential confounders.
Limitations: Medical conditions of the controls were not
clearly described.
Glickman, Neutered status versus intact was a risk factor for TCC of the urinary Strengths: Cases were histologically confirmed.
200444 bladder among Scottish terriers. Limitations: Small study size did not permit for analyses by
age at neutering.
Dias Pereira, No overall association was observed for COMT genotype and canine mam- Strengths: Strong biologic rationale for research hypothesis;
2008114 mary cancer risk. Older age at mammary cancer diagnosis cases were histologically confirmed.
was observed by COMT genotype. Limitations: Very small numbers in subgroup analyses;
selection methods were not provided; information was
not collected on potential confounders (e.g., hormone-
related exposures).
Cooley, 200272 Neutering before 1 year of age increased risk of canine osteosarcoma Strengths: Radiographic or histologic confirmation of cases;
among Rottweilers, regardless of gender. Incidence rates decreased retrospective cohort study design.
with later age at neutering. Reproductive factors (number of litters, Limitations: Low participation rate.
number of live births, age at first pregnancy) were not associated
with osteosarcoma among female dogs.
Stovring, MPA use was positively associated with canine mammary cancer. Strengths: Population-based study design; histologic
199742 confirmation of cases.
Limitations: Information was not collected on details of MPA
use (e.g., frequency, dose, age at first use).
Teske, 200277 Castration was positively associated with canine prostate cancer Strengths: Strong biologic plausibility.
risk, compared with intact status. Limitations: Only cytology was used to make cancer
diagnosis.
Bryan, 200778 Neutered versus intact status was a risk factor for the following Strengths: Histopathologic confirmation of cases; included
canine cancers: TCC of the urinary bladder, prostate carcinoma, analyses by histologic subtype.
prostate adenocarcinoma, and TCC of the prostate. Limitations: Statistically nonsignificant measures were not
presented.
Misdorp, Ovariectomy was inversely associated with feline mammary cancer Strengths: Histologic confirmation of cases; collection of
1991115 risk. Regular administration of progestogens increased risk. detailed exogenous progestogens (frequency, brand,
No association was observed for irregular progestogen type); large study size.
administration or for parity. Limitations: Cases and controls were selected over different
time periods.
Overley, Intact versus neutered status was a risk factor for feline mammary Strengths: Histologic confirmation of cases; large study
2005116 cancer. Cats spayed before 1 year of age were at lower risk of size.
mammary cancer than those spayed after 6 months of age. Limitations: Univariate analyses were performed, although
There was no risk benefit in cats spayed after 2 years of age. detailed information was collected on exogenous hor-
mone use, parity, and number of litters; large amount of
missing data because of veterinarian nonresponse.
Torres de la Male golden retrievers were more likely to develop lymphoma Strengths: Large study size.
Riva, 201379 when neutered <1 year of age compared with intact dogs; Limitations: Cases were not histologically confirmed, only
Female golden retrievers were more likely to develop one breed of dog included.
hemangiosarcoma when neutered at >1 year of age
Continued
Zink, 201480 Neutering increased risk of developing mast cell tumors, Strengths: Large study size, multivariable analyses were
hemangiosarcoma, lymphoma, and all other cancers in Vizslas performed.
compared with intact dogs. Neutering at <6 months of age was Limitations: Cases were not histologically confirmed, only
not associated with increased risk of developing cancer compared one breed of dog included.
with neutering at older ages.
Hart, 201481 Neutering male golden retrievers and male and female Labrador Strengths: Large study size, Cox proportional hazard
retrievers or at any age had no significant effect on cancer models were used.
occurrence compared with intact dogs; Neutering female golden Limitations: Population limited to dogs visiting a single
retrievers at any point beyond 6 months of age was associated Veterinary Teaching Hospital, only two breeds of dogs
with increased risk of developing any cancer with the exception of included, age categories at time of neutering were not
lymphoma which was associated with an increased risk if consistent among analyses.
neutering occurred before 6 months of age.
Hart, 2016117 Cancer occurrences in this population of German shepherd dogs were Strengths: Large study size, Cox proportional hazard
rare and there were no associations reported between neutering models were used.
at any age and cancer outcomes. Limitations: Population limited to dogs visiting a single
veterinary teaching hospital, only one breed of dog
included, potential data quality issues.
Diet
Perez Alenza, Higher intake of red meat (as percentage of total calories) was Strength: Used biomarkers of exposure; multivariable
1998118 positively associated with canine mammary carcinoma risk. analyses included covariates for body conformation.
No differences were observed for intake of fruits and vegetables, Limitations: Use of a retrospective study design is not
or biomarker levels of selenium, retinol, or individual fatty acids. recommended for evaluating biomarkers of exposure
and cancer risk because of possible disease and/or
treatment effects on the biomarker measurement.
Sonnenschein, Higher intake of fat and table food (as percentage of total calories) Strengths: Cases and controls were matched by age, spay
199175 was inversely associated with canine mammary carcinoma. No status, and breed size, thus reducing the opportunity for
associations were observed with protein or carbohydrates intake. confounding by these factors; the dietary assessment
tool was validated using a 7-day food record,
Limitations: Study size was too small to evaluate diet-
cancer associations in subgroups.
Raghavan, Vegetable intake (≥3 versus 0 times/week) was inversely associated Strengths: Histopathology and/or cytology confirmation; use
2005119 with TCC of the urinary bladder in Scottish terriers. A trend was of a comprehensive dietary questionnaire; multivariable
observed with greater servings of vegetables per week and analyses.
decreased risk of TCC. No association was observed for weekly Limitations: Used a volunteer population; 61% of the cases
vitamin supplement intake, compared with no intake. were deceased at the start of the study; 24% of control
dogs had neoplastic diseases.
Body Size
Perez Alenza, Obese body condition at 1 year of age and at 1 year before diagnosis Strengths: Objective measurements of weight and height
1998118 was positively associated with canine mammary cancer, compared were collected at presentation; body conformation was
with normal or underweight body condition at the same time determined by a clinician at presentation.
points. Limitations: Height, weight, and body conformation at 1
year of age and 1 year before diagnosis were based on
owners’ reports.
Sonnenschein, Spayed dogs that were thin at 9–12 months of age had a lower Strengths: Cases were limited to mammary carcinoma or
199175 risk of mammary cancer. Intact dogs that were not overweight adenocarcinoma; designed to assess a timely hypothesis
in adulthood had a lower risk of mammary cancer. that early life factors are related to mammary cancer risk.
Limitations: Controls may not be representative of the
base population; subgroup analyses did not have ample
statistical power to calculate precise measures.
Weeth, 200745 BCSs ≥6 were inversely associated with canine cancer risk (all Strengths: Very large study size; 9-point BCS determined
cancers, sarcomas, and carcinomas), compared with BCSs of by physical examination; analyses were conducted by
4–6. BCSs <3 were inversely associated with canine sarcoma risk, cancer type (sarcoma, carcinoma, round cell tumors)
compared with scores of 4–6. Limitations: Selection of cases and controls depended
on availability of BCS in medical records; the inverse
associations with higher BCS may be a result of reverse
causation.
Ru, 199876 Height (>61 versus <35.5 cm) and weight (>45 versus <23 kg) Strengths: Histologic or radiologic confirmation; large study
were positively associated with canine osteosarcoma risk, after size; collected information on potential confounders.
adjusting for age and standard weight and height, respectively. Limitations: Medical conditions of the controls were not
Longer length of hind limbs and front limbs was positively clearly described; a proxy measure for height was used;
associated with canine osteosarcoma risk, compared with there was a large percentage (22.5%) with missing
shortest length. weight information.
Glickman, Greater weight was positively associated with TCC urinary bladder risk Strengths: Cases were histologically confirmed.
200444 in Scottish terriers, comparing third versus first tertile. Greater weight- Limitations: Weight and height information was based on
to-height ratio was also a risk factor for TCC. owners’ reports.
Vaccines/Injection Site
Kass, 200391 Cats with sarcomas at a vaccine injection site (n = 662) were Strengths: Histologic confirmation of cases and controls;
compared with cats with basal cell tumors or noninjection site collected extensive vaccine information (date of injec-
sarcomas (n = 473). Univariate analyses showed no difference tion, manufacturer, type, brand, site of injection).
in the vaccine type (FVRCP, rabies, FeLV) between cases and Limitations: Cases were identified on a volunteer basis from
controls. There were no differences between time at vaccination participating clinics; heterogeneous sarcoma case group.
and tumor diagnosis between the two groups.
Kass, 199390 In 345 cats diagnosed with fibrosarcoma, 53.6% had tumors at the Strengths: Biopsy-confirmed diagnoses; vaccination history
vaccine injection site. The time from FeLV vaccination to tumor diagno- was validated by veterinarian; collected vaccination
sis was significantly shorter among cats that had tumors at the cervical/ details allowing for analyses by type of vaccine, time
interscapular region than cats that had tumors at noninjection sites. since vaccination and location of injection site.
Limitations: Differential missing data by exposure status.
FeLV/FIV
Hutson, Among 1160 cats identified from an oncology referral and a general Strengths: Descriptive information of neoplasia among
1991120 practice clinic, 2.5% were FIV positive. Of the FIV-positive cats, 62% FIV-positive cats.
were diagnosed with neoplasia (myeloproliferative disease, lymphoma, Limitations: No evident population base; only count data
and SCC). were presented.
Gabor, 2001121 Among 101 cats with lymphosarcoma, 50% were FIV positive. These cats Strengths: Histopathologic confirmation of cases; FIV anti-
were more likely to be male domestic crossbreeds. bodies were determined using Western blot.
Limitations: Convenience study population was used.
Shelton, 199096 Coinfection with FIV and FeLV was present in 14.4% of 353 cats collected Strengths: FIV antibodies were determined using ELISA and
in several US cities. FIV and FeLV infection were strongly associated Western blot.
with risk of leukemia or lymphoma. A very imprecise positive associa- Limitations: Base population and subject recruitment meth-
tion was also reported for coinfection and leukemia/lymphoma risk. ods were not well defined; low prevalence of coinfection
among controls limited statistical power.
Solar Irradiation
Dorn, 1971122 Among white cats, the observed incidence of SCC of the skin was greater Strengths: Population-based study population.
than the expected incidence (p < 0.001). For SCC of the mouth–pharynx, Limitations: Amount of sun exposure was not quantified;
white cats had no difference between observed and expected incidence. the number of cats with SCC of the mouth–pharynx was
small (n = 29).
BCS, Body condition score; COMT, catechol-O-methyltransferase; ELISA, enzyme-linked immunosorbent assay; ETS, environmental tobacco smoke; FeLV, feline leukemia virus; FIV, feline immunodefi-
ciency virus; FVRCP, feline viral rhinotracheitis-calicivirus-panleukopenia; MPA, medroxyprogesterone acetate; SCC, squamous cell carcinoma; TCC, transitional cell carcinoma.

