Deepak Kademani - Improving Outcomes in Oral Cancer - A Clinical and Translational Update-Springer International Publishing (2020)

Improving Outcomes
in Oral Cancer
A Clinical and Translational
Update
Deepak Kademani
Editor
123
Improving Outcomes in Oral Cancer
Deepak Kademani
Editor
Improving Outcomes in
Oral Cancer
A Clinical and Translational Update
Editor
Deepak Kademani
Department of Oral and Maxillofacial Surgery
North Memorial Medical Center
Minneapolis, MN
USA
ISBN 978-3-030-30093-7 ISBN 978-3-030-30094-4 (eBook)

https://doi.org/10.1007/978-3-030-30094-4
© Springer Nature Switzerland AG 2020

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, expressed or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To my patients who have given me the privilege to care for them and taught
me so much about the human spirit
Contents
1 Oral Epithelial Dysplasia�� 1

Kenneth Wan and Deepak Kademani
2 Chemoprevention in Oral Cancer�� 13
Jeffrey Y. Tsai and Jasjit K. Dillon
3 Diagnostic Adjuncts for Screening and Surveillance
in Head and Neck Cancer�� 23
James Murphy and Mohammed Qaisi
4 Current Updates in Staging and Prognosis in Oral Cancer�� 33
Paul Covello and D. David Kim
5 Surgical Factors Affecting Outcomes
in Oral Squamous Cell Carcinoma�� 45
Justine Moe, Andrew Baker, and Brent Ward
6 Pathological Factors Affecting Outcomes in Oral Cancer �� 65
Eric R. Carlson and J. Michael McCoy
7 Genetics of Oral Cancer�� 97
Anthony Morlandt and Hope Amm
8 Factors Affecting Response and Survival in Radiotherapy�� 105
Michael Awadallah, Kurt Nisi, and Ketan J. Patel
9 Novel and Emerging Chemotherapeutic Agents
in Head and Neck Cancer�� 117
Arielle Elkrief, Nicholas Makhoul, and Nathaniel Bouganim
10 Immunotherapy in Oral Cancer: A Fourth Dimension
of Cancer Treatment�� 129
Marcus A. Couey, Rom S. Leidner, Simon W. Young,
and R. Bryan Bell
11 Recurrent Oral Cancer and Salvage Options�� 155
Joshua E. Lubek and Michael Nagai
12 Oral Reconstruction�� 165
J. Collin, B. Turner, and R. Fernandes
13 Predicting Quality of Life (QoL) of Oral Cancer�� 181
Roderick Y. Kim, Jason N. Burkes, and Fayette C. Williams
vii
Oral Epithelial Dysplasia
1
Kenneth Wan and Deepak Kademani
1.1 Introduction in the oral cavity compared to normal mucosa in

a healthy patient. Recently, the term, “potentially
In general histopathology terms, dysplasia is a premalignant oral epithelial lesions” (PPOEL),
disordered growth that encompasses an abnor- has been described in the literature to replace
mality in the maturation of cells within tissues OPMD. For a lesion to be described as an oral
and the development of cytological atypia within epithelial dysplasia, there must be a biopsied and
cells. When dysplasia occurs in the epithelium of histopathologically reported foci of dysplasia.
the oral cavity, the WHO have termed it oral epi-
thelial dysplasia (OED), defining it as a precan-
cerous lesion of stratified squamous epithelium, 1.2 Grading and Classification
characterized by cellular atypia and loss of normal of OED
maturation and stratification short of carcinoma
in situ. It is a histologically proven oral prema- OED is a condition comprising of a spectrum
lignant lesion that is associated with a signifi- of tissue changes, with several grading systems
cant higher risk of malignant transformation. An established to classify into arbitrary levels of
OED may be part of a clinically apparent lesion, severity, hence diagnosis is extremely subjective
such as leukoplakia, eythroplakia, erytholeuko- [1, 2]. The relevant diagnostic criteria have been
plakia, lichen planus and submucosal fibrosis, revised several times and many systems of clas-
actinic chelitis, and chronic hyperplastic candi- sification exist, each with their own biases [3].
diasis. These lesions are termed “oral potentially These are generally based on the histopathologi-
malignant disorders” (OPMD) by the 2005 WHO cal classification of premalignant lesions of other
workshop and are referred to a variety of clinical mucosal sites, which frequently develop SCC.
lesions, conditions, or systemic disorders, which For example, squamous intraepithelial neopla-
result in an increased risk of cancer development sia (SIN) is an oral adaptation of a system used
for classifying precursor lesions of the uterine
K. Wan cervix and have been used for grading OED in the
Department of Oral and Maxillifacial Surgery, North older literature [3]. While the SIN system has its
Memorial Health, Robbinsdale, MN, USA advantages, it has been rejected for use in the oral
D. Kademani (*) cavity and oropharynx due to the emphasis placed
Oral and Maxillofacial Surgery and Hospital on tissue thickness due to hyperkeratinization,
Dentistry, Head and Neck Oncologic and which is not considered to carry a higher risk of
Reconstructive Surgery, North Memorial Medical
Center, Minneapolis, MN, USA malignancy than normal tissue in the oral cavity
e-mail: Deepak.Kademani@northmemorial.com [4, 5]. Furthermore, the SIN system suggests an
© Springer Nature Switzerland AG 2020 1
D. Kademani (ed.), Improving Outcomes in Oral Cancer,
https://doi.org/10.1007/978-3-030-30094-4_1
2 K. Wan and D. Kademani
i nevitable progression to malignancy, which is not Table 1.1 List of architectural and cytological changes
associated with oral epithelia dysplasia, 2005 WHO
the case of OEDs in the oral cavity [6].
Classification
In the Ljubljana grading system, lesions are cat-
Architecture Cytology
egorized into simple hyperplasia, basal/parabasal
Irregular epithelial Abnormal variation in nuclear
hyperplasia, atypical hyperplasia, or carcinoma in stratification size (anisonucleosis)
situ. It is an alternative system based on another Loss of polarity of Abnormal variation in nuclear
anatomical site but adapted for the oral cavity and basal cells shape (nuclear pleomorphism)
oropharynx [7]. Originally utilized in the context of Drop-shaped rete Abnormal variation in cell size
ridges (anisocytosis)
laryngeal precursor lesions, it is considered beyond
Increased number of Abnormal variation in cell shape
the scope of the histopathological changes which mitotic figures (cellular pleomorphism)
occur in the oral cavity and oropharynx [3]. Abnormal superficial Increased nuclear-cytoplasmic
Another grading systems include the Smith mitoses ratio
and Pindborg, which utilizes 13 histological Dyskeratosis Increase nuclear size
features that are standardized by a set of photo- Keratin pearls within Atypical mitotic figures
rete pegs Increase number and size of
graphs. After comparing with the photographic
nucleoli
standard, the feature is graded as none, slight, or Hyperchromasia
marked and given a score. The scores are added
to achieve the epithelial atypia index (EAI) score
(maximum possible is 75), and depending upon Table 1.2 Classification of oral epithelial dysplasia,
2005 WHO classification
the EAI, the dysplasia is graded as no dysplasia,
mild, moderate, or severe. Hyperplasia Increased cell number; the architecture
shows regular stratification without
Currently, the 2005 WHO Classification is most cellular atypia
widely used for classification of tissue dysplasia. Mild Architectural disturbance limited to the
A range of cellular and architectural changes in epithelial lower third of the epithelium
the tissue is assessed and classified into a specific dysplasia accompanied by cytological atypia
Moderate Architectural disturbance extending
grade of dysplasia (Tables 1.1 and 1.2).
epithelial into the middle third of the epithelium
Issues with intraobserver reproducibility and dysplasia with consideration of the degree of
interobserver agreement plagues all the afore- cytologic atypia
mentioned grading system, with the Ljubljana Severe Greater than two-thirds of the
and Smith & Pindborg system faring worse than epithelial epithelium showing architectural
dysplasia disturbance with associated cytologic
the WHO system. Merely determining the pres- atypia or architectural disturbance
ence of OED appears to be a challenge, with one extended into the middle-third of the
US study reporting a Kappa value of 0.51 (mod- epithelium with sufficient cytologic
erate strength of agreement) between three oral atypia
Carcinoma in Full-thickness architectural
pathologists when asked to assess OED presence situ abnormalities in the viable cellular
and absence [8]. Intraexaminer reliability varied layer accompanied by pronounced
greatly among the pathologists, with one scoring cytologic atypia; atypical mitotic
a Kappa value as low as 0.22 (slight strength of figures and abnormal superficial
mitoses
agreement) [8]. Another study reported Kappa
agreement scores of 0.15 and 0.41 between six
pathologists in determining the presence of OED
among 120 slides [9]. Lumerman et al. reports an 1.3 Clinical Presentation of OED
interexaminer reliability of only 54% [10].
Considering the low consistency between OED within the oral cavity may present in a
diagnoses, it is expected that this would be a range of clinical lesions, rendering it not possible
major limitation among most studies, and this to diagnose without invasive biopsy. Clinically,
has resulted in remaining controversy surround- OEDs may appear as homogenous lesion (clini-
ing the predictive value of OED. cally provisionally diagnosed as homogenous
1 Oral Epithelial Dysplasia 3
leukoplakia or keratosis), nonhomogenous (clini- reported to undergo malignant transformation

cally provisionally diagnosed as nonhomogenous [17]. Krutchkoff et al. argue that OLP in itself is
leukoplakia, eythroplakia, speckled leukoplakia), not potentially malignant, and that associations
lichenoid (clinically provisionally diagnosed as with transformation are due to inaccurate and
oral lichen planus or oral lichenoid tissue reac- overdiagnosis [16].
tion), or others (lesions which are diagnosed as
nonspecific ulcerations/erosions/atrophies, angio-
granuoloma, frictional keratosis, leukoedema). In 1.4 Risk Factors for OED
several studies, nonhomogenous clinical appear-
ance was highly associated with dysplasia, and OED has a high association with the male gender
over 80% of provisional nonhomogenous lesions [18–21]. Studies have demonstrated that males
were dysplastic or malignant on biopsy. are more at risk because of greater levels of expo-
Lesions that display redness or surface irreg- sure to risk habits such as alcohol and tobacco
ularity are more likely to be dysplastic [11]. consumption [4].
Erythroplakia is reported to carry the greatest I
It is well established in the literature that
rate of OED of any oral mucosal lesion, with smoking is highly associated with the develop-
greater than 90% exhibiting dysplastic character- ment of PPOELs and malignancies in the oral
istics on biopsy [12], and a vast majority of these cavity [4]. In respect to the development of
undergo malignant transformation [2]. In a study OED, the exposure and the level of exposure of
of 166 leukoplakias, a nonhomogeneous clini- tobacco to the oral epithelium is significantly
cal appearance was found to be associated with associated with the development of dysplastic
presence of OED on histopathological assess- tissue changes. In one study, those who were
ment, and they were more likely to develop oral identified as current smokers had an odds ratio
SCC on follow-up [13]. of 4.1 for developing OED when compared to
In respect to the clinical features of OED, those who never smoked [22]. Of 173 OED
one study has found that all lesions that displays cases in a retrospective study, the author found
any degree of OED were associated with some that half of the patients who reported tobacco
form of leukoplakic appearance [14]. In the same usage presented with some degree of OED on
study, severe dysplasia was diagnosed mostly in biopsy [14]. Another study reported that severe
mixed red and white lesions; however, this was and moderate dysplasia in particular arose at a
not statistically significant. Comparably, lesions higher rate among smokers, with approximately
which exhibited redness had a greater tendency 77% of severe dysplasia occurring in tobacco
to present with moderate dysplasia in contrast users [23].
to clinically white lesions; however, the rate of It is recognized in the literature that alcohol
severe dysplasia was equal between white and and tobacco act synergistically as a risk factor for
mixed red and white lesions, indicating that these oral SCC [24–26], but conflicting evidence exists
findings may be due to sample variation [11]. to support alcohol’s role in the development of
Tissue redness as a feature of malignant progres- OED. A paper by Morse et al. reported that they
sion can also be appreciated in relation to OLP, did not find any significant association between
where it has been reported that erosive and ulcer- alcohol consumption and development of OED
ative types are at risk of malignancy [15]. [27] while a previous study by the same groups
Lichenoid dysplasia is a term to describe of authors observed that consumption of seven or
lesions that on histopathology are primarily dys- more alcoholic beverages a week increased the
plastic in nature but exhibit some features of risk of detecting OED on biopsy of a PPOEL
OLP [16]. Oral lichen planus is assumed to be by two times [27]. In another study, 50% the
potentially malignant and may of undertaking PPOELs presenting with dysplasia occurred in
malignant transformation; however, controversy individuals who reported regular consumption
does exist. Up to 3% of OLP cases have been [28]. The carcinogenicity of alcohol is thought to
be due to the metabolism of ethanol to mutogenic of cases that will undergo malignant transforma-
acetaladehyde in the oral cavity. tion is 100% [2, 4, 12].
The most high-risk area for development Oral lichen planus (OLP) is considered by
of oral SCC and OED, as agreed on by many some authors to be a PPOEL; however, contro-
authors, is the floor of the mouth and the versy does exist about this inclusion [13]. OLP
tongue, particularly the lateral border [1, 29, is a chronic inflammatory disease thought to be
30]. This is owing to the fact that a greater level immune-mediated and have some genetic predis-
of carcinogenic exposure is present as tobacco, position; however, the exact etiology is not known.
and alcoholic products dissolve in the saliva The proportion of biopsied OLP cases positive for
and settle on the floor of the oral cavity [29]. In OED that will undergo malignant transformation
addition, due to the thinner and nonkeratinized has been reported as 1–3.2% [2, 4].
epithelium of these sites, tissue penetration and Oral submucous fibrosis (OSF) is a chronic
a more potent level of carcinogenic exposure is mucosa disease of the upper digestive tract [11,
possible. Also contributing may be the differing 20]. Fibrosis of the lamina propria and submuco-
embryonic origins of these site and response to sal layers of the mucosal lining of the oral cavity,
carcinogens [29]. In a study by Barnes et al. oropharynx, and, at times, the esophagus, result-
that examined the clinical features of OED, it ing in loss of tissue mobility and limited oral
was reported that severe dysplasia was more opening, is its histopathological characteristic.
likely to form on the lateral tongue and floor of The proportion of biopsied OSF cases positive
the mouth, compared to the rest of the oral cav- for OED is 7–25%, and the proportion of cases
ity [4]. Pereira et al. also has a similar finding, that will undergo malignant transformation is
with severe dysplasia occurring most often on 1–8% [2, 4, 31].
the floor of mouth and tongue. Actinic cheilitis is a keratotic condition of
the lip vermillion which is considered to be
potentially malignant. Patients may present with
1.5 elationship Between OED
R various clinical signs and symptoms, the most
and PPOEL commonly reported being dryness of the lip, atro-
phy, erythema, ulceration, edema, and blurring
Leukoplakia, erythroplakia, oral lichen planus, of the vermillion border. Proportion of biopsied
oral submucosa fibrosis, and actinic chelitis are OSF cases positive for OED is 100% [4, 32].
recognized potential premalignant oral epithelial
lesion.
Leukoplakia is defined as any “white plaque of 1.6 Detection and Diagnosis
questionable risk having excluded other known
disease and disorders that carry no increased risk The gold standard test for diagnosis of OED is
of cancer”. The can be grouped in to homoge- histopathological from specimens taken from
nous or nonhomogenous (erythro-leukoplakia). a formal tissue biopsy [33–35]. There are no
A subtype of leukoplakia, proliferative verrucous reported chairside adjunctive tests currently that
leukoplakia has the highest rate of malignant have reported higher sensitivity and specificity
transformation of any oral white patch lesion. numbers that trumps the combination of clinical
The proportion of biopsied leukoplakia cases examination and tissue biopsy [33]. There is a
positive for OED has been reported as 15%, and myriad of adjunctive tests available for the cli-
the proportion of cases that will undergo malig- nician’s armamentarium, with the pros of being
nant transformation is 1% [2, 4]. non/minimally invasive and causing little to no
Erythroplakia is defined as a fiery red patch morbidity but with a tradeoff for giving apprecia-
that cannot be characterized clinically or patho- ble levels of false positives and negatives. These
logically as any other definable disease. The pro- adjunctive tests should only be used in situations
portion of biopsied erythroplakia cases positive where the clinician is unsure after clinical exami-
for OED is reported as 91%, and the proportion nation whether to go ahead with a tissue biopsy
or to find areas within a large homogenous lesion lesions that were clinically obvious by routine
to incisional biopsy that will yield a specimen visual examination [40, 41]. From a specificity
with more advanced atypia or dysplasia. They standpoint, both autofluorescence and chemilu-
should not be used when the lesion is clinically minescence performed suboptimally in differen-
frankly dysplastic or cancerous where a formal tiating dysplasia/malignancy, inflammation, and
biopsy would be indicated. We describe several reactive from each other. An alternative, modified
more common adjunctive tests below. chemilumninescence method that takes advan-
Metachromatic dyes that have high affinity tage of dysplastic and malignant cells expressing
for nucleic acids, such as toluidine blue can be a different glycan residue, which can then be con-
used as an aid to detect high-grade dysplasia and jugated with a proprietary fluorescent lectin has
malignant lesion based on the premise that they shown promising results, with an in vivo study
produce higher levels of nucleic acids compared yielding sensitivity and specificity of 89% and
with normal tissue. The intention is to guide the 82%, respectively [42–44].
clinician to areas for biopsy, but overall, toludine Dysplastic and malignant cells generally have
blue has poor sensitivity and specificity (57–81% depleted or negligible glycogen content com-
and 56–67%, respectively) [33, 36]. The figures pared to healthy mucosa. Lugol’s iodine solution,
are better with increasing severity of dysplasia, which contains iodine and potassium iodide in an
such as severe dysplasia/carcinoma in situ but are aqueous solvent will bind to normal mucosa and
poor in mild and moderate dysplasia. have low affinity for dysplastic/malignant tissue.
Another minimally invasive detection tech- The literature has shown that it is a useful adjunct
nique is brush cytology/biopsy that required the in obtaining clear margins for dysplasia/intraepi-
clinician to use a dedicated brush to collect a thelial neoplasm during tumor resections (32%
transepithelial and exfoliated cell sample, which clear margins in the control group versus 4% in
is then fixed on a histology slide and submit- the Lugol’s solution group) [45].
ted for specialized computer-aided scanning
analysis. As the transepithelial array of cells are
architecturally disordered, it can only detect the 1.7 Human Papillomavirus
presence of cellular atypia and not able to dif- and Dysplasia
ferentiate invasive carcinoma from carcinoma in
situ. Apart from its high cost, it is in limited use There are over 160 genotypes of human papil-
in clinical practice owing to inconsistent sensitiv- loma virus and some subtypes are risk factors
ity and specificity figures in the literature (range for development of oropharyngeal and oral squa-
73–100% and 32–94%, respectively) [37, 38]. mous cell carcinoma. Specifically, 16 and 18
There is evidence to suggest that light- have been described as high risk for the develop-
reflecting properties of oral mucosal changes in ment of oropharyngeal carcinoma [46, 47]. HPV-
a progressive and predictable manner from the derived oncogenes, E6 and E7, causes epithelial
spectrum of normal to frankly malignant oral malignant transformation by repressing p53 and
epithelial tissue. Autofluorescence and chemi- Rb tumor suppressor gene functions [46]. The
luminescence diagnostic/screening tools take prevalence of HPV is 0.9–12% in clinical normal
advantage of this assumption for them to be mucosa, and in an immunocompetent host, the
marketed to clinicians for use, with a myriad of infection is usually cleared within 2 years [48].
commercial brands available for sale. The tools’ Perseverance of the virus past the 2 year mark
specificity and sensitivity functioning statistics augments the likelihood of malignant genetic
are generally not promising with one systemic mutation and transformation. Controversy exist
review’s reporting rates of 0–100% and 0–75% on the association and prevalence of HPV in
for chemiluminescence and 30–100% and 15.3– PPOELs. A systematic review described an over-
100% for autofluorescence, respectively [33, all odds ratio of 3.87 between all PPOELs and
39]. It was purported that sensitivities of 100% aggregate HPV-DNA, and when dysplasia was
published in some of the papers where owing to the specific variable, the OR raised to 5.10 [46].
The prevalence of HPV subtypes 16 and 18 was the higher end of the MTR spectrum along with
reported to be 25% in oral and oropharyngeal the grade of the dysplasia [1]. There are conflict-
dysplasia in one meta-analysis [49]. HPV-driven ing reports with respect to grading severity being
dysplasia have been described as being unique correlated with MTR [23, 53, 54]. A predomi-
in histopathological studies as they are char- nance of the contemporary literature supports the
acterized by karyorrhexis and apoptosis [47]. hypothesis that MTR is correlated with the pres-
Chemoprevention and HPV vaccination is antici- ence of OED and it severity [1]. On the other-
pated to reduce the prevalence and incidence of hand, there are some studies not supporting the
oropharyngeal/oral squamous cell carcinoma. relationship between MTR and grade of dyspla-
sia, such as Dost et al.’s paper, involving biopsy-
proven OED in 368 individuals, which came to a
1.8 Field Cancerization and OED conclusion that the severity of dysplasia, graded
according the 2005 WHO classification and the
OED presents the initial steps of field cancer- Kujan et al. binary system, was not correlated
ization, when early cellular and architectural with the risk of malignant transformation [23].
changes affect the mucosal epithelium. Field can- In a retrospective study of biopsy specimens
cerization describes the multistep and sequential collected over 20 years, Cowen et al. [55] dem-
process of carcinogenesis of epithelial tumors. onstrated that a relationship existed between the
This process was first described by Slaughter presence of OED and malignancy. However, the
et al. after microscopic examination of almost authors failed to undertake a statistical analysis
800 oral and pharyngeal cancers revealed that tis- of their findings. A similar retrospective study
sue abnormalities extended beyond the clinically was conducted in the UK, and a significant rela-
obvious tumor [50]. This suggested that cancers tionship between OED grading and oral SCC
arise from patches or fields of genetically abnor- development was found [56]. The annual trans-
mal cells which display features of malignancy formation rate of severe dysplasia was 5.6%,
but remain noninvasive. These fields develop compared with 0.3% of nondysplastic PMDs.
from a single mutated stem cell which divides This is further corroborated by Schepman et al.
and differentiates to produce similarly abnor- [13], who concluded that leukoplakia which
mal daughter cells. Uncontrolled cell division presented with moderate to severe OED, had a
allows for the growth and development of this significantly higher predisposition to developing
field which replaces the overlying normal tissue a malignancy. Silverman et al. [17] reported a
[51]. Histopathologically, this field is diagnosed malignant transformation rate of 36% for lesions
as OED and is considered potentially malignant. which present with OED. Small sample size
however is a shortcoming of this study, which
puts into doubt the validity of these results; only
1.9 Malignant Transformation 22 lesions which presented with some degree of
OED were included in the aforementioned study.
There is a myriad of widely varying figures in the Another study from Australia has reported that
literature relating to the malignant transforma- 4.7% of OEDs progressed to oral cancer in a
tion rate of OED to OSCC; this may be owing to mean time of 3.3 years, and it also suggested that
when effect of confounding factors such as expo- mild grades of OED were just as likely to trans-
sure to risk factors not considered, classification form into OSCC as severe-grade OEDs. This is
of clinical lesions being varied between studies, in stark contrast to Mehanna’s meta-analysis that
and, as previously outlined, the classification of showed mild/moderate OEDs had a malignant
dysplasia is not an exact science. The malignant transformation of 10% versus severe OED/CIS
transformation rate of OED varies vastly in the which has a rate of 24%.
literature, with a range of 6–36% [52]. Current These findings differ from those from Asian
variables in the literature that affect the MTR are countries such as Taiwan and India, which tend
the site of the lesion; tongue and FOM being at to conclude that OED does not affect malig-
nant potential. Prospective evaluation of 1458 of clinically applicable prognostic and diagnostic
Taiwanese patients presenting with a PPOEL markers and targeted therapies that eventuate in
revealed that in over 10 years of follow-up, no improved prognosis and survival of head and neck
cases of severe OED developed a malignancy [57]. cancer patients has not been fruitful to date.
While results show that those patients presenting A comprehensive description of all mole-
with OED transformed at a higher rate than those cules studied is beyond the scope of this chap-
without, this was not statistically significant, indi- ter. The majority of the molecules explored are
cating SCC development is unaffected by dys- associated with critical cellular and molecular
plastic features. The estimated annual malignant oncogenic processes or the term “hallmarks of
transformation rate of 3.02% is considered partic- cancer” coined by Hanahan and Weinberg [61].
ularly low in light of other research; however, the These processes involve sustained cell prolifera-
authors suggest this to be due to broader inclusion tion, evasion of growth suppression, resistance
criteria, incorporating a wider variety of PPOELs. to apoptosis, replicative immortality, angiogen-
Comparative studies tend to limit analysis to a par- esis, invasion, and metastasis as well as emerg-
ticular type of lesion, such as leukoplakia [13, 17, ing “hallmarks” such as evasion of immune
29, 58]. A smaller scale study also conducted in system surveillance, reprogramming of cellular
Taiwan found similar results [59]. metabolism, and enabling molecular character-
Difficulty arises when comparing the results of istics (genomic instability and tumor-promoting
these studies, as differing study design, inclusion inflammation) [61].
criteria, and statistical analyses affect the findings. Of note are markers relating to epigentetic
Several studies restrict inclusion criteria to certain events, which is an emerging area in research.
types of PPOELs, most commonly leukoplakias These epigenetic events include histone
[13, 17, 58], which limit the generalisability of sumoylation and acetylation, microRNA and long
malignant transformation rates, which themselves encoding RNA post-transcriptional regulation
are calculated via differing means. Varying defi- (upregulation, downregulation, or overprogression
nitions of PPOELs also affect selection criteria, or underexpression) and DNA methylation [60].
particularly those with a focus on leukoplakia, the DNA Hypermethylation. In approximately
diagnostic criteria of which has been revised sev- 40% of OED cases, p16 gene hypermethyl-
eral times. Older studies tend to follow the classifi- ation is detected and a corresponding propor-
cation of the time, so conditions such as frictional tion progresses to OSCC. In addition, during the
keratosis, which have no risk of malignancy above progression of mild to severe OED, hypermeth-
normal healthy mucosa, were included as leuko- ylation of the p15 and 16 gene has been docu-
plakias, affecting overall study outcomes [2]. mented. MGMT gene methylation is described
to be greater than 50–80% of OL. Oral lichen
planus without dysplastic features can be distin-
1.10 M
olecular Markers Associated guished from those with dysplasia by detection
with Development of methylation of TSPYLS5, NKX2-3, RBP4,
and Progression of OED/ TRPC4, CMTM3, CLDN11, and MAP6 genes.
PPOEL Methylated HOXA9, EDNRB, and DCC (deleted
in colorectal cancer) were correlated with malig-
Research of complex molecular mechanisms nant or premalignant oral lesions. Methylated zinc
underpinning oral behavior, development, and finger protein 582 (ZNF582, transcription factor
progression of oral cancers has been vast and pro- on chromosome 19) has also been suggested as a
gressing at a rapid pace in the past several decades. biomarker for oral dysplasia and cancer [60].
Despite this, our current cognizance of the Histone Modification. Tumor invasion and
critical molecular process that heralds and drives oncological transformation can be the result of
dysplastic or potentially premalignant epithelial histone modification that has triggered deregu-
lesions’ progression to oral squamous cell carci- lation of chromatic-based process. An example
noma is still lacking [60]. As such, the development of this is lysine modification on H3 histone at
specific position Lys9, Lys4, Lys18, and Lys27 study involving 5338 patients who received sur-
that become methylated and/or acetylated as is gical excision/resection of oral squamous cell
observed in some oral squamous cell carcinoma carcinoma, 30% were found positive for all HPV
lesion. Papillon-Cavanagh et al. [62] have dem- subtypes, 18% for HPV18, and 25% for HPV16
onstrated diminished H3-K36 methylation char- [65]. Although the link between HPV as a caus-
acterizes a subset of head and neck SCC, but all ative factor in oral squamous cell carcinoma is
studies have thus far addressed only SCCs, and not as strong as for oropharyngeal carcinoma, the
data on PPOELs or dysplasia are deficient. HPV vaccine may play a role in secondary pre-
Micro-RNA. Cellular noncoding mirco-RNA, vention. Other prevention strategies include early
in concert with other factors, regulates the cel- detection of PPOELs and prevention of malig-
lular protein expression and functions. Reports nant transformation [33].
of association between miRNA profiles and oral
premalignant/dysplasia are few. MiR-31 was
reported as being augmented in some poten- 1.11.2 Surgical Management
tially premalignant oral epithelial lesions, such
as hyperkeratotic and hyperplastic lesions, which Surgery management of OED involving excision
are deemed less likely to progress to SCC com- using scalpel, excision, or ablation using laser or
pared to OED. Cervigne et al. have reported that cryosurgery is reported. There are no RCTs com-
overexpression of miRNA-345, miRNA-21, and paring the efficacies of these in respect to recur-
miRNA-181b was essential to malignant trans- rence, progression to malignancy. Surgery in the
formation. Increase in lesion severity during pro- form of excision and/or laser ablation is at present,
gression was associated with elevated expression the most accepted mode of treatment [54, 66–68].
of miRNA-345, miR-181b, and miR-21 [63]. Cryosurgery have limited use in treatment of
OED and have been reported to yield higher rates
of recurrence and malignant transformation [69].
1.11 Management Surgical excision with a scalpel blade is a consis-
tent modality and common in surgical practice as
1.11.1 Prevention it is cost effective, simple to use, and provides a
surgical specimen with margins that is undamaged
In the management of premalignancy, primary by heat of a laser, which allow for accurate histo-
prevention should be the first armamentarium pathological examination [52]. Excision of large
utilized, and any modifiable risk factors for OED OED lesion with a blade may produce undesirable
should be eliminated in order to prevent and cicatricial healing, this can be overcome by placing
arrest the progression of premalignancy to malig- a split thickness skin graft in the surgical bed [69].
nancy. Patients should be counseled on tobacco CO2 lasers are used frequently in the surgical
use cessation and limit alcohol intake. Risk strati- management of OED. The mechanism of CO2
fying is extremely important in order to identify laser involved applied focal, collimated energy
high-risk individuals and then to provide appro- that augments the temperature of the target tissue
priate screening and counseling. From a systemic to greater than 100 °C, culminating in the phase
review, the predicted attributable lifetime risk change of water to steam. Adjustable power of
for developing oral squamous cell carcinoma if laser permits its use as a surgical knife or ablative
an individual smoked solely, consumed alcohol agent (5–25 W). Laser can be used defocused to
solely, or in used tobacco and alcohol in com- ablate the tissue and permit hemostasis, and the
bination was 25%, 18%, and 40%, respectively site is left deepithelialized to heal by secondary
[64]. The correlation between development of intention. CO2 laser creates a unique wound, in
OED that may progress to oral squamous cell that it is only a few tenth of mm deep with limited
carcinoma with tobacco use and alcohol con- removal of healthy tissue. Meltzer suggests recur-
sumption risk is dose-dependent and cumulative rence of leukoplakia with laser is only 10% com-
over an individual’s lifetime. A meta-analysis pared with scalpel at 34% [70]. Other advantages
include cellular destruction by ablation minimizes [1, 6]. This is somewhat supported by the litera-
release of inflammatory mediators compared with ture, which reports the risk of a mildly dysplastic
a scalpel, hence patient is reported to have less lesion progressing to cancer being less than 5%
pain and swelling; blood vessels with diameter of [6]. More severe tissue changes are reported to
the lumen less than half of a millimeter are sealed progress to SCC in as low as 7% of cases [6].
off, producing a less bloody field; and limited Distinguishing between these levels of severity
wound contracture. The main criticism is that the in itself presents a challenge, with subjectivity
vaporized tissue is not available for histological unavoidable in the process of classifying a con-
exam, but this can be overcome somewhat by tak- tinuous scale of tissue change. In contrast, there
ing multiple incision biopsy specimens prior to is literature to support the notion that irrespec-
lasering. Another disadvantage of laser ablation is tive of the grade of OED, all biopsy-proven OED
that epithelial migration is delayed, and the surgi- should be treated by excision or laser ablation,
cal wound may take longer to heal [70]. instead of the “wait and watch” approach some
The general consensus is that the presence clinicians take for mild dysplasia. The manage-
of OED predisposes a lesion to undergo malig- ment of mild and moderate OED remains con-
nant transformation; logic would follow that troversial, and there is no concrete well-designed
the severity would have an impact, as the more RCTs that give support either way. Owing to the
severe the dysplasia is, the more genetically aber- higher risk of malignant transformation of severe
rant and therefore histopathologically similar to OEDs and CIS, the accepted convention treat-
an SCC. Regardless of this supposed multistep ment is surgical excision with or without recon-
progression model, current practice sees some struction. Diagram 1 depicts our departments
clinicians forgoing active treatment of milder protocol in the management of mild, moderate,
lesions, which are monitored rather than excised and severe/CIS OEDs [71] (Fig. 1.1).
Biopsy Proven
Oral Epithelia
Dysplasia
Mild Moderate Severe