and weaknesses. Studies with the strongest level of evidence Highlighted Findings from Observational
included several characteristics related to study design (e.g., Studies
hypothesis-driven, population-based, large study size, vali-
dated exposure assessment) and results (e.g., a precise measure In this section, we discuss risk factors for which there is relatively
of association, a modest-to-strong magnitude of association, strong evidence, those that relate to key issues in animal or human
statistically significant measure of association, statistically oncology, and those for which important controversies need to be
significant trend between exposure level and magnitude of addressed by further research. These categories coincide with those
association). shown in Table 4.5.
Environmental Exposures Exposure misclassification is a primary limitation of using geo-

The identification of environmental exposures that are related to graphic proximity as a marker of exposure to an industrial or waste
canine cancer risk have a broad public health interest, given the site because it may or may not be a good proxy for individual-
shared environments of companion animals and their owners, level exposure. For example, a validation study would need to be
and they have similar etiology of some cancers.49 In a compre- conducted that provides information on whether dogs that live
hensive review, a historic perspective is provided on how stud- close to an industrial site are necessarily exposed at higher levels to
ies in pet populations have informed human health with respect environmental hazards compared with dogs that live further from
to the specific exposures of air pollution, environmental tobacco the site. Misclassification of exposure that does not differ by dis-
smoke (ETS), and pesticides.50 The shared etiologic characteristics ease status (e.g., nondifferential misclassification) typically results
of cancers such as lymphoma, OSA, and mammary cancer also in an underestimate of the exposure–cancer association, although
support the utility of looking to both pet and human populations there are situations when the observed association results in an
to investigate environmental–cancer associations.51 overestimate of the true association.70,71
There is experimental evidence for an underlying biologic
mechanism for the compounds of cigarette smoke to have a Hormones and Neuter Status
causal relationship with canine carcinogenesis.52,53 There are few Hormones may act as either growth factors or inhibitors, depend-
observational studies that were designed to specifically evaluate ing on the sex of the dog and the tissue type.72–74 For some
associations between ETS exposure and canine cancer risk.54,55 cancers, such as breast cancer, less exposure to sex hormones is
There is support for a positive association (3.4-fold increased risk) protective; whereas for others, such as OSA, lymphoma, and pros-
between ETS and lymphoma43 and sinonasal cancers,54 but not tate cancer, less exposure has been reported to increase risk.72,75–82
for lung cancer.55 In a clinic-based case-control study, household Neuter/spay status and age at neuter/spay are the most commonly
ETS exposure was strongly associated with feline lymphoma.56 used measures of endogenous hormone exposure. In spite of some
The OR for any exposure, compared with no exposure, was 2.4, newer studies, there is limited evidence that the age of the dog at
and statistically significant trends were reported for more years of the time of neuter/spay can have an effect on the risk of develop-
ETS exposure, more smokers in the household, and number of ing cancer in certain breeds of dogs.79,81 Given the widespread rec-
cigarettes smoked per day in the household. In contrast, there is ommendation for early spay/neuter, especially in North America,
only weak observational evidence for ETS as a risk factor for oral this is a topic in need of further study.
squamous cell carcinoma in cats.57 In summary, avoiding ETS In a mammary cancer case-control study, there was clear evi-
exposure may reduce the risk of lymphoma in cats and dogs and dence that spayed dogs were at lower risk of mammary cancer.75
the risk of sinonasal cancers in dogs. In particular, the earlier age at which dogs were spayed the lower
Pesticides are a heterogeneous group of chemicals, some of their mammary cancer risk compared with dogs that were not
which are known human and canine carcinogens.58–60 Dogs may spayed. This finding has been supported by other observational
be exposed to pesticides in the home, in the yard/garden, and studies of spay status and mammary cancer risk.83
on application of flea and tick treatments. The most consistent Contrary to human epidemiologic and experimental evi-
observational evidence for pesticide exposure as a cancer risk fac- dence,84,85 exposure to sex hormones such as androgens may be
tor is for phenoxy acid–containing herbicides and lymphoma risk, protective for canine prostate cancer.82 From two case-control
both in humans and dogs.61 These data, however, have not been studies using large veterinary teaching hospital databases, neu-
deemed strong enough to establish causality.62 In a large case- tered dogs had a 2.8- and 3.4-fold increased risk of prostate cancer
control study (n = 491 cases and n = 945 controls), any use of compared with intact dogs.77,78 The apparent opposite associa-
pesticides that contained dichlorophenoxyacetic acid (2,4-D) was tions between hormone exposure and prostate cancer risk in men
associated with a 30% increased risk of lymphoma compared with and dogs are likely a result of the higher rate of androgen-indepen-
no use.63,64 Although modest, the positive association also dem- dent tumors in dogs than in men.86,87
onstrated a dose-dependent effect in which more frequent use of Neuter status is also a risk factor for OSA and transitional cell
2,4-D pesticides resulted in a stronger positive association with carcinoma of the urinary bladder,44,72,76,78 regardless of sex.72,76
lymphoma risk (p for trend <0.02). Additional support for 2,4-D Cooley et al conducted a retrospective cohort study in 1999
and canine bladder cancer risk is from the result of a small case- among 683 Rottweilers and used a self-administered question-
control study in Scottish terriers.42 naire to test the hypothesis that neuter/spay status was related to
Residential proximity to environmental hazard–containing the development of OSA.72 The owners were identified through