Dysplasia Dysplasia Dysplasia/CIS
Low Suspicion High Excision with Excision and

Close surveillance Suspicion Laser ablation laser ablation
Conservation Mx Excision +/- laser (<2mm margin) (<5mm margin)
ablation
6 monthly review 6 montly review Severe Malignancy

for 12 months 6 monthly review for 24 months Proceed with
Dysplasia/CIS
then yearly for 12 months then yearly work-up and Mx
3 monthly review
(lifelong with then yearly (lifelong with as cancerous
for 24 month and
clinical photos) (lifelong with clinical photos) lesion
then 6 monthly
clinical photos)
(lifelong with
clinical photos)
Fig. 1.1 Authors’ algorithm for management of oral epithelial dysplasia

1.11.3 Medical Treatments after surgical excision of OEDs. In cohort study

by Thomson et al. that looked at PPOEL excised
Medical treatments have no supporting evidence with CO2 lasers, of the 590 patients, there was
at this time. Topical bleomycin, systemic retinoic a persistence rate of 9% and malignant transfor-
acid, and lycopene may help to resolve OED in mation rate of 16%. The mean surveillance dura-
the short term, but there is no evidence exists to tion was 7.3 years, and mean time for malignant
support that they prevent malignant transforma- transformation was 7 years [68].
tion [35]. There is also poor evidence in support
of photodynamic therapy or using COX inhibi- Acknowledgement Disclaimer: All figures and tables
submitted are original. We report no conflict of interest
tors or attenuated virus as mouthwash therapy to
and no financial support was received.
treat OED [35].
References
1.12 Follow-Up and Recurrence
After Excision or Ablation 1. Speight PM, Speight PM. Update on oral epithe-
of OED lial dysplasia and progression to cancer. Head Neck
Pathol. 2007;1:61–6.
2. Van Der Waal I. Potentially malignant disorders of the
For those individuals diagnosed with OED, at oral and oropharyngeal mucosa; terminology, clas-
present, widely accepted standardized or guide- sification and present concepts of management. Oral
lines on the frequency of surveillance, and clini- Oncol. 2009;45:317–23.
cal follow-up do not exist [35]. Despite this, 3. Warnakulasuriya S, Reibel J, Bouquot J, Dabelsteen
E. Oral epithelial dysplasia classification systems:
clinical surveillance ought to be sufficiently fre- predictive value, utility, weaknesses and scope for
quent; especially in moderate and severely dys- improvement. J Oral Pathol Med. 2008;37:127–33.
plastic lesions [54, 67]. This should be contingent 4. Barnes L, Eveson JW, Reichart P, Sidransky D, edi-
from the notion of field cancerization, variable tors. World Health Organisation classification of
tumours pathology and genetics of head and neck
interval to malignant transformation, and rate tumours. Lyon: IARC Press; 2005.
of recurrence. Surveillance of a patient should 5. Warnakulasuriya S, Johnson NW, van der Waal
be individualized according to the lesion (site I. Nomenclature and classification of potentially
of lesion, degree of dysplasia) and the patient’s malignant disorders of the oral mucosa. J Oral Pathol
Med. 2007;36:575–80.
risk factors (alcohol/tobacco uses, age, gender) 6. Bouquot JE, Speight PM, Farthing PM. Epithelial
[1, 53]; some authors recommend that continued dysplasia of the oral mucosa—diagnostic prob-
surveillance is mandatory long term, as long as lems and prognostic features. Curr Diagn Pathol.
20 years to life-long after excision [35, 71]. 2006;12:11–21.
7. Hellquist H, Cardesa A, Gale N, Kambic V, Michaels
Despite surgical modalities of treatment, L. Criteria for grading in the Ljubljana classifica-
an appreciable risk of recurrence and malig- tion of epithelial hyperplastic laryngeal lesions.
nant transformation still persists. Excision of A study by members of the Working Group on
OEDs is reported to diminish the rate of recur- Epithelial Hyperplastic Laryngeal Lesions of the
European Society of Pathology. Histopathology.
rence by close to 50% but does not eradicate the 1999;34:226–33.
risk [72]. Holmstup et al. reported a recurrence 8. Brothwell DJ, Lewis DW, Bradley G, et al. Observer
rate of 13% and malignant transformation rate agreement in the grading of oral epithelial dysplasia.
of 12% in a retrospective study of 94 patients, Commun Dent Oral. 2003;31:300–5.
9. Abbey LM, Kaugars GE, Gunsolley JC, et al.
who had nonhomogenous or homogenous oral Intraexaminer and interexaminer reliability in
leukoplakias excised; their clinical surveillance the diagnosis of oral epithelial dysplasia. Oral
period ranged from 1.5 to 18 years. Seventy-one Surg Oral Med Oral Pathol Oral Radiol Endod.
percentage of the excised leukoplakias exhib- 1995;80:188–91.
10. Lumerman H, Freedman P, Kerpel S. Oral epithelial
ited oral epithelial dysplasia, and 7 years was dysplasia and the development of invasive squamous
the mean time to malignant transformation [72]. cell carcinoma. Oral Surg Oral Med Oral Pathol Oral
Nankivell reported a lower recurrence rate of 5% Radiol. 1995;79:321–9.
11. Jaber MA, Porter SR, Speight P, Eveson JW, Scully 27. Morse DE, Psoter WJ, Cleveland D, et al. Smoking
C. Oral epithelial dysplasia: clinical characteris- and drinking in relation to oral cancer and oral epithe-
tics of western European residents. Oral Oncol. lial dysplasia. Cancer Cause Contr. 2007;18:919–29.
2003;39:589–96. 28. Dost F, Cao KA, Ford P, Farah C, et al. A retrospec-
12. Hashibe M, Mathew B, Kuruvilla B, et al. Chewing tive analysis of clinical features of oral malignant and
tobacco, alcohol, and the risk of erythroplakia. Cancer potentially malignant disorders with and without oral
Epidem Biomar. 2000;9:639–45. epithelial dysplasia. Oral Surg Oral Med Oral Pathol
13. Schepman KP, van der Meij EH, Smeele LE, van Oral Radiol Endod. 2013;116:725–33.
der Waal I. Malignant transformation of oral leuko- 29. Zhang L, Cheung KJ Jr, Lam WL, et al. Increased
plakia: a follow-up study of a hospital-based popula- genetic damage in oral leukoplakia from high risk
tion of 166 patients with oral leukoplakia from The sites: potential impact on staging and clinical manage-
Netherlands. Oral Oncol. 1998;34:270–5. ment. Cancer. 2001;91:2148–55.
14. Pereira J, Carvalho Mde V, Henriques AC, et al.
30. Napier SS, Speight PM. Natural history of potentially
Epidemiology and correlation of the clinicopathologi- malignant oral lesions and conditions: an overview of
cal features in oral epithelial dysplasia: analysis of the literature. J Oral Pathol Med. 2008;37:1–10.
173 cases. Ann Diagn Pathol. 2011;15:98–102. 31. Cox SC, Walker DM. Oral submucous fibrosis. A
15. Barnard NA, Scully C, Eveson JW, Cunningham S, review. Aust Dent J. 1996;41:294–9.
Porter SR. Oral cancer development in patients with 32. Savage NW, McKay C, Faulkner C. Actinic cheilitis in
oral lichen planus. J Oral Pathol Med. 1993;22:421–4. dental practice. Aust Dent J. 2010;55(Suppl 1):78–84.
16. Krutchkoff DJ, Eisenberg E. Lichenoid dysplasia: a 33. Rhodus NL, Kerr AR, Patel K. Oral cancer: leuko-
distinct histopathologic entity. Oral Surg Oral Med plakia, premalignancy, and squamous cell carcinoma.
Oral Pathol Oral Radiol Endod. 1985;60:308–15. Dent Clin N Am. 2014;58(2):315–40.
17. Silverman S Jr, Bahl S. Oral lichen planus update: clin- 34. Panwar A, Lindau R, Wieland A. Management for
ical characteristics, treatment responses, and malig- premalignant lesions of the oral cavity. Expert Rev
nant transformation. Am J Dent. 1997;10:259–63. Anticancer Ther. 2014;14(3):349–57.
18. Epstein JB, Wan LS, Gorsky M, Zhang L. Oral lichen 35. Brennan M, et al. Management of oral epithelial dys-
planus: progress in understanding its malignant plasia: a review. Oral Surg Oral Med Oral Pathol Oral
potential and the implications for clinical manage- Radiol Endod. 2007;103(Suppl):S19.e1–12.
ment. Oral Surg Oral Med Oral Pathol Oral Radiol 36. Portugal LC, Wilson KM, Biddinger PW, et al. The
Endod. 2003;96:32–7. role of toluidine blue in assessing margin status after
19. Silverman S Jr, Gorsky M, Lozada-Nur F, Giannotti resection of squamous cell carci-nomas of the upper
K. A prospective study of findings and management aerodigestive tract. Arch Otolaryngol Head Neck
in 214 patients with oral lichen planus. Oral Surg Oral Surg. 1996;122:517–9.
Med Oral Pathol Oral Radiol Endod. 1991;72:665–70. 37. Gupta S, et al. Clinical correlative study on early
20. Silverman S Jr, Gorsky M, Lozada-Nur F. A prospec- detection of oral cancer and precancerous lesions
tive follow-up study of 570 patients with oral lichen by modified oral brush biopsy and cytology fol-
planus: persistence, remission, and malignant asso- lowed by histopathology. J Cancer Res Ther.
ciation. Oral Surg Oral Med Oral Pathol Oral Radiol 2014;10(2):232–8.
Endod. 1985;60:30–4. 38. Babshet M, et al. Efficacy of oral brush cytology in
21. Silverman S, Bhargava K, Smith LW, Malaowalla
the evaluation of the oral premalignant and malignant
AM. Malignant transformation and natural history lesions. J Cytol. 2011;28(4):165–72.
of oral leukoplakia in 57,518 industrial workers of 39. Rashid A, Warnakulasuriya S. The use of light-based
Gujarat, India. Cancer. 1976;38:1790–5. (optical) detection systems as adjuncts in the detec-
22. Morse DE, Katz RV, Pendrys DG, et al. Smoking tion of oral cancer and oral potentially malignant
and drinking in relation to oral epithelial dysplasia. disorders: a systematic review. J Oral Pathol Med.
Cancer Epidem Biomar. 1996;5:769–77. 2015;44(5):307–28.
23. Dost F, Cao KA, Ford P, Farah C, et al. Malignant 40. Liviu Feller JL. Oral squamous cell carcinoma epide-
transformation rate of oral epithelial dysplasia: a miology, clinical presentation and treatment. Cancer
real-world evaluationof histopathological grading. Ther. 2012;3:263–8.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 41. Gillenwater A, Papadimitrakopoulou V, Richards-

2014;117:323–52. Kortum R. Oral premalignancy: new methods
24. Ogden GR, Wight AJ. Aetiology of oral cancer: alco- of detection and treatment. Curr Oncol Rep.
hol. Brit J Oral Max Surg. 1998;36:247–51. 2006;8(2):146–54.
25. Zygogianni AG, Kyrgias G, Karakitsos P, et al. Oral 42. McCullough MJ, Farah CS. The role of alcohol

squamous cell cancer: early detection and the role of in oral carcinogenesis with particular reference
alcohol and smoking. Head Neck Oncol. 2011;3:2. to alcohol-containing mouthwashes. Aust Dent J.
26. Znaor A, Brennan P, Gajalakshmi V, et al. Independent 2008;53:302–5.
and combined effects of tobacco smoking, chewing 43. Baeten J, et al. Chairside molecular imaging of aber-
and alcohol drinking on the risk of oral, pharyngeal rant glycosylation in subjects with suspicious oral
and esophageal cancers in Indian men. Int J Cancer. lesions using fluorescently labeled wheat germ agglu-
2003;105:681–6. tinin. Head Neck. 2018;40(2):292–301.
44. Baeten J, et al. Molecular imaging of oral premalig- 59. Ho PS, Chen PL, Warnakulasuriya S, et al. Malignant
nant and malignant lesions using fluorescently labeled transformation of oral potentially malignant disorders
lectins. Transl Oncol. 2014;7(2):213–20. in males: a retrospective cohort study. BMC Cancer.
45. McMahon J, et al. Use of Lugol’s iodine in the resec- 2009;9:260.
tion of oral and oropharyngeal squamous cell carci- 60. Nikitakis N, Pentenero M, Georgaki M, Poh C,

noma. Br J Oral Maxillofac Surg. 2010;48(2):84–7. et al. Molecular markers associated with develop-
46. Syrjanen S, et al. Human papillomaviruses in oral ment of potentially premalignant oral epithelial
carcinoma and oral potentially malignant disorders: a lesions: current knowledge and future implications.
systematic review. Oral Dis. 2011;17(Suppl 1):58–72. Oral Surg Oral Med Oral Pathol Oral Radiol Endod.
47. Woo SB, Cashman EC, Lerman MA. Human
2008;125:650–69.
papillomavirus-associated oral intraepithelial neopla- 61. Hanahan D, Weinberg RA. Hallmarks of cancer: the
sia. Mod Pathol. 2013;26(10):1288–97. next generation. Cell. 2011;144:646–74.
48. Lerman MA, et al. HPV-16 in a distinct sub-
62. Papillon-Cavanagh S, Lu C, Gayden T, et al.

set of oral epithelial dysplasia. Mod Pathol. Impaired H3 K36 methylation defines a subset of
2017;30(12):1646–54. head and neck squamous cell carcinomas. Nat Genet.
49. Jayaprakash V, et al. Human papillomavirus types 2017;49:180–5.
16 and 18 in epithelial dysplasia of oral cavity and 63. Cervigne NK, Reis PP, Machado J, et al. Identification
oropharynx: a meta-analysis, 1985–2010. Oral Oncol. of a microRNA signature associated with progression
2011;47(11):1048–54. of leukoplakia to oral carcinoma. Hum Mol Genet.
50. Slaughter DP, Southwick HW, Smejkal W. Field
2009;18:4818–29.
cancerization in oral stratified squamous epithelium; 64. Radoi L, Luce D. A review of risk factors for oral
clinical implications of multicentric origin. Cancer. cavity cancer: the importance of a standardized
1953;6:963–8. case definition. Community Dent Oral Epidemiol.
51. Braakhuis BJ, Tabor MP, Kummer JA, Leemans CR, 2013;41(2):97–109, e78–91.
Brakenhoff RH. A genetic explanation of Slaughter's 65. Kansy K, Thiele O, Freier K. The role of human papil-
concept of field cancerization: evidence and clinical lomavirus in oral squamous cell carcinoma: myth and
implications. Cancer Res. 2003;63:1727–30. reality. Oral Maxillofac Surg. 2014;18(2):165–72.
52. Mehanna HM, et al. Treatment and follow-up of oral 66. Mogedas-Vegara A, et al. Oral leukoplakia treat-

dysplasia – a systematic review and meta-analysis. ment with the carbon dioxide laser: a systematic
Head Neck. 2009;31:600–9. review of the literature. J Craniomaxillofac Surg.
53. Amagasa T, Yamashiro M, Uzawa N. Oral premalig- 2016;44(4):331–6.
nant lesions: from a clinical perspective. Int J Clin 67. Field EA, et al. The management of oral epithe-

Oncol. 2011;16(1):5–14. lial dysplasia: the Liverpool algorithm. Oral Oncol.
54. Goodson ML, et al. Oral precursor lesions and malig- 2015;51(10):883–7.
nant transformation—who, where, what, and when? 68. Thomson PJ, et al. Interventional laser surgery for
Br J Oral Maxillofac Surg. 2015;53(9):831–5. oral potentially malignant disorders: a longitudi-
55. Cowan CG, Gregg TA, Napier SS, McKenna SM, nal patient cohort study. Int J Oral Maxillofac Surg.
Kee F. Potentially malignant oral lesions in northern 2017;46(3):337–42.
Ireland: a 20-year population-based perspective of 69. Kademani D. Surgical management of oral and muco-
malignant transformation. Oral Dis. 2001;7:18–24. sal dysplasia: the case for surgical excision. J Oral
56. Warnakulasuriya S, Kovacevic T, Madden P, et al. Maxillofac Surg. 2007;65:287–92.
Factors predicting malignant transformation in oral 70. Meltzer C. Surgical management of oral and muco-
potentially malignant disorders among patients sal dysplasia: the case for laser excision. J Oral
accrued over a 10-year period in South East England. Maxillofac Surg. 2007;65:293–5.
J Oral Pathol Med. 2011;40(9):677–83. 71.
Awadallah M, Idle M, Patel K, Kademani
57. Hsue SS, Wang WC, Chen CH, et al. Malignant trans- D. Management update of potentially premalignant
formation in 1458 patients with potentially malignant oral epithelial lesions. Oral Surg Oral Med Oral
oral mucosal disorders: a follow-up study based in a Pathol Oral Radiol. 2018;125(6):628–36.
Taiwanese hospital. J Oral Pathol Med. 2007;36:25–9. 72. Holmstrup P, et al. Long-term treatment out-

58. Banoczy J. Follow-up studies in oral leukoplakia. J come of oral premalignant lesions. Oral Oncol.
Maxillofac Surg. 1977;5:69–75. 2006;42(5):461–74.
Chemoprevention in Oral Cancer
2
Jeffrey Y. Tsai and Jasjit K. Dillon
2.1 Introduction allowing for the identification of biomarkers that

can be targeted by chemoprevention. Based on
Chemoprevention is the concept of preventing, the multistep carcinogenesis model of progres-
delaying, or reversing the progression of prema- sion from hyperplasia to invasive cancer (hyper-
lignant lesions to invasive cancer. This can be plasia, to dysplasia, to carcinoma in situ, to
achieved through the use of natural, synthetic, or cancer), the identification of the molecular driv-
biologic agents [1]. Studies have been reported in ers of these pathways could reveal molecular tar-
the literature regarding the utility of agents such gets that can halt this stepwise progression.
as systemic and topical retinoids, bleomycin, To date, there have been many compounds
cyclooxygenase inhibitors, and phytochemical- that have been studied for their potential in pre-
enriched products in decreasing the clinical pres- venting progression of premalignant lesions to
ence of oral premalignant lesions, as well as in cancer in the oral cavity and/or prevention of sec-
preventing second primary tumors that may arise ond primary head and neck tumors. Further, with
through field cancerization of the oral cavity [2, the advent of the identification of molecular bio-
3]. Population-based studies have revealed pos- markers of oral cancer risk, molecularly targeted
sible reduction in the relative risk of developing agents (e.g., epidermal growth factor [EGFR]
head and neck cancers with the ingestion of cer- inhibitors) have additionally been evaluated [7].
tain nutrients (e.g., carotenoids [4]) or medica- This chapter aims to provide the reader with an
tions (e.g., NSAIDs [5]). Thus, the idea that overview of the present literature of several che-
chemoprevention can be utilized to prevent or moprevention agents that have been studied in
even reverse premalignant disease of the oral clinical trials and provide commentary on poten-
cavity in high-risk individuals has attracted a tial future directions of research in this field.
great deal of scientific inquiry. While chemoprevention of oral cancer has not
As described by Bauman in 2016 [6], finding yet entered the domain of routine clinical prac-
effective chemoprevention strategies against tice, it is an evolving field that may yet yield
head and neck squamous cell carcinoma has been future therapeutics.
made more difficult by several factors, including
the lack of preclinical oral carcinogenesis models
2.2 Antimutagens
J. Y. Tsai · J. K. Dillon (*) Mutagens lead to the development of cancer
Department of Oral and Maxillofacial Surgery,
University of Washington, Seattle, WA, USA through permanent changes in a cell’s DNA,
e-mail: tsaijeff@uw.edu; dillonj5@uw.edu resulting in alterations to a gene product, defects
https://doi.org/10.1007/978-3-030-30094-4_2
14 J. Y. Tsai and J. K. Dillon
in cellular functions, and potentially the loss of form of randomized controlled trials (RCTs) has
cell growth control. In the case of oral squamous been less confirmatory. A large RCT performed
cell carcinoma, known mutagens include lifestyle in 2002 by the European Organization for
agents (tobacco, alcohol, betel nut) and infec- Research and Treatment of Cancer (EORTC)
tious agents (bacterial, candidiasis, human papil- investigated the effectiveness of Vitamin A and
loma virus, herpes simplex virus). Compounds NAC as chemoprevention agents that may
with antimutagenic properties exert their effect improve the prognosis of patients that were
by either preventing the mutagen’s effect on already treated for head and neck cancer or for
DNA (desmutagens) or by suppressing the pro- lung cancer by preventing second primary
cess of mutation after genes are damaged by tumors. Focusing on the arm that received NAC
mutagens (bio-antimutagens). The major mecha- (600 mg daily for 2 years), no statistically signifi-
nisms of mutagenesis include chemical or enzy- cant difference in survival or event-free survival
matic inactivation, prevention of formation of were found between these subjects as compared
active species, scavenging of mutagens, and anti- to subjects who did not receive NAC during the
oxidant or scavenging of free radicals [8]. study time period of 2 years [12]. To date, no fur-
ther RCTs evaluating the use of NAC in oral can-
cer chemoprevention have been published.
2.2.1 N-Acetyl-l-Cysteine
A variant of the amino acid l-cysteine, N-acetyl- 2.2.2 Topical Bleomycin

L-cysteine (NAC) is widely used for the treatment
of acetaminophen toxicity via repletion of gluta- Bleomycin is an antibiotic that has been found to
thione reserves. This increase in glutathione levels exhibit antitumoral effects through several mech-
leads to support of the body’s antioxidant system. anisms, including the scission of DNA via acti-
As such, NAC has been found to have anti-inflam- vated oxygen and inhibition of DNA ligase. The
matory effects beneficial for the treatment of drug is often incorporated into the chemotherapy
chronic obstructive pulmonary disease (COPD), regimen for the treatment of Hodgkin’s lym-
influenza, and idiopathic pulmonary fibrosis [9]. phoma and non-Hodgkin’s lymphoma and can be
NAC has been also investigated for its poten- administered intravenously, subcutaneously,
tial anticancer effects and has been demonstrated intramuscularly, intralesionally, or topically [13].
to suppress epidermal growth factor (EGF)- Potential adverse effects include interstitial pneu-
induced EGF receptor (EGFR) phosphorylation. monitis and mucocutaneous toxicity. The advan-
Overexpression of EGFR has been demonstrated tage of a locally administered formulation of
in greater than 80% of head and neck squamous bleomycin (and conceivably of any chemopre-
cell carcinoma and is linked with high recurrence vention agent) is the ability to deliver a high local
and low survival rates [10, 11]. An in vitro study dose with low total systemic dose which may
performed by Lee et al. in 2012 [11] demon- minimize toxicity [2].
strated that NAC induces cell cycle arrest and The effectiveness of topical bleomycin in the
apoptosis in a human tongue squamous carci- treatment of oral premalignant lesions has been
noma cell line that expresses higher EGFR levels documented in investigational clinical studies. A
than other such cell lines. A murine model study systematic review of the literature on topical
performed by the same group demonstrated agents for oral cancer chemoprevention by Chau
reduction in mean tumor volume relative to con- et al. in 2017 found five studies that utilized topi-
trol animals. Taken together, NAC could be a cal bleomycin in the treatment of oral premalig-
potential anticancer adjuvant in EGFR overex- nant lesions [2]. With a mean treatment duration
pressing oral cancers. of 2 weeks, the mean complete response rate
Evidence demonstrating the effectiveness of (defined as no clinical and/or histopathologic evi-
NAC in chemoprevention of oral cancer in the dence of leukoplakia) was 40.2%, 25%, and 8%
2 Chemoprevention in Oral Cancer 15
for patients receiving 1% topical bleomycin in sible dose–response to green tea extract, whereby
DMSO, 0.5% topical bleomycin in DMSO, and subjects administered higher doses of green tea
placebo, respectively, thereby demonstrating a extract had higher rates of clinical response. Of
dose-dependent complete response rate. In one of the biomarkers examined in this study, stromal
the included studies by Epstein et al. in 1994, VEGF expression was downregulated in subjects
participants were randomized to 1% topical bleo- receiving higher doses of green tea extract, sug-
mycin in DMSO versus DMSO alone (placebo) gesting a possible mechanistic action of green tea
[14]. These authors found that 100% of those extract causing inhibition of angiogenesis [17].
receiving the 1% topical bleomycin in DMSO While the study did demonstrate possible effect
agent had an adverse effect, which included ery- of green tea extract on clinical response of oral
thema, erosion, and discomfort at the site of premalignant lesions, this clinical response was
application. Future clinical studies that evaluate not associated with a decreased risk of progres-
the highest effective dose while producing the sion to oral cancer. Further studies demonstrating
fewest adverse effects are needed. the effectiveness and therapeutic dosing of green
tea extract in preventing oral cancer are needed.
2.2.3 Polyphenols 2.2.3.2 Curcumin