sites has been used to estimate chemical exposure and canine can- eight national Rottweiler breed specialty clubs and had a purebred
cer risk in several observational studies.43,65,66 A 2.4-fold increase Rottweiler that was alive on January 1, 1995. The participation
in risk of lymphoma was observed among dogs living in the cit- rate ([number of participants] ÷ [total number of invited owners]
ies containing illegal waste sites compared with dogs living in × 100) was 49%. This low participation rate suggests that selection
other cities.43 No association was observed with mast cell tumors bias may have influenced the results of this study. In other words,
(MCTs) or breast cancer. Mortality caused by cancer is also higher the participants of the study are likely to have systematic differ-
among human populations living near the same waste sites com- ences compared with those who did not participate. However, a
pared with the general population.67 Chemical mixtures that have strength of this study is the ability to calculate incidence because
been identified at hazard waste landfills include organic solvents, the total number of dog-months of observation were estimated
polychlorinated biphenyls, and heavy metals. These can reach retrospectively among dogs that were neutered/spayed and those
human and pet populations through contaminated air, water, that were not. During a total of 71,004 dog-months of obser-
and/or soil,68 and have been causally related to adverse human vation, there were 86 cases of OSA. Collectively, the findings of
health effects, including childhood lymphoma.69 The biologic a positive association between neuter/spay status and OSA from
plausibility and the observational findings from Marconato et al43 both case-control and cohort studies, and the biologic plausibility
both help strengthen the evidence that living near the waste sites of the association, provide strong evidence that neutering/spaying
increases risk of canine lymphoma. dogs, regardless of sex, increases risk of OSA.
A number of reports have linked the age of dogs at the time development,95 and coinfection with FIV and FeLV may have
of neuter/spay to increased risk of developing hemangiosarcoma, synergistic effects on feline neoplasia risk.96
lymphoma, and MCTs in select breeds of dogs.79–81 Using a retro-
spective analysis of medical records including dogs younger than Diagnosis and Screening
8 years of age from the Veterinary Medical Teaching Hospital at
the University of California, Davis, an increased risk of develop- As mentioned previously, numerous reporting guidelines have
ing hemangiosarcoma and MCTs was identified in female golden been produced for the human medical literature (e.g., http://meth-
retrievers neutered at 12 months of age or older compared with ods.cochrane.org/mecir). One of these describes an approach to
intact female golden retrievers.79 The same analysis revealed an complete and accurate reporting of studies of diagnostic accuracy
increased risk of developing lymphoma in male golden retriev- (Standards for Reporting of Diagnostic Accuracy [STARD]).97,98
ers neutered before 12 months of age compared with intact male The application of this and another instrument in the veterinary
golden retrievers.79 A subsequent report that used a larger data set field has been discussed.99 Unfortunately, relatively few diagnos-
from the same location reported similar risks in golden retrievers, tic interventions in veterinary medicine have been examined or
but showed there was no difference in risk of cancer occurrence evaluated as fully, in terms of reliability, accuracy, efficacy, and
in Labrador retrievers associated with neuter status.81 Research effectiveness, as is needed to support EBM practices. Guidelines
from the same institution that examined cancer-related mortality for prognostic studies in veterinary oncology have also been
in golden retrievers found that increasing age increased the odds reported.100
of cancer-related mortality regardless of neuter status.88 Neutered It may also be appropriate to use clinical trial methodology to
female golden retrievers were found to live statistically signifi- evaluate the outcomes of diagnostic tests, for example, whether
cantly longer than intact females and it was this increase in lon- the animal is better off for having had the test performed.101 In
gevity that affected the rate of cancer occurrence rather than the addition, recommendations about diagnostic tests and screening
hormonal differences caused by neutering. This conflicting result programs may have both positive and negative effects beyond any
may be because of the fact that dogs of any age were included individual, on populations of animals and owners. Although new,
in this study whereas the previous reports limited results to dogs sophisticated diagnostic tests used in humans are being evaluated
aged 8 years or younger.79,81,88 Results of a survey of Vizsla owners for use in companion animals, it is important to remember that
indicate that neutering at any age was associated with increased beyond the benefit in a specific case, efficacy and effectiveness
risk of developing lymphoma and MCTs regardless of sex.80 of tests should consider the broadest aspects of cost–benefit. In
Increased risk associated with development of hemangiosarcoma human oncology, there has been much discussion about the prob-
was also indicated in both sexes regardless of the timing of neuter- lems inherent in certain widely applied screening processes and
ing with the exception of males neutered at less than 12 months of the consequences of false positives and negatives (e.g., prostate-
age, which had no difference in risk compared with intact males.80 specific antigen test for prostate cancer102). Owing to space con-
Prospective research conducted to examine associations between straints, we cannot expand on this crucially important area of
age at neutering and cancer outcomes is needed in a wider variety cancer epidemiology.
of dog breeds before current recommendations regarding spay/
neuter can be modified. Therapeutic Interventions
Risk Factors in Cats A review published in the Journal of Veterinary Internal Medicine
In cats, the epidemiology of injection-site sarcomas has been well highlighted that the quality of reporting of oncology studies in
studied. A review from 2011 provides information on the current dogs and cats has not improved appreciably over time and that
epidemiology, etiology, and clinical knowledge of feline injection- quality of reporting is highly correlated with the rate of positive