A yellow coloring agent found in turmeric, cur-
As a category, polyphenols are a broad group of cumin has been studied as a potential agent in the
phytochemicals found in plants, particularly chemoprevention of colon, breast, prostate, and
fruits, seeds, and leaves. Polyphenols have been oral cancers. In vitro studies have demonstrated
shown to exhibit a protective effect against reac- its ability to downregulate nuclear factor-kappa B
tive oxygen species and have been utilized in (NF-κB) and cyclooxygenase-2 (COX-2) in oral
in vitro and clinical studies to examine their premalignant and cancer cells [18, 19]. Chronic
potential antimutagenic effect [15]. Within the exposure to carcinogens causes inflammation and
broad category of plant-derived dietary polyphe- exposure to reactive oxygen species that contrib-
nols, tea and tea extracts (e.g., from green tea) ute to aberrant activation of NF-κB and develop-
have been utilized in epidemiological and clinical ment of squamous cell carcinoma, thus making
studies examining their potential to suppress the curcumin a potential agent of study. A recent pro-
development of oral cancer. spective cohort study by Rai in 2010 [18] demon-
strated possible antioxidant effect of curcumin by
2.2.3.1 Green Tea Extract increasing serum and salivary concentrations of
Preclinical models have demonstrated that the vitamin C and E in patients with oral leukoplakia,
polyphenol epigallocatechin 3-gallate (EGCG), lichen planus, and submucous fibrosis. The poor
found in high amounts in green tea extract, arrests bioavailability of curcumin due to biotransforma-
cells in the G0/G1 phase, regulates apoptosis, and tion in the gut and enterohepatic cycling of
blocks angiogenesis through the phosphorylation metabolites has made the clinical study of cur-
of vascular endothelial growth factor receptor cumin as a chemoprevention agent more difficult,
(VEGFR) and inhibition of VEGF secretion in and high-quality randomized trials are necessary
tumor cells [15, 16]. A phase II randomized, to determine curcumin’s potential use as a sys-
placebo-controlled trial of green tea extract in temic agent of chemoprevention in oral cancer.
patients with high-risk oral premalignant lesions
was published in 2009 by Tsao et al. [17], which
demonstrated a potential partial or complete 2.3 Antiproliferatives
regression of oral premalignant lesions in sub-
jects administered oral green tea extract over a Antiproliferative agents exert their chemoprotec-
period of 12 weeks. While the results of the study tive effect by preventing the proliferation of cells.
did not reach statistical significance (perhaps due A select review of antiproliferative agents that
to small sample sizes), they demonstrated a pos- have been utilized in clinical studies is described
and includes retinoids, carotenoids, anti- ineffective in reducing recurrence rates of head
inflammatories, and ligands of peroxisome and neck squamous cell carcinoma [24]. Thus, the
proliferator-activated receptor gamma (PPAR-γ). local delivery of isotretinoin in the oral cavity has
attracted consideration for its potential as a che-
moprevention agent against oral cancer. While
2.3.1 Retinoids studies demonstrating the effectiveness of topical
retinoids in preventing oral cancer are scarce, a
Included in this group are vitamin A (retinol) and recent study by Kadakia et al. in 2016 [25] sum-
vitamin A analogues (e.g., isotretinoin), which marized a single institution’s 15-year experience
exert their effects through the modulation of gene with isotretinoin rinse as a chemoprophylaxis
expression, apoptosis, and cell proliferation [20, against recurrence for patients previously treated
21]. Retinol enters a cell and is eventually con- for oral cavity squamous cell carcinoma, in situ
verted to retinoic acid via oxidation, which then disease, or dysplasia. All subjects were treated
binds to nuclear receptors that regulate the with reconstituted 0.2% isotretinoin rinses twice
expression of genes. Retinoids have been used in daily, for 1 min duration each, following comple-
the induction treatment of malignancies such as tion of ablative therapy. Patients who used the
lymphoma. medication for a minimum of 12 months were
The effects of vitamin analogs on the remission included in the analysis. The difference between
of oral leukoplakia was clinically demonstrated recurrence among subjects using isotretinoin
first by Koch et al. in 1978 [22], with a confirma- rinses compared to the control group was statisti-
tory randomized controlled trial by Hong et al. in cally significant, whereby subjects treated with
1986 [23] verifying the effectiveness of oral isotretinoin rinse had lower rates of recurrence of
isotretinoin in the treatment of oral premalignant disease. Individual group analysis revealed that
lesions. This randomized controlled trial also dem- the recurrence rate was statistically decreased in
onstrated relapse of oral premalignant lesions with subjects with multiple oral cavity early-stage can-
discontinuation of isotretinoin therapy. A follow- cerous foci or multifocal dysplasia, whereas
up study on high-dose isotretinoin by Hong et al. patients with squamous cell carcinoma in situ or
in 1990 [3] demonstrated a significant reduction in patients with dysplasia following treatment of
the development of second primary head and neck oral cavity squamous cell carcinoma did not have
squamous cell carcinoma tumors in subjects this response to treatment. This finding suggests a
receiving high-dose isotretinoin (50–100 mg/m2 of potential benefit of topical isotretinoin use in
body-surface area per day). As with the previous patients with multiple lesions (consistent with
clinical trial, toxicity with high doses of oral field cancerization), with less or no benefit to
isotretinoin was a barrier to therapy compliance, those with a single isolated lesion [25].
with severe skin dryness, cheilitis, hypertriglyceri- While these results are promising, caution
demia, and conjunctivitis leading to reduction or should be taken prior to the widespread adminis-
discontinuation of isotretinoin. The results from a tration of retinoids for the chemoprophylaxis of
randomized clinical trial published by Khuri et al. cancer. A randomized Phase III trial of isotreti-
in 2006 [24] could not demonstrate the efficacy of noin investigating its effects in preventing second
low-dose (30 mg/day) isotretinoin in reducing the primary tumors in non-small-cell lung cancer
rates of second primary tumors, suggesting that patients demonstrated a higher risk of mortality
combination therapies or retinoid receptor ago- among those with squamous first primaries and
nists may be avenues for further research. in current smokers [26]. Further, a systematic
As highlighted by the challenges associated review of antioxidants for primary and secondary
with the systemic administration of isotretinoin, prevention by Bjelakovic et al. in 2007 [27] dem-
long-term patient adherence to therapy may be onstrated that Vitamin A supplementation could
hampered by toxic effects associated with ther- increase mortality. Further investigations regard-
apy. In addition, low-dose isotretinoin appears ing the safety and efficacy of retinoids are needed,
particularly in light of evidence suggesting the and neck squamous cell carcinoma, exhibit
potential for harm. increased prostaglandin synthesis [29]. These
prostaglandins facilitate the pathogenesis of
malignancies by affecting cell proliferation,
2.3.2 Carotenoids angiogenesis, immune surveillance, and apopto-
sis. The increased synthesis of prostaglandins
Carotenoids are a broad group of organic pig- may be a consequence of increased expression of
ments found in various plants. Various carot- cyclooxygenase-2 (COX-2) in tumor cells.
enoids have Vitamin A activity and can be Experimental data have previously demonstrated
converted to retinol, thus providing an antipro- that COX-2 is involved in several mechanisms
liferative effect. Others exhibit antioxidant important to carcinogenesis, including apoptosis,
effects, decreasing DNA damage from reactive inflammation and immunosuppression, angio-
oxygen species [4]. These properties have genesis, and metastasis. Increased amounts of
prompted investigations into carotenoids and COX-2 are also seen in oral leukoplakia and head
their potential chemopreventive effects in vari- and neck squamous cell carcinoma [29]. Thus,
ous cancers, including head and neck cancers. A COX-2 inhibition (as with NSAIDs) is a potential
systematic review of the effect of dietary carot- strategy in preventing head and neck cancer.
enoid intake and the risk of head and neck can- A metanalysis of prospective cohort and case-
cer by Leonici et al. in 2015 [4] demonstrated a control studies published by Shi et al. in 2017
reduction in the rate various head and neck can- [30] examined the association of NSAID use
cers. Specifically, dietary beta-carotene intake with the relative risk of developing head and neck
resulted in a relative risk reduction of 46% for cancers, including oral cancer. Their analysis
oral cavity cancer. Dietary lycopene intake dem- revealed a statistically significant reduction in the
onstrated a 26% reduction in the rate of oral and relative risk of head and neck cancers in the
pharyngeal cancer [28]. included populations taking NSAIDs, and further
Conversely, potential harm has been demon- subgroup analysis revealed that this relationship
strated from the administration of beta-carotene was seen with aspirin, COX-2 inhibitor, ibupro-
in populations at high risk for lung cancer. Two fen, and other NSAID use. Further, the results of
randomized clinical trials in particular demon- their metanalysis also suggest a dose–response
strated an increased rate of lung cancer in sub- relationship between NSAID use and relative risk
jects administered beta-carotene: the reduction of head and neck cancer, whereby
Beta-Carotene and Retinol Efficiency Trial increasing NSAID use was associated with a
(CARET) and the Alpha-Tocopherol, Beta- greater reduction in relative risk.
Carotene cancer prevention study (ATBC study), Randomized clinical trials assessing the effect
both published in 1996 [26]. These findings are of topical or systemic COX-2 inhibition on oral
possibly attributed to the high doses of supple- premalignant lesions have thus far have been
ments provided. Further study is required to unable to demonstrate a clinical or histologic
understand why dietary carotenoid intake (includ- response to treatment versus placebo [31, 32].
ing beta-carotene) may have a protective effect Further, these medications have also been dem-
against oral cancer, whereas supplementation onstrated to cause potential harm. High use of
with beta-carotene appears to cause harm. regular strength aspirin was associated with an
elevated risk of small-cell lung cancer, a finding
not seen with low-dose aspirin or nonaspirin
2.3.3 Nonsteroidal Anti- NSAIDs [26]. Use of selective COX-2 inhibitors
inflammatory Drugs (NSAIDs) may be associated with an increased risk of car-
diovascular events [33]. Thus, the potential ben-
The rationale for using NSAIDs for chemopre- efit of COX-2 inhibition in preventing head and
vention of oral cancer developed from the real- neck cancer must be balanced with the potential
ization that many malignancies, including head for adverse effects.
2.3.4 Ligands of Peroxisome data [6]. Once selected, these agents are then
Proliferator-Activated studied on populations believed to be at high risk
Receptor gamma (PPAR-γ) of developing oral cancer. Clinical trials have
therefore focused on patients with oral premalig-
Upon activation via ligand binding, PPAR-γ het- nant lesions as these patients have an elevated
erodimerizes with retinoic acid receptors and risk of developing invasive cancer and have
controls the expression of genes involved in met- lesions that are accessible for clinical and histo-
abolic pathways, including for lipid biosynthesis logical examination.
and glucose metabolism [34]. Synthetic ligands Evidence of clinical and histological reversal
of PPAR-γ (such as the thiazolidinedione class of of oral premalignant lesions has often been the
medications) have been developed for the treat- desired outcome of chemoprevention trials.
ment of type II diabetes mellitus as they increase However, oral premalignant lesions as a broad
insulin sensitivity. Activation of PPAR-γ may category encompass pathologies with varying
have an antiproliferative and proapoptotic action. natural histories that can include spontaneous
In vitro studies have demonstrated a proportional regression. Further, clinical or histological
decrease in COX-2 expression with increased response of oral premalignant lesions does not
PPAR-γ expression, suggesting an anti-necessarily correlate with the prevention of inva-
inflammatory mechanism as well [34]. Such pre- sive cancer [6]. Thus, the identification of molec-
clinical data suggest that ligands of PPAR-γ may ular markers that can be measured and can
serve a role in the chemoprevention of various correlate with disease risk and progression may
cancers, including squamous cell carcinoma of aid in selecting high-risk populations and thera-
the head and neck. peutics suitable for clinical trial.
A retrospective cohort analysis by Such an approach to a randomized clinical trial
Govindarajan and Siegel in 2017 [35] examined was utilized by William et al. in 2016 [7] for the
the incidence of head and neck cancers across Erlotinib Prevention of Oral Cancer (EPOC)
diabetic male veterans taking thiazolidinediones study. Erlotinib is an inhibitor of epidermal
(TZDs) across ten Veterans Affairs medical cen- growth factor receptor (EGFR) tyrosine kinase,
ters. When adjusted for several confounding vari- therefore inhibiting the action of EGFR upon
ables, individuals with TZD exposure showed a binding. As discussed previously, overexpression
large and statistically significant reduction in the of EGFR has been demonstrated in many head
incidence of developing head and neck cancer and neck squamous cell carcinomas, and EGFR
compared to individuals managing their diabetes amplification is associated with oral premalignant
through diet alone. This relationship remained lesion transformation to invasive cancer. The
statistically significant if the patient was taking investigators chose to select subjects at high risk
TZD in addition to other oral antidiabetic agents. of oral premalignant lesion transformation to
Importantly, this study population included only invasive cancer based on specific loss of heterozy-
diabetic male veterans, thus potentially limiting gosity (LOH) patterns associated with increased
the generalizability of these findings. A random- risk of such transformation [7, 36]. Rather than
ized clinical trial assessing the effects of the TZD selecting clinical and/or histologic regression of
pioglitazone on the clinical and histologic the lesions as the primary endpoint, the authors
response of oral leukoplakia in nondiabetic sub- chose to focus on cancer-free survival. While the
jects is currently underway [34]. trial did not yield an improvement in cancer-free
survival in subjects treated with erlotinib, this
study is notable for being the first chemopreven-
2.4 Future Directions tion trial to prospectively validate LOH as a
molecular marker for oral premalignant lesion
The selection of oral cancer chemoprevention progression to malignant disease [6, 7].
agents for clinical investigation has been histori- While much focus has been placed on the
cally based on epidemiological and laboratory genomic sequencing and molecular profiling of
head and neck invasive squamous cell carcinoma, more high-quality, long-term randomized clinical
less of this attention has been placed on the identi- trials are necessary [38]. Several factors make the
fication of molecular profiles of oral premalignant clinical study of chemoprevention agents diffi-
lesions. Identifying such profiles serially from pre- cult, including toxicity of therapy, limited bio-
malignancy to invasive carcinoma would aid in the availability of agents, and still-unknown
development of predictive molecular markers and molecular basis of the disease and its progression
targeted therapies [37]. Possible molecular path- to invasive cancer.
ways that may warrant further study include The recurrence of oral premalignant lesions
Notch-1 (second-most commonly mutated gene in with discontinuation of the chemoprevention
head and neck SCC), Stat-3 (found to be a consti- agent (as seen with green tea extract and reti-
tutively activated oncogenic transcription factor in noids) suggests a transitory prevention benefit of
head and neck SCC), chemokine receptor 7 such agents. Thus, long-term chemoprevention
(CCR7, upregulated expression associated with might be necessary in order to prevent progres-
tumor cell survival), nuclear factor-kappa B (NF- sion to invasive cancer. In order for patients to
κB), and cyclooxygenase-2 (COX-2) [37]. tolerate such long-term therapy, the ideal agent
should have low systemic toxicity with minimal
side effects. Many of the chemoprevention agents
2.5 Conclusions discussed in this article present with evidence of
potential harm with prolonged use at high doses
Despite the identification of several agents that (see Table 2.1). Reducing the systemic dose of
may prevent the development of oral squamous the agent (as with isotretinoin) seems ineffective
cell carcinoma via in vitro, animal, and epidemi- in preventing disease.
ological studies, evidence of the clinical effec- Topical therapy may be a strategy to avoid
tiveness of these agents remains limited, and systemic toxicity, but this is not without its own
Table 2.1 List of chemoprevention agents discussed in this review, purported mechanism of action, clinical evidence
for efficacy, and potential harm
Possible mechanism of
Agent action Clinical evidence Potential harm
N-acetylcysteine • Support of body’s • RCT in 2002 unable to • At high doses, can cause
antioxidant system demonstrate effectiveness in headache, urticaria, fever,
via restoration of preventing second primary anaphylactoid reaction [9]
glutathione reserves tumors in subjects previously • Potentiates the effect of
[9] treated for primary head and nitroglycerin [9]
• Anti-inflammatory neck cancer [12]
effect [9]
• Suppression of EGFR
[11]
Topical bleomycin • Scission of DNA and • Systematic review • With systemic
inhibition of DNA demonstrated complete (40% administration, can cause
ligase of subjects) or partial (25% of interstitial pneumonitis and
subjects) response to treatment mucocutaneous toxicity
[2] • With topical administration,
can cause localized
erythema, erosion, and
discomfort [13]
Green tea extract • Regulation of • Phase II trial in 2009 • Insomnia and
containing EGCG apoptosis demonstrated possible partial gastrointestinal symptoms
• Inhibition of VEGFR or complete regression of oral [17].
action [15, 16] premalignant lesions [17]
• Downregulation of stromal
VEGFR expression [17]
Table 2.1 (continued)
Possible mechanism of
Agent action Clinical evidence Potential harm
Curcumin • Downregulation of • Prospective cohort study in • None reported in a 2003
NF-κB and COX-2 2010 demonstrated possible review of curcumin use in
antioxidant effect by human trials [39]
increasing serum and salivary
concentrations of Vitamin E
and C [18]
Retinoids • Modulation of gene • Randomized clinical trials in • High systemic doses
expression, apoptosis, 1986 and 1990 demonstrate associated with severe skin
and cell proliferation efficacy of high-dose dryness, cheilitis,
[20, 21] isotretinoin in treating oral hypertriglyceridemia, and
premalignant lesions and conjunctivitis [3, 23]
preventing second primary • Supplementation with
head and neck tumors [3, 23] Vitamin A may increase
• Low-dose isotretinoin has not risk of mortality [26, 27]
been demonstrated to reduce
rate of developing second
primary tumors [24]
• Possible efficacy as a topical
rinse in preventing recurrence
of disease [25]
Carotenoids • Some carotenoids • Systematic review in 2015 • Possible increase in lung
undergo conversion demonstrated reduction in rate cancer rate in high-risk
to retinol, providing of head and neck cancer with populations taking
antiproliferative dietary intake of carotenoids beta-carotene [26].
effect (including beta-carotene and
• Antioxidant effect lycopene) [4]
[4].
NSAIDs • COX-2 inhibition • Metanalysis in 2017 shows • High use of regular strength
leading to anti- possible reduction in relative aspirin is associated with
inflammatory effect risk of head and neck cancer in elevated risk of small-cell
and inhibition of populations taking NSAIDs in a lung cancer [26]
prostaglandin dose–response relationship [30] • Selective COX-2 inhibitors
synthesis [29] • Randomized clinical trials of may be associated with
topical and systemic NSAIDs increased risk of
unable to demonstrate a cardiovascular events [33]
response in treating oral
premalignant lesions [31, 32]
Ligands of • Antiproliferative and • Retrospective cohort study in • Abnormalities in liver
PPAR-γ proapoptotic effects 2017 demonstrated that function tests
• Anti-inflammatory subjects with TZD exposure • Possible cardiovascular
effect via suppression showed a reduction in the effects with glitazones
of COX-2 [34] incidence of developing head • Possible increased
and neck cancer [35] incidence of bladder cancer
[35]
Erlotinib • Inhibitor of EGFR [7] • Randomized clinical trial in • Dermatologic effects,
2016 demonstrated no including mucositis [7]
improvement in cancer-free
survival [7]
EGFR epidermal growth factor receptor, NSAIDs nonsteroidal anti-inflammatory drugs, PPAR-γ peroxisome prolifera-
tor-activated receptor gamma, EGCG epigallocatechin 3-gallate, VEGFR vascular endothelial growth factor receptor,
NF-κB nuclear factor-kappa B, COX-2 cyclooxygenase-2, TZD thiazolidinediones
challenges. Topical agents may still enter sys- 10. Lee MF, Chan CY, Hung HC, et al. N-acetylcysteine
temic circulation and cause side effects [2]. They (NAC) inhibits cell growth by mediating the EGFR/
Akt/HMG box-containing protein 1 (HBP1) sig-
can produce localized side effects that can limit naling pathway in invasive oral cancer. Oral Oncol.
compliance, as seen with topical bleomycin and 2013;49:129–35.
retinoids. Selective topical application may not 11. Temam S, Kawaguchi H, El-Naggar AK, et al.

target additional at-risk mucosa from field can- Epidermal growth factor receptor copy number altera-
tions correlate with poor clinical outcome in patients
cerization [21]. with head and neck squamous cancer. J Clin Oncol.
Despite the difficulties in demonstrating clini- 2007;25(16):2164–70.
cal effectiveness of these agents in the chemopre- 12. van Zandwijk N, Dalesio O, Pastorino U, et al.

EUROSCAN, a randomized trial of vitamin A
vention of oral cancer, further elucidation of the and N-acetylcysteine in patients with head and
molecular pathways responsible for the stepwise neck cancer or lung cancer. J Natl Cancer Inst.
progression from hyperplasia to carcinoma may 2000;92(12):977–86.
reveal new therapeutic targets and the specific 13. Surber C, Elsner P, Farage MA. Topical applications
and the mucosa. Curr Probl Dermatol. 2011;40:71–82.
populations that may be responsive to such treat- 14. Epstein JB, Wong FL, Millner A, Le ND. Topical
ments. With this deeper understanding of the dis- bleomycin treatment of oral leukoplakia: a ran-
ease, the field of chemoprevention in oral cancer domized double-blind clinical trial. Head Neck.
may enter into an era of precision therapy. 1994;16(6):539–44.
15. Ding Y, Yao H, Yao Y, et al. Protection of dietary
polyphenols against oral cancer. Nutrients.
2013;5:2173–91.
16. Gomes de Moura CF, Noguti J, Pacheco de Jesus GP,
References et al. Polyphenols as chemopreventive agent in oral
carcinogenesis: putative mechanisms of action using
1. Fotedar V, Fotedar S, Seam RK, Gupta MK. Oral in-vitro and in-vivo test systems. Eur J Cancer Prev.
cancer and chemoprevention. Int J Pharm Sci Invent. 2013;22:467–72.
2013;2(2):16–20. 17. Tsao AS, Liu D, Martin J, et al. Phase II randomized,
2. Chau L, Jabra JT, Lai W, et al. Topical agents for oral placebo-controlled trial of green tea extract in patients
cancer chemoprevention: a systematic review of the with high-risk oral premalignant lesions. Cancer Prev
literature. Oral Oncol. 2017;67:153–9. Res (Phila). 2009;2(11):931–41.
3. Hong WK, Lippman SM, Itri LM, et al. Prevention of 18. Rai B, Kaur J, Jacobs R, Singh J. Possible action
second primary tumors with isotretinoin in squamous- mechanism for curcumin in pre-cancerous lesions
cell carcinoma of the head and neck. N Engl J Med. based on serum and salivary markers of oxidative
1990;323(12):795–801. stress. J Oral Sci. 2010;52(2):251–6.
4. Leoncini E, Nedovic D, Panic N, et al. Carotenoid 19. Patel JB, Shah FD, Joshi GM, Patel PS. Clinical sig-
intake from natural sources and head and neck c ancer: nificance of inflammatory mediators in the pathogen-
a systematic review and meta-analysis of epidemio- esis of oral cancer. J Can Res Ther. 2016;12:447–57.
logical studies. Cancer Epidemiol Biomark Prev. 20. Sheth SH, Johnson DE, Kensler TW, Bauman
2015;24(7):1003–11. JE. Chemoprevention targets for tobacco-related head
5. Wilson JC, Murray LJ, Hughes CM, et al. Non- and neck cancer: past lessons and future directions.
steroidal anti-inflammatory drug and aspirin use Oral Oncol. 2015;51:557–64.
and the risk of head and neck cancer. Br J Cancer. 21. Foy JP, Bertolus C, William W, Saintigny P. Oral
2013;108:1178–81. premalignancy: The roles of early detection and
6. Bauman JE, Grandis J. Oral cancer chemopreven- chemoprevention otolaryngol. Clin North Am.
tion – the end of EPOC, the beginning of an epoch of 2013;46(4):579–97.
molecular selection. JAMA Oncol. 2016;2(2):178–9. 22. Koch HF. Biochemical treatment of precancerous oral
7. William W, Papadimitrakopoulou V, Lee JJ, et al. lesions: the effectiveness of various analogues of reti-
Erlotinib and the risk of oral cancer: the erlotinib pre- noic acid. J Maxillofac Surg. 1978;6:59–63.
vention of oral cancer (EPOC) randomized clinical 23. Hong WK, Endicott J, Itri LM, et al. 13-cis-retinoic
trial. JAMA Oncol. 2016;2(2):209–16. acid in the treatment of oral leukoplakia. N Engl J
8. Bhattacharya S. Natural antimutagens: a review. Res J Med. 1986;315(24):1501–5.
Med Plant. 2011;5(2):116–26. 24. Khuri FR, Lee JJ, Lippman SM, et al. Randomized
9. Millea PJ. N-Acetylcysteine: multiple clinical applica- phase III trial of low-dose isotretinoin for preven-
tions. Am Fam Physician. 2009;80(3):265–9. tion of second primary tumors in stage I and II
head and neck cancer patients. J Natl Cancer Inst. 32. Papadimitrakopoulou VA, William W, Dannenberg

2006;98(7):441–50. AJ, et al. Pilot randomized phase II study of cele-
25. Kadakia S, Badhey A, Milam M, et al. Topical oral coxib in oral premalignant lesions. Clin Cancer Res.
cavity chemoprophylaxis using isotretinoin rinse: a 2008;14(7):2095–101.
15-year experience. Laryngoscope. 2017;127:1595–9. 33. Bertagnolli MM. Chemoprevention of colorectal can-
26. Potter JD. The failure of cancer chemoprevention. cer with cyclooxygenase-2 inhibitors: two steps for-
Carcinogenesis. 2014;35(5):974–82. ward, one step back. Lancet Oncol. 2007;8:439–43.
27. Bjelakovic G, Nikolova D, Gluud LL, Simonetti RG, 34. Burotto M, Szabo E. PPARγ in head and neck cancer
Gluud C. Mortality in randomized trials of antioxi- prevention. Oral Oncol. 2014;50(10):924–9.
dant supplements for primary and secondary preven- 35. Govindarajan R, Siegel ER. The effect of exposure to
tion systematic review and meta-analysis. JAMA. thiazolidinediones on the development of head-and-
2007;297(8):842–57. neck cancer in patients with diabetes mellitus. Transl
28. Lu R, Dan H, Wu R, et al. Lycopene: features and Res Oral Oncol. 2017;2:1–5.
potential significance in the oral cancer and precan- 36. Rosin MP, Cheng X, Poh C, et al. Use of allelic loss
cerous lesions. J Oral Pathol Med. 2011;40:361–8. to predict malignant risk for low-grade oral epithelial
29.
Lin DT, Subbaramaiah K, Shah JP, et al. dysplasia. Clin Cancer Res. 2000;6:357–62.
Cyclooxygenase-2: a novel molecular target for the 37. Bhatia A, Burtness B. Novel molecular targets for
prevention and treatment of head and neck cancer. chemoprevention in malignancies of the head and
Head Neck. 2002;24:792–9. neck. Cancers. 2017;9:113.
30. Shi J, Leng W, Zhao L, et al. Nonsteroidal anti-
38. Lodi G, Franchini R, Warnakulasuriya S, et al.

inflammatory drugs using and risk of head and neck Interventions for treating oral leukoplakia to prevent
cancer: a dose–response meta analysis of prospective oral cancer. Cochrane Database Syst Rev. 2016;7:Art.
cohort studies. Oncotarget. 2017;8(58):99066–74. No.: CD001829.
31.
Mushine JL, Atkinson JC, Greer RO, et al. 39. Chainani-Wu N. Safety and anti-inflammatory activ-
Randomized, double-blind, placebo-controlled phase ity of curcumin: a component of tumeric (Curcuma
IIB trial of the cyclooxygenase inhibitor ketorolac longa). J Alternat Compl Med. 2003;9(1):161–8.
as an oral rinse in oropharyngeal leukoplakia. Clin
Cancer Res. 2004;10:1565–73.
Diagnostic Adjuncts for Screening
and Surveillance in Head and Neck
3
Cancer
James Murphy and Mohammed Qaisi
3.1 Introduction be diagnosed on average at a statistically signifi-

cant earlier clinical and pathological stage [4]. In
Head and neck cancer is the seventh most com- contradiction to this, though, the United States
mon cancer worldwide and the incidence is Preventative Services Task Force (USPSTF) cur-
increasing in the developed world largely due to rently states evidence is insufficient to assess the
the human papilloma virus (HPV) epidemic and balance of benefits and harms of screening for
its effect on oropharyngeal carcinoma [1]. The oral cancer in asymptomatic adults [5]. The
majority of head and neck cancers are diagnosed American Academy of Family Physicians has
at a late stage which has implications for progno- stated it concurs with the USPSTF recommenda-
sis. The eighth edition of the American Joint tion [6]. The USPSTF position statement, how-
Commission on Cancer staging for head and ever, does not apply to dental professionals or
neck cancer factored HPV status into the staging otolaryngologists. The American Cancer Society,
algorithm, which would lead to downstaging of American Head & Neck Society, and the
some p16-positive tumors, but despite this, a sig- American Dental Association all stress the impor-
nificant proportion of head and neck cancers tend tance of regular oral cavity examination looking
to present at later stages [2]. Screening programs for suspicious lesions without advocating for any
have been shown to be valuable in certain cancers diagnostic screening adjuncts [7, 8].
like cervical carcinoma and colon cancer. With The difficulty that arises with visual examina-
regards to head and neck cancer and specifically tion is potentially malignant lesions can look non
oral cancer, screening by visual exam has only worrisome on clinical exam, and benign lesions
been shown to be of epidemiological benefit can look suspicious, especially to the nontrained
when undertaken in high-risk populations such as eye or those who infrequently see head and neck
in the Indian subcontinent [3]. In lower risk pop- cancer. The gold standard to diagnose head and
ulations, malignant lesions detected during a neck cancer is biopsy and histopathological eval-
non-symptom-driven exam have been shown to uation. Tissue biopsy is not without its inherent
issues, among which include sampling error by
the clinician obtaining the biopsy and interpretive
J. Murphy (*) · M. Qaisi error by the pathologist [9]. Pathological inter-
Division of Oral and Maxillofacial Surgery, pretive error is particularly high when dysplastic
Department of Surgery, John H. Stroger, Jr. Hospital lesions are being evaluated [10]. Furthermore,
of Cook County, Chicago, IL, USA
e-mail: james.murphy@cookcountyhhs.org; mqaisi@ patients tend to resent tissue biopsy and may be a
cookcountyhhs.org factor in late presentation of head and neck can-

https://doi.org/10.1007/978-3-030-30094-4_3
24 J. Murphy and M. Qaisi
cer for fear of having to undergo tissue biopsy. As have good sensitivity rates but questionable speci-
a means of mitigating this, various diagnostic and ficity rates in the detection of oral cancer and oral
screening adjuncts have been created to diagnose premalignant lesions [11, 12]. Even if it has a
and facilitate surveillance of the head and neck potential benefit as a diagnostic aid in malignant
cancer patient. lesion detection, research data suggest it is of very
A diagnostic and screening adjunct needs to questionable benefit in detecting dysplasia [13].
conform to a number of elements if it is to be The application method of toluidine blue to
considered useful and viable. The most important evaluate for suspicious lesions within the oral
ones are: cavity differs slightly between practitioners. The
authors follow the following regimes depending
–– Simple to use. on if an isolated area is being evaluated or the
–– Inexpensive. entire oral cavity is being assessed. With regards
–– Minimally invasive. to when an isolated area or lesion of the oral cav-
–– Detects early disease that is likely to ity is being evaluated, the authors wipe the area
progress. gently with a moist 0.9% normal saline-soaked
–– High sensitivity with low false-negative gauze to remove any surface debris. Acetic acid
results. 1% is used to gently wipe the area with a cotton
tip applicator for 20 s, and then 1% toluidine blue
Unfortunately, the current gold standard of the is subsequently applied with a cotton-tipped
scalpel biopsy does not meet a number of these applicator for 20 s. The area is then assessed with
criteria. From the physician’s perspective, its areas of greater intensity of blue staining being
accuracy is overshadowed by patients’ misgiv- regarded as more suspicious and potentially
ings about its invasiveness. It is hoped that the deserving of tissue biopsy. Toludine blue can also
diagnostic and screening adjuncts described be used to evaluate the entire oral cavity. The
below may conform more to the elements of an authors’ regime for this is the patient rinses with
ideal test, and that combined with a thorough 15 ml of 0.9% normal saline for 30 s to remove
clinical exam, they may be helpful in early detec- debris. Following this, the patient rinses with 1%
tion and the decision making of obtaining a scal- acetic acid 15 ml for 30 s and then after expecto-
pel biopsy. These techniques will be discussed in rating the patient rinses with 15 ml of 1% tolu-
the following sections. idine blue for 30 s. Finally the patient once again
rinses 1% acetic acid for 30 s to remove mechani-
cally retained stain. The oral cavity is then
3.2 Vital Staining assessed with areas of more intense blue staining
deserving of greater evaluation and consideration
3.2.1 Toludine Blue for tissue biopsy.
From a practical perspective, toluidine blue
Toludine blue, also known as tolonium chloride, may be of benefit as a diagnostic adjunct when
is a vital dye of the thiazine group which has an used to direct biopsy of a clinically evident suspi-
affinity to bind to the nucleic acids of DNA. This cious lesion isolating areas of more intense stain-
premise forms the logic of using it as a diagnostic ing and theoretically minimizing selection error
and screening aid in head and neck cancer, in that or for monitoring for recurrent disease in altered
it preferentially stains neoplastic cells with their oral mucosa tissue. Epstein et al. recommended
comparatively high turnover rate. Unfortunately, the use of toluidine blue as a diagnostic screening
therein lies the main disadvantage of toludine adjunct in high-risk populations when used by
blue as it cannot reliably distinguish between expert providers [14]. This caveat limits its use-
inflammatory, regenerative epithelium, and fulness as a diagnostic screening adjunct among
exposed connective tissue, which also have high the majority of community-based practitioners as
cell turnover rates. Studies show toluidine blue to it is of questionable value in detecting oral
3 Diagnostic Adjuncts for Screening and Surveillance in Head and Neck Cancer 25
p remalignant disease in this population [15]. The a

reported ranges of sensitivity and specificity of
toluidine blue in the literature is not ideal at
57–81% and 56–67%, respectively [16]. It has
also been noted that toluidine blue becomes more
sensitive and specific as lesion severity of dyspla-
sia increases or in frankly malignant lesions. This
b
is a limitation that needs to be appreciated when
used as a diagnostic adjunct. Toludine blue use as
an adjunct does not eliminate the need for scalpel
biopsy.
3.2.2 Lugol’s Iodine
Another screening adjunct is Lugol’s iodine. Fig. 3.1 (a) Lesion floor of mouth before lugols applica-
tion. (b) Lesion floor of mouth after lugols application
Lugol’s iodine aids in detecting suspicious
lesions in nonkeratinized mucosa based on the
fact that glycogen within the cell binds to the with 0.9% normal saline for 30 s. Cotton-tipped
iodine component of this vital stain. Preneoplastic applicators are then used to apply 1% lugol’s
cells tend to be more metabolically active with a iodine to individual nonkeratinized anatomic
high cell turnover, thus leaving these cells with subunits of the oral cavity. Multiple passes of
depleted glycogen stores when compared to nor- lugol’s iodine-soaked cotton-tipped applicators
mal tissues. Because of the depleted glycogen, are generally used on an individual anatomic sub-
these cells do not stain with the lugol’s iodine and unit. Following application of the lugol’s iodine,
remain relatively white compared to normal tis- suction is used to remove any excess pooled
sue which stains mahogany brown. lugol’s iodine. The anatomic subunit is then eval-
Lugol’s iodine has been shown to be useful as uated. A mahogany brown color is considered
an intraoperative guide in determining where to nonsuspicious but an off-white area should be
resect head and neck malignant and dysplastic viewed with suspicion and consideration given to
lesions [17] (Fig. 3.1a, b). In that study by tissue biopsy.
McMahon et al., the rate of carcinoma, carci- Lugol’s iodine is relatively inexpensive and
noma in situ (CIS), or dysplasia present at mar- easy to use and may be considered as a diagnostic
gins was 32% following resection with a 1 cm adjunct in conjunction with a thorough exam for
margin based solely on visual clues. When lugol’s screening and surveillance of head and neck can-
iodine was used to guide the resection margins, cer. Like toluidine blue, lugol’s iodine does not
the rate of dysplasia or CIS at the margin dropped replace a thorough clinical exam or the need for a
to 4%. In addition to aiding with margins, lugol’s scalpel biopsy for clinically suspicious lesions.
iodine may be helpful as a screening adjunct in
the identification and diagnosis of oral malignant
and premalignant lesions [18]. In a comparative 3.3 Visual Light-Based Adjuncts
study, the sensitivity/specificity of toluidine blue
was 0.93/0.63 compared to 0.88/0.84 with lugol’s 3.3.1 Chemiluminescence
iodine [19]. This higher specificity of lugol’s
iodine is a noteworthy finding. ViziLite® is the most well-known within this cat-
Like toludine blue, there is practitioner varia- egory, although it now comes as ViziLite Plus®
tion in the application of lugol’s iodone. The which requires its use in combination with tolu-
authors’ protocol involves rinsing the oral cavity idine blue. Microlux DL® is another example,
although it is not well researched in the peer- better at detecting and evaluating suspicious
reviewed literature. white lesions as opposed to red lesions [23]. For
The method to use chemiluminescene in the many, chemiluminescence is currently not a ben-
evaluation of the oral cavity for suspicious lesions eficial diagnostic adjunct [24].
is as follows: The oral cavity is rinsed with 1%
acetic acid for 1 min to remove surface glycopro-
tein and cause cellular dehydration which facili- 3.3.2 Tissue Fluorescence Imaging
tates exposure of the cellular elements to the
luminescence. Upon completion of the prerinse, The ideology behind this diagnostic adjunct is
the chemiluminescent blue-white light in the that abnormal tissues of the oral cavity have an
spectral wavelength of 490–510 nm is applied to altered structure (e.g., hyperkeratosis, hyperchro-
evaluate the tissues. The blue-white light is gen- matin, increased cellular/nuclear pleomorphism),
erated either via an interaction between acetyl- altered metabolism, and potentially changes in
salicylic acid and hydrogen peroxide (ViziLite the subepithelial stroma which subsequently alter
Plus®) or by a battery-powered light-emitting the interaction of the tissue with light compared
diode (Microlux DL®) (Fig. 3.2a, b). Upon exam- to normal oral mucosal tissues. Various systems
ination of the oral mucosa in a darkened room are on the market including VELscope®, Sapphire
using manufacturer specific eyewear, normal epi- Plus®, Identafi®, Bio/Screen®, DOE SE Kit®,
thelium appears light blue, whereas abnormal OrallD®, and ViziLite PRO®, with VELscope®
epithelium appears white. Toludine blue is then being the most studied [25] (Fig. 3.3a, b).
applied to the lesion that appears abnormal with The process to use tissue fluorescence imag-
the ViziLite Plus® system. ing to detect suspicious lesions of the oral cavity
Unfortunately, the sensitivity and specificity involves the application of an intense blue light in
of chemiluminescence is relatively poor with a the wavelength of 390–460 nm and viewing the
lack of good evidence to support the use of this effect on tissue through the device viewer or spe-
adjunct in screening and surveillance of head and cial eyewear to create a narrow band filter.
neck cancer at present [20, 21]. Mehrotra et al. Normal oral mucosa appears pale green using
noted 102 innocuous clinically evident lesions on this technique, but abnormal tissues appear dark
conventional oral exam. These lesions were then due to a loss of fluorescence. The phenomenon of
examined with ViziLite following which the diascopic fluorescence has been described with
lesions were biopsied. Histopathological exami- this modality [26]. Tissues with loss of fluores-
nation revealed three dysplasias and one malig- cence but which completely blanch on the
nancy, none of which were identified with the application of pressure and return to normal fluo-
ViziLite [22]. Chemiluminescence appears to be rescence pattern should be considered negative
a b
Fig. 3.2 (a) Clinical suspicious lesion before Microlux Dx use. (b) Microlux Dx highlighting suspicious lesion.
(Courtesy of Addent Inc.)
a b
Fig. 3.3 (a) Clinical photo of palate with possible suspicious lesion. (b) VELscope highlighting suspicious lesion of
palate. (Courtesy of LED Dental)
for loss of fluorescence. Therefore, a degree of screening technique, is routinely heralded as the
experience and interpretive acumen is necessary champion screening test for malignancy. The
in utilizing this diagnostic and screening adjunct. principle behind this screening method has been
Tissue fluorescence imaging has been shown adapted for the oral cavity to obtain transepithe-
in a retrospective case–control observational lial tissue samples using a specialized brush
study to significantly reduce the rate of local which is subsequently analyzed for cytomorpho-
recurrence when used as an intraoperative surgi- logical abnormalities. Early following the intro-
cal aid to determine resection margins for high- duction of this diagnostic adjunct, there were
grade dysplastic lesions and early-stage oral multiple peer-reviewed articles extolling its vir-
squamous cell carcinoma [27]. Encouraging data tue. Among its fanfare introduction, it even fea-
are present in the literature with regard to using tured on the cover of the Journal of the American
this technique in screening and surveillance of Dental Association in October 1999. In the
oral cancer; however, it is limited by false- accompanying article, Sciubba reported a 100%
positive results [11, 28, 29]. More worrisome sensitivity and specificity with this diagnostic
than the false positives with this diagnostic adjunct following a multicenter trial [31]. The
adjunct is the fact that keratin is autofluorescent titles of letters to the Editor of the Journal of the
which has implications that potentially malignant American Dental Association in 2002 responding
and frankly malignant lesions may be missed in to negative criticism of the technique included
hyperkeratotic tissues such as proliferative ver- “Brush Biopsy ‘Saves Lives,” “Brush Biopsy
rucous leukoplakia [30]. The importance of sub- ‘Bridges the Gap,” and “Overwhelmingly
jective interpretation and the recognition that it is Positive” [32–34]. Unfortunately, the initial hype
a diagnostic adjunct to be used in combination masked a significant flaw that permeated early
with a thorough clinical oral exam cannot be studies looking at this adjunct, in that a scalpel
overstated. biopsy was only performed on those with an
abnormal or positive result. Therefore, confi-
dence in the negative predictive value of this
3.4 Exfoliative-Based adjunct is compromised. Currently it is believed
Techniques the sensitivities and specificities vary between
71–100% and 27–100%, respectively, for this
3.4.1 Brush Biopsy diagnostic adjunct [35].
The OralCDx Brush Test® system obtains a
The screening of cervical cancer with the Pap full transepithelial specimen sample using a sam-
Smear test, a brush biopsy-based cytology pling brush. The harvested specimen can be
scanned and analyzed microscopically by means is obtained, but unlike the previously discussed
of a computer-based imaging system, OraScan®, cytological evaluation of the sample with the aid
to get one of the following results—“negative or of Papanicolaou staining, Feulgen staining is
benign”, “positive”, or “atypical”. The test has used for DNA-image cytometry. The assessment
also been adapted for the evaluation of laryngo- of the stained specimen with 566 nm monochro-
pharyngeal lesions based on the same principles matic light allows an analysis of DNA-aneuploidy
(EndoCDx LP—Laryngeal®). Abnormal or posi- of the specimen. The findings of the test may
tive results should undergo scalpel biopsy and document DNA-diploidy, DNA-polyploidy, or
histopathological evaluation. DNA-aneuploidy, with DNA-aneuploidy needing
When properly performed, this diagnostic and further evaluation and workup. A recent meta-
screening adjunct has the potential to be helpful analysis reported relatively good sensitivity
and accurate. Research experience and results (89%) and specificity (99%) along with an
suggest that the technique is good at diagnosing impressive diagnostic odds ratio of 446, but the
clinically suspicious lesions correctly which will fact only five DNA-image cytometry studies met
ultimately need a scalpel biopsy, but the fact the the inclusion criteria of the meta-analysis cannot
majority of studies do not take scalpel biopsy of be ignored [38]. Therefore, even though the ide-
samples taken from questionable sites or sites of ology behind this technique makes logical sense,
low clinical suspicion, for which this technique it is understudied at present to be recommended
should be primarily used for, means an element as a reliable diagnostic and screening adjunct for
of doubt over the true usefulness of this product head and neck cancer.
remains. The encouraging results for brush
biopsy in a recent meta-analysis from the
Cochrane Collaboration cannot fully allay this 3.5 Molecular Biology
concern [36]. The technique may have usefulness
but further rigorous study with comparison to the Advancements in molecular biology and data
gold standard of scalpel biopsy and histopatho- from the Human Genome Project and Cancer
logical examination from more innocuous lesions Genome Atlas open the possibility of diagnosis
needs to be performed and reported before fully and screening for head and neck cancer with the
embracing this technique as a diagnostic and application of this knowledge. Saliva is seen as a
screening adjunct. readily available source from which to obtain
samples which can be tested for potential bio-
markers for oral malignant and premalignant
3.4.2 DNA-Image Cytometry conditions. Potential biomarkers include but are
not limited to proteinaceous material, DNA, mes-
Like OralCDx Brush Test®, DNA-image cytom- senger RNA, microRNA, long noncoding RNA,
etry is a computer-assisted analysis of exfoliative nonorganic compounds, and metabolites. To date,
cells. DNA-image cytometry has been proposed over 100 potential biomarkers have been reported
as a diagnostic adjunct capable of very early in the literature that may be useful in detecting
detection of malignant transformation of squa- oral/oropharyngeal malignant and premalignant
mous epithelial cells. The method behind this conditions [39]. Unfortunately a lack of standard-
involves the assessment of exfoliative cells for ization with regards to collection, storage, and
chromosomal DNA-aneuploidy, an abnormal processing of samples as well as the natural vari-
nuclear DNA finding. It is theorized that the find- ability of potential salivary biomarkers between
ing of DNA-aneuploidy is a sensitive and effec- subjects with oral/oropharyngeal malignant and
tive method of detecting very early malignant premalignant conditions compared to those with-
transformation within head and neck epithelium out, together with a lack of validity of this method
before cytology and histopathology can detect mean that at present salivary biomarkers remain a
any such changes [37]. A brush biopsy specimen focus of research interest without any evidence to
support clinical application [40]. Screening of ders of the oral cavity. The accompanying article
head and neck cancer has been performed on presented one good practice statement and six
exfoliated cells and biopsy samples using molec- clinical recommendations formulated by an
ular biology techniques analyzing them for loss expert panel convened by the American Dental
of heterozygosity at various chromosome loca- Association Council on Scientific Affairs and the
tions using microsatellite markers [41]. Center for Evidence-Based Dentistry [44]. The
Unfortunately despite some promising candidate panel’s good practice statement advocated that
biomarkers being identified, this method at pres- clinicians should obtain an updated medical,
ent is not clinically useful as a diagnostic or social, and dental history and perform an intra-
screening adjunct in head and neck cancer. oral and extraoral conventional visual clinical
exam in all adult patients. The synopsis of the
clinical recommendations is that no available
3.6 In Vivo Microscopy diagnostic adjuncts demonstrated sufficient diag-
nostic test accuracy to support their use in evalu-
This modality, which is a potential diagnostic and ating oral cavity lesions. The panel did state that
screening adjunct for head and neck cancer, is in the exceptional circumstance of a patient
currently undergoing scientific investigation. It refusing a scalpel biopsy or in cases where geo-
incorporates some of the features of existing graphical limitations hinder access to care, cyto-
adjuncts with high-resolution microscopic imag- logical testing may be used to initiate the
ing. Issues with the various modalities within this diagnostic process until a biopsy can be per-
potential group of adjuncts include minimal data formed. A flow chart presented by the American
in the peer-reviewed literature and expense. Dental Association Council on Scientific Affairs
Examples include multimodal imaging, optical and the Center for Evidence-Based Dentistry,
coherence tomography, reflectance confocal which can be used as a guideline, of how best to
microscopy, and multiphoton microscopy [42]. incorporate the panel’s good practice statement
Multimodal imaging, reflectance confocal and clinical recommendations in the evaluation
microscopy, and multiphoton microscopy make of the patient with a potentially malignant condi-
use out of fluorescence in order to highlight and tion of the oral cavity (Fig. 3.4).
evaluate areas/cells of interest. Optical coherence Similarly the American Head & Neck Society
tomography is a technique in clinical use in oph- position statement on early detection of prema-
thalmology [43]. The technique may have value lignant oral cancer states that the gold standard
in looking at epithelial architecture changes such for detecting potentially premalignant lesions is a
as epithelial thickness, basement membrane con- thorough examination combined with biopsy
tinuity, and rete ridge arrangement that may be where needed [45]. Their conclusion with regards
used to detect abnormalities suggestive of dys- to the different screening adjuncts discussed in
plasia and/or carcinoma. No in vivo microscopy this chapter is that they have not yet become
technique is currently in clinical use as a diagnos- widely adopted as part of the existing standard of
tic and screening adjunct for head and neck can- care, and they may warrant additional consider-
cer, but they are the source of intense research ation in the future.
interest. The authors’ thoughts are consistent in that
there is no substitute for a good oral head and
neck exam; however, if screening adjuncts help
3.7 Putting It All Together or motivate dental providers to examine patients,
then they may be utilized. Caution has to be taken
In October 2017, the Journal of the American when interpreting the results of these diagnostic
Dental Association once again featured a cover adjuncts due to the limitations discussed and with
story on evaluating potentially malignant disor- the understanding that an exam and biopsy super-
B. A clinically evident, seemingly