site sarcomas (FISSs).89 Kass et al conducted one of the first epi- outcomes (i.e., well-described studies are more likely to report
demiologic studies investigating the hypothesis that vaccinations positive effects of a treatment).103 This may also be exacerbated
were related to feline fibrosarcoma risk in the early 1990s.90 A by the fact that the profession increasingly depends on corporate
main finding of this study was the shorter time interval from vac- contracts for funding of research, and, in addition to this having a
cination to FISS compared with the interval from vaccination to major effects on which treatments are investigated, there may also
non-FISSs. This finding was not supported by a second, larger be underreporting of studies in which either beneficial effects were
case-control study.91 Although there is no doubt that the phenom- not seen or where there were deleterious side effects. As mentioned
enon of FISS exists, the administration of an injection itself is not previously, efforts are underway to produce reporting guidelines
sufficient to cause development of an FISS. The component causes for companion animal intervention studies. However, guidelines
(e.g., the nature of vaccines and adjuvants in the injected mate- for appropriate study design for clinical trials have been widely
rial and the role of the resulting inflammatory reaction), in addi- available for many decades, and the need for appropriate trials
tion to the physical injection that leads to the development of an in oncology has been specifically advocated (see Chapter 18).104
FISS, are not well characterized. Further epidemiologic research Longer-term analyses of survival after diagnosis and treatment,
designed with due consideration of the challenging methodologi- including both outcome and cost–benefit analysis, are needed to
cal issues is needed to identify the various factors associated with provide the information that owners and veterinarians need to
FISSs.92 choose the best options, with due consideration of quality-of-life
Cats infected with the feline immunodeficiency virus (FIV), issues.
the feline analog to the human immunodeficiency virus (HIV), Although there are good examples of randomized, controlled,
are at increased risk of certain cancers.93 FIV is a lentivirus typi- blinded trials in veterinary oncology,105,106 essentially all studies have
cally transmitted by biting.94 Lymphomas, particularly those some limitations in terms of either quality (validity) or relevance
of B-cell origin, are the most commonly diagnosed neoplasia (extrapolation to other situations). For example, because many trials
among FIV-infected cats. Persistent feline leukemia virus (FeLV) are performed on clients at specialty practices or veterinary teach-
infection is also known to have a strong role in feline neoplasia ing hospitals, animals have passed through numerous filters to be
available for the study (e.g., referral, have a willing and capable 6. Hinchcliff KW, DiBartola SP: Quality matters: publishing in the
owner, live long enough to have a confirmed diagnosis). Although era of CONSORT, REFLECT, and EBM, J Vet Intern Med 24(8–
this may improve the validity of the study (e.g., by increasing com- 9), 2010.
pliance and reducing loss to follow-up), it reduces the relevance to, 7. Sargeant JM, O’Connor AM, Dohoo IR, et al.: Methods and
processes of developing the strengthening the reporting of obser-
for example, primary practice. Therefore clinicians must be able to
vational studies in epidemiology – veterinary (STROBE-Vet) state-
apply the rules of evidence to determine both the quality and rel- ment, J Vet Intern Med 30:1887–1895, 2016.
evance of information for their specific situation and patients. Other 8. Dorn CR, Taylor DO, Frye FL, et al.: Survey of animal neoplasms
authors in this text will present current information on treatments in Alameda and Contra Costa Counties, California. I. Methodol-
for cancer, and a further review of the literature is beyond the scope ogy and description of cases, J Natl Cancer Inst 40:295–305, 1968.
of this chapter. Hopefully, the quality of the veterinary literature will 9. Dorn CR, Taylor DO, Schneider R, et al.: Survey of animal neo-
continue to improve over time, and the application of appropriate plasms in Alameda and Contra Costa Counties, California. II.
reporting guidelines and production of evidence-based reviews will Cancer morbidity in dogs and cats from Alameda County, J Natl
assist clinicians to interpret and apply published information. Cancer Inst 40:307–318, 1968.
10. MacVean DW, Monlux AW, Anderson Jr PS, et al.: Frequency
of canine and feline tumors in a defined population, Vet Pathol
Knowledge Gaps and Future Directions 15:700–715, 1978.
11. Moe L, Gamlem H, Dahl K, et al.: Canine neoplasia–population-
Even though systematic reviews and meta-analyses may not be based incidence of vascular tumours, APMIS Suppl 125:63–68,
currently possible, evidence-based reviews of the existing oncol- 2008.
ogy literature in dogs and cats are needed to further elucidate what 12. Nodtvedt A, Gamlem H, Gunnes G, et al.: Breed differences in the
we know about breed risks, other risk factors, appropriate use of proportional morbidity of testicular tumours and distribution of
diagnostic and screening aids, therapies, prognoses, and so forth, histopathologic types in a population-based canine cancer registry,
and to identify the most crucial gaps in our knowledge. Appro- Vet Comp Oncol 9:45–54, 2011.
priate assessment of existing oncology prevention and treatment 13. Merlo DF, Rossi L, Pellegrino C, et al.: Cancer incidence in pet
strategies is also needed. dogs: findings of the Animal Tumor Registry of Genoa, Italy, J Vet
Intern Med 22:976–984, 2008.
With increasingly available genomic information, our under-
14. Vascellari M, Baioni E, Ru G, et al.: Animal tumour registry of two
standing of breeds and breed risk may change.107,108 It is espe- provinces in northern Italy: incidence of spontaneous tumours in
cially important to recognize the value of studying populations dogs and cats, BMC Vet Res 5(39), 2009.
from different areas or countries. There are important differences 15. Grüntzig K, Graf R, Hässig M, et al.: The Swiss canine cancer reg-
and similarities in genetics (across and within breeds), environ- istry: a retrospective study on the occurrence of tumours in dogs in
ments, diets, and activities that will inform cancer etiology and Switzerland from 1955 to 2008, J Comp Path 152:161–171, 2015.
management. Such complex relationships will be fully understood 16. Bronden LB, Nielsen SS, Toft N, et al.: Data from the Danish vet-
only by a multidisciplinary approach using various methodolo- erinary cancer registry on the occurrence and distribution of neo-
gies and study designs. These should include more population- plasms in dogs in Denmark, Vet Rec 166:586–590, 2010.
based, longitudinal observational studies and outcomes-based 17. Baioni E, Scanziani E, Vincenti MC, et al.: Estimating canine can-
cer incidence: findings from a population-based tumour registry in
approaches.56,57,109 Although these may not yield results for many
northwestern Italy, BMC Vet Res 13:203, 2017.
years, relying solely on traditional approaches (case-control studies 18. Moe L, Bredal WP, Glattre E: Census of dogs in Norway, Oslo,
and clinical trials of invasive or risky treatments) is not an effec- 2001, Norwegian School of Veterinary Science.
tive strategy to reduce the population burden of cancer. In addi- 19. Priester WA, McKay FW: The occurrence of tumors in domestic
tion and beyond the scope of this chapter, there are important animals, Natl Cancer Inst Monogr 54:1–210, 1980.
complex issues of human–animal interactions, in general and spe- 20. Bronden LB, Flagstad A, Kristensen AT: Veterinary cancer regis-
cifically related to the field of oncology. Thus there is a need for tries in companion animal cancer: a review, Vet Comp Oncol 5:133–
an increased understanding of the social and emotional factors 144, 2007.
underpinning many aspects of this diverse field.110 21. Nødtvedt A, Berke O, Bonnett BN, et al.: Current status of canine
cancer registration: report from an international workshop, Vet
Comp Oncol 10:95–101, 2011.
References 22. Craig LE: Cause of death in dogs according to breed: a nec-
ropsy survey of five breeds, J Am Anim Hosp Assoc 37:
1. Sargeant JM, Thompson A, Valcour R, et al.: Quality of reporting 438–443, 2001.
of clinical trials of dogs and cats and associations with treatment 23. Fleming JM, Creevy KE, Promislow DE: Mortality in North
effects, J Vet Intern Med 24:44–50, 2010. American dogs from 1984 to 2004: an investigation into age-, size-,
2. Sargeant JM, O’Connor AM, Gardner IA, et al.: The REFLECT and breed-related causes of death, J Vet Intern Med 25:187–198,
statement: reporting guidelines for randomized controlled trials in 2011.
livestock and food safety, J Food Prot 73:579–603, 2010. 24. Bonnett BN, Egenvall A, Hedhammar A, et al.: Mortality in over
3. O’Connor AM, Sargeant JM, Gardner IA, et al.: The REFLECT 350,000 insured Swedish dogs from 1995-2000: I. breed-, gender-,
statement: methods and processes of creating reporting guidelines age- and cause-specific rates, Acta Vet Scand 46:105–120, 2005.
for randomized controlled trials for livestock and food safety, J Food 25. Guy MK, Page RL, Jensen WA, et al.: The golden retriever lifetime
Prot 73:132–139, 2010. study: establishing an observational cohort study with translational
4. Rishniw M, Pion PD, Herndon WE, et al.: Improving reporting of relevance for human health, Philos Trans R Soc Lond B Biol Sci
clinical trials in veterinary medicine, J Vet Intern Med 24:799–800, 370:20140230, 2015.
2010. 26. Veterinary Medical Database (VMDB) www.vmdb.org. accessed:
5. O’Connor AM, Sargeant JM, Gardner IA, et al.: The REFLECT 08/08/2017.
statement: methods and processes of creating reporting guidelines 27. Bartlett PC, Van Buren JW, Neterer M, et al.: Disease surveillance
for randomized controlled trials for livestock and food safety, J Food and referral bias in the veterinary medical database, Prev Vet Med
Prot 73:132–139, 2010. 94:264–271, 2010.
28. BSAVA: SAVSNET The next stage, BSAVA Companion 4–5, 2013. 52. Hernandez JA, Anderson AE, Holmes WL, et al.: Pulmonary
29. Veterinary Companion Animal Surveillance System (VetCompass) parenchymal defects in dogs following prolonged cigarette smoke
www.rvc.ac.uk/vetcompass. accessed; 08/13/2017. exposure, Am Rev Respir Dis 93:78–83, 1966.
30. Latter M: VetCompass Australia – a leap forward for companion 53. Cross FT, Palmer RF, Filipy RE, et al.: Carcinogenic effects of
animals, Aust Vet J 94:N14, 2016. radon daughters, uranium ore dust and cigarette smoke in beagle
31. O’Neill DG, Church DB, McGreevy PD, et al.: Approaches to dogs, Health Phys 42:33–52, 1982.
canine health surveillance, Canine Genet Epidemiol 1:1–13, 2014. 54. Reif JS, Bruns C, Lower KS: Cancer of the nasal cavity and para-
32. Egenvall A, Hedhammar A, Bonnett BN, et al.: Survey of the nasal sinuses and exposure to environmental tobacco smoke in pet
Swedish dog population: age, gender, breed, location and enroll- dogs, Am J Epidemiol 147:488–492, 1998.
ment in animal insurance, Acta Vet Scand 40:231–240, 1999. 55. Reif JS, Dunn K, Ogilvie GK, et al.: Passive smoking and canine
33. Egenvall A, Bonnett BN, Hedhammar A, et al.: Mortality in over lung cancer risk, Am J Epidemiol 135:234–239, 1992.
350,000 insured Swedish dogs from 1995-2000: II. breed-specific 56. Bertone ER, Snyder LA, Moore AS: Environmental tobacco smoke
age and survival patterns and relative risk for causes of death, Acta and risk of malignant lymphoma in pet cats, Am J Epidemiol
Vet Scand 46:121–136, 2005. 156:268–273, 2002.
34. Egenvall A, Nodtvedt A, Penell J, et al.: Insurance data for research 57. Bertone ER, Snyder LA, Moore AS: Environmental and lifestyle
in companion animals: benefits and limitations, Acta Vet Scand risk factors for oral squamous cell carcinoma in domestic cats, J Vet
51:42, 2009. Intern Med 17:557–562, 2003.
35. Dobson JM, Samuel S, Milstein H, et al.: Canine neoplasia in the 58. Dich J, Zahm SH, Hanberg A, et al.: Pesticides and cancer, Cancer
UK: estimates of incidence rates from a population of insured dogs, Causes Control 8:420–443, 1997.
J Small Anim Pract 43:240–246, 2002. 59. Hardell L: Pesticides, soft-tissue sarcoma and non-Hodgkin lym-
36. Egenvall A, Nodtvedt A, von Euler H: Bone tumors in a population phoma—historical aspects on the precautionary principle in cancer
of 400 000 insured Swedish dogs up to 10 y of age: incidence and prevention, Acta Oncol 47:347–354, 2008.
survival, Can J Vet Res 71:292–299, 2007. 60. Andrade FH, Figueiroa FC, Bersano PR, et al.: Malignant mam-
37. Egenvall A, Bonnett BN, Ohagen P, et al.: Incidence of and sur- mary tumor in female dogs: environmental contaminants, Diagn
vival after mammary tumors in a population of over 80,000 insured Pathol 5(45), 2010.