A. No clinically evident C. A clinically evident, suspicious lesion
innocuous lesion (not suspected
lesion or symptoms (suspected to be either a PMD or malignant disorder)
to be malignant)
Periodically follow up
with patient to Perform a biopsy of lesion
determine or provide immediate
the need for referral to a specialist Should a patient decline
further evaluation a biopsy or referral
No further action is Lesion Lesion persists Definitive Use cytologic adjunct to

necessary resolves or progresses Definitive triage patient and
diagnosis of
at this time (either sponta- or clinical diagnosis of provide additional
no malignancy
neously or diagnosis of PMD or lesion assessment
after PMD cannot malignancy
treatment) be ruled out
Positive or
Negative
atypical test
test result
result
Fig. 3.4 American Dental Association Council on Scientific Affairs and the Center for Evidence-Based Dentistry flow
chart
Patient undergoes
conventional visual and
tactile examination
Adjuncts used to evaluate

the lesion clinically and
triage patient
Positive result: Higher Negative result: Lower

probability of having a probability of having a
potentially malignant disorder. potentially malignant disorder.
Biopsy and histopathological

Follow up to confirm the
assessment performed to
negative result.
provide definitive diagnosis
Positive diagnosis Negative diagnosis
Fig. 3.5 Algorithm for the Incorporation of Diagnostic Adjuncts into Clinical Practice
sede diagnostic adjuncts. A suggestive algorithm adjunct being used, and if any doubt exists
for the incorporation of diagnostic adjuncts is regarding a suspicious oral lesion, an incisional
shown in Fig. 3.5. It is important to understand biopsy with histopathological examination is the
the risk of false negatives with any diagnostic current standard of care.
3.8 Conclusion cal recommendations regarding screening for oral

squamous cell carcinomas. J Am Dent Assoc.
2010;141(5):509–20.
To the head and neck surgeon, the potential util- 9. Lee JJ, Hung HC, Cheng SJ, Chiang CP, Liu BY, Yu
ity of a reliable, valid, and accurate diagnostic CH, Jeng JH, Chang HH, Kok SH. Factors associated
and screening adjunct for the head and neck can- with underdiagnosis from incisional biopsy of oral
leukoplakic lesions. Oral Surg Oral Med Oral Pathol
cer patient is clear. Current adjuncts suffer from Oral Radiol Endod. 2007;104:217–25.
suboptimal sensitivity, poor negative predictive 10. Abbey LM, Kaugars GE, Gunsolley JC, Burns JC,
values, expense, and lack of quality supportive Page DG, Svirsky JA, Eisenberg E, Krutchkoff DJ,
research in the peer-reviewed literature corrobo- Cushing M. Intraexaminer and interexaminer reli-
ability in the diagnosis of oral epithelial dysplasia.
rating manufacturer or marketing claims to allow Oral Surg Oral Med Oral Pathol Oral Radiol Endod.
them to receive a strong recommendation. Some 1995;80:188–91.
have merits, but unanswered questions remain 11. Rosenberg D, Cretin S. Use of meta-analysis to evalu-
that need elucidating before they can be advo- ate tolonium chloride in oral cancer screening. Oral
Surg Oral Med Oral Pathol. 1989;67:621–7.
cated for. At present a thorough head and neck
12. Warnakulasuriya KA, Johnson NW. Sensitivity and
history and examination by a competent, experi- specificity of Orascan® toluidine blue mouthrinse
enced physician/dentist with selective scalpel in the detection of oral cancer and precancer. J Oral
biopsy of suspicious lesions is the best method Pathol Med. 1996;25:97–103.
13. Onofre MA, Sposto MR, Navarro CM. Reliability
we have for diagnosing and screening of head
of toluidine blue application in the detection of oral
and neck cancer. epithelial dysplasia and in situ and invasive squamous
cell carcinomas. Oral Surg Oral Med Oral Pathol Oral
Acknowledgement The authors have no conflict of inter- Radiol Endod. 2001;91:535–40.
est to declare. 14. Epstein JB, Güneri P. The adjunctive role of toluidine
blue in detection of oral premalignant and malignant
lesions. Curr Opin Otolaryngol Head Neck Surg.
2009;17:79–87.
References 15. Awan K, Yang Y, Morgan P, Warnakulasuriya

S. Utility of toluidine blue as a diagnostic adjunct in
1. WHO. World cancer report 2014. Geneva: World the detection of potentially malignant disorders of the
Health Organization; 2014. oral cavity—a clinical and histological assessment.
2. AJCC. AJCC cancer staging manual. 8th ed. Oral Dis. 2012;18:728–33.
New York, NY: Springer; 2016. 16.
Awadallah M, Idle M, Patel K, Kademani
3. Sankaranarayanan R, Ramadas K, Thomas G, D. Management update of potentially premalignant
Muwonge R, Thara S, Mathew B, Rajan B, oral epithelial lesions. Oral Surg Oral Med Oral
Trivandrum Oral Cancer Screening Study Group. Pathol Oral Radiol. 2018;125:628–36.
Effects of screening on oral cancer mortality in 17. McMahon J, Devine JC, McCaul JA, McLellan DR,
Kerala, India: a cluster randomized controlled trial. Farrow A. Use of Lugol’s iodine in the resection of
Lancet. 2005;365:1927–33. oral and oropharyngeal squamous cell carcinoma. Br
4. Holmes JD, Dierks EJ, Homer LD, Potter BE. Is J Oral Maxillofac Surg. 2010;48:84–7.
detection of oral and oropharyngeal squamous cell 18. Petruzzi M, Lucchese A, Baldoni E, Grassi FR,

cancer by a dental care provider associated with a Serpico R. Use of Lugol’s iodine in oral cancer diag-
lower stage at diagnosis. J Oral Maxillofac Surg. nosis: an overview. Oral Oncol. 2010;46:811–3.
2003;61:285–91. 19. Epstein JB, Scully C, Spinelli J. Toluidine blue and
5. https://www.uspreventiveservicestaskforce.org/ Lugol’s iodine application in the assessment of oral
Page/Document/RecommendationStatementFinal/ malignant disease and lesions at risk of malignancy. J
oral-cancer-screening1. Oral Pathol Med. 1992;21:160–3.
6. h t t p s : / / w w w. a a f p . o rg / p a t i e n t - c a r e / c l i n i c a l - 20. Rashid A, Warnakulasuriya S. The use of light-based
recommendations/all/oral-cancer.html. (optical) detection systems as adjuncts in the detec-
7. American Cancer Society. American Cancer Society tion of oral cancer and oral potentially malignant
guidelines for the early detection of cancer. Atlanta, disorders: a systematic review. J Oral Pathol Med.
GA: American Cancer Society; 2013. Accessed at 2015;44:307–28.
http://www.cancer.org/healthy/findcancerearly/can- 21. Lingen MW, Kalmar JR, Karrison T, Speight

cerscreeningguidelines/american-cancer-society- PM. Critical evaluation of diagnostic aids for the
guidelines-for-the-early-detection-of-cancer. detection of oral cancer. Oral Oncol. 2008;44:10–22.
8. Rethman MP, Carpenter W, Cohen EE, Epstein J, 22. Mehrotra R, Singh M, Thomas S, Nair P, Pandya S,
Evans CA, Flaitz CM, et al. Evidence-based clini- Nigam NS, Shukla P. A cross-sectional study evalu-
ating chemiluminescence and autofluorescence in nosis. A systematic review of the literature. JADA.
the detection of clinically innocuous precancer- 2008;139:896–905.
ous and cancerous oral lesions. J Am Dent Assoc. 36. Macey R, Walsh T, Brocklehurst P, Kerr AR, Liu JL,
2010;141:151–6. Lingen MW, Ogden GR, Warnakulasuriya S, Scully
23. Epstein JB, Gorsky M, Lonky S, Silverman S Jr,
C. Diagnostic tests for oral cancer and potentially
Epstein JD, Bride M. The efficacy of oral lumenos- malignant disorders in patients presenting with clini-
copy (ViziLite) in visualizing oral mucosal lesions. cally evident lesions. Cochrane Database Syst Rev.
Spec Care Dentist. 2006;26:171–4. 2015;5:CD010276.
24. Oh ES, Laskin DM. Efficacy of the ViziLite system 37. Kämmerer PW, Koch FP, Santoro M, Babaryka G,
in the identification of oral lesions. J Oral Maxillofac Biesterfeld S, Brieger J, Kunkel M. Prospective,
Surg. 2007;65:424–6. blinded comparison of cytology and DNA-image
25. Huber MA. Adjunctive diagnostic techniques for oral cytometry of brush biopsies for early detection of oral
and oropharyngeal cancer discovery. Dent Clin N Am. malignancy. Oral Oncol. 2013;49:420–6.
2018;62:59–75. 38. Ye X, Zhang J, Tan Y, Chen G, Zhou G. Meta-analysis
26. Bhatia N, Matias MAT, Farah CS. Assessment of a of two computer-assisted screening methods for
decision making protocol to improve the efficacy of diagnosing oral precancer and cancer. Oral Oncol.
VELscope™ in general dental practice: a prospective 2015;51:966–75.
evaluation. Oral Oncol. 2014;50:1012–9. 39. Cheng YS, Rees T, Wright J. A review of research on
27. Poh CF, Anderson DW, Durham JS, Chen J, Berean salivary biomarkers for oral cancer detection. Clin
KW, MacAulay CE, Rosin MP. Fluorescence Transl Med. 2014;3:3.
visualization-guided surgery for early-stage oral 40. Kaur J, Jacobs R, Huang Y, Salvo N, Politis C. Salivary
cancer. JAMA Otolaryngol Head Neck Surg. biomarkers for oral cancer and pre-cancer screening:
2016;142:209–16. a review. Clin Oral Investig. 2018;22:633–40.
28. Leston JS, Dios PD. Diagnostic clinical aids in oral 41. Graveland AP, Bremmer JF, de Maaker M, Brink

cancer. Oral Oncol. 2010;46:418–22. A, Cobussen P, Zwart M, Braakhuis BJ, Bloemena
29.
Awan KH, Morgan PR, Warnakulasuriya E, van der Waal I, Leemans CR, Brakenhoff
S. Evaluation of an autofluorescence based imaging RH. Molecular screening of oral precancer. Oral
system (VELscope™) in the detection of oral poten- Oncol. 2013;49:1129–35.
tially malignant disorders and benign keratoses. Oral 42. Yang EC, Tan MT, Schwarz RA, Richards-Kortum
Oncol. 2011;47:274–7. RR, Gillenwater AM, Vigneswaran N. Noninvasive
30. Pavlova I, Williams M, El-Naggar A, Richards-
diagnostic adjuncts for the evaluation of potentially
Kortum R, Gillenwater A. Understanding the bio- premalignant oral epithelial lesions: current limita-
logical basis of autofluorescence imaging for oral tions and future directions. Oral Surg Oral Med Oral
cancer detection: high-resolution fluorescence Pathol Oral Radiol. 2018;125:670–81.
microscopy in viable tissue. Clin Cancer Res. 43. Adhi M, Duker JS. Optical coherence tomogra-

2008;14:2396–404. phy – current and future applications. Curr Opin
31. Sciubba JJ. Improving detection of precancerous and Ophthalmol. 2013;24:213–21.
cancerous oral lesions. Computer-assisted analysis 44. Lingen MW, Abt E, Agrawal N, Chaturvedi AK,

of the oral brush biopsy US Collaborative OralCDx Cohen E, D’Souza G, Gurenlian J, Kalmar JR, Kerr
Study Group. J Am Dent Assoc. 1999;130:1445–57. AR, Lambert PM, Patton LL, Sollecito TP, Truelove
32. Eisen D. Brush biopsy ‘saves lives’. J Am Dent Assoc. E, Tampi MP, Urquhart O, Banfield L, Carrasco-
2002;133:688. Labra A. Evidence-based clinical practice guideline
33. Sciubba JJ. Brush biopsy ‘bridges the gap’. J Am for the evaluation of potentially malignant disorders
Dent Assoc. 2002;133:692. in the oral cavity: a report of the American Dental
34. Farran H. Overwhelmingly positive. J Am Dent
Association. J Am Dent Assoc. 2017;148:712–27.
Assoc. 2002;133:692. 45. https://www.ahns.info/wp-content/uploads/2017/05/
35. Patton LL, Epstein JB, Kerr AR. Adjunctive tech- Early-Detection-of-Pre-Malignant-Oral-Cancer-
niques for oral cancer examination and lesion diag- FINAL.pdf.
Current Updates in Staging
and Prognosis in Oral Cancer
4
Paul Covello and D. David Kim
4.1 Introduction 4.2 Changes to the American

Joint Committee on Cancer
Cancer staging helps define tumor burden, pre- (AJCC) Staging System
dict prognosis, direct treatment, evaluate out-
comes, and guide research initiatives. Ultimately, The TNM cancer classification system was devel-
the system is used to improve provider-to-patient oped in the 1940s by Pierre Denoix. Under his
and provider-to-provider communication. Oral leadership, the Union for International Cancer
squamous cell carcinoma (OSCC) has been reli- Control (UICC) established the Committee on
ably staged using the TNM system, which has Clinical Stage Classification, which continued to
been developed and maintained by the American develop the system. Since its foundation in 1959,
Joint Committee on Cancer (AJCC) and the the American Joint Committee on Cancer (AJCC)
Union for International Cancer Control (UICC). has also focused on defining and standardizing
In this update, data supporting the recent changes cancer classification. The first cancer staging
to the eighth edition of the AJCC staging system manual from the AJCC was released in 1977. In
for OSCC, as well as other important prognostic 1987, the UICC and AJCC unified their TNM
considerations, will be presented. classification systems. Revisions are made every
6–8 years to accommodate advances in cancer
research. In 2017, the eighth edition of the Cancer
Staging Manual by the AJCC was released.
P. Covello In the latest edition, depth of invasion (DOI)
and extranodal extension (ENE) are used to fur-
Geisinger Health, Wilkes-Barre, PA, USA
ther define staging categories. Essentially, the T
Division of Head and Neck Surgical Oncology and
stage increases by one for every 5 mm of tumor
Microvascular Reconstruction, Department of Oral
and Maxillofacial Surgery, Louisiana State University DOI until ≥10 mm, and the pathologic N stage
Health Sciences Center, Shreveport, LA, USA increases by one with ENE. Of note, infiltration
e-mail: pcovello@geisinger.edu of tumor cells into the extrinsic tongue muscles is
D. D. Kim (*) no longer used as a criterion for T4 staging, as
Division of Head and Neck Surgical Oncology and DOI supersedes it. A comparison of the AJCC 7
Microvascular Reconstruction, Department of Oral
and AJCC 8 TNM staging systems for OSCC is
and Maxillofacial Surgery, Louisiana State University
Health Sciences Center, Shreveport, LA, USA detailed in Table 4.1 [1]. Case examples are pro-
e-mail: DKim1@lsuhsc.edu vided in Fig. 4.1.

https://doi.org/10.1007/978-3-030-30094-4_4
34 P. Covello and D. D. Kim
Table 4.1 The TNM staging system for oral squamous cell carcinoma, as described by the seventh and eighth editions
of the AJCC cancer staging manual
T category AJCC 7 criteria AJCC 8 criteria
TX Primary cannot be assessed.
T0 Primary tumor cannot be assessed N/A
Tis Carcinoma in situ
T1 Size ≤2 cm Size ≤2 cm with DOI ≤ 5 mm
T2 Size >2–4 cm Size < 2 cm with DOI 5–10 mm or
Size >2–4 cm with DOI ≤ 10 mm
T3 Size >4 cm Size >4 cm or DOI > 10 mm
T4a Lip: Tumor invades through cortical bone, inferior alveolar nerve, floor of mouth, or skin of face
Oral Cavity: Tumor invades through cortical bone of Oral Cavity: Tumor invades through
maxilla or mandible, into deep muscle of tongue cortical bone of maxilla or mandible,
(genioglossus, hyoglossus, palatoglossus, or styloglossus), maxillary sinus, or skin of the face.
maxillary sinus, or skin of the face
T4b Invasion of masticator space, pterygoid plates, skull base, and/or encases the internal carotid artery.
N
category AJCC 7 criteria AJCC 8 clinical criteria AJCC 8 pathological criteria
NX Regional lymph nodes cannot be assessed
N0 No regional lymph node metastasis
N1 Single, ipsilateral ≤3 cm Single, ipsilateral ≤3 cm with ENE(−)
N2a Single, ipsilateral >3 to Single, ipsilateral – Single, ipsilateral ≤ 3 cm with ENE(+)
≤6 cm >3 - ≤6 cm with ENE(−) – Single, ipsilateral >3 - ≤6 cm with
ENE(−)
N2b Multiple, ipsilateral ≤6 cm Multiple, ipsilateral ≤6 cm with ENE(−)
N2c Multiple, bilateral, or Multiple, bilateral, or contralateral ≤6 cm with ENE(−)
contralateral ≤6 cm
N3 Any >6 cm N/A
N3a N/A Any >6 cm with ENE(−)
N3b N/A Any with clinically overt – Single, ipsilateral >3 cm with ENE(+)
ENE(+) – Multiple, ipsilateral, bilateral, or
contralateral any size with ENE(+)
M category Criteria
M0 No distant metastasis
M1 Distant metastasis
Stage T N M
I T1 N0 M0
II T2 N0 M0
III T3 N0 M0
T1–3 N1 M0
IVA T4a N0–1 M0
T1–4a N2
IVB Any T N3 M0
T4b Any N
IVC Any Any M1
4.3 Depth of Invasion exophytic lesions, thickness may be greater than

DOI, and in ulcerative lesions, DOI may be greater
Depth of invasion (DOI) is defined as the deepest than thickness. An example is depicted in Fig. 4.2.
point of tumor invasion, as measured from the Clinically, DOI is estimated to the examiner’s best
basement membrane of adjacent normal mucosa. judgment as superficial (≤5 mm), moderate (>5 to
Tumor thickness, on the other hand, is measured ≤10 mm), or deep (>10 mm). If uncertainty arises,
from the mucosal surface of the tumor. Thus, in the lesser depth is assigned, according to the rec-
4 Current Updates in Staging and Prognosis in Oral Cancer 35
a b
Fig. 4.1 (a) Case example. A 65-year-old male who pre- sion to perform adjuvant radiation therapy. (b) Case
sented with biopsy-proven squamous cell carcinoma of example. A 68-year-old female who presented with
the right ventral tongue, extending into the floor of mouth biopsy-proven squamous cell carcinoma of the right buc-
without evidence of nodal disease. Final pathological cal mucosa with level 1B lymphadenopathy. Right neck
evaluation demonstrated a 5.3 × 3.0 × 1.4 cm3 moderately dissection, levels I–V, depicted with 4 × 4 × 3 cm3 lymph
differentiated squamous cell carcinoma with a DOI of 15 node indicated with dotted ink line in right level IB. 11/36
and tumor extension into the glossus muscle, but no nodes found to be positive for carcinoma with evidence of
involvement of mandibular bone. Final pathological stag- ENE. Final pathological staging was noted to be pN2b
ing was noted to be pT4a according to the AJCC 7, but according to the AJCC 7, but pN3b according to the AJCC
pT3 according to the AJCC 8. Difference in the T-staging 8. Difference in the N-staging system was due to ENE. The
system was due to the tumor extension into the extrinsic presence of ENE encouraged the use of adjuvant chemo-
muscles of the tongue. Final staging did not alter the deci- therapy in addition to postoperative radiation
ommendations of the AJCC/UICC TNM staging biopsies, exact quantitative measurements of DOI
manual [1]. Pathologically, DOI is assessed by do not correlate well with those of the final post-
establishing a “plumb line” from the horizon of the operative specimens. However, when categorized
basement membrane of the adjacent squamous as superficial (<4 mm) and deep (≥4 mm), suffi-
mucosa to the deepest point of tumor invasion, cient agreement between pre- and postoperative
measured in millimeters [1]. samples allow for prognostication and treatment
When implemented retrospectively, incorpo- planning [4, 5]. For OSCC, a DOI of 4 mm or
rating DOI results in a pathologic upstaging of the greater was found to be a strong pathologic pre-
T category in as many as 30% of patients, while dictor of local recurrence and mortality [6].
demonstrating a significant correlation with Critical primary tumor DOI in OSCC with a 20%
5-year disease-specific survival that was not noted or greater risk of occult nodal metastasis in clini-
using the former staging system [2, 3]. Due to the cal N0 disease was found to be 2 mm in the
limited sample size attained from preoperative tongue, 3 mm in the floor of the mouth, and 4 mm
Fig. 4.2 Photomicrograph demonstrating the difference line depicts tumor thickness. Photomicrograph by Ashley
between DOI and tumor thickness in an exophytic lesion. Flowers, MD, from Louisiana State University—
The vertical black line depicts DOI, and the vertical blue Shreveport department of Pathology
in the retromolar trigone, alveolus, and hard pal- deep tissues, and adjacent nerve dysfunction, in
ate [7]. Patients with a DOI > 4 mm have a nearly conjunction with strong radiographic evidence,
sixfold higher risk of lymph node metastasis [8]. should be used for clinical staging (Fig. 4.3) [1].
Similarly, medullary bone invasion, but not corti- The two most common imaging modalities for the
cal invasion alone, has been associated with poor evaluation of cervical lymph node metastases are
local control, as well as decreased disease-specific computed tomography (CT) and magnetic reso-
and overall survival [9, 10]. Thus, erosion through nance imaging. Using the definition of “ill-defined
cortical bone classifies a tumor as T4a, while nodal borders” for ENE, CT has a sensitivity of
superficial erosion of the alveolus does not. 61% and specificity of 95%, while MRI has a sen-
sitivity of 40% and a specificity of 97% with an
acceptable intrarater reliability [11]. With the pres-
4.4 Extranodal Extension ence of three or more ENE imaging criteria,
including indistinct nodal margins, infiltration into
Extranodal extension (ENE) is defined as tumor adjacent tissues, irregular nodal enhancement,
infiltration that extends from the confines of the matted nodes, and central necrosis, improved
lymph node through the lymph node capsule into specificity and PPV are possible (Fig. 4.4) [12].
the surrounding connective tissue. Only unques- Pathologically, only ENE >2 mm beyond the
tionable evidence of ENE, such as matted nodes, lymph node capsule microscopically is used to
skin invasion, gross infiltration into surrounding define nodal status [1]. Lewis et al. described a
a b
Fig. 4.3 Right level 1b lymphadenopathy, unquestionably matted to surrounding tissues with ENE confirmed on
pathologic evaluation
a b
Fig. 4.4 (a) Left level 1b lymph node, 1.3 cm in greatest in greatest diameter with pathologically evident
diameter with no evidence of ENE. Clinically, mobile ENE. Clinically, fixed lymphadenopathy reported.
lymphadenopathy reported. Radiographically, only cen- Radiographically, infiltration into adjacent tissues (arrow),
tral necrosis noted. (b) Right level 1b lymph node, 3.8 cm irregular nodal enhancement, and central necrosis noted
histological grading system for ENE, divided included in the current AJCC cancer staging
into four major groups [13]. A summary can be system.
found in Table 4.2. However, internationally ENE in patients with OSCC is associated
standardized histological grading criteria have with a significant decrease in 5-year disease-
not yet been defined or validated and are not free survival, disease-specific survival, and
Table 4.2 Histological grading system for extranodal nation of the margin, measured from the tumor
extension (ENE)
bed to the mucosal edge without stretching, has
Extranodal been reported to be similar to intraoperative fro-
extension Description
zen section with no difference in terms of disease-
Grade 0 Tumor confined to the lymph node
(surrounded by lymphoid tissue) free survival and overall survival [21]. In fact,
Grade 1 Tumor reaching lymph node capsule patients with involved intraoperative frozen mar-
(no intervening lymphoid tissue) with gins in which an additional resection was per-
thickening of the overlying capsule formed demonstrated a 27% local recurrence
Grade 2 Tumor in perinodal tissue, limited to
rate, which was not statistically different from
≤1 mm beyond the capsule
Grade 3 Tumor in perinodal tissue, extending microscopically positive margins that were not
>1 mm beyond the capsule cleared [19]. Conversely, intraoperative assess-
Grade 4 Soft tissue metastasis. No residual ment of bony resection margins by cytological
nodal tissue or architecture assessment has a sensitivity of 94% and specific-
ity of 97%, resulting in a 60% relative risk reduc-
tion of residual carcinoma tissue in the final
overall survival, as well as an increase in locore- resection specimen that correlated with higher
gional recurrence when compared to node-neg- disease-free and overall survival [22].
ative (N0) and ENE-negative patients of a A “close” margin is most commonly defined
similar cohort [14, 15]. Patients with ENE have as being within 5 mm of the invasive tumor bed
also been found to have as high as a threefold [23]. Regardless of additional adverse tumor fea-
increase in the incidence of distant metastases tures, the local control rate with surgery alone in
[16]. When implemented retrospectively, incor- patients with a margin <5 mm was found to be
porating ENE has been shown to result in a 91% with an 84% disease-specific survival at
pathologic upstaging of the N category in as 5 years [24]. Stratification by categorization of
many as 30% of patients, while demonstrating a margin status by 1 mm subunits was not found to
significant correlation with 5-year disease-spe- be statistically significant. Patients with 0, 1, 2,
cific and overall survival that was not noted and 3 additional adverse features (i.e., T3/T4
using the former staging system [17]. Of note, tumors, PNI, LVI, or multimodal involvement)
ENE is not associated with a negative impact in had 5-year local control rates of 100%, 96%,
HPV-positive oropharyngeal squamous cell 83%, and 71%, respectively [24].
carcinoma [18]. Submillimeter margins, however, have been
associated with 28% rate of local recurrence [25].
Nevertheless, a close margin (<5 mm) alone,
4.5 Resection Margins without other negative prognostic indicators,
does not warrant postoperative adjuvant therapy.
Most ablative surgeons perform wide local exci-
sion of the primary OSCC lesion and subse-
quently sample the remaining tumor bed margins 4.6 Sentinel Node Biopsy
via frozen section. Concordance between final
and frozen specimen samples is 99% with a false- Intraoperative lymphatic mapping with biopsy of
negative rate of 3.5% [19]. However, the concor- the sentinel node in the regional basin is a mini-
dance between tumor bed margins and the main mally invasive manner of detecting metastatic
specimen are low, likely due to sampling error. disease [26]. The sentinel node is identified visu-
Thus, the margin obtained from the en-bloc spec- ally, following intralesion injection of methylene
imen remains the only prognostically relevant blue dye, or via gamma probe, following intrale-
margin in terms of local control [20]. sion injection of Technetium-99m-labeled human
Using a cutoff of 7 mm, the incidence of inad- serum (Fig. 4.5). The goal of sampling the SLN is
equate margins with intraoperative gross exami- to detect occult nodal metastasis with the inten-
a b
Fig. 4.5 Sentinel lymph node biopsy in left level 1b utilizing a combination of methylene blue (a) and gamma probe
localization with Technetium-99m-labeled human serum (b)
tion of sparing the patient from the morbidity of rounding blood vessels and lymphatic channels.
an elective neck dissection or the potential In clinically N0 patients, LVI is associated with
locoregional recurrence during the postoperative locoregional recurrence and decreased overall
observation period. SLNB for early AJCC Stage survival [38–40]. Studies have yet to further elu-
I and II OSCCs have demonstrated a detection cidate the role of adjuvant therapy when LVI
rate of 98% with an overall sensitivity of 92%, alone is detected.
specificity of 100%, and NPV of 96% [27–30].
Moreover, concordance of the SN status during
intraoperative frozen sections with permanent 4.7.2 Human Papilloma Virus
histopathologic specimens is 97% [31].
Squamous cell carcinoma that is related to tran-
scriptionally active Human Papilloma Virus
4.7 Other Tumor-Specific (HPV) has been shown to have distinct character-
Prognostic Factors istics from HPV-negative carcinomas, particu-
larly in the oropharynx. With the advent of
4.7.1 Perineural various testing methods, it has been elucidated
and Lymphovascular Invasion that 90% of OPSCC are caused by HPV, specifi-
cally subtype 16 [41]. Other high-risk subtypes
Perineural invasion is defined as the presence of include 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59,
tumor cells within any of the three layers of nerve 68, 69, 73, and 82. Furthermore, patients with
sheath or surrounding 33% of the nerve circum- oropharyngeal tumors in which HPV DNA is
ference [32]. PNI has been associated with a detected are generally younger with a lower num-
decrease in disease-specific and overall survival ber of cumulative pack-years of tobacco smok-
and increases in regional lymph node and distant ing, carrying a small primary tumor burden and
metastasis, as well as locoregional recurrence [8, better disease-specific and overall survival pat-
33, 34]. In the absence of other adverse patho- terns [42]. Among HPV-positive OPSCC patients
logic features, however, postoperative adjuvant who receive radiochemotherapy, treatment out-
radiation therapy was not found to significantly comes are significantly better than those com-
reduce the incidence of recurrence in PNI- pared to the HPV-negative counterparts [43]. The
positive patients [35–37]. impact that HPV testing has had on OPSCC has
Lymphovascular invasion is the presence of prompted a separate staging system, as defined
neoplastic cells in the wall and/or lumen of sur- by the eighth edition of the AJCC cancer staging
manual, which is beyond the scope of this Table 4.3 Histological patterns of invasion
chapter. Pattern of
HPV detection in sampled tissue can be invasion Definition
accomplished using a variety of techniques, Type 1 Broad, pushing invasion with well-
delineated infiltrating borders
including polymerase chain reaction (PCR) Type 2 Broad, pushing “fingers,“cords, bands,
assays, in situ hybridization (ISH), or immuno- and/or strands of infiltration with a
histochemistry (IHC) for the p16 protein [41, 44, stellate appearance
45]. At this time, no consensus on which testing Type 3 Invasive islands of tumor cells (>15 per
group)
method should be used has been reached. PCR
Type 4 Invasive islands of tumor cells (<15 per
for HPV DNA is highly sensitive, but expensive group) and/or single cells
and cannot distinguish transcriptionally active
from inactive HPV. Cross-contamination during
PCR may also result in a relatively low specific- 4.7.3 Pattern of Invasion
ity. PCR for the transcription-active oncogenes
E6/E7 mRNA is highly sensitive and highly spe- Byrne et al. previously described a histological
cific but requires fresh frozen tissue, remains malignancy grading system, in which a 4-point
technically challenging to perform, and is not pattern of invasion was defined [53]. A brief sum-
readily available in all laboratories. ISH for DNA mary can be found in Table 4.3. As the POI wors-
is highly specific but demonstrates a low sensitiv- ens, the risk for lymph node metastasis increases.
ity for detection, while ISH for and E6/E7 mRNA Type 3 POI predicts lymph node metastasis with
is still in development. IHC for p16 has demon- a sensitivity and specificity of 87% and 86%,
strated good concordance with ISH studies, par- respectively [8]. Brandwein-Gensler et al. subse-
ticularly in oropharyngeal carcinomas, while quently introduced Type 5, a widely dispersed
being easier to interpret, highly sensitive, and pattern of tumor infiltrate with ≥1 mm of normal
more cost-effective [46]. To be interpreted as tissue between tumor satellites and collapsed the
positive, p16 immunostaining must be nuclear grading system into two groups, cohesive (Types
(not cytoplasmic) with a intensity +2/+3 or 1–3) and infiltrative (Types 4 and 5) [54].
greater with a distribution 75% or greater [1]. Infiltrative POI is a strong pathologic predictor of
In early studies, HPV has been shown to be locoregional recurrence and mortality from
2–3 times more likely to be detected in precan- OSCC with a hazard ratio of 1.5 and 2.34, respec-
cerous oral mucosa and nearly five times more tively [6].
likely to be detected in OSCC than in normal
mucosa [47]. Recently, as many as 30% of
OSCCs are HPV-positive, as elucidated by PCR 4.7.4 Lymph Node Ratio
and ISH techniques, with high-risk subtypes 16
and 18 found in 25% and 18% of samples, respec- The ratio of tumor-laden nodes to the total num-
tively [48]. Despite the presence of HPV DNA, ber of nodes resected and examined has shown
active mRNA expression seems to be limited in prognostic significance in OSCC. Generally, a
OSCC [45]. Furthermore, p16 overexpression in selective neck dissection should include ≥10
IHC is not a reliable marker for the presence of lymph nodes, while a comprehensive neck dis-
transcriptionally active HPV in OSCC [49–51]. section should include ≥15 lymph nodes. A
Unlike OPSCC, HPV-positive and HPV-negative lymph node ratio, >6% in tongue OSCC and >7%
OSCCs have not been shown to differ signifi- in buccal mucosa OSCC, correlates to a 4.8- and
cantly in terms of pathogenesis, survival, or sen- 10.3-fold increase in the risk of locoregional
sitivity to radiation therapy [43, 52]. To date, recurrence, respectively [55, 56]. Estimated
HPV has been consistently reported to have a 5-year overall survival rates were 65.3%, 49.9%,
minor role in oral oncogenesis. 41.1%, 29.7%, 18.5%, and 9.7% for groups with
0, 1, 2, 3, 4–6, 7–9, and 10 or more metastatic 5. Tan WJ, Chia CS, Tan HK, Soo KC, Iyer
NG. Prognostic significance of invasion depth
lymph nodes [16]. In these studies, the number of in oral tongue squamous cell carcinoma. ORL.
metastatic nodes demonstrated more importance 2012;74(5):264–70.
than size or contralaterality. 6. Almangush A, Bello IO, Coletta RD, Makitie AA,
Makinen LK, Kauppila JH, et al. For early-stage oral
tongue cancer, depth of invasion and worst pattern
of invasion are the strongest pathological predictors
4.8 Conclusion for locoregional recurrence and mortality. Virchows
Archiv. 2015;467(1):39–46.
The eighth edition of the Cancer Staging System 7. Brockhoff HC 2nd, Kim RY, Braun TM, Skouteris C,
Helman JI, Ward BB. Correlating the depth of inva-
by the AJCC appropriately emphasizes DOI and sion at specific anatomic locations with the risk for
ENE as prognostic factors for OSCC, as both have regional metastatic disease to lymph nodes in the
been shown to correlate to locoregional control neck for oral squamous cell carcinoma. Head Neck.
and 5-year survival. Factors that are not included 2017;39(5):974–9.
8. Arora A, Husain N, Bansal A, Neyaz A, Jaiswal R,
in the staging system, but must be considered, Jain K, et al. Development of a new outcome predic-
include resection margins, PNI, LVI, WPOI, and tion model in early-stage squamous cell carcinoma of
LNR. Intraoperative margin evaluation and senti- the oral cavity based on histopathologic parameters
nel lymph node biopsy techniques are being fur- with multivariate analysis: the Aditi-Nuzhat Lymph-
node Prediction Score (ANLPS) system. Am J Surg
ther developed to aid in improving outcomes while Pathol. 2017;41(7):950–60.
minimizing surgical morbidity. Despite its impact 9. Li C, Lin J, Men Y, Yang W, Mi F, Li L. Does med-
on oropharyngeal cancer, HPV seems to play a ullary versus cortical invasion of the mandible affect
minor role in the oncogenesis of OSCC. prognosis in patients with oral squamous cell carci-
noma? J Oral Maxillofac Surg. 2017;75(2):403–15.
10. Fives C, Nae A, Roche P, O'Leary G, Fitzgerald