female dogs in Sweden from 1995 to 2002, Prev Vet Med 69:109– 61. Zahm SH, Blair A: Pesticides and non-Hodgkin’s lymphoma, Can-
127, 2005. cer Res 52:5485s–5488s, 1992.
38. Egenvall A, Nodtvedt A, Haggstrom J, et al.: Mortality of life- 62. Garabrant DH, Philbert MA: Review of 2,4-dichlorophenoxy-
insured Swedish cats during 1999-2006: age, breed, sex, and diag- acetic acid (2,4-D) epidemiology and toxicology, Crit Rev Toxicol
nosis, J Vet Intern Med 23:1175–1183, 2009. 32:233–257, 2002.
39. Egenvall A, Bonnett BN, Haggstrom J, et al.: Morbidity of insured 63. Hayes HM, Tarone RE, Cantor KP, et al.: Case-control study of
Swedish cats during 1999-2006 by age, breed, sex, and diagnosis, canine malignant lymphoma: positive association with dog owner’s
J Feline Med Surg 12:948–959, 2010. use of 2,4-dichlorophenoxyacetic acid herbicides, J Natl Cancer Inst
40. Rissetto K, Villamil JA, Selting KA, et al.: Recent trends in feline 83:1226–1231, 1991.
intestinal neoplasia: an epidemiologic study of 1,129 cases in the 64. Hayes HM, Tarone RE, Cantor KP: On the association
veterinary medical database from 1964 to 2004, J Am Anim Hosp between canine malignant lymphoma and opportunity for
Assoc 47:28–36, 2011. exposure to 2,4-dichlorophenoxyacetic acid, Environ Res 70:
41. Hayes HM, Milne KL, Mandell CP: Epidemiological features of 119–125, 1995.
feline mammary carcinoma, Vet Rec 108:476–479, 1981. 65. Glickman LT, Schofer FS, McKee LJ, et al.: Epidemiologic study
42. Stovring M, Moe L, Glattre E: A population-based case-control of insecticide exposures, obesity, and risk of bladder cancer in
study of canine mammary tumours and clinical use of medroxy- household dogs, J Toxicol Environ Health 28:407–414, 1989.
progesterone acetate, APMIS 105:590–596, 1997. 66. Gavazza A, Presciuttini S, Barale R, et al.: Association between
43. Marconato L, Leo C, Girelli R, et al.: Association between waste canine malignant lymphoma, living in industrial areas, and use of
management and cancer in companion animals, J Vet Intern Med chemicals by dog owners, J Vet Intern Med 15:190–195, 2001.
23:564–569, 2009. 67. Comba P, Bianchi F, Fazzo L, et al.: Cancer mortality in an area
44. Glickman LT, Raghavan M, Knapp DW, et al.: Herbicide expo- of Campania (Italy) characterized by multiple toxic dumping sites,
sure and the risk of transitional cell carcinoma of the urinary blad- Ann N Y Acad Sci 1076:449–461, 2006.
der in Scottish Terriers, J Am Vet Med Assoc 224:1290–1297, 2004. 68. Upton AC, Kneip T, Toniolo P: Public health aspects of
45. Weeth LP, Fascetti AJ, Kass PH, et al.: Prevalence of obese dogs in toxic chemical disposal sites, Annu Rev Public Health 10:
a population of dogs with cancer, Am J Vet Res 68:389–398, 2007. 1–25, 1989.
46. Paluszkiewicz P, Smolinska K, Debinska I, et al.: Main dietary 69. Vrijheid M: Health effects of residence near hazardous waste land-
compounds and pancreatic cancer risk. The quantitative analy- fill sites: a review of epidemiologic literature, Environ Health Per-
sis of case-control and cohort studies, Cancer Epidemiol 36: spect 108:101–112, 2000.
60–67, 2012. 70. Dosemeci M, Wacholder S, Lubin JH: Does nondifferential mis-
47. Sun CL, Yuan JM, Koh WP, et al.: Green tea, black tea and breast classification of exposure always bias a true effect toward the null
cancer risk: a meta-analysis of epidemiological studies, Carcinogen- value? Am J Epidemiol 132:746–748, 1990.
esis 27:1310–1315, 2006. 71. Wacholder S, Hartge P, Lubin JH, et al.: Non-differential misclas-
48. Aune D, Chan DS, Lau R, et al.: Dietary fibre, whole grains, and sification and bias towards the null: a clarification, Occup Environ
risk of colorectal cancer: systematic review and dose-response meta- Med 52:557–558, 1995.
analysis of prospective studies, BMJ 343, 2011. 72. Cooley DM, Beranek BC, Schlittler DL, et al.: Endogenous
49. Backer LC, Grindem CB, Corbett WT, et al.: Pet dogs as sentinels gonadal hormone exposure and bone sarcoma risk, Cancer Epide-
for environmental contamination, Sci Total Environ 274:161–169, miol Biomarkers Prev 11:1434–1440, 2002.
2001. 73. Millanta F, Calandrella M, Bari G, et al.: Comparison of steroid
50. Reif JS: Animal sentinels for environmental and public health, Pub- receptor expression in normal, dysplastic, and neoplastic canine
lic Health Rep 126:50–57, 2011. and feline mammary tissues, Res Vet Sci 79:225–232, 2005.
51. Kelsey JL, Moore AS, Glickman LT: Epidemiologic stud- 74. Rhodes L, Ding VD, Kemp RK, et al.: Estradiol causes a dose-
ies of risk factors for cancer in pet dogs, Epidemiol Rev 20: dependent stimulation of prostate growth in castrated beagle dogs,
204–217, 1998. Prostate 44:8–18, 2000.
75. Sonnenschein EG, Glickman LT, Goldschmidt MH, et al.: Body 98. Bossuyt PM, Reitsma JB, Bruns DE, et al.: Towards complete and
conformation, diet, and risk of breast cancer in pet dogs: a case- accurate reporting of studies of diagnostic accuracy: the STARD
control study, Am J Epidemiol 133:694–703, 1991. initiative. Standards for reporting of diagnostic accuracy, Clin
76. Ru G, Terracini B, Glickman LT: Host related risk factors for Chem 49:1–6, 2003.
canine OSA, Vet J 156:31–39, 1998. 99. Gardner IA: Quality standards are needed for reporting of test
77. Teske E, Naan EC, van Dijk EM, et al.: Canine prostate carcinoma: accuracy studies for animal diseases, Prev Vet Med 97:136–143,
epidemiological evidence of an increased risk in castrated dogs, Mol 2010.
Cell Endocrinol 197:251–255, 2002. 100. Webster JD, Dennis MM, Dervisis N, et al.: Recommended guide-
78. Bryan JN, Keeler MR, Henry CJ, et al.: A population study of lines for the conduct and evaluation of prognostic studies in veteri-
neutering status as a risk factor for canine prostate cancer, Prostate nary oncology, Vet Pathol 48:7–18, 2011.
67:1174–1181, 2007. 101. Lawrence J, Rohren E, Provenzale J: PET/CT today and tomorrow
79. Torres de la Riva G, Hart BL, Farver TB, et al.: Neutering dogs: in veterinary cancer diagnosis and monitoring: fundamentals, early
effects on joint disorders and cancers in Golden Retrievers, PLoS results and future perspectives, Vet Comp Oncol 8:163–187, 2010.
One 8:e55937, 2013. 102. Chou R, Croswell JM, Dana T, et al.: Screening for prostate cancer:
80. Zink MC, Farhoody P, Elser SE, et al.: Evaluation of the risk and a review of the evidence for the U.S. Preventive Services Task Force,
age of onset of cancer and behavioral disorders in gonadectomized Ann Intern Med 155:762–771, 2011.