Acknowledgements Photomicrograph in Fig. 4.2 by B, Feeley L, et al. Impact of mandibular inva-
Ashley Flowers, MD—Assistant Professor, Department sion on prognosis in oral squamous cell carcinoma
of Pathology, Louisiana State University Health Sciences four centimeters or less in size. Laryngoscope.
Center, Shreveport, LA. 2017;127(4):849–54.
Conflicts of interest: There are no conflicts of interest 11. Almulla A, Noel CW, Lu L, Xu W, O'Sullivan B,
to report. Goldstein DP, et al. Radiologic-pathologic correlation
of extranodal extension in patients with squamous cell
carcinoma of the oral cavity: implications for future
editions of the TNM classification. Int J Radiat Oncol
References Biol Phys. 2018;102(4):698–708.
12. Carlton JA, Maxwell AW, Bauer LB, McElroy SM,
1. Ridge JA, Lydiatt WM, Patel SG, Glastonbury CM, Layfield LJ, Ahsan H, et al. Computed tomography
Brandwein-Gensler MS, Ghossein RA, et al. Lip and detection of extracapsular spread of squamous cell
oral cavity. In: Amin MB, editor. AJCC cancer stag- carcinoma of the head and neck in metastatic cervical
ing manual. 8th ed. NewYork, NY: Springer; 2017. lymph nodes. Neuroradiol J. 2017;30(3):222–9.
p. 79–94. 13. Lewis JS Jr, Carpenter DH, Thorstad WL, Zhang Q,
2. Tirelli G, Gatto A, Boscolo Nata F, Bussani R, Haughey BH. Extracapsular extension is a poor pre-
Piccinato A, Marcuzzo AV, et al. Prognosis of oral can- dictor of disease recurrence in surgically treated oro-
cer: a comparison of the staging systems given in the pharyngeal squamous cell carcinoma. Modern Pathol.
7th and 8th editions of the American Joint Committee 2011;24(11):1413–20.
on cancer staging manual. Br J Oral Maxillofac Surg. 14. Suton P, Salaric I, Granic M, Mueller D, Luksic

2018;56(1):8–13. I. Prognostic significance of extracapsular spread
3. Ebrahimi A, Gil Z, Amit M, Yen TC, Liao CT, of lymph node metastasis from oral squamous cell
Chaturvedi P, et al. Primary tumor staging for oral carcinoma in the clinically negative neck. Int J Oral
cancer and a proposed modification incorporating Maxillofac Surg. 2017;46(6):669–75.
depth of invasion: an international multicenter retro- 15. Jose J, Coatesworth AP, Johnston C, MacLennan

spective study. JAMA Otolaryngol Head Neck Surg. K. Cervical node metastases in squamous cell carci-
2014;140(12):1138–48. noma of the upper aerodigestive tract: the significance
4. Almangush A, Leivo I, Siponen M, Sundquist E, of extracapsular spread and soft tissue deposits. Head
Mroueh R, Makitie AA, et al. Evaluation of the bud- Neck. 2003;25(6):451–6.
ding and depth of invasion (BD) model in oral tongue 16. Ho AS, Kim S, Tighiouart M, Gudino C, Mita A,
cancer biopsies. Virchows Archiv. 2018;472(2):231–6. Scher KS, et al. Metastatic lymph node burden
and survival in oral cavity cancer. J Clin Oncol. European Node Trial (SENT): 3-year results of
2017;35(31):3601–9. sentinel node biopsy in oral cancer. Eur J Cancer.
17. Matos LL, Dedivitis RA, Kulcsar MAV, de Mello ES, 2015;51(18):2777–84.
Alves VAF, Cernea CR. External validation of the 30. Civantos FJ, Zitsch RP, Schuller DE, Agrawal A,

AJCC cancer staging manual, 8th edition, in an inde- Smith RB, Nason R, et al. Sentinel lymph node
pendent cohort of oral cancer patients. Oral Oncol. biopsy accurately stages the regional lymph nodes
2017;71:47–53. for T1–T2 oral squamous cell carcinomas: results of
18. Mermod M, Tolstonog G, Simon C, Monnier
a prospective multi-institutional trial. J Clin Oncol.
Y. Extracapsular spread in head and neck squa- 2010;28(8):1395–400.
mous cell carcinoma: a systematic review and meta- 31. Miura K, Hirakawa H, Uemura H, Yoshimoto S,

analysis. Oral Oncol. 2016;62:60–71. Shiotani A, Sugasawa M, et al. Sentinel node biopsy
19. Buchakjian MR, Tasche KK, Robinson RA, Pagedar for oral cancer: a prospective multicenter Phase II
NA, Sperry SM. Association of main specimen and trial. Auris Nasus Larynx. 2017;44(3):319–26.
tumor bed margin status with local recurrence and 32. Liebig C, Ayala G, Wilks JA, Berger DH, Albo

survival in oral cancer surgery. JAMA Otolaryngol D. Perineural invasion in cancer: a review of the lit-
Head Neck Surg. 2016;142(12):1191–8. erature. Cancer. 2009;115(15):3379–91.
20. Maxwell JH, Thompson LD, Brandwein-Gensler
33. Tai SK, Li WY, Chu PY, Chang SY, Tsai TL, Wang
MS, Weiss BG, Canis M, Purgina B, et al. Early oral YF, et al. Risks and clinical implications of perineu-
tongue squamous cell carcinoma: sampling of mar- ral invasion in T1-2 oral tongue squamous cell carci-
gins from tumor bed and worse local control. JAMA noma. Head Neck. 2012;34(7):994–1001.
Otolaryngol Head Neck Surg. 2015;141(12):1104–10. 34. Jardim JF, Francisco AL, Gondak R, Damascena A,
21. Mair M, Nair D, Nair S, Dutta S, Garg A, Malik A, Kowalski LP. Prognostic impact of perineural inva-
et al. Intraoperative gross examination vs frozen sec- sion and lymphovascular invasion in advanced stage
tion for achievement of adequate margin in oral can- oral squamous cell carcinoma. Int J Oral Maxillofac
cer surgery. Oral Surg Oral Med Oral Pathol Oral Surg. 2015;44(1):23–8.
Radiol. 2017;123(5):544–9. 35. Chatzistefanou I, Lubek J, Markou K, Ord RA. The
22. Nieberler M, Haussler P, Kesting MR, Kolk A, Deppe role of neck dissection and postoperative adjuvant
H, Weirich G, et al. Clinical impact of intraoperative radiotherapy in cN0 patients with PNI-positive squa-
cytological assessment of bone resection margins in mous cell carcinoma of the oral cavity. Oral Oncol.
patients with head and neck carcinoma. Ann Surg 2014;50(8):753–8.
Oncol. 2016;23(11):3579–86. 36. Bur AM, Lin A, Weinstein GS. Adjuvant radiotherapy
23. Meier JD, Oliver DA, Varvares MA. Surgical mar- for early head and neck squamous cell carcinoma with
gin determination in head and neck oncology: cur- perineural invasion: a systematic review. Head Neck.
rent clinical practice. the results of an International 2016;38(Suppl 1):E2350–7.
American Head and Neck Society Member Survey. 37. Liu T, Chua B, Batstone M. Postoperative radiother-
Head Neck. 2005;27(11):952–8. apy for oral squamous cell carcinoma with histologic
24. Ch'ng S, Corbett-Burns S, Stanton N, Gao K, Shannon risk factors: are we over-treating? J Oral Maxillofac
K, Clifford A, et al. Close margin alone does not war- Surg. 2018;76(7):1565–70.
rant postoperative adjuvant radiotherapy in oral squa- 38. Cassidy RJ, Switchenko JM, Jegadeesh N, Sayan

mous cell carcinoma. Cancer. 2013;119(13):2427–37. M, Ferris MJ, Eaton BR, et al. Association of lym-
25. Buchakjian MR, Ginader T, Tasche KK, Pagedar
phovascular space invasion with locoregional fail-
NA, Smith BJ, Sperry SM. Independent predictors of ure and survival in patients with node-negative oral
prognosis based on oral cavity squamous cell carci- tongue cancers. JAMA Otolaryngol Head Neck Surg.
noma surgical margins. Otolaryngol Head Neck Surg. 2017;143(4):382–8.
2018;2018:194599818773070. 39. Fives C, Feeley L, O'Leary G, Sheahan P. Importance
26. Ross MI, Reintgen D, Balch CM. Selective lymphad- of lymphovascular invasion and invasive front
enectomy: emerging role for lymphatic mapping and on survival in floor of mouth cancer. Head Neck.
sentinel node biopsy in the management of early stage 2016;38(Suppl 1):E1528–34.
melanoma. Semin Surg Oncol. 1993;9(3):219–23. 40. Hosni A, McMullen C, Huang SH, Xu W, Su J, Bayley
27. Yang Y, Zhou J, Wu H. Diagnostic value of sentinel A, et al. Lymph node ratio relationship to regional
lymph node biopsy for cT1/T2N0 tongue squamous failure and distant metastases in oral cavity cancer.
cell carcinoma: a meta-analysis. Eur Archiv Oto- Radiother Oncol. 2017;124(2):225–31.
rhino-laryngol. 2017;274(11):3843–52. 41. Bishop JA, Lewis JS Jr, Rocco JW, Faquin WC. HPV-
28. Lin R, Yan J, Liu W, Fan T, Tang L. The predic- related squamous cell carcinoma of the head and
tive value of cervical lymph node metastasis neck: An update on testing in routine pathology prac-
through sentinel lymph node biopsy in patients with tice. Semin Diagn Pathol. 2015;32(5):344–51.
oral cancer: a meta-analysis. J Cancer Res Ther. 42. Ang KK, Harris J, Wheeler R, Weber R, Rosenthal
2016;12(Suppl):C256–c9. DI, Nguyen-Tan PF, et al. Human papillomavirus and
29. Schilling C, Stoeckli SJ, Haerle SK, Broglie
survival of patients with oropharyngeal cancer. N
MA, Huber GF, Sorensen JA, et al. Sentinel Engl J Med. 2010;363(1):24–35.
43. Wang F, Zhang H, Xue Y, Wen J, Zhou J, Yang X, pression with improved outcomes in young patients
et al. A systematic investigation of the association with squamous cell cancers of the oral tongue. Head
between HPV and the clinicopathological parameters Neck. 2011;33(11):1622–7.
and prognosis of oral and oropharyngeal squamous 51. Lingen MW, Xiao W, Schmitt A, Jiang B, Pickard R,
cell carcinomas. Cancer Med. 2017;6(5):910–7. Kreinbrink P, et al. Low etiologic fraction for high-
44. Westra WH. Detection of human papillomavirus
risk human papillomavirus in oral cavity squamous
(HPV) in clinical samples: evolving methods and cell carcinomas. Oral Oncol. 2013;49(1):1–8.
strategies for the accurate determination of HPV 52. Kruger M, Pabst AM, Walter C, Sagheb K, Gunther
status of head and neck carcinomas. Oral Oncol. C, Blatt S, et al. The prevalence of human papil-
2014;50(9):771–9. loma virus (HPV) infections in oral squamous cell
45. Mirghani H, Amen F, Moreau F, Lacau St Guily
carcinomas: a retrospective analysis of 88 patients
J. Do high-risk human papillomaviruses cause and literature overview. J Cranio-maxillo-fac Surg.
oral cavity squamous cell carcinoma? Oral Oncol. 2014;42(7):1506–14.
2015;51(3):229–36. 53. Bryne M, Koppang HS, Lilleng R, Kjaerheim

46. Thomas J, Primeaux T. Is p16 immunohistochem- A. Malignancy grading of the deep invasive margins
istry a more cost-effective method for identifica- of oral squamous cell carcinomas has high prognostic
tion of human papilloma virus-associated head and value. J Pathol. 1992;166(4):375–81.
neck squamous cell carcinoma? Ann Diagn Pathol. 54. Brandwein-Gensler M, Teixeira MS, Lewis CM,

2012;16(2):91–9. Lee B, Rolnitzky L, Hille JJ, et al. Oral squamous
47. Miller CS, Johnstone BM. Human papillomavirus as a cell carcinoma: histologic risk assessment, but
risk factor for oral squamous cell carcinoma: a meta- not margin status, is strongly predictive of local
analysis, 1982-1997. Oral Surg Oral Med Oral Pathol disease-free and overall survival. Am J Surg Pathol.
Oral Radiol Endod. 2001;91(6):622–35. 2005;29(2):167–78.
48. Kansy K, Thiele O, Freier K. The role of human papil- 55. Safi AF, Grandoch A, Nickenig HJ, Zoller JE,

lomavirus in oral squamous cell carcinoma: myth and Kreppel M. The importance of lymph node ratio
reality. Oral Maxillofac Surg. 2014;18(2):165–72. for locoregional recurrence of squamous cell car-
49. Poling JS, Ma XJ, Bui S, Luo Y, Li R, Koch WM, cinoma of the tongue. J Cranio-maxillo-fac Surg.
et al. Human papillomavirus (HPV) status of non- 2017;45(7):1058–61.
tobacco related squamous cell carcinomas of the lat- 56. Safi AF, Grandoch A, Nickenig HJ, Zoller JE, Kreppel
eral tongue. Oral Oncol. 2014;50(4):306–10. M. Importance of lymph node ratio for locoregional
50. Harris SL, Thorne LB, Seaman WT, Hayes DN, Couch recurrence of squamous cell carcinoma of the buccal
ME, Kimple RJ. Association of p16(INK4a) overex- mucosa. Head Neck. 2017;39(12):2488–93.
Surgical Factors Affecting
Outcomes in Oral Squamous
5
Cell Carcinoma
Justine Moe, Andrew Baker, and Brent Ward
5.1 Introduction This chapter discusses the factors that can allow
the surgeon to optimize outcomes in the surgical
Oral cancer is the 16th most common cancer treatment of OCSCC.
worldwide with an estimated incidence of
350,000 new cases per year [1]. Oral cavity squa-
mous cell carcinoma (OCSCC) is the most com- 5.2 Care Team Factors
mon type of oral cancer and a cause of significant
morbidity and mortality worldwide. The treat- Multidisciplinary care is essential in the manage-
ment of OCSCC follows the clinical practice ment of patients with OCSCC. Multimodality
guidelines established by the National therapy is integral to the restoration of function,
Comprehensive Cancer Network [2]. Surgery is quality of life, and survival of these patients.
the primary treatment modality when possible, Multidisciplinary team composition varies
although definitive radiotherapy may be used for between institutions but typically include abla-
a select group of early stage OCSCC or for tive surgeons, reconstructive surgeons, and mem-
patients unable to undergo surgery. Adjuvant bers of medical oncology, radiation oncology,
radiation, with or without chemotherapy, has radiology, pathology, dentistry, speech and lan-
been shown to improve disease-free survival and guage pathology, nutrition, rehabilitation therapy,
overall survival in advanced stages and in the palliative care, and social work. Multidisciplinary
presence of high-risk histological features. tumor board conference allows for a comprehen-
The treatment for OCSCC should be person- sive discussion of the optimal treatment for indi-
alized to some degree for each patient. There are vidual patients.
multiple determinants that affect surgical out- In addition, the institution at which surgery is
comes, as well as prognosis, including a patient’s performed should be well equipped to manage
surgical candidacy, the type and extent of sur- head and neck cancer patients in all phases of
gery required, the mode and timing of recon- care. The institution should have microvascular
struction, and the use of adjunct surgical tools. capability as well as intensive care, floor, and
operating room staff trained to care for head and
neck cancer patients. The treatment of oral can-
J. Moe · A. Baker · B. Ward (*) cer at high volume centers is associated with
Department of Oral and Maxillofacial Surgery, improved survival for all stages of cancer. A ret-
University of Michigan, Ann Arbor, MI, USA
e-mail: jusmoe@med.umich.edu; abakes@med. rospective cohort study of 13,655 patients found
umich.edu; bward@med.umich.edu a higher incidence of positive margins at nonaca-

https://doi.org/10.1007/978-3-030-30094-4_5
46 J. Moe et al.
demic facilities and low volume centers [3]. considered as a prolonged recovery may delay or
Trends toward the regionalization of care of oral prevent a patient from receiving adjunctive ther-
cancer patients may be a factor in improved apy. Patients with severe comorbidities unable to
prognosis. tolerate general anesthesia may not be candidates
for surgery, and primary radiation with or with-
out chemotherapy may be indicated.
5.3 Patient Factors Patients with recurrent OCSCC pose chal-
lenges for surgical treatment. Recurrent OCSCC
Patients with OCSCC often have significant med- often heralds an aggressive tumor biology with
ical comorbidities because smoking, excessive an associated poor prognosis. Additionally, in
alcohol consumption, poor dietary habits, and patients with previous head and neck radiation,
decreased oral intake are common in this popula- radiation-induced inflammation and fibrosis dis-
tion. A preoperative assessment of the patient’s tort normal anatomy and tissue planes. Previous
functional status and comorbidities should be surgeries also alter anatomy, potentially limiting
completed by using one of many comorbidity reconstructive options.
indexes available. The Adult Comorbidity
Evaluation (ACE-27) is a validated instrument
that grades the severity of comorbidities of 5.4 Surgical Principles
patients with cancer at the time of diagnosis [4].
The American Society of Anesthesiologists The primary goal of oncologic surgery is the
(ASA) physical status classification predicts peri- complete tumor extirpation with a cuff of sur-
operative risk and may be used as a proxy to rounding normal tissue and negative margins.
evaluate comorbidity [5, 6]. The goal of reconstructive surgery is the restora-
In head and neck cancer patients, age alone is tion of form and function. The schema of the
not a predictor of complication rates in the surgi- “reconstructive ladder” has traditionally been
cal treatment of OCSCC [7, 8]. Rather, a high used to describe the spectrum of options for oro-
comorbid burden is associated with increased facial reconstruction, based on the principle of
surgical and anesthesia complications, prolonged selecting the least complex treatment required for
hospital stay, and reduced functional outcome the defect (Fig. 5.1). The surgeon should instead
following treatment [5]. Comorbidity is also consider the “reconstructive toolbox,” in which
associated with increased perioperative mortality, complex procedures such as free flaps should not
increased short-term mortality, and decreased be thought of as a last resort, but as one of the
overall survival [5]. In early years following tools to be used when necessary to restore both
curative-intent surgery, a reported 16–40% of form and function (Fig. 5.2).
deaths is secondary to comorbid conditions and
non-cancer-related causes, particularly cardio-
vascular, respiratory, gastrointestinal comorbid-
ity, and diabetes [5, 9]. Severe comorbidity has
been found to have comparable survival impact
to a T4 tumor or N2 neck [9].
While surgery is essential to cure in most
cases of OCSCC, the perioperative risks associ-
ated with a patient’s medical status need to be
weighed. Patients with high comorbid burden or
poor functional status may be poor candidates for
free flap reconstruction, and an alternative recon-
structive method may be considered. Predictors
of poor postoperative functional status need to be Fig. 5.1 Reconstructive ladder
5 Surgical Factors Affecting Outcomes in Oral Squamous Cell Carcinoma 47
When reconstructing orofacial defects, the addition, the success of corticocancellous bone
component tissue types and structural subunits grafts is associated with graft length. Failure
involved in the defect should be identified, and rates have been reported as high as 17% for grafts
consideration should be made to reconstruct the 6 cm in length or less, increasing significantly for
critical components. The type of reconstruction grafts greater than 9 cm to a failure rate of 75%
should be chosen based on the size and composi- for grafts greater than 12 cm [10].
tion of the defect. Soft tissue defects may be
reconstructed using simple techniques or local,
regional or free flaps; however, bony and com- 5.5 Management of the Primary
posite defects most often require free flap recon- Tumor
struction. Corticocancellous bone grafting is
often not an option in the oncologic patient due to 5.5.1 Mandible
the need for primary reconstruction, the concom-
itant loss of soft tissue, large bony defects, the OCSCC of the gingival, floor of mouth, buccal
risk of graft loss with adjuvant therapy, or poor mucosa, and retromolar trigone can invade the
recipient bed vascularity following radiation. In mandible secondarily. Patterns of mandibular
invasion include through periosteum, foramina,
attached mucosa, cortical bone defects in the
edentulous mandible, and periodontal ligament
in the dentate mandible [11–13]. Mandibular
involvement should be suspected in gingival SCC
even in the absence of gross bone involvement
clinically or on imaging.
For mandibular SCC with early bone involve-
ment and in nonatrophic mandibles, a marginal
mandibulectomy may be satisfactory to achieve
negative margins while maintaining mandibular
continuity. In such cases, prophylactic placement
of a reconstruction plate may be required to pre-
vent fracture. Segmental mandibulectomy is nec-
essary for gross invasion of cancellous bone.
Fig. 5.2 Reconstructive toolbox Classification systems of mandibular defects
Fig. 5.3 H-C-L classification of mandibular defects. H L implies a lateral segment not inducing the condyle.
implies a hemimandibulectomy defect including the con- (Adapted from Jewer et al. [14])
dyle. C implies a central segment including both canines.
48 J. Moe et al.
based on location and extent such as the HCL cludes mandibular reconstruction. The radial
classification have been described to reflect the forearm (RF) osteocutaneous free flap provides
complexity of reconstruction (Fig. 5.3) [14]. poor-quality bone stock with significant donor
When possible, immediate reconstruction site morbidity and is a poor choice for mandibu-
with an osteocutaneous free flap is an optimal lar reconstruction.
approach for the reconstruction of mandibular While the restoration of mandibular continu-
segmental defects. Corticocancellous grafts are ity is ideal, soft tissue-only reconstruction in
often not amenable for use due to multiple rea- large composite defects has been described using
sons as listed above. The fibular free flap is the pectoralis major myocutaneous (PMMC) pedi-
workhorse for mandibular reconstruction and has cled flaps or anterolateral thigh (ALT) free flaps
minimal donor site morbidity. Up to 25 cm of have been described with acceptable functional
bone length may be harvested for the reconstruc- and cosmetic results [16].
tion of long-span mandibular defects and a reli-
able skin paddle allows for soft tissue
reconstruction in composite defects. 5.5.2 Maxilla
The deep circumflex iliac artery (DCIA) free
flap may be used to reconstruct the anterior or OCSCC of the maxilla can involve the alveolus,
posterior mandible including the ramus, as the palate, maxillary sinus, nasal cavity, orbit, eth-
natural curvature of the iliac crest allows for a moid and sphenoid sinuses, and base of skull
replication of the natural mandibular form often depending on the extent of disease. Various clas-
without osteotomies [15]. Internal oblique mus- sification systems of maxillary defects have been
cle may be harvested for soft tissue reconstruc- described, including the Brown and Okay classi-
tion. However, the poor reliability of the fication systems, which are used to assess the
perforator skin paddle as well as the potential for functional outcome of rehabilitation and to deter-
significant donor site morbidity limits the utility mine the extent and type of reconstruction
of the DCIA flap. The scapular free flap provides (Figs. 5.4 and 5.5) [17, 18].
10–14 cm of bone length and large amount of soft Approaches to the maxilla vary based on dis-
tissue for the reconstruction of composite defects; ease extent. Most Brown class I or II defects can
however, limited bone stock and length often pre- be approached transorally. The midfacial deglov-
I II III IV V VI
a b c d
Fig. 5.4 Brown classification of maxillectomy defects. (Adapted from Brown et al. [17])
Ia
Ib
II
III
Fig. 5.5 Okay classification of maxillectomy defects. (Adapted from Okay et al. [18])
ing incision which includes sublabial and rhino- mary goals of maxillectomy defect closure and
plasty incisions improves access to the bilateral separation of the oral cavity from the sinonasal
anterior maxilla and paranasal sinuses without cavities (Fig. 5.6) [18]. Obturation allows for
the need for facial incisions [19]. The Weber– shorter operative time, shorter hospital stay and
Ferguson incision allows wide access to the direct visualization of the defect for oncologic
entire maxilla and orbital floor. A lip split man- surveillance [20]. However, maxillectomy site
dibulotomy improves access to tumors of the hygiene, placement and removal can be challeng-
posterior maxilla with extension into the ptery- ing, particularly in the setting of trismus, and fre-
goid plates or infratemporal fossa. Additional quent adjustments are required during the acute
approaches to the pterygoid region and base of healing period. Poorly retentive or unstable obtu-
skull are described elsewhere. rators may be associated with hypernasal speech
Obturation has traditionally been the standard and regurgitation into the nasal cavity [21].
method of maxillary rehabilitation, with the pri- Endosseous and zygomatic implants can facili-
50 J. Moe et al.
defects can be reconstructed using an obturator

or soft tissue local or free flap; Class II defects
can be reconstructed with an obturator or vascu-
larized bone flap; and Class III defects have
improved functional outcomes when recon-
structed with a vascularized bone flap [24].
Orbital floor and zygomatic defects should be
restored with bony reconstruction because obtu-
rators inadequately restore cosmesis to the mid-
face in these instances [18].
5.5.3 Oral Tongue
The oral tongue is the most common site of

OCSCC. Local excision, partial glossectomy,
hemiglossectomy, subtotal, or total glossectomy
may be required based on the disease extent.
Fig. 5.6 Maxillectomy defect with a temporary maxil- Early stage lesions can be excised transorally
lary obturator in place
with the assistance of retraction sutures placed in
the anterior tongue. The lip split mandibulotomy
tate retention and support of large obturators; this approach may facilitate access for larger or pos-
topic is discussed below. terior tongue lesions or those involving the floor
In contrast, free tissue transfer allows for pri- of mouth. A transcervical approach may be used
mary reconstruction with abundant tissue with for subtotal or total glossectomy defects, in which
relative freedom of orientation and shape. Fibula, the resection and majority of the reconstruction is
scapula, DCIA, medial femoral condyle, RF, rec- completed through the neck incision without the
tus, and ALT free flaps have been described in need for lip split [25].
maxillary reconstruction. Fibula and DCIA free Clear histological margins can be achieved
flaps provide adequate bone stock for dental with 95% confidence interval if the surgeon uti-
rehabilitation with osseointegrated implants lizes 1.5–2 cm surgical margins [26]. However,
while scapula flaps often do not [22]. excess resection of normal tissue should be
Disadvantages of free flaps in maxillary recon- avoided on the tongue in order to preserve func-
struction include donor site morbidity, longer and tional tissue, and therefore margins of 1.0–1.5 cm
more complex surgeries, and a prolonged hospi- are standardly utilized.
tal course. Immediate reconstruction with a free Reconstruction of small tongue defects may
flap precludes direct inspection for cancer be completed by primary closure, healing by sec-
surveillance; however, no study to date has dem- ondary intention, split-thickness skin grafts, or
onstrated a delay in the detection of local recur- similar substitutes. Larger tongue defects require
rence in patients with immediate reconstruction reconstruction with regional pedicled flaps or
[23]. In these patients, computed tomography or free tissue transfer in order to restore tongue
magnetic resonance imaging as well as endos- mobility, such as the PMMC flap, RF fasciocuta-
copy allow for adequate assessment of recurrence neous free flap and ALT free flap. Free flap recon-
without direct inspection [22]. struction of the tongue has shown superior
Reconstruction of the palate and alveolar arch functional results related to swallowing and
is critical to restore speech and swallowing, while speech as compared to myocutaneous pedicled
the maxillary superstructure has little effect on flaps and should thus be used over pedicled flaps
these functional outcomes [24]. Okay Class I when possible [27].
The utility of sensate flaps in tongue recon- it lies within the resection margins. Free flap
struction remains a point of debate. Microsurgical reconstruction of larger excisions of the buccal
reinnervation of flaps has been shown in small mucosa allows restoration of function and pre-
studies to improve sensory recovery [28]. vents trismus. The RF flap is often used to
However, there is insufficient evidence to discern reconstruct the buccal mucosa as it is thin and
any benefit from sensate flaps in improving pliable. More extensive lesions involving the
speech and swallowing; these findings highlight skin of the cheek, mandible, maxilla, and infra-
the fact that functional recovery in these patients temporal fossa require composite resections
is complex and multifactorial [29]. including a marginal or segmental mandibulec-
Reconstruction of total glossectomy defects tomy or partial maxillectomy and free flap
remains challenging, and functional outcomes reconstruction flap.
are variable. The need for total laryngectomy
with total glossectomy is controversial and may 5.5.4.1 Retromolar Trigone
be advocated for the prevention of aspiration. Tumors in the retromolar trigone may involve the
However, total glossectomy with laryngeal pres- lingual nerve, submandibular duct, and palato-
ervation has been found to be associated with glossus with the possible need to sacrifice these
favorable swallowing and speech outcomes and structures with tumor extirpation. Small superfi-
meaningful long-term quality of life. Also, feed- cial defects can be resected transorally with
ing tube dependence has not been found to be reconstruction by primary closure, skin grafts,
associated with laryngeal preservation or recon- local and regional flaps, and buccal fat. More
structive technique [30, 31]. Laryngeal preserva- extensive lesions may involve the mandible,
tion is a valid option with total glossectomy, and maxilla, soft palate, and lateral pharyngeal wall,
postoperative rehabilitation is essential to opti- requiring a composite resection and free flap
mize functional outcomes. reconstruction with lateral, paramedian, or
median mandibulotomy approaches.
5.5.4 Buccal Mucosa

5.5.5 Floor of Mouth
OCSCC of the buccal mucosa is an aggressive
tumor, which may be due to an intrinsic aggres- Tumors of the floor of mouth may involve the
sive biology, early invasion into the buccal fat sublingual glands, submandibular duct, and lin-
pad, lack of a substantial anatomic barrier in this gual nerve. Early lesions of the floor of mouth are
location, and the difficulty of achieving clear sur- amenable to simple excisions. Sialodochoplasty
gical margins without full-thickness resection of and stenting of the submandibular duct is required
the cheek [32]. High incidence of regional metas- if the duct lies within the resection margins and if
tasis have been reported, up to 28% in the clini- the submandibular gland is not planned for
cally negative (cN0) neck [33], as have high removal such as with a neck dissection (Fig. 5.7)
locoregional recurrence rates, ranging from 30% [37]. The stent is kept in place for 2–4 weeks and
to 80% [34–36]. Due to high recurrence rates, allows for formation of a neo-ostium in the floor
adjuvant radiotherapy should be considered in of mouth. More extensive lesions may involve
even early stage lesions [34]. the tongue, mandible, and floor of mouth muscu-
Small buccal mucosal lesions may be excised lature and require composite resection and recon-
with wide margins via a transoral approach struction of these structures. Floor of mouth
with reconstruction by primary closure, healing tumors invading the lingual periosteum or lingual
by secondary intention, local rotational flaps, or cortex may be managed with a lingual corticot-
with buccal fat advancement. Sialodochoplasty omy. Midline or paramedian mandibulotomy can
or stenting of Stenson’s duct may be required if improve access to large floor of mouth lesions.
52 J. Moe et al.
function of tumor aggressiveness rather than

inadequate surgical treatment, and highly aggres-
sive tumors have a shorter time to recurrence
regardless of margin status. In a series of 292
patients with OCSCC, margin status was not an
independent predictor of local recurrence (LR) or
OS but the histological features of worst pattern
of invasion, perineural invasion, and lymphocytic
response were [50]. Another study found
advanced tumor stage to be a predictor of locore-
gional recurrence regardless of tumor margin sta-
tus [38]. Both stage and histologic grade are
independent factors on survival and should be
considered when evaluating the need for adjuvant
therapy regardless of surgical margins.
Fig. 5.7 Stent placed in right submandibular duct during The definition of a negative margin in OCSCC
resection of floor of mouth cancer. (Courtesy of Moe and
Helman [37])
in terms of the perpendicular distance from the
tumor to the resection margin has not been stan-
dardized and remains a point of controversy.
Tumor excision with a cuff of normal adjacent
5.6 Evaluation of Margin Status tissue is the standard of care, but excessive resec-
tion of normal anatomy should be avoided in
Complete tumor resection is a fundamental prin- order to preserve functionality. Pathologic nega-
ciple in oncologic surgery. Many studies have tive margins are generally defined as greater than
shown an improvement in both disease-free sur- or equal to 5 mm, close margins are less than
vival (DFS) and overall survival (OS) when 5 mm, and a positive margin is defined as carci-
negative margins are achieved [38–49]. A series noma in situ or invasive cancer at the margin
of 148 patients with OCSCC found margin status [51]. More recently, this tenet has been ques-
to be an independent predictor of DFS with tioned, with some evidence that maintaining a
5-year local control rates of 91.0% for clear mar- narrower margins of 2.2 mm is safe and confers a
gins, 80.4% for close margins, 81.8% for dyspla- survival benefit [52]. The clinical margins should
sia at surgical margins, and 43.8% for positive be greater than the anticipated pathologic margin
margins [38]. to account for tumor shrinkage. Retraction of the
The ablative head and neck surgeon strives for mucosal margin following resection of OCSCC
curative-intent surgery, as debulking procedures has been found to be 20–25% [53]. An additional
have not shown to be worthwhile. The surgeon 10% of tissue shrinkage occurs with formalin
should identify risk factors which make complete fixation and paraffin embedding [54]. Margin
tumor extirpation more challenging. Late tumor discrepancy following resection and processing
stages involve more anatomic structures which has been reported to reach up to 75% in one study
increase the complexity of the tumor and rela- [55].
tionship to its surrounding structures; in these
cases, greater surgical complexity increases the
risk of positive margins. 5.7 argin Status on Frozen
M
However, the prognostic significance of nega- Section
tive margins remains controversial, and a number
of studies have not found an association between Margin analysis on frozen section provides abla-
tumor margin status and prognosis. It is sug- tive surgeons the opportunity to assess the ade-
gested that positive margin status may in fact be a quacy of resection in real time. Margin assessment
with frozen sections have been shown to be

highly accurate (96.7–98.4%), sensitive (72.0–
88.8%), and specific (94.4–98.9%) with a posi-
tive predictive value ranging from 77.9% to
95.7% and a negative predictive value ranging
from 96.0% to 99.2% [56–58]. Intraoperative
margin analysis can be used to guide the a dequacy
of resection and allows for positive margins to be
revised immediately to negative margins.
However, the prognostic and therapeutic impli-
cations of such revisions remain inconclusive
[59–61]. The revision of positive to negative mar- Fig. 5.8 Tumor bed margin assessment with mucosal
gins has been found in some studies to be a nega- shave margins (yellow) and deep margin sampling (red)
tive predictor of locoregional control. A cohort
study of 156 patients with OCSCC found positive and to select the locations for sampling and allow
to negative revised margins to be an independent the surgeon to sample multiple areas of the tumor
risk factor for local recurrence (LR), with a 42.2% bed quickly (Fig. 5.8). However, margins from
rate of LR as compared to 16.2% in those with the tumor bed do not assess distance of the mar-
negative margins [59]. Other studies have found gin to the tumor, and identification of small clus-
improved prognosis with re-resection of positive ters of tumor cells can be difficult because the
to negative margins. A retrospective review of 547 known position of tumor bulk is absent from the
patients found that positive to negative revised specimen [63]. In contrast, margins from the
margins in the presence of regional disease was resected specimen allow for the evaluation of
associated with poorer local control but also found tumor distance to the specimen edge but require
that revisions in the absence of regional disease orientation of the specimen by the pathologist,
lead to DSS similar to those with initially negative which is often challenging with complex oral
margins and no regional disease [60]. cavity tumors. Direct communication between
Errors in relocating the location of the positive the surgeon and pathologist is crucial to allow for
margin on frozen section on the tumor bed may an understanding of specimen orientation.
lead to sampling error during re-resection. A pro- Margin assessment from the resected speci-
spective study of one surgeon in 14 cases found a men has been shown to have improved accuracy
mean error in relocating a sample site of 9 mm and a better correlation with final margin status
for mucosal margins and 12 mm for deep mar- and patient survival [43, 64]. A retrospective
gins [62]. This would suggest that re-resection to study of 126 patients with oral tongue SCC found
negative margins may be done in the erroneous that specimen margin status was a predictor of
location and may not necessarily treat the loca- local recurrence, while tumor bed margin status
tion of the positive margin. To prevent this sam- was not, and a positive specimen margin con-
pling error, sites of frozen section should be ferred a relative risk of 2.5 for local recurrence
marked with ink, suture or staple, and wide re- [43]. Tumor-driven specimens more accurately
excisions should be completed [62]. Nevertheless, predict the completion of resection and local
it is critical that a clear surgical resection with recurrence and should thus be taken whenever
negative margins be completed on the initial possible.
attempt. Two methods for retrieving sections to assess
Margins on frozen section can be taken from margins have been described (Fig. 5.9). A radial
the tumor bed in a defect-driven approach or or perpendicular margin includes a portion of the
from the resection specimen in a tumor-driven tumor with the margin and is useful when the
approach. Defect-driven margins obviate the margin is clinically near the tumor. While it
need for the pathologist to orient the specimen allows the distance between the tumor and mar-
54 J. Moe et al.
a b
Fig. 5.9 Specimen margin assessment with (a) shave and (b) radial margins
gin to be measured, only a smaller amount of the be present in up to 34–44% of all stages and
margin is sampled. In comparison, shave margins 20–30% of early stage OCSCC [70, 71]. T stag-
include a large portion of the margin without the ing has been shown to be an independent risk fac-
tumor and are useful for tumors far from the mar- tor of nodal involvement. Elective neck dissection
gin. While shave margins allow a larger area to be (END) allows for both staging and treatment and
examined microscopically, the distance between is generally indicated in cN0 patients who are
the tumor and margin is not assessed. Shave and deemed to have a 15–20% risk of occult nodal
radial margins may both be inaccurate in tumors disease based on features of the primary tumor.
with discontinuous growth [65]. Multiple studies have found improved overall
Three types of margins should be considered: survival and decreased relapse rates with END
mucosal margins, soft tissue or deep margins, [72–78]. However, the current available evidence
and osseous margins. Deep margins include all for END is not definitive as prospective trials
connective tissue components including skeletal have produced conflicting evidence [79–82]. The
muscle, adipose tissue and neurovascular bun- treatment of all cN0 patients with END would
dles. While bone is not amenable to rapid tissue result in the overtreatment of a percentage of
analysis due to its high mineral content and need patients resulting in unnecessary cost and mor-
for decalcification, intraoperative bone margins bidity without evidence of a meaningful survival
assessment via cancellous bone sampling, mar- benefit. While END is accepted for late-stage
row cytologic assessment, cortical bone osteot- tumors, management of the cN0 neck in early
omy, and trephination have been described with stage OCSCC remains an area of debate.
high sensitivity (79–89%), specificity (98– Treatment options for T1T2N0 OCSCC include
100%), and accuracy (94–100%) [54, 66–69]. END, sentinel node biopsy (SNB), radiation, or
watchful waiting with therapeutic neck dissec-
tion in the case of nodal relapse.
5.8 Management A 2015 randomized controlled clinical trial of
of the Clinically Negative 500 patients with T1T2N0 OCSCC found
Neck improved 3-year OS (80.0% vs. 67.5%) and DFS
(69.5% vs. 45.9%) in patients that underwent an
Cervical node status is an independent predictor elective neck dissection versus those treated with
of survival in OCSCC, and nodal metastasis surveillance and possible therapeutic neck dis-
decreases 5-year survival by approximately 50%. section at a later date [83]. Multiple limitations of
Clinically overt nodal disease is present in 30% this study have been cited, including short fol-
of all patients with OCSCC. However, in those low-up duration, the omission of radiation as an
with a clinically N0 neck, occult metastasis may option in the therapeutic neck dissection group,
Table 5.1 Cutoff DOI for oral cavity subsites over which Table 5.2 Classification of neck dissection
neck dissection is indicated
Nonlymphatic
Oral cavity Recommendation for neck Type of neck structures
subsite dissection, mm dissection Levels removed removed
Tongue 2 Radical I, II, III, IV, V SCM, IJV,
Floor of mouth 2–3 SAN
Retromolar 3–4 Modified radical, I, II, III, IV, V SCM, IJV
trigone type I (SAN spared)
Alveolus/hard 3–4 Modified radical, I, II, III, IV, V SCM (IJV,
palate type II SAN spared)
All sites 2–4 Modified radical, I, II, III, IV, V (SCM, IJV,
type III SAN spared)
Adapted from Brockhoff et al. [85]
(Functional)
Selective neck
and the inclusion of primarily high-risk patients dissection
Supraomohyoid I–III
[tongue tumors and primarily tumors with a
Lateral II–IV
depth of invasion (DOI) of greater than 3 mm] Posterolateral II–V
who would benefit from a neck dissection regard- Anterior VI, VII
less. In addition, on multivariate analysis, while Extended radical I, II, III, IV, V, Additional
there was a survival benefit with END noted for additional nonlymphatic
T2 tumors, there was no survival benefit seen for lymphatic groups structures (e.g.,
(e.g., skin, platysma,
T1 tumors or for tumors less than 3 mm in thick- retropharyngeal, digastric
ness. While END is generally indicated for T2 central muscle, carotid
and above, the higher risk cohort of patients compartment, artery,
within the T1 N0 group has not yet been identi- mediastinal hypoglossal
nodes) nerve, vagus
fied who would benefit from END. nerve)
Histologic features including DOI and tumor
SCM sternocleidomastoid muscle, IJV internal jugular
grade have been used to determine the need for vein, SAN spinal accessory nerve
END in early-stage SCC. DOI is an independent
predictor of occult nodal metastasis; however, the
cutoff DOI at which an END is indicated has not In the clinically negative neck, the selective neck
been standardized and ranges from 2 to 5 mm dissection (SND) has been shown to be effective,
[84]. A retrospective review of 286 patients with but the extent of dissection has not been agreed
OCSCC identified critical tumor depths at which upon. The goal of the SND is to achieve similar
the risk of occult metastasis exceeded 20% for rates of regional control with less morbidity and
various oral cavity subsites based on the correla- operative time as compared with the RND.
tion between tumor DOI and nodal positivity in The pattern of cervical node metastasis in
surgical specimens [85]. The threshold DOIs at OCSCC have been established. OCSCC typically
which END is indicated at various oral cavity follows a sequential metastatic pattern with the
subsites are shown in Table 5.1. involvement of successive anatomic nodal levels
from level I to V, with levels I, II, and III at greatest
risk for nodal metastasis (Fig. 5.10). The risk of
5.9 Extent of Neck Dissection lower neck level metastasis and the risk of skip
metastases, or the involvement of higher level
The extent and type of neck dissection required nodes without the involvement of first echelon or
as a staging and therapeutic procedure remains a intermediary node groups, threaten the efficacy of
point of controversy (Table 5.2). The radical neck a selective neck dissection in regional control [87].
dissection (RND) described by Crile in 1906 has The reported incidence of skip metastasis past
been traditionally used to treat nodal metastasis levels I and II is rare, ranging from 0% to 2%,
but is associated with significant morbidity [86]. with no skip metastasis to level V and an inci-
56 J. Moe et al.
limited N+ OCSCC, with adjuvant radiotherapy

essential for disease control in these cases [91].
5.10 Sentinel Node Biopsy

IB IIB
IA IIA
SNB is an alternative management option of the
neck in T1T2N0 OCSCC. The procedure relies
III on the theory that a primary oral cavity tumor
VA
drains first to a sentinel lymph node in the neck
before subsequently spreading to the remaining
IV
VI VB lymph node basin. SNB has the advantages of
being less invasive, more cost effective and
results in a better patient quality of life; however,
the diagnostic efficacy of SNB in early OCSCC
VII is still debated [71].
Multiple techniques of SNB have been
described. In general, a radioactive tracer is
Fig. 5.10 Levels of the neck
injected around the tumor preoperatively.
Preoperative lymphoscintigraphy, intraoperative
dence of 0–1.9% to level IV [87]. A supra- gamma probe, and/or blue dye are used for senti-
omohyoid neck dissection (levels I–III) is thus nel node localization. Following sentinel lymph
indicated for most N0 OCSCC. Some authors node removal, serial sectioning and immunohisto-
advocate for the inclusion of level IV in SND for chemical staining are completed for the detection
all OCSCC as the procedure has minimal mor- of micrometastasis. Superselection of the nodes
bidity and does not significantly increase operat- most likely reflecting the disease status of the rest
ing time [88]. In oral tongue cancer, while many of the neck reduces the number of lymph nodes
studies have found no level IV involvement, up to for pathologic evaluation and allows for more in-
8% incidence of skip metastasis or subsequent depth evaluation of the small number of sentinel
recurrence in level IV has been reported. Many nodes [92]. While pathologic evaluation of the
authors advocate for SND including levels I–IV sentinel node currently is done in a delayed fash-
for OCSCC of the tongue. ion, early trials of intraoperative sentinel node
In the N+ neck, the incidence of level IV and V evaluation techniques show promise [93, 94].
involvement ranges between 6.5% and 15% and A meta-analysis of 66 studies of T1T2N0
between 2% and 6.9%, respectively [87, 89]. OCDCC found that SNB had a pooled identifi-
Therefore, modified radical neck dissection cation rate of 96.3%, a sensitivity of 87%, a
(MRND) or RND with removal of neck levels I to negative predictive value of 94%, and an overall
V is indicated in the N+ neck. When possible, the diagnostic efficacy of 94% [71]. The addition of
MRND has been shown to have similar oncologi- immunohistochemistry was found to improve
cal efficacy as compared to the RND, with compa- SNB diagnostic sensitivity. SNB has high diag-
rable number of recovered lymph nodes and with nostic accuracy in T1T2N0 OCSCC and is an
less morbidity including shoulder dysfunction acceptable alternative to END. However, senti-
[90]. In a recent meta-analysis, SND was found to nel node analysis is currently completed by
have similar locoregional control rates as com- postoperative pathologic procedure, and the
pared to RND or MRND for select patients with clinical applicability of SNB by frozen section
has not yet been validated. Furthermore, the

implementation of SNB requires protocolized
co-ordination between surgery, pathology, and
radiology departments, as well as hospital ser-
vices; thus, SND has not been broadly adopted
at present.
5.11 Dental Rehabilitation
Dental rehabilitation has become an integral

aspect of the reconstructive plan following abla-
tive and reconstructive surgery for patients with
OCSCC. Dental implants are often necessary for Fig. 5.11 Maxillectomy defect with zygomatic implants
prosthetic rehabilitation due to severe alterations in place
of normal anatomy of the dental arches preclud-
ing traditional prosthetic options. In OCSCC,
dental implants are used for both the retention of Bony free flap reconstruction of the maxilla
palatal obturators as well as for dental rehabilita- and mandible allows for dental rehabilitation
tion following bony reconstruction of the maxilla with endosseous implants. Realistically, only a
or mandible. Dental prosthetic design and fabri- small percentage of patients will ultimately
cation are discussed elsewhere. receive implant-supported prostheses due to sev-
Palatal obturators are the traditional method of eral contributing factors including financial con-
reconstructing maxillary defects, but retention straints, disease progression, flap complications,
and support by the remaining dentition, remain- implant failure, trismus, and an inability to toler-
ing palate, and maxillectomy cavity can be com- ate a prosthesis. The reported rates of dental
promised in extensive defects. Implant-retained rehabilitation following free flap reconstruction
palatal obturators increase prosthetic retention are low, ranging from 2% to 46% [98, 99]. The
and support, minimize unfavorable forces on the effect of irradiation on implant success is contro-
remaining dentition, and distribute force to the versial. The success rate of endosseous implants
facial skeleton [95]. in irradiated native bone ranges between 74%
Dental implants may be placed in remaining and 97% [100]. Implant success in osseous free
alveolar bone or bone surrounding the maxillec- flaps has not been shown to be significantly dif-
tomy site. Zygomatic implants are an effective ferent with irradiation or with the timing of irra-
alternative when conventional implant placement diation, with success rates ranging from 72.5%
is not possible due to extensive bony resection, to 97.5% [101–103].
and they allow obturation in extensive defects When implant restoration is planned, the
otherwise only amenable to free flap reconstruc- prosthetic plan should be discussed with the
tion (Fig. 5.11). A retrospective review of 28 restoring dentist or prosthodontist prior to sur-
zygomatic implants in nine patients following gery. The DCIA and fibula free flap both provide
maxillectomy found a 79% success rate of zygo- adequate bone stock for implant placement. The
matic implants, with failures attributed to radia- DCIA free flap provides adequate bone volume
tion therapy [96]. Other studies have reported and height to restore the alveolus but has limited
greater success rates of zygomatic implant reha- bone length, a shorter vascular pedicle, and
bilitation in maxillectomy defects ranging from higher risk of significant donor site morbidities.
94.1% to 100% [95–97]. The fibula free flap provides a long bony seg-
58 J. Moe et al.
Fig. 5.12 Virtual surgery planning for resection of an anterior mandibular squamous cell carcinoma and reconstruction
with fibular free flap
ment with a long pedicle, easy to harvest, but segments (Fig. 5.12) [105, 106]. The use of vir-
lacks vertical bone height with the fibula averag- tual planning and guided surgery have been
ing 13–15 mm in height. To improve dental shown to be financially favorable with costs off-
implant placement, discrepancies between the set by decreased operative time and subsequently
height of the dentate mandible and fibula may be decreased operating room cost [107].
overcome by fixating the fibula superiorly to the Surgical navigation is a useful tool to guide
inferior mandibular border, distraction osteogen- oncologic resections and to improve the preci-
esis of fibular segments, or performing a double sion of reconstruction (Fig. 5.13) [108]. Image-
barrel technique [104]. guided resection has been described for
resection of advanced tumors involving the skull
base, sinuses, and infratemporal fossa. It has
5.12 Adjunct Surgical Tools been suggested to improve the accuracy and
safety in these cases, potentially leading to bet-
Virtual surgery planning (VSP) and guided sur- ter local disease control [109]. Surgical naviga-
gery using three-dimensional printing technol- tion has also been described for the
ogy have allowed for patient-specific, highly reconstruction of orbital floor defects with max-
precise bony ablation, and reconstruction in illectomy [110]. The utility of navigation in
OCSCC. VSP confers reproducible accuracy for OCSCC has only started to be explored, and
maxillary and mandibular reconstructions, with a navigation may be a useful adjunct in the man-
great benefit for reconstructions with multiple agement of OCSCC in the future.
a b
Fig. 5.13 Surgical navigation patient tracking system (a) and display (b) used in the resection of a maxillary tumor
5.13 Conclusion nccn.org/professionals/physician_gls/default.aspx.