Vizslas, J Am Vet Med Assoc 244:309–319, 2014. 103. Sargeant JM, Thompson A, Valcour J, et al.: Quality of reporting
81. Hart BL, Hart LA, Thigpen AP, et al.: Long-term health effects of of clinical trials of dogs and cats and associations with treatment
neutering dogs: comparison of Labrador Retrievers with Golden effects, J Vet Intern Med 24:44–50, 2010.
Retrievers, PLoS One 9:e102241, 2014. 104. Vail DM: Cancer clinical trials: development and implementation,
82. Johnston SD, Kamolpatana K, Root-Kustritz MV, et al.: Prostatic Vet Clin North Am Small Anim Pract 37:1033–1057, 2007.
disorders in the dog, Anim Reprod Sci, 2000. 60-61:405–415. 105. London CA, Malpas PB, Wood-Follis SL, et al.: Multi-center, placebo-
83. Perez Alenza MD, Pena L, del Castillo N, et al.: Factors influencing controlled, double-blind, randomized study of oral toceranib phosphate
the incidence and prognosis of canine mammary tumours, J Small (SU11654), a receptor tyrosine kinase inhibitor, for the treatment of
Anim Pract 41:287–291, 2000. dogs with recurrent (either local or distant) mast cell tumor following
84. Heinlein CA, Chang C: Androgen receptor in prostate cancer, surgical excision, Clin Cancer Res 15:3856–3865, 2009.
Endocr Rev 25:276–308, 2004. 106. Rau SE, Barber LG, Burgess KE: Efficacy of maropitant in the
85. Gann PH, Hennekens CH, Ma J, et al.: Prospective study of sex prevention of delayed vomiting associated with administration of
hormone levels and risk of prostate cancer, J Natl Cancer Inst doxorubicin to dogs, J Vet Intern Med 4:1452–1457, 2010.
88:1118–1126, 1996. 107. Scott MC, Sarver AL, Gavin KJ, et al.: Molecular subtypes of osteo-
86. Sorenmo KU, Goldschmidt M, Shofer F, et al.: Immunohisto- sarcoma identified by reducing tumor heterogeneity through an
chemical characterization of canine prostatic carcinoma and cor- interspecies comparative approach, Bone 49:356–367, 2011.
relation with castration status and castration time, Vet Comp Oncol 108. Modiano JF, Breen M, Burnett RC, et al.: Distinct B-cell and T-cell
1:48–56, 2003. lymphoproliferative disease prevalence among dog breeds indicates
87. Navarro D, Luzardo OP, Fernandez L, et al.: Transition to andro- heritable risk, Cancer Res 65:5654–5661, 2005.
gen-independence in prostate cancer, J Steroid Biochem Mol Biol 109. Akesson A, Julin B, Wolk A: Long-term dietary cadmium intake
81:191–201, 2002. and postmenopausal endometrial cancer incidence: a popula-
88. Kent MS, Burton JH, Dank G, et al.: Association of cancer- tion-based prospective cohort study, Cancer Res 68:6435–6441,
related mortality, age and gonadectomy in golden retriever dogs 2008.
at a veterinary academic center (1989-2016), PLoS One 13: 110. Shaw J: Relationship-centered approach to cancer communication.
e0192578, 2018. In Withrow SJ, MacEwen EG, Page RL, editors: Small animal clini-
89. Martano M, Morello E, Buracco P: Feline injection-site sarcoma: cal oncology, ed 5, St. Louis, 2012, Elsevier, pp 272–279.
past, present and future perspectives, Vet J 188:136–141, 2011. 111. Raghavan M, Knapp DW, Dawson MH, et al.: Topical flea and
90. Kass PH, Barnes WG, Spangler WL, et al.: Epidemiologic evidence tick pesticides and the risk of transitional cell carcinoma of the uri-
for a causal relation between vaccination and fibrosarcoma tumori- nary bladder in Scottish Terriers, J Am Vet Med Assoc 225:389–394,
genesis in cats, J Am Vet Med Assoc 203:396–405, 1993. 2004.
91. Kass PH, Spangler WL, Hendrick MJ, et al.: Multicenter case-con- 112. Bettini G, Morini M, Marconato L, et al.: Association between
trol study of risk factors associated with development of vaccine- environmental dust exposure and lung cancer in dogs, Vet J
associated sarcomas in cats, J Am Vet Med Assoc 223:1283–1292, 186:364–369, 2010.
2003. 113. Bukowski JA, Wartenberg D, Goldschmidt M: Environmental
92. Kass PH: Methodological issues in the design and analysis of epi- causes for sinonasal cancers in pet dogs, and their usefulness as sen-
demiological studies of feline vaccine-associated sarcomas, Anim tinels of indoor cancer risk, J Toxicol Environ Health A 54:579–591,
Health Res Rev 5:291–293, 2004. 1998.
93. Magden E, Quackenbush SL, Vandewoude S: FIV associated neo- 114. Dias Pereira P, Lopes CC, Matos AJ, et al.: Influence of catechol-
plasms-A mini-review, Vet Immunol Immunopathol 143:227–234, O-methyltransferase (COMT) genotypes on the prognosis of
2011. canine mammary tumors, Vet Pathol 46:1270–1274, 2009.
94. Yamamoto JK, Hansen H, Ho EW, et al.: Epidemiologic and clini- 115. Misdorp W, Romijn A, Hart AA: Feline mammary tumors: a case-
cal aspects of feline immunodeficiency virus infection in cats from control study of hormonal factors, Anticancer Res 11:1793–1797,
the continental United States and Canada and possible mode of 1991.
transmission, J Am Vet Med Assoc 194:213–220, 1989. 116. Overley B, Shofer FS, Goldschmidt MH, et al.: Association
95. Rezanka LJ, Rojko JL, Neil JC: Feline leukemia virus: pathogenesis between ovarihysterectomy and feline mammary carcinoma, J Vet
of neoplastic disease, Cancer Invest 10:371–389, 1992. Intern Med 19:560–563, 2005.
96. Shelton GH, Grant CK, Cotter SM, et al.: Feline immunodefi- 117. Hart BL, Hart LA, Thigpen AP, et al.: Neutering of German Shep-
ciency virus and feline leukemia virus infections and their rela- herd dogs: associated joint disorders, cancers and urinary inconti-
tionships to lymphoid malignancies in cats: a retrospective study nence, Vet Med and Sci 2:191–199, 2016.
(1968-1988), J Acquir Immune Defic Syndr 3:623–630, 1990. 118. Perez Alenza D, Rutteman GR, Pena L, et al.: Relation between
97. Bossuyt PM, Reitsma JB: The STARD initiative, Lancet 361:71, habitual diet and canine mammary tumors in a case-control study,
2003. J Vet Intern Med 12:132–139, 1998.
119. Raghavan M, Knapp DW, Bonney PL, et al.: Evaluation of the 121. Gabor LJ, Love DN, Malik R, Canfield PJ: Feline immunodefi-
effect of dietary vegetable consumption on reducing risk of tran- ciency virus status of Australian cats with lymphosarcoma, Aust Vet
sitional cell carcinoma of the urinary bladder in Scottish Terriers, J 79:540–545, 2001.
J Am Vet Med Assoc 227:94–100, 2005. 122. Dorn CR, Taylor DO, Schneider R: Sunlight exposure and risk of
120. Hutson CA, Rideout BA, Pedersen NC: Neoplasia associated with developing cutaneous and oral squamous cell carcinomas in white
feline immunodeficiency virus infection in cats of southern Califor- cats, J Natl Cancer Inst 46:1073–1078, 1971.
nia, J Am Vet Med Assoc 199:1357–1362, 1991.