Accessed 23 Sep 2018.
3. Schwam ZG, Judson BL. Improved prognosis for
Surgery remains the primary treatment modality patients with oral cavity squamous cell carcinoma:
for OCSCC, and as such, the role of the surgeon analysis of the National Cancer Database 1998–2006.
is pivotal in the management of patients with Oral Oncol. 2016;52:45.
4. Piccirillo JF. Importance of comorbidity in head and
OCSCC. By considering complex patient-related, neck cancer. Laryngoscope. 2000;110:593.
tumor-related, and procedure-related factors, the 5. Paleri V, Wight RG, Silver CE, Haigentz M, Takes RP,
surgeon can optimize the surgical management of Bradley PJ, Rinaldo A, Sanabria A, Bień S, Ferlito
OCSCC and can improve surgical outcomes and A. Comorbidity in head and neck cancer: a critical
appraisal and recommendations for practice. Oral
overall prognosis. Oncol. 2010;46:712.
6. Ferrier MB, Spuesens EB, Le Cessie S, Baatenburg
Acknowledgements The authors would like to acknowl- de Jong RJ. Comorbidity as a major risk factor
edge Raea Bucher, illustrator, for providing Figs. 5.1–5.5 for mortality and complications in head and neck
and 5.10 for this chapter. surgery. Arch Otolaryngol Head Neck Surg. 2005;
131:27.
7. Jun MY, Strong EW, Saltzman EI, Gerold FP. Head and
neck cancer in the elderly. Head Neck. 1983;5:376.
References 8. Yamada S-I, Kurita H, Tomioka T, Ohta R, Yoshimura
N, Nishimaki F, Koyama Y, Kondo E, Kamata
1. Global Cancer Observatory. Cancer today. http://gco. T. Healthy life expectancy of oral squamous cell car-
iarc.fr/today/fact-sheets-populations. Accessed 23 Sep cinoma patients aged 75 years and older. Oral Oncol.
2018. 2017;64:22.
2. National Comprehensive Cancer Network. NCCN 9. Datema FR, Ferrier MB, van der Schroeff MP, de Jong
clinical practice guidelines in oncology. https://www. RJB. Impact of comorbidity on short-term mortality
60 J. Moe et al.
and overall survival of head and neck cancer patients. 25. Li W, Li R, Safdar J, Huang S, Xu Z, Tan X, Sun
Head Neck. 2010;32:728. C. Modified visor approach applied to total or subto-
10. Pogrel MA, Podlesh S, Anthony JP, Alexander J. A tal glossectomy and reconstruction: avoidance of lip
comparison of vascularized and nonvascularized bone splitting and mandibulotomy and cutting off mental
grafts for reconstruction of mandibular continuity nerve. Tumour Biol. 2014;35:7847.
defects. J Oral Maxillofac Surg. 1997;55:1200. 26. Yuen PW, Lam KY, Chan AC, Wei WI, Lam
11. McGregor AD, MacDonald DG. Patterns of spread LK. Clinicopathological analysis of local spread of
of squamous cell carcinoma to the ramus of the man- carcinoma of the tongue. Am J Surg. 1998;175:242.
dible. Head Neck. 1993;15:440. 27. Guerin-Lebailly C, Mallet Y, Lambour V, Fournier C,
12. He Y, Zhang ZY, Zhu HG, Sader R, He J, Kovács Bedoui El S, Ton Van J, Lefebvre JL. Functional and
AF. Free fibula osteocutaneous flap for primary recon- sensitive outcomes after tongue reconstruction: about
struction of T3–T4 gingival carcinoma. J Craniofac a series of 30 patients. Oral Oncol. 2012;48:272.
Surg. 2010;21:301. 28. Kuriakose MA, Loree TR, Spies A, Meyers S, Hicks
13. Pandey M, Rao LP, Das SR, Mathews A, Chacko EM, WL. Sensate radial forearm free flaps in tongue
Naik BR. Patterns of mandibular invasion in oral squa- reconstruction. Arch Otolaryngol Head Neck Surg.
mous cell carcinoma of the mandibular region. World J 2001;127:1463.
Surg Oncol. 2007;5:12. 29. Namin AW, Varvares MA. Functional outcomes of
14. Jewer DD, Boyd JB, Manktelow RT, Zuker RM, Rosen sensate versus insensate free flap reconstruction in
IB, Gullane PJ, Rotstein LE, Freeman JE. Orofacial oral and oropharyngeal reconstruction: a systematic
and mandibular reconstruction with the iliac crest free review. Head Neck. 2016;38:1717.
flap: a review of 60 cases and a new method of clas- 30. Dziegielewski PT, Ho ML, Rieger J, Singh P, Langille
sification. Plast Reconstr Surg. 1989;84:391. M, Harris JR, Seikaly H. Total glossectomy with
15. Urken ML, Vickery C, Weinberg H, Buchbinder D, laryngeal preservation and free flap reconstruction:
Lawson W, Biller HF. The internal oblique-iliac crest objective functional outcomes and systematic review
osseomyocutaneous free flap in oromandibular recon- of the literature. Laryngoscope. 2013;123:140.
struction: report of 20 cases. Arch Otolaryngol Head 31. Lin DT, Yarlagadda BB, Sethi RKV, Feng AL,
Neck Surg. 1989;115:339. Shnayder Y, Ledgerwood LG, Diaz JA, Sinha P,
16. Kekatpure VD, Manjula BV, Mathias S, Trivedi NP, Hanasono MM, Yu P, Skoracki RJ, Lian TS, Patel UA,
Selvam S, Kuriakose MA. Reconstruction of large Leibowitz J, Purdy N, Starmer H, Richmon JD. Long-
composite buccal defects using single soft tissue term functional outcomes of total glossectomy with or
flap—analysis of functional outcome. Microsurgery. without total laryngectomy. JAMA Otolaryngol Head
2013;33:184. Neck Surg. 2015;141:797.
17. Brown JS, Rogers SN, McNally DN, Boyle M. A 32. Lubek JE, Clayman L. An update on squamous car-
modified classification for the maxillectomy defect. cinoma of the oral cavity, oropharynx, and maxil-
Head Neck. 2000;22:17. lary sinus. Oral Maxillofac Surg Clin North Am.
18. Okay DJ, Genden E, Buchbinder D, Urken 2012;24:307.
M. Prosthodontic guidelines for surgical reconstruc- 33. Niu LX, Feng Z, Li JN, Li CZ, Peng X, Guo
tion of the maxilla: a classification system of defects. J CB. Prognostic factors of squamous cell carcinoma
Prosthet Dent. 2001;86:352. of the buccal mucosa: a retrospective study of 168
19. Maniglia AJ. Indications and techniques of midfacial cases in North China. J Oral Maxillofac Surg. 2014;
degloving: a 15-year experience. Arch Otolaryngol 72:2344.
Head Neck Surg. 1986;112:750. 34. Lin C-S, Jen Y-M, Cheng M-F, Lin Y-S, Su W-F,
20. Davison SP, Sherris DA, Meland NB. An algorithm for Hwang J-M, Chang L-P, Chao H-L, Liu D-W, Lin
maxillectomy defect reconstruction. Laryngoscope. H-Y, Shum W-Y. Squamous cell carcinoma of the buc-
1998;108:215. cal mucosa: an aggressive cancer requiring multimo-
21. Sharaf MY, Ibrahim SI, Eskander AE, Shaker dality treatment. Head Neck. 2006;28:150.
AF. Prosthetic versus surgical rehabilitation in patients 35. Jan J-C, Hsu W-H, Liu S-A, Wong Y-K, Poon C-K,
with maxillary defect regarding the quality of life: sys- Jiang R-S, Jan J-S, Chen IF. Prognostic factors in
tematic review. Oral Maxillofac Surg. 2018;22:1. patients with buccal squamous cell carcinoma: 10-year
22. Futran ND, Wadsworth JT, Villaret D, Farwell experience. J Oral Maxillofac Surg. 2011;69:396.
DG. Midface reconstruction with the fibula free flap. 36. Chiou W-Y, Lin H-Y, Hsu F-C, Lee M-S, Ho H-C,
Arch Otolaryngol Head Neck Surg. 2002;128:161. Su Y-C, Lee C-C, Hsieh C-H, Wang Y-C, Hung
23. Uglešić V, Virag M, Varga S, Knežević P, Milenović S-K. Buccal mucosa carcinoma: surgical margin less
A. Reconstruction following radical maxillectomy than 3 mm, not 5 mm, predicts locoregional recur-
with flaps supplied by the subscapular artery. J Cranio- rence. Radiat Oncol. 2010;5:79.
Maxillofac Surg. 2000;28:153. 37. Moe J, Helman JI. Surgical techniques for the man-
24. Moreno MA, Skoracki RJ, Hanna EY, Hanasono agement of submandibular salivary duct strictures.
MM. Microvascular free flap reconstruction versus Atlas Oral Maxillofac Surg Clin. 2018;26:99.
palatal obturation for maxillectomy defects. Head 38. Kurita H, Nakanishi Y, Nishizawa R, Xiao T, Kamata
Neck. 2010;32:860. T, Koike T, Kobayashi H. Impact of different surgical
margin conditions on local recurrence of oral squa- 52. Zanoni DK, Migliacci JC, Xu B, Katabi N, Montero
mous cell carcinoma. Oral Oncol. 2010;46:814. PH, Ganly I, Shah JP, Wong RJ, Ghossein RA, Patel
39. Loree TR, Strong EW. Significance of positive mar- SG. A proposal to redefine close surgical margins in
gins in oral cavity squamous carcinoma. Am J Surg. squamous cell carcinoma of the oral tongue. JAMA
1990;160:410. Otolaryngol Head Neck Surg. 2017;143:555.
40. Ravasz LA, Slootweg PJ, Hordijk GJ, Smit F, van der 53. Mistry RC, Qureshi SS, Kumaran C. Post-resection
Tweel I. The status of the resection margin as a prog- mucosal margin shrinkage in oral cancer: quantifica-
nostic factor in the treatment of head and neck carci- tion and significance. J Surg Oncol. 2005;91:131.
noma. J Cranio-Maxillofac Surg. 1991;19:314. 54. Hinni ML, Ferlito A, Brandwein-Gensler MS, Takes
41. Jacobs JR, Ahmad K, Casiano R, Schuller DE, Scott RP, Silver CE, Westra WH, Seethala RR, Rodrigo JP,
C, Laramore GE, Sarraf Al M. Implications of positive Corry J, Bradford CR, Hunt JL, Strojan P, Devaney
surgical margins. Laryngoscope. 1993;103:64. KO, Gnepp DR, Hartl DM, Kowalski LP, Rinaldo A,
42. Binahmed A, Nason RW, Abdoh AA. The clinical sig- Barnes L. Surgical margins in head and neck cancer: a
nificance of the positive surgical margin in oral cancer. contemporary review. Head Neck. 2013;35:1362.
Oral Oncol. 2007;43:780. 55. Cheng A, Cox D, Schmidt BL. Oral squamous cell
43. Chang AMV, Kim SW, Duvvuri U, Johnson JT, Myers carcinoma margin discrepancy after resection and
EN, Ferris RL, Gooding WE, Seethala RR, Chiosea pathologic processing. J Oral Maxillofac Surg.
SI. Early squamous cell carcinoma of the oral tongue: 2008;66:523.
comparing margins obtained from the glossectomy 56. Du E, Ow TJ, Lo Y-T, Gersten A, Schiff BA, Tassler
specimen to margins from the tumor bed. Oral Oncol. AB, Smith RV. Refining the utility and role of frozen
2013;49:1077. section in head and neck squamous cell carcinoma
44. Anderson CR, Sisson K, Moncrieff M. A meta- resection. Laryngoscope. 2016;126:1768.
analysis of margin size and local recurrence in 57. DiNardo LJ, Lin J, Karageorge LS, Powers
oral squamous cell carcinoma. Oral Oncol. 2015; CN. Accuracy, utility, and cost of frozen section mar-
51:464. gins in head and neck cancer surgery. Laryngoscope.
45. Byers RM, Bland KI, Borlase B, Luna M. The prognos- 2000;110:1773.
tic and therapeutic value of frozen section determina- 58. Ord RA, Aisner S. Accuracy of frozen sections in
tions in the surgical treatment of squamous carcinoma assessing margins in oral cancer resection. J Oral
of the head and neck. Am J Surg. 1978;136:525. Maxillofac Surg. 1997;55:663.
46. Chen TY, Emrich LJ, Driscoll DL. The clinical signifi- 59. Ettl T, El-Gindi A, Hautmann M, Gosau M, Weber
cance of pathological findings in surgically resected F, Rohrmeier C, Gerken M, Müller S, Reichert T,
margins of the primary tumor in head and neck carci- Klingelhöffer C. Positive frozen section margins pre-
noma. Int J Rad Oncol Biol Phys. 1987;13:833. dict local recurrence in R0-resected squamous cell car-
47. Jones AS. Prognosis in mouth cancer: tumour factors. cinoma of the head and neck. Oral Oncol. 2016;55:17.
Eur J Cancer B Oral Oncol. 1994;30:8. 60. Patel RS, Goldstein DP, Guillemaud J, Bruch GA,
48. Scholl P, Byers RM, Batsakis JG, Wolf P, Santini Brown D, Gilbert RW, Gullane PJ, Higgins KM, Irish
H. Microscopic cut-through of cancer in the surgi- J, Enepekides DJ. Impact of positive frozen section
cal treatment of squamous carcinoma of the tongue: microscopic tumor cut-through revised to negative on
prognostic and therapeutic implications. Am J Surg. oral carcinoma control and survival rates. Head Neck.
1986;152:354. 2010;32:1444.
49. Sieczka E, Datta R, Singh A, Loree T, Rigual N, Orner 61. Varvares MA, Poti S, Kenyon B, Christopher K,
J, Hicks W Jr. Cancer of the buccal mucosa: are mar- Walker RJ. Surgical margins and primary site resec-
gins and T-stage accurate predictors of local control? tion in achieving local control in oral cancer resec-
Am J Otolaryngol. 2001;22:395. tions. Laryngoscope. 2015;125:2298.
50. Brandwein-Gensler M, Teixeira MS, Lewis CM, Lee 62. Kerawala CJ, Ong TK. Relocating the site of frozen
B, Rolnitzky L, Hille JJ, Genden E, Urken ML, Wang sections? Is there room for improvement? Head Neck.
BY. Oral squamous cell carcinoma: histologic risk 2001;23:230.
assessment, but not margin status, is strongly predic- 63. Shapiro M, Salama A. Margin analysis: squamous cell
tive of local disease-free and overall survival. Am J carcinoma of the oral cavity. Oral Maxillofac Surg
Surg Pathol. 2005;29:167. Clin. 2017;29:259.
51. Colevas AD, Yom SS, Pfister DG, Spencer S, 64. Yahalom R, Dobriyan A, Vered M, Talmi YP, Teicher
Adelstein D, Adkins D, Brizel DM, Burtness B, Busse S, Bedrin L. A prospective study of surgical margin
PM, Caudell JJ, Cmelak AJ, Eisele DW, Fenton M, status in oral squamous cell carcinoma: a preliminary
Foote RL, Gilbert J, Gillison ML, Haddad RI, Hicks report. J Surg Oncol. 2008;98:572.
WL, Hitchcock YJ, Jimeno A, Leizman D, Maghami 65. Chiosea SI. Intraoperative margin assessment in early
E, Mell LK, Mittal BB, Pinto HA, Ridge JA, Rocco J, oral squamous cell carcinoma. Surg Pathol Clin.
Rodriguez CP, Shah JP, Weber RS, Witek M, Worden 2017;10:1.
F, Zhen W, Burns JL, Darlow SD. NCCN guidelines 66. Arick Forrest L, Schuller DE, Karanfilov B, Lucas
insights: head and neck cancers, version 1.2018. J Natl JG. Update on intraoperative analysis of mandibular
Compr Canc Netw. 2018;16:479. margins. Am J Otolaryngol. 1997;18:396.
62 J. Moe et al.
67. Mahmood S, Conway D, Ramesar KCRB. Use of Lam AKY, Yuen NWF, Smith NJT, Chan YW, Wong
intraoperative cytologic assessment of mandibular BYH, Li GKH, Ho ACW, Ho WK, Wong SY, Yao
marrow scrapings to predict resection margin sta- TJ. Prospective randomized study of selective neck
tus in patients with squamous cell carcinoma. J Oral dissection versus observation for N0 neck of early
Maxillofac Surg. 2001;59:1138. tongue carcinoma. Head Neck. 2009;31:765.
68. Oxford LE, Ducic Y. Intraoperative evaluation of 83. D'Cruz AK, Vaish R, Kapre N, Dandekar M, Gupta
cortical bony margins with frozen-section analysis. S, Hawaldar R, Agarwal JP, Pantvaidya G, Chaukar
Otolaryngol Head Neck Surg. 2016;134:138. D, Deshmukh A, Kane S, Arya S, Ghosh-Laskar S,
69. Wysluch A, Stricker I, Hölzle F, Wolff K-D, Maurer Chaturvedi P, Pai P, Nair S, Nair D, Badwe R, Head
P. Intraoperative evaluation of bony margins with fro- and Neck Disease Management Group. Elective ver-
zen-section analysis and trephine drill extraction tech- sus therapeutic neck dissection in node-negative oral
nique: a preliminary study. Head Neck. 2010;32:1473. cancer. N Engl J Med. 2015;373:521.
70. Greenberg JS, Naggar El AK, Mo V, Roberts D, 84. Pentenero M, Gandolfo S, Carrozzo M. Importance of
Myers JN. Disparity in pathologic and clinical lymph tumor thickness and depth of invasion in nodal involve-
node staging in oral tongue carcinoma. Cancer. ment and prognosis of oral squamous cell carcinoma:
2003;98:508. a review of the literature. Head Neck. 2005;27:1080.
71. Liu M, Wang SJ, Yang X, Peng H. Diagnostic efficacy 85. Brockhoff HC, Kim RY, Braun TM, Skouteris C,
of sentinel lymph node biopsy in early oral squamous Helman JI, Ward BB. Correlating the depth of invasion
cell carcinoma: a meta-analysis of 66 studies. PLoS at specific anatomic locations with the risk for regional
ONE. 2017;12:e0170322. metastatic disease to lymph nodes in the neck for oral
72. Lydiatt DD, Robbins KT, Byers RM, Wolf squamous cell carcinoma. Head Neck. 2017;39:974.
PF. Treatment of stage I and II oral tongue cancer. 86. Crile G. Landmark article Dec 1, 1906: excision of
Head Neck. 1993;15:308. cancer of the head and neck. With special reference to
73. Yuen APW, Wei WI, Wong YM, Tang KC. Elective the plan of dissection based on one hundred and thirty-
neck dissection versus observation in the treatment of two operations. JAMA. 1987;258:3286–93.
early oral tongue carcinoma. Head Neck. 1997;19:583. 87. Patel S, Sakthivel P, Singh I, Gulati A, Gupta
74. Haddadin KJ, Soutar DS, Oliver RJ, Webster MH, D. Frequency of skip metastases in oral cancer: an
Robertson AG, MacDonald DG. Improved survival overview. Int J Head Neck Surg. 2015;6:80.
for patients with clinically T1/T2, N0 tongue tumors 88. Cariati P, Cabello Serrano A, Fernandez Solis J,
undergoing a prophylactic neck dissection. Head Martinez Lara I. Distribution of cervical metastasis
Neck. 1999;21:517. in tongue cancer: are occult metastases predictable?
75. Keski-Säntti H, Atula T, Törnwall J, Koivunen P, A retrospective study of 117 oral tongue carcinomas. J
Mäkitie A. Elective neck treatment versus observation Craniomaxillofac Surg. 2018;46:155.
in patients with T1/T2 N0 squamous cell carcinoma of 89. Feng Z, Li JN, Niu LX, Guo CB. Supraomohyoid neck
oral tongue. Oral Oncol. 2006;42:95. dissection in the management of oral squamous cell
76. Capote A, Escorial V, Guerra MFM, Campo FJR, carcinoma: special consideration for skip metastases
Gamallo C, Naval L. Elective neck dissection in early- at level IV or V. J Oral Maxillofac Surg. 2014;72:1203.
stage oral squamous cell carcinoma—does it influence 90. Kohler HF, Cunha IWD, Kowalski LP. Impact of
recurrence and survival? Head Neck. 2007;29:3. modified radical neck dissections on the number
77. Huang SF, Kang C-J, Lin CY, Fan KH, Yen T-C, Wang of retrieved nodes, recurrence and survival. Braz J
H-M, Chen IH, Liao C-T, Cheng AJ, Chang JTC. Neck Otorhinolaryngol. 2010;76:374.
treatment of patients with early stage oral tongue can- 91. Rodrigo JP, Grilli G, Shah JP, Medina JE, Robbins KT,
cer. Cancer. 2008;112:1066. Takes RP, Hamoir M, Kowalski LP, Suárez C, López F,
78. Huang SF, Tung-Chieh Chang J, Liao C-T, Kang C-J, Quer M, Boedeker CC, de Bree R, Coskun H, Rinaldo
Lin C-Y, Fan KH, Wang H-M, Chen IH. The role of A, Silver CE, Ferlito A. Selective neck dissection in
elective neck dissection in early stage buccal cancer. surgically treated head and neck squamous cell carci-
Laryngoscope. 2014;125:128. noma patients with a clinically positive neck: system-
79. Vandenbrouck C, Garnier HS, Chassagne D, Saravane atic review. Eur J Surg Oncol. 2018;44:395.
D, Cachin Y, Micheau C. Elective versus therapeutic 92. Alkureishi LWT, Ross GL, Shoaib T, Soutar DS,
radical neck dissection in epidermoid carcinoma of Robertson AG, Thompson R, Hunter KD, Sorensen
the oral. Cavity results of a randomized clinical trial. JA, Thomsen J, Krogdahl A, Alvarez J, Barbier L,
Cancer. 1980;46:386. Santamaria J, Poli T, Sesenna E, Kovács AF, Grünwald
80. Fakih AR, Rao RS, Borges AM, Patel AR. Elective F, Barzan L, Sulfaro S, Alberti F. Sentinel node biopsy
versus therapeutic neck dissection in early carcinoma in head and neck squamous cell cancer: 5-year follow-
of the oral tongue. Am J Surg. 1989;158:309. up of a European Multicenter Trial. Ann Surg Oncol.
81. Kligerman J, Lima RA, Soares JR, Prado L, Dias FL, 2010;17:2459.
Freitas EQ, Olivatto LO. Supraomohyoid neck dissec- 93. Ferris RL, Stefanika P, Xi L, Gooding W, Seethala
tion in the treatment of T1/T2 squamous cell carci- RR, Godfrey TE. Rapid molecular detection of meta-
noma of oral cavity. Am J Surg. 1994;168:391. static head and neck squamous cell carcinoma as an
82. Yuen APW, Ho CM, Chow TL, Tang LC, Cheung WY, intraoperative adjunct to sentinel lymph node biopsy.
Ng RWM, Wei WI, Kong CK, Book KS, Yuen WC, Laryngoscope. 2012;122:1020.
94. Shaw R, Christensen A, Java K, Maddani El R, 103. Jackson RS, Price DL, Arce K, Moore EJ. Evaluation
Liloglou T, Asterios T, Buchwald von C, Wessel of clinical outcomes of osseointegrated dental
I, Kiss K, Kjaer A, Lelkaitis G, Long A, Risk J, implantation of fibula free flaps for mandibular recon-
Robinson M. Intraoperative sentinel lymph node struction. JAMA Facial Plast Surg. 2016;18:201.
evaluation: implications of cytokeratin 19 expres- 104. He Y, Zhang ZY, Zhu HG, Wu YQ, Fu HH. Double-
sion for the adoption of OSNA in oral squamous cell Barrel fibula vascularized free flap with dental
carcinoma. Ann Surg Oncol. 2016;23:4042. rehabilitation for mandibular reconstruction. J Oral
95. Boyes-Varley JG, Howes DG, Davidge-Pitts KD, Maxillofac Surg. 2011;69:2663.
Branemark I, McAlpine JA. A protocol for maxillary 105. Zweifel DF, Simon C, Hoarau R, Pasche P, Broome
reconstruction following oncology resection using M. Are virtual planning and guided surgery for head
zygomatic implants. Int J Prosthodont. 2007;20:521. and neck reconstruction economically viable? J Oral
96. Schmidt BL, Pogrel MA, Young CW, Sharma Maxillofac Surg. 2015;73:170.
A. Reconstruction of extensive maxillary defects 106. Metzler P, Geiger EJ, Alcon A, Ma X, Steinbacher
using zygomaticus implants. J Oral Maxillofac Surg. DM. Three-dimensional virtual surgery accuracy
2004;62:82. for free fibula mandibular reconstruction: planned
97. Zwahlen RA, Grätz KW, Oechslin CK, Studer versus actual results. J Oral Maxillofac Surg.
SP. Survival rate of zygomatic implants in atrophic 2014;72:2601.
or partially resected maxillae prior to functional 107. Roser SM, Ramachandra S, Blair H, Grist W,
loading: a retrospective clinical report. Int J Oral Carlson GW, Christensen AM, Weimer KA, Steed
Maxillofac Implants. 2006;21:413. MB. The accuracy of virtual surgical planning in
98. van Gemert JTM, van Es RJJ, Rosenberg AJWP, van free fibula mandibular reconstruction: comparison
der Bilt A, Koole R, Van Cann EM. Free vascular- of planned and final results. J Oral Maxillofac Surg.
ized flaps for reconstruction of the mandible: com- 2010;68:2824.
plications, success, and dental rehabilitation. J Oral 108. Azarmehr I, Stokbro K, Bell RB, Thygesen
Maxillofac Surg. 2012;70:1692. T. Surgical navigation: a systematic review of
99. Barrowman RA, Wilson PR, Wiesenfeld D. Oral indications, treatments, and outcomes in oral and
rehabilitation with dental implants after cancer treat- maxillofacial surgery. J Oral Maxillofac Surg.
ment. Aust Dent J. 2011;56:160. 2017;75:1987.
100. Shugaa-Addin B, Al-Shamiri H-M, Al-Maweri S, 109. Feichtinger M, Pau M, Zemann W, Aigner RM,
Tarakji B. The effect of radiotherapy on survival of Kärcher H. Intraoperative control of resection
dental implants in head and neck cancer patients. J margins in advanced head and neck cancer using
Clin Exp Dent. 2016;8:e194. a 3D-navigation system based on PET/CT image
101. Sclaroff A, Haughey B, Gay WD, Paniello fusion. J Cranio-Maxillofac Surg. 2010;38:589.
R. Immediate mandibular reconstruction and place- 110. Zhang W-B, Mao C, Liu X-J, Guo CB, Yu G-Y, Peng
ment of dental implants: at the time of ablative sur- X. Outcomes of orbital floor reconstruction after
gery. Oral Surg Oral Med Oral Pathol Oral Radiol. extensive maxillectomy using the computer-assisted
1994;78:711. fabricated individual titanium mesh technique. J
102. Salinas TJ, Desa VP, Katsnelson A, Miloro Oral Maxillofac Surg. 2015;73:20653.e1.
M. Clinical evaluation of implants in radiated fibula
flaps. J Oral Maxillofac Surg. 2010;68:524.
Pathological Factors Affecting
Outcomes in Oral Cancer
6
Eric R. Carlson and J. Michael McCoy
It has been estimated that 300,000 new cases of with a synergistic association with alcohol con-
oral cancer are diagnosed internationally on an sumption and a clear dose–response relation-
annual basis [1]. These cancers represent the ship [2]. Human papillomavirus is noted in
sixth most common cancer and account for association with approximately 1–10% of cases
approximately 140,000 deaths each year [1]. of oral squamous cell carcinoma [3]. While
Squamous cell carcinoma accounts for 90% of impressive improvement has occurred in our
oral cancer cases. The major risk factors associ- comprehension of the molecular biology associ-
ated with squamous cell carcinoma of the oral ated with the development of oral cancer and the
cavity include smoked and smokeless tobacco staging of these malignancies, only modest
improvements in outcomes have been realized
E. R. Carlson (*) over the past 50 years. In fact, patients with sim-
Department of Oral and Maxillofacial Surgery, ilar stages of disease may demonstrate very dif-
University of Tennessee Graduate School of ferent clinical courses even when treated with
Medicine, Knoxville, TN, USA identical regimens [1]. As such, the search for
Oral/Head and Neck Oncologic Surgery Fellowship reliable and uniform prognostic indices is an
Program, University of Tennessee Cancer Institute, important venture to elucidate our true under-
Knoxville, TN, USA
standing of these complex malignancies and to
Oral and Maxillofacial Surgery Residency Program, guide treatment accordingly. To this end, the
University of Tennessee Graduate School of poor prognosis of oral cavity squamous cell car-
Medicine, Knoxville, TN, USA cinoma is seen related to high-grade cancers,
e-mail: ecarlson@mc.utmck.edu increased depth of invasion, perineural invasion,
J. M. McCoy noncohesive patterns of invasion, high-grade
Department of Oral and Maxillofacial Surgery, dysplasia at the surgical margins, positive mar-
University of Tennessee Graduate School of gins, and cervical lymph node metastases with
Medicine, Knoxville, TN, USA
or without extracapsular extension and soft tis-
Oral and Maxillofacial Surgery Residency Program, sue spread. It is the purpose of this chapter to
University of Tennessee Graduate School of discuss these and other unfavorable pathologic
Medicine, Knoxville, TN, USA features of oral squamous cell carcinoma while
Department of Pathology, University of Tennessee citing the international literature that represents
Graduate School of Medicine, Knoxville, TN, USA a great source of prognostic information. Where
Department of Radiology, University of Tennessee appropriate, adverse feature driven recommen-
Graduate School of Medicine, Knoxville, TN, USA dations for the administration of adjuvant ther-
e-mail: mmccoy@mc.utmck.edu apy will be discussed.
https://doi.org/10.1007/978-3-030-30094-4_6
66 E. R. Carlson and J. M. McCoy
6.1 merican Joint Committee

A Table 6.1 The eighth edition of the AJCC staging of oral
cancer
on Cancer (AJCC)/Union
Internationale Contre le TX Primary tumor cannot be assessed
Tis: Carcinoma in situ
Cancer (UICC) Tumor Node
T1: Tumor <2 cm; less than or equal to 5 mm depth of
Metastasis (TNM) Staging invasion (DOI)
System T2: Tumor less than or equal to 2 cm, DOI > 5 mm and
less than or equal to 10 mm, or tumor >2 cm but
The tumor node metastasis (TNM) staging for <4 cm, and less than or equal to 10 mm DOI
T3: Tumor >4 cm or any tumor >10 mm DOI
classification of human cancers was initially
T4: Moderately advanced or very advanced local
established and published by Pierre Denoix in disease
1952 [4], and the first edition of the Union T4a: M oderately advanced local disease: (lip)
Internationale Contre le Cancer (UICC) TNM tumor invades through cortical bone or
classification of malignant tumors was published involves the inferior alveolar nerve, floor of
mouth, or skin of face (i.e., chin or nose);
in 1953. The American Joint Committee on (oral cavity) tumor invades adjacent structures
Cancer (AJCC) was established in 1959 and pub- only (e.g., through cortical bone of the
lished its first cancer staging manual in 1977. The mandible or maxilla, or involves the maxillary
collaboration of the UICC and the AJCC resulted sinus or skin of the face); note that superficial
erosion of bone/tooth socket (alone) by a
in the creation of a universal system for the clas- gingival primary is not sufficient to classify a
sification of tumors of epithelial origin. The tumor as T4
eighth edition was published in 2016 for use T4b: Very advanced local disease; tumor invades
effective in 2017. Of note are the designations of masticator space, pterygoid plates, or skull
tumor depth of invasion (DOI) in the T classifica- base and/or encases the internal carotid artery
NX Regional lymph nodes cannot be assessed
tion and extranodal extension (ENE) in the N
N0: No regional lymph node metastasis
classification (Table 6.1). It has been recognized N1: Metastasis in a single ipsilateral lymph node, 3 cm
for decades that the prognosis of oral cancer or less in greatest dimension and ENE-negative
worsens when the tumor is thicker [5]. More N2: Metastasis in a single ipsilateral lymph node, 3 cm
recent data suggest that depth of invasion is a bet- or less in greatest dimension and ENE-positive; or
more than 3 cm but not more than 6 cm in greatest
ter prognostic index than tumor thickness and dimension and ENE-negative; or metastases in
cancers demonstrating higher grade and depths multiple ipsilateral lymph nodes, none more than
of invasion of 5 mm or greater should be consid- 6 cm in greatest dimension and ENE-negative; or
ered for adjuvant radiation therapy [6]. Depth of metastasis in bilateral or contralateral lymph
nodes, none more than 6 cm in greatest dimension,
invasion will adjust the T category, emphasizing ENE-negative
the distinction between superficial or exophytic N2a: Metastasis in a single ipsilateral or
tumors (Fig. 6.1) and those that are more invasive contralateral lymph node 3 cm or less in
or endophytic (Fig. 6.2). Staging no longer greatest dimension and ENE-positive; or
metastasis in a single ipsilateral lymph node
depends solely upon greatest surface dimension.
more than 3 cm but not more than 6 cm in
Since data reported from a large international greatest dimension and ENE-negative
collaborative study of oral cancer demonstrated a N2b: Metastasis in multiple ipsilateral lymph
significant difference in outcomes between T1 nodes, none more than 6 cm in greatest
tumors with more than 5 mm DOI and T2 through dimension and ENE-negative
N2c: Metastasis in bilateral or contralateral lymph
T4 tumors with greater than 10 mm DOI, the T
nodes, none more than 6 cm in greatest
category for OCC has been modified in the eighth dimension and ENE-negative
edition to improve hazard discrimination [7]. N3: Metastasis in a lymph node more than 6 cm in
Therein, for every 5-mm increase in DOI, both greatest dimension and ENE-negative; or
cT and pT categories will increase one level metastasis in a single ipsilateral lymph node more
than 3 cm in greatest dimension and ENE-positive;
according to the following: 5 mm, >5 mm but or metastasis in multiple ipsilateral, contralateral,
<10 mm, and >10 mm. Pathologically, DOI is or bilateral lymph nodes, with any ENE-positive
6 Pathological Factors Affecting Outcomes in Oral Cancer 67
Table 6.1 (continued) 6.2 Grade

N3a: Metastasis in a lymph node more than 6 cm
in greatest dimension and ENE-negative The grade of a human cancer has frequently been
N3b: Metastasis in a single ipsilateral node more
applied to a diagnosis to predict that cancer’s
than 3 cm in greatest dimension and
ENE-positive; or metastasis in multiple prognosis. Grading systems are thought to repre-
ipsilateral, contralateral, or bilateral lymph sent practical prognostic indices that merely
nodes, with any ENE-positive require tumor staining with hematoxylin and
eosin (H&E) and a seasoned pathologist.
Historical examples include the Nottingham his-
tologic grade for breast cancer [8] and the
Gleason grading scale for prostate cancer [9]. In
terms of oral/head and neck cancer, Broders eval-
uated 537 cases of squamous cell carcinoma of
the lip in 1920 and was the initial author to rec-
ommend a grading system that stratified patients
based on the degree of differentiation of neoplas-
tic cells and mitoses, with a special emphasis
placed on differentiation [10]. The grading sys-
tem was established with a score of 1–4. A grade
of 1 was created for those tumors that showed
three fourths of its structure to be differentiated
Fig. 6.1 An exophytic, stage II squamous cell carcinoma epithelium and one fourth undifferentiated
of the left tongue that exhibited a tumor thickness of
(Fig. 6.3). A grade of 2 was assigned to those
approximately 1 cm and a depth of invasion of 3 mm
tumors in which the differentiated and undiffer-
entiated epithelium were essentially equivalent
(Fig. 6.4). A grade of 3 was assigned to tumors in
which the undifferentiated epithelium formed
about three fourths and the differentiated tumor
one fourth (Fig. 6.5), and a grade 4 tumor was
one in which there was no tendency of the cells to
differentiate (Fig. 6.6). The number of mitotic
figures and cells with single large deeply staining
nucleoli played an integral but secondary role in
the grading of Broders’ lip cancers. In terms of
prognosis, 40 of 45 (88.88%) of patients with
grade 1 tumors were alive, 128 of 192 (66.66%)
patients with grade 2 tumors were alive, 16 of 65
Fig. 6.2 An endophytic, stage II squamous cell carci- (24.6%) of patients with grade 3 tumors were
noma of the left tongue that exhibited a tumor thickness of alive, and 0 patients with grade 4 tumors were
approximately 1 cm and a depth of invasion of 8 mm. alive during Broders’ period of study.
Compared to the tongue cancer illustrated in Fig. 6.1, this
endophytic cancer with a larger depth of invasion has a
The 2005 World Health Organization (WHO)
more ominous prognosis, even with an identical tumor classification [11] of oral squamous cell carci-
thickness noma places tumors into one of three categories,
and the 2017 classification is unchanged in this
measured from the level of the basement mem- regard [12]. Well-differentiated cancers resemble
brane of the closest adjacent normal mucosa. The normal squamous epithelium (Fig. 6.7).
depth of invasion is measured as a perpendicular Moderately differentiated cancers contain obvi-
distance of the deepest point of tumor invasion ous nuclear pleomorphic and mitotic activity,
from this line. including abnormal mitoses and less keratiniza-
Fig. 6.3 Broders grade

1 squamous cell
carcinoma. Near normal
maturation of the
epithelium is noted.
(Hematoxylin & eosin,
original magnification
×100)

2 squamous cell
carcinoma. Multiple
atypical cells exist, but
some similarity of
normal squamous
maturation is noted in
this specimen.
×200)
tion (Fig. 6.8). In poorly differentiated tumors, that a lack of correlation between these classifi-
immature cells are most common, and abundant cations and prognosis exists and limits their prog-
typical and atypical mitoses are present with nostic utility [13]. As such, in 1973, Jacobsson
minimal or no keratinization (Fig. 6.9). The et al. [14] and, in 1987, Anneroth et al. [15]
shortcoming of the Broders and WHO grading developed a multifactorial grading system that
systems is that oral squamous cell carcinoma is assesses the histologic grading of the cancer
most commonly a nonhomogenous tumor that based on three morphologic parameters of the
presents in multiple stages of differentiation such tumor cells including the degree of k eratinization,

3 squamous cell
carcinoma. Only a few
malignant cells (arrows)
resemble their squamous
origin. (Hematoxylin &
eosin, original
magnification ×100)