4 - Epidemiology and The Evidence Bas - 2019 - Withrow and MacEwen S Small Anima

Uploaded by

Copyright:

Available Formats

4 - Epidemiology and The Evidence Bas - 2019 - Withrow and MacEwen S Small Anima

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

4 - Epidemiology and The Evidence Bas - 2019 - Withrow and MacEwen S Small Anima

Uploaded by

Copyright:

Available Formats

4

Epidemiology and the Evidence-Based

TABLE 4.1 Glossary of Terms

1 1 1 3 3,4 4,5 5,6 ,6 ,7 7

TABLE 4.4 Commonly Diagnosed Cancers and Suspected Risk Factors

Cutaneous squamous cell Solar irradiation

Environmental Exposures Exposure misclassification is a primary limitation of using geo-

You might also like

4 - Epidemiology and The Evidence Bas - 2019 - Withrow and MacEwen S Small Anima

Uploaded by

Copyright:

Available Formats

4 - Epidemiology and The Evidence Bas - 2019 - Withrow and MacEwen S Small Anima

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

4 - Epidemiology and The Evidence Bas - 2019 - Withrow and MacEwen S Small Anima

Uploaded by

Copyright:

Available Formats

4

Epidemiology and the Evidence-Based

TABLE 4.1 Glossary of Terms

1 1 1 3 3,4 4,5 5,6 ,6 ,7 7

TABLE 4.4 Commonly Diagnosed Cancers and Suspected Risk Factors

Cutaneous squamous cell Solar irradiation

Environmental Exposures Exposure misclassification is a primary limitation of using geo-

You might also like

TABLE 4.1 Glossary of Terms

TABLE 4.4 Commonly Diagnosed Cancers and Suspected Risk Factors