4 squamous cell
carcinoma. Tumor cells
no longer appear
squamous in origin.
×100)
nuclear polymorphism, and number of mitoses added. A grade is assigned as follows: grade I
per high-power field. In addition, the histologic (6–12), grade II (13–18), and grade III (19–24).
grading of malignancy of the tumor–host rela- In 1989 Bryne [13] introduced the concept that
tionship is assessed by three anatomic parameters the more invasive front of the tumor should be
including the pattern of invasion, the stage of examined since this area contains the part of the
invasion (depth), and lymphoplasmocytic infil- tumor that contains cells likely to determine the
tration (Table 6.2). Scores of 1–4 are assigned to clinical behavior of the malignancy. The histo-
each category, and the scores are subsequently logically invasive areas might be responsible for
Fig. 6.7 Well-
differentiated squamous
cell carcinoma. Note the
keratin-producing
squamous cells that
resemble normal
squamous cell growth.
×100)
Fig. 6.8 Moderately
cell carcinoma. Each
island of tumor still
produces keratin but
contains many more
atypical cells.
×100)
metastases, therefore being most prognostically more accurate prognostic information than the
significant, and therefore being of importance in corresponding incisional biopsy specimens. The
terms of specific therapy for the malignancy. A dilemma is obviously seen in that the nonrepre-
small biopsy to merely establish the diagnosis of sentative nature of incisional biopsies forms the
the tumor might not include the metastatic phe- basis for surgical treatment that may introduce
notype within the tumor such that the evaluation errors in surgical treatment under the circum-
of definitively resected malignancies provides stances. Bryne stated that this grading system
Fig. 6.9 Poorly
cell carcinoma. Each
malignant spindle cell
no longer has the
microscopic appearance
of its squamous origin.
×200)
Table 6.2 Points attributed to each category of the different parameters used in Anneroth’s and Bryne’s grading
systems
Points
Parameter 1 2 3 4
Degree of Highly keratinized Moderately Minimal keratinization No keratinization
keratinization (>50% of the cells) keratinized (5–20% of cells) (0–5% of cells)
(20–50% of cells)
Nuclear Little nuclear Moderately nuclear Abundant nuclear Extreme nuclear
pleomorphism pleomorphism pleomorphism pleomorphism (25–50% pleomorphism (0–25%
(>75% of mature (50–75% of mature of mature cells) of mature cells)
cells) cells)
Number of mitosis/ 0–1 2–3 4–5 >5
HPFa
Pattern of invasion Pushing, well- Infiltrating solid Small groups or cords Marked and wide
delineated cords of infiltrating cells spread cellular
infiltrating borders (n > 15) dissociation in small
groups (n < 15) and/or
in single cells
Stage of invasionb Carcinoma in situ Distinct invasion Invasion below lamina Extensive and deep
and/or but involving propria adjacent to invasion replacing most
questionable lamina propria only muscles, salivary gland of the stromal tissue
invasion tissues, and periosteum and infiltrating jaw
bone
Lymphoplasmocytic Marked Moderate Slight None
infiltration
HPF high-power field
Note: The score recorded for each morphologic feature was summed into a total malignancy score
a
Excluded from Bryne [16] system
b
Excluded from Bryne [13, 16] systems
was better than Broders’ system for prognosticat- nodal metastases, and recurrence. The 1992 histo-
ing oral squamous cell carcinoma, and that large logic grading system of Bryne et al. [16] was the
and representative incisional biopsies should be only system predictive of patient survival
procured to examine the underlying connective (p = 0.01). A statistically significantly greater per-
tissue as well as the tumor. In 1992, Bryne et al. centage of deceased patients were noted in Bryne
[16] determined that the accuracy of the grading [16] grade III cases (p < 0.05). Cox univariate sur-
system could be improved by eliminating the vival analyses demonstrated that the 1992 Bryne
mitotic count from grade designation while the grade III cases were significantly associated with
prognostic value remained highly significant. poor survival rates (p = 0.02). This grading sys-
This study retrospectively examined 61 cases of tem was subjected to a multivariate analysis,
floor of mouth squamous cell carcinoma that including age, gender, the cancer’s clinical stage,
were independently graded by two pathologists. and the type of treatment. The results indicated
The authors adopted their 1989 grading system that the 1992 Bryne grading system was an inde-
and its five morphological features with scoring pendent prognostic factor for squamous cell carci-
from 1 to 4 according to the definitions of noma of the oral cavity (p = 0.03) even after
Anneroth et al. [15]. These authors identified that considering these factors. The multivariate analy-
the cancers were often less differentiated in the sis indicated that patients classified as grade III
most invasive aspects of the tumor compared to demonstrated a sixfold higher likelihood of dying
their central parts. They confirmed their previous in the follow-up period compared to patients clas-
findings that invasive cell grading is highly prog- sified as grade I. No association was noted
nostic while the conventional Broders grading of between the grading systems of the World Health
the entire tumor is not prognostically significant. Organization [11], Anneroth et al. [15], and Bryne
The authors estimated that 15 percent of oral [13] and patient survival.
cancer biopsies cannot be assessed with invasive In 2015, Sawazaki-Calone et al. [17] evalu-
cell grading when clear invasion of tumor cells ated the prognostic significance of additional his-
into the connective tissue is absent. They there- topathological grading systems including the
fore recommended that larger incisional biopsies histological risk (HR) system of Brandwein-
be taken from the tumors, and that a biopsy mea- Gensler [18] that evaluated tumor specimens uti-
suring 15 × 5 × 5 mm would be sufficient for lizing three histopathological parameters
invasive cell grading. including the worst pattern of invasion (WPOI),
In 2017 Wagner et al. [1] performed a retro- lymphocytic host response (LHR), and perineu-
spective study of surgical specimens from 85 ral invasion (PNI), and Almangush et al. [19]
cases of primary oral squamous cell carcinoma who first proposed the BD model based on tumor
diagnosed between 1996 and 2010 at the budding (B) and depth of tumor invasion (D).
Pathology Laboratory of the Clinics Hospital of Sawazaki-Calone et al. evaluated 113 oral squa-
Porto Alegre, Brazil. Glass slides of the surgical mous cell carcinoma patients and assessed the
specimens stained with H&E were acquired for outcomes of these patients as a function of the
histologic grading by three expert pathologists grading systems of the World Health Organization
who were blinded to the clinicopathologic factors [11], Bryne [16], Brandwein-Gensler et al. [18],
and patient outcomes. Specifically, the cases were and Almangush et al. [19]. The 5-year disease-
graded by the criteria of the World Health specific survival and disease-free survival were
Organization [11], Anneroth et al. [15], and 46% and 62%, respectively. Disease-specific sur-
Bryne’s 1989 and 1992 classifications. The vival was directly influenced by T stage
authors identified no association between the four (p = 0.001), lymph node status (p = 0.001), WHO
grading systems and clinical features including grading system (p = 0.01), and BD model
alcohol consumption, the use of tobacco, the ana- (p = 0.009). The disease-free survival was only
tomic site of the cancer, the presence or absence correlated with the BD model (p = 0.005). The
of pain, TNM stage, the presence of absence of adjusted multivariate analysis based on Cox pro-
portion regression demonstrated that age and local invasion and can be noted without lym-
lymph node metastases were significantly corre- phatic or vascular invasion. In the case of a neu-
lated with disease-specific survival, whereas T rotropic squamous cell carcinoma of the oral
stage and BD model were the features signifi- cavity, perineural spread in a retrograde fashion
cantly associated with disease-free survival. toward the skull base or in an antegrade fashion
along smaller peripheral branches can occur.
Antegrade perineural spread is the more com-
6.3 Perineural Invasion monly observed pattern of spread in oral cancer
patients [21]. Perineural invasion can also be
Perineural invasion is a histopathologic feature classified as clinical or subclinical depending on
of some squamous cell carcinomas of the oral/ the presence of absence of pain, hypesthesia,
head and neck region associated with aggressive dysesthesia, or motor deficits. As many as 40%
tumor behavior, disease recurrence, and of patients with perineural invasion are without
increased morbidity and mortality. Squamous clinical symptoms as the onset of nerve dys-
cell carcinoma of the oral cavity is a well-recog- function may be delayed [21].
nized neurotropic malignancy with perineural While there is no internationally agreed upon
involvement by the tumor identified in 6–30% definition of perineural invasion, the criteria pro-
of cases [20]. Cruveilheir initially described posed by Liebig et al. [22] is most widely utilized
perineural involvement in the head and neck and referenced. According to these criteria, peri-
region in 1835, and surgeons, medical oncolo- neural invasion is diagnosed when tumor cells are
gists, and radiation oncologists continue to dis- noted in association with any of the three layers
cuss and debate the significance of this finding of the nerve sheath (endoneurium, epineurium,
in terms of adjuvant treatment of oral squamous perineurium), and tumor is identified in close
cell carcinoma [21]. Perineural invasion repre- proximity to the nerve, involving greater than
sents a distinct means of cancer cell dissemina- one-third of its circumference (Fig. 6.10).
tion in and along nerve bundles as noted by the Involvement less than one-third is considered
development of disease beyond the extent of abutment rather than invasion (Fig. 6.11).
Fig. 6.10 Perineural
invasion. Near total
encasement of the nerve
bundle (arrows) by
tumor is present.
×400)
Fig. 6.11 Nerve
abutment by tumor. Note
that less than one-third
of the nerve bundle
(arrow) is involved by
tumor. (Hematoxylin &
eosin, original
Tarsitano et al. [20] retrospectively studied failures for perineural invasion [1/51 (1.9%)] vs.
236 consecutive patients with oral squamous cell no perineural invasion [0/185 (0%)].
carcinoma and specifically investigated the Cracchiolo et al. [23] retrospectively reviewed
impact of perineural invasion as an independent 381 patients treated with primary surgery for
prognostic factor for local and regional failure. squamous cell carcinoma of the oral tongue over
Extirpative surgery was the primary method of a 13-year period of time. One hundred five
treatment offered to all patients with elective patients (28%) demonstrated perineural invasion
neck dissection (I–III) performed in 116 patients in their primary cancer specimens. There were
(49.1%), modified radical neck dissection per- 96 deaths with 55 deaths related to disease.
formed in 48 patients (20.4%), and no neck dis- Recurrent disease was noted in 97 patients
section in 72 patients (30.5%). One hundred including 58 local recurrences, 53 regional
fifty-six patients (66.1%) were treated with sur- recurrences, and 23 distant recurrences. Patients
gery alone while 80 patients (33.9%) received whose specimens identified perineural invasion
postoperative radiation therapy. Fifty-one of the were more likely to demonstrate a higher T clas-
236 patients’ tumors (21.6%) demonstrated peri- sification and lymph node metastasis. Fifty
neural invasion. Cancers of the tongue and floor patients with perineural invasion (13.1%) dem-
of mouth were the most commonly associated onstrated microscopically positive lymph nodes
with perineural invasion in 38% and 47% of compared to 35 patients (9.18%) without peri-
cases, respectively. Overall failure was noted in neural invasion demonstrating microscopically
71 of the 236 patients (30%) and a significant dif- positive lymph nodes. In this study, perineural
ference (p = 0.007) in local failure was seen in invasion was associated with a worse disease-
patients with perineural invasion [31/51 (60.7%)] specific survival (DSS) on univariate analysis.
compared to patients without perineural invasion On multivariate analysis, while adjusting for
[22/185 (11.9%)]. A statistically significant dif- tumor size, adjuvant therapy, and lymph node
ference (p = 0.041) was also noted in regional status, patients with perineural invasion demon-
failures with perineural invasion [10/51 (19.6%)] strated a decreased DSS. Although perineural
compared to those without perineural invasion invasion predicted local and regional recurrence
[7/185 (3.8%)] in this study. No statistically sig- on univariate analysis, it was not predictive on
nificant difference (p = 1.0) was noted in distant multivariate analysis.
Fig. 6.12 Treatment recommendations for T1–2, N0 Cancer Network, Inc. All rights reserved. The NCCN
squamous cell carcinoma of the buccal mucosa, floor of Guidelines® and illustrations herein may not be repro-
mouth, anterior tongue, alveolar ridge, retromolar tri- duced in any form for any purpose without the express
gone, and hard palate according to the National written permission of NCCN. To view the most recent
Comprehensive Cancer Network (NCCN). Adjuvant and complete version of the NCCN Guidelines, go online
treatment for the adverse features of extranodal exten- to NCCN.org. The NCCN Guidelines are a work in prog-
sion, positive surgical margins, perineural invasion, and ress that may be refined as often as new significant data
others are discussed. Reproduced with permission from becomes available. NCCN makes no warranties of any
the NCCN Clinical Practice Guidelines in Oncology kind whatsoever regarding their content, use, or applica-
(NCCN Guidelines®) for Head and Neck Cancer tion and disclaims any responsibility for their application
V.2.2018, page OR-2. © 2018 National Comprehensive or use in any way
The microscopic presence of perineural inva- chemoradiation therapy for patients with peri-
sion is well accepted as an adverse feature in neural invasion based on the results of the
oral squamous cell carcinoma and is associated European Organization for Research and
with increased recurrence and a decrease in sur- Treatment of Cancer 22,931, a prospective ran-
vival. The 2018 National Comprehensive Cancer domized study comparing postoperative radia-
Network (NCCN) guidelines (Figs. 6.12 and tion therapy alone vs. chemoradiation therapy in
6.13) recommend adjuvant radiation therapy high-risk squamous cell carcinomas of the head
when perineural invasion is identified in the and neck that showed a survival advantage for
cancer specimen [24]. In addition the NCCN patients receiving adjuvant cisplatin concur-
guidelines recommend considering adjuvant rently with radiation therapy [25].
Fig. 6.13 Treatment recommendations for T3, N0; T1–3, and illustrations herein may not be reproduced in any
N1–3; T4a, any N squamous cell carcinoma of the buccal form for any purpose without the express written permis-
mucosa, floor of mouth, anterior tongue, alveolar ridge, sion of NCCN. To view the most recent and complete ver-
retromolar trigone, and hard palate. The same adverse fea- sion of the NCCN Guidelines, go online to NCCN.org.
tures are considered for adjuvant therapy as the smaller, The NCCN Guidelines are a work in progress that may be
less advanced cancers. Reproduced with permission from refined as often as new significant data become available.
the NCCN Clinical Practice Guidelines in Oncology NCCN makes no warranties of any kind whatsoever
(NCCN Guidelines®) for Head and Neck Cancer V.2.2018, regarding their content, use, or application and disclaims
page OR-3. © 2018 National Comprehensive Cancer any responsibility for their application or use in any way
Network, Inc. All rights reserved. The NCCN Guidelines®
6.4 Status of the Surgical removing oral squamous cell carcinoma. To this

Margins end, therefore, the specimen handoff from sur-
geon to pathologist is paramount to the identifica-
The surgical margin is a clinical and quantified tion of true surgical margins, clinically concerning
linear margin that separates the advancing tumor margins, and those margins that ought to be evalu-
front from the inked margin on the specimen. The ated by frozen sections. The handoff procedure
inclusion of this linear margin at the periphery of represents a specimen orientation that occurs in a
the specimen is with intentionality and recognizes face-to-face fashion. While this procedure has not
the ability of neoplasms to spread beyond their been studied scientifically, experience indicates
clinical demarcations. The surgical margin there- that the handoff process minimizes ambiguity in
fore represents the only prognostic factor over margin determination [26]. The discussion of ana-
which the ablative surgeon has influence when tomic landmarks in the specimen, oncologic lev-
els of lymph nodes in a composite resection withstanding, local tumor recurrences are noted
specimen, measurement of tumor in the unfixed to occur even when pathologists declare that all
state, gross assessment of margins, directionality surgical margins are negative for cancer. Byers
of the surgical specimen, and the identification of et al. [30] identified a 12% incidence of local
areas of concern to the surgeon represent topics of recurrence of oral squamous cell carcinoma when
discussion during the handoff orientation. surgical margins were noted to be negative and an
The traditionally accepted definition of an 80% incidence of recurrence when margins were
oncologic surgical margin is an anatomic clear- positive. Dillon et al. [29] performed a retrospec-
ance of all malignant cells in a three-dimensional tive cohort study that examined 174 patients with
orientation [27]. There is agreement by the oral squamous cell carcinoma of whom 54 met
National Comprehensive Cancer Network the study inclusion criteria. Of these 54 patients,
(NCCN), the American College of Pathologists 9 patients (17%) demonstrated greater than or
(ACP), and the Royal College of Pathologists equal to 5 mm (negative) surgical margins, 21
(RCP) that a negative margin is defined by the (39%) demonstrated 1–5 mm (close) surgical
presence of at least 5 mm of normal tissue in the margins, and 24 (44%) demonstrated less than
margin [28, 29] (Fig. 6.14). A close margin is 1 mm (positive) surgical margins. In all, 83% of
defined as negative but one where the distance patients had close or positive margins that speak
from the invasive tumor to the specimen’s margin to the difficulty of obtaining negative margins in
is less than 5 mm (Fig. 6.15), and a positive mar- this patient population, despite good intentions.
gin is variably defined. The ACP defines a posi- The 2-year survival rates were 78%, 62%, and
tive margin as invasive cancer less than 1 mm 50%, respectively, for the negative, close, and
from the surgical margin (Fig. 6.16) while both positive margins. Seventy-nine percent of patients
the RCP and the NCCN define a positive margin with positive margins received adjuvant chemo-
as invasive cancer, carcinoma-in-situ, or high- therapy, radiation therapy, or combined chemora-
grade dysplasia present at the microscopic mar- diation therapy, and 42% of patients developed
gins [28] (Fig. 6.17). Therein, a negative margin locoregional recurrences at 2 years. Sixty-two
implies that the entirety of the malignancy is percent of patients with close margins received
encased within the resection specimen. This not- adjuvant therapy, with 31% of patients develop-
Fig. 6.14 Surgical
margin greater than
5 mm. Note the
discontinuance of tumor
growth (arrow) far from
the surgical margin that
is unequivocally
considered a negative
margin. (Hematoxylin &
eosin, original
magnification ×40)
Fig. 6.15 Surgical
margin less than 5 mm.
The tumor (arrow) is
within 3–4 mm of the
surgical margin. This
distance connotes a
close margin.
×100)
Fig. 6.16 Surgical
margin 1 mm or less.
Note the proximity of
the tumor (arrow) to the
surgical margin.
×100)
ing locoregional recurrences at 2 years. Seventy- Tissue shrinkage associated with formalin
eight percent of patients with negative margins fixation may falsely proclaim truly negative mar-
received adjuvant therapy with 29% of patients gins as close and close surgical margins as posi-
developing locoregional recurrences at 2 years. tive. In 1997, Johnson et al. [31] reviewed the
The authors concluded their study by indicating shrinkage associated with margins in the oral
that the presence of close surgical margins mucosa in dogs. They indicated that a 30–50%
(1–5 mm) is an adverse feature that is similar to a discrepancy exists in margins measured in situ
positive margin. and following processing. They noted that most
Fig. 6.17 High-grade
dysplasia at the surgical
margin. Note the
atypical maturation of
the epithelium at the
surgical margin.
×100)
of the shrinkage occurred immediately after Intraoperative frozen sections represent one
excision. The labiobuccal margin demonstrated additional method to avoid the dilemma of close
38.3% loss after excision and an additional or positive margins, a technique employed by
10.5% loss following fixation. Tongue margins greater than 97% of ablative surgeons managing
showed 24.8% loss after excision and an addi- cancer of the oral cavity [28]. Frozen sections
tional 7.6% loss after fixation. The authors involve procuring tissue from the periphery of the
attributed the shrinkage to unopposed contractil- excised specimen or the remaining tissue bed,
ity of the underlying muscles in the surgical embedding these soft tissue specimens in optimal
specimen and the release from surrounding tis- cutting temperature compound, and freezing
sues. In 2005, Mistry et al. [32] determined that using a cryostat machine. Specimens are then
the mean shrinkage of tumor margins for buccal thinly sectioned to an average thickness of
mucosal and tongue specimens were 21.1% and approximately 7 μm, affixed to glass slides,
23.5%, respectively. They also determined that stained with H&E, and microscopically evalu-
tumor shrinkage was less for T3 and T4 tumors ated by a pathologist. Ord et al. [34] retrospec-
(9.2%) than T1 and T2 tumors (25.6%). The tively evaluated 49 consecutive patients with oral
authors suggested that increased tumor burdens cancer. Arbitrary areas of anterior, posterior,
translate to less contractility of the margins. medial, and lateral margins were subjected to fro-
Cheng et al. [33] evaluated oral mucosal shrink- zen section analysis that yielded a total of 307
age as a function of anatomic site and noted that frozen sections, and an average of 6.2 frozen sec-
retromolar trigone, mandibular alveolar ridge, tions per patient. When compared to their perma-
and buccal mucosa exhibited the greatest tissue nently stained counterparts, 304 of the 307
shrinkage (71%) compared to hard palate and sections showed a concordant accuracy of 99%.
maxillary alveolar ridge (53%) and oral tongue Two false-negative frozen sections were noted
mucosa (42%). All of these studies indicate the and one false-positive frozen section existed that
need for ablative surgeons to increase the width resulted in a sensitivity of 86.6% and a specificity
of the soft tissue linear margin on the specimen of 99.6%. The authors regarded the presence of
by 25–50% in order to avoid the dilemma of dysplasia, carcinoma in situ, or invasive cancer
close or positive margins [27]. within 5 mm of the margin as a positive margin.
Thirty-nine of the 49 patients (79.6%) showed geon not being reticent to include a larger linear
clear surgical margins in the final analysis of margin at the periphery of the specimen [36].
their specimens. Thirty-eight of these patients Ellis et al. [35] found no statistical difference in
had specimens with negative margins on initial the surgical margin when a free-flap procedure
frozen sections, whereas one patient had a speci- was performed. Further, no statistically meaning-
men with an initial positive margin on frozen sec- ful improvements in the surgical margins were
tion that required additional excision that was identified using frozen sections, surgical access
negative for cancer. Four patients demonstrated procedures, or tumor size. The authors concluded
cancer within 5 mm of the final surgical margin their study by emphasizing that improved patient
and one patient (25%) recurred. Four patients outcomes are realized if ablative oral cancer sur-
demonstrated dysplasia at the surgical margins gery is performed at high-volume centers by
and one patient (25%) recurred. Two patients high-volume surgeons.
demonstrated cancer at the surgical margins and Genetic analysis of surgical margins in head
both (100%) recurred. In all, seven patients and neck cancer was introduced to the interna-
(14.5%) did not benefit by the implementation of tional literature by Brennan [37] in 1995. The
frozen sections. In addition, only one patient was presence of the p53 gene mutation in a histologi-
benefitted by the use of frozen sections to clear a cally negative margin was associated with local
positive margin. recurrence in patients who had undergone surgi-
Ellis et al. [35] retrospectively evaluated a cal resection. Specifically, in their study, Brennan
cohort of 250 patients with an analysis of five et al. [37] identified 30 of 69 patients with muta-
surgical and seven histologic variables to deter- tions of the p53 gene in their head and neck can-
mine their effects on the surgical margins in oral cers. Seventy-eight surgical margins and 33
cancer resections. The authors concluded six sta- cervical lymph nodes were obtained from these
tistically meaningful study variables in terms of 30 patients. Five patients had positive surgical
surgical margins. The results of the study indi- margins on final histopathological analysis of
cated that surgeons who had resected higher vol- their specimens and were eliminated from further
umes of oral cavity cancer had larger surgical analysis. Seventy-two margins containing no
margins than those with lower surgical volumes. microscopic evidence of cancer in 25 patients
High-volume surgeons were those who treated therefore comprised the study group. These 72
more than 40 patients, although the time period negative margins were probed with the p53
for this volume of work was not specified in the mutant oligonucleotide derived from the primary
study. Low volume surgeons were defined as tumors. In 13 of the 25 patients (52%), the ampli-
those who performed fewer than 40 cases, fied p53 region from at least one surgical margin
although this volume was similarly not catego- hybridized to the tumor-specific probe, exhibit-
rized as to its time period. The variable of suring the presence of mutated neoplastic cells
geon volume was statistically the most important within these negative margins. The estimated
variable of the study. The second variable was the percentage of cells with mutations in the surgical
cancer’s location that demonstrated an associa- margins was from 0.05% to 28%. The PCR prod-
tion between shorter surgical margins and tumors ucts from the margins of the remaining 12
located in the retromolar trigone. Thirdly, the patients did not hybridize to the mutant-specific
study showed an association between smaller probes, indicating that those margins did not har-
surgical margins and perineural invasion. Among bor neoplastic cells and were truly negative mar-
the nonstatistically meaningful variables included gins. Sections from 33 cervical lymph nodes in
free-flap reconstructive surgery of the ablative six patients identified metastatic squamous cell
defect. The commitment to free-flap surgical carcinoma in five lymph nodes (15%). However,
reconstruction of ablative defects related to oral molecular analysis identified mutant p53 genes in
cancer at least theoretically reduces the likeli- the PCR products of 11 nodes (33%). Of the 28
hood of positive margins due to the ablative sur- negative lymph nodes by light microscopy, 6
(21%) were identified to contain neoplastic cells. those without; 13 of 98 (13%), vs. 35 of 553
On follow-up, 5 of the 13 patients (38%) with (6.3%), respectively.
positive margins by molecular analysis, yet nega- de Carvalho et al. [39] retrospectively studied
tive surgical margins by light microscopy, devel- the specimens of 55 patients who underwent
oped biopsy-proven recurrences of carcinoma. tumor ablation for head and neck squamous cell
None of the 12 patients whose surgical margins carcinoma with curative intent. Surgical margins
were negative for p53 mutations developed recur- that were histologically negative were evaluated
rent disease. The results of the lymph node exam- from these patients and primary SCCA samples
ination by molecular analysis were felt to be were obtained from a subset of 23 patients. The
noteworthy by the authors. Based on the d iscovery resected specimens demonstrated no invasive
of p53 mutations in seemingly benign lymph cancer or dysplasia at the surgical margins.
nodes, four of these six patients would have been Twenty-five oral mucosal tissue samples from
upstaged in terms of their N designation. This healthy donors were included in the study. The
upstaging would have been significant in terms of conclusion of the study was that the identification
the delivery of adjuvant therapy and would have of molecular factors in oral cancer specimens
been negatively impacting from a prognostic may provide useful prognostic information and
perspective. influence the management of patients. Ferris
Liu et al. [38] prospectively examined 168 et al. [40] demonstrated that PTHLH (parathy-
patients with primary oral cavity cancer who roid hormone-like hormone, also known as
underwent surgical ablation of their cancers. One PTHrP) and EPCAM (epithelial cell adhesion
hundred forty-five patients satisfied the author’s molecule) expression could distinguish positive
study protocol including histologically negative and negative lymph nodes with high accuracy. In
margins of at least 5 mm. Six hundred fifty-one addition, the immunohistochemical identification
surgical margins were analyzed in the 145 study of MMP9 (matrix metalloproteinase-9) in surgi-
patients. Forty-two patients (29%) developed cal margins demonstrates a positive association
local recurrence, and seven patients (48%) devel- with the risk of recurrence in head and neck squa-
oped distant metastatic disease. Six markers were mous cell carcinoma patients. The study’s goal
utilized for microsatellite alteration analysis was to identify the expression of PTHLH,
including D9S1748, THRB, D3S1300, IFNA. EPCAM, MMP9, LGALS1 (lectin, galactoside-
PCR2, D2S206, and D21S236. Microsatellite binding, soluble, 1), and MET (MET proto-
alteration was identified in 100 specimens from oncogene, receptor tyrosine kinase) expression in
145 patients. Fifty-five patients had microsatel- histologically negative margins as a useful
lite instability for one or more markers in the marker for the detection of molecular alterations
tumor specimen, and 85 patients had loss of het- associated with local disease control in these
erozygosity for one or more markers in malignant patients. Gene expression was assessed with
tissues. Of the 55 patients with microsatellite quantitative reverse transcription polymerase
instability in the specimen, 41 demonstrated mic- chain reaction (qRT-PCR). MMP9 (91%),
rosatellite instability at the surgical margins and LGALS1 (83%), PTHLH (74%), MET (48%),
14 showed no microsatellite instability at the and EPCAM (30%) were commonly overex-
margins. Recurrence was noted in 22 of the 41 pressed in the 23 SCCA samples (high sensitiv-
(54%) patients in the former group and 4 of the ity). MMP9, PTHLH, and EPCAM were rarely
14 patients (29%) in the latter group. Those with overexpressed in the 15 healthy oral mucosal
microsatellite instability in the surgical margins samples (0%, 7%, 0%, respectively), confirming
demonstrated a higher rate of local recurrence their overexpression as highly specific. MET and
than those without; 18 of 55 patients (33%) vs. 30 LGALS1 were overexpressed in 14% and 27%,
of 596 (5%), respectively. Patients with loss of respectively, of the normal controls, indicative of
heterozygosity in the surgical margin also dem- their lack of specificity. Based on the high speci-
onstrated a higher rate of local recurrence than ficity and sensitivity of MMP9, EPCAM, and
PTHLH, the expression of these genes was pression of PTHLH and MMP9 in negative
assessed in the 55 negative surgical margins. SCCA margins is directly correlated with a high
MMP9 was overexpressed in 23.6% (13/55) of risk of local recurrence and the development of
the surgical margins evaluated, EPCAM in 10.9% secondary primary tumors.
(6/55), and PTHLH in 9.1% (5/55). Thirty-six
percent of the negative margins demonstrated
overexpression of at least one of the 3 selected 6.5 Depth of Invasion
genes, MMP9, EPAM, and PTHLH. Despite neg-
ative surgical margins, 11/55 (20%) patients in Surgical decision-making regarding the clinically
the study presented with local recurrences, negative neck in patients with oral squamous cell
suggesting that the molecular changes present in carcinoma represents a formidable discipline,
the margins, undetected by microscopic analysis, and assessment of tumor depth of invasion repre-
could directly be responsible for malignant trans- sents the greatest histologic predictor of occult
formation of this normal tissue and the poor out- cervical metastases [41]. To that end, tumor depth
comes realized by the patients in this study. In of invasion (Fig. 6.18) is distinguished from
particular, the study indicated that the overex- tumor thickness (Fig. 6.19), with depth of inva-
Fig. 6.18 Measuring
depth of invasion (DOI).
The horizontal line
connects the basement
membrane of the closest
intact squamous mucosa
on each side of the
cancer. A “plumb” line
is then dropped from the
horizontal line to the
deepest tumor cells. The
measurement of this line
represents the DOI.
×40)
Fig. 6.19 The tumor

thickness is measured
from the most superior
portion of the tumor to
the deepest portion. This
is often quite different
than the DOI.
×40)
sion being prognostically more reliable than surgical therapy of the primary cancer since biop-
tumor thickness. That said, some authors incor- sies are typically not representative of the entire
rectly equate or interchangeably utilize the terms thickness of an oral cancer. Nonetheless, palpa-
depth of invasion and tumor thickness [5, 42]. tion of the primary tumor may result in approxi-
Determining the exact metric for the depth of mation of the thickness to determine the utility of
invasion that would dictate an elective neck dis- elective neck dissection. In the Spiro et al. [5]
section due to a threshold level of occult neck study, treatment failure occurred in 32 determi-
disease continues to represent a controversial nate patients (35%). The primary site was
issue in oral/head and neck oncologic surgery. An involved in eight patients, the neck in 18 patients,
elective neck dissection is commonly performed and both sites in four patients. Neck recurrence
in patients who demonstrate at least a 20% risk was noted in 2 of the 29 patients who underwent
for occult cervical lymph node metastases [43]. elective neck dissection. Of the 63 patients whose
Thin and superficially invasive squamous cell necks were observed, metastases subsequently
carcinomas of the oral cavity have a lower risk of developed in 17 patients (27%) and 8 patients
regional lymph node metastases compared to (47%) died of poorly controlled disease in the
thicker cancers that are deeply invasive of the neck despite 15 patients undergoing radical neck
underlying soft tissues. Although equating tumor dissection for salvage.
thickness and depth of invasion, Spiro et al. [5] Brockhoff et al. [41] performed a retrospec-
quantified the risk of occult neck disease for 105 tive review of their database at the University of
T1 and T2 primary squamous cell carcinomas of Michigan and identified 286 patients who had
the tongue and floor of mouth with no evidence undergone excisions of primary cancers and elec-
of cervical metastases at the time of their primary tive neck dissections. There were 105 cancers
surgical treatment. There were 43 patients in the located in the oral tongue, 91 cancers of the alve-
group of patients with tumor thicknesses of 2 mm olus/hard palate, 39 cancers of the floor of mouth,
or less, 43 patients had a tumor thickness between 25 cancers of the retromolar trigone, and 24 can-
3 and 8 mm, and 19 patients had tumor thick- cers located at other sites. Sixty-six patients had
nesses of 9 mm or greater. Univariate survival stage I disease, 54 patients had stage II disease,
analysis indicated that tumor thickness and stage 39 patients had stage III disease, and 127 patients
were the most important prognostic indices. had stage IV disease. The shallowest depth of
Multivariate analysis confirmed that tumor thick- invasion where at least 20% of the neck dissec-
ness had the greatest impact on survival. Tumors tions had histologically positive lymph nodes
2 mm or less in thickness predicted a subthresh- was 2 mm for tongue, 3 mm for floor of mouth,
old risk of occult nodal metastases of 7.5%, over- 3 mm for retromolar trigone, and 4 mm for alveo-
all incidence of nodal metastases of 13%, and 3% lus/hard palate. This study therefore answered
of patients died from their disease. Tumor thick- the question of threshold depth of invasion and its
nesses between 3 and 8 mm demonstrated a correlation with occult nodal metastases as a
threshold risk of occult nodal metastases of function of anatomic site of the primary cancer of
25.7%, 46% overall incidence of nodal metasta- the oral cavity.
ses, and 17% of patients died from their disease. Masood et al. [44] retrospectively analyzed 67
Tumors thicker than 9 mm show a threshold patients with T1 N0 (n = 30) and T2 N0 (n = 37)
41.2% risk of occult nodal metastases, 65% over- HPV-negative squamous cell carcinoma of the
all incidence of nodal metastases, and 35% of tongue. All patients underwent elective neck dis-
patients died from their disease. When treatment sections. Thirty-five, twenty, and twelve patients
planning patients with oral squamous cell carci- demonstrated tumor thickness of less than or
noma, it would be ideal to know the approximate equal to 5 mm, between 5 and 10 mm, and greater
tumor thickness within these three categories. than 10 mm, respectively. Thirty-seven, sixteen,
That said, it is not clinically or practically possi- and fourteen patients demonstrated primary
ble to possess that information prior to definitive tumor depth of invasion of less than or equal to
5 mm, between 5 and 10 mm, and greater than Histologic evaluation of the radical neck dissec-
10 mm, respectively. Five recurrences were noted tion specimens permitted the authors to catego-
among the 67 patients (7.46%). Two of the recur- rize patients into four groups: group 1 consisted
rences were local, two were regional, and one of patients with histologically negative neck dis-
was distant. In the group with depth of invasion section specimens; group 2 consisted of patients
less than or equal to 5 mm (n = 37), there were with one positive lymph node in the neck dissec-
two recurrences with one being local and one tion specimen; group 3 consisted of patients with
being regional. In the group with depth of inva- two positive lymph nodes; and group 4 consisted
sion greater than 10 mm (n = 14), there were of patients with three or more positive lymph
three recurrences, with one being local, one nodes. Of the 340 patients, 213 had histologically
regional, and one distant. No recurrences were positive lymph nodes. Anatomic location of posi-
noted in the group with depth of invasion of tive lymph nodes was determined for 108 patients
between 5 and 10 mm. Depth of invasion was demonstrating positive lymph nodes confined to
associated with occult nodal metastases and lym- the superior or suprahyoid region of the neck; 84
phovascular invasion. Tumor thickness was also a patients with positive lymph nodes in the middle
significant predictor of lymphovascular invasion, third of the neck between the hyoid bone and the
but not occult cervical metastases. omohyoid muscle with or without involvement of
the upper third of the neck; and 21 patients had
positive lymph nodes in the lower third of the
6.6 tatus of the Cervical Lymph
S neck defined as inferior to the omohyoid muscle,
Nodes with or without involvement of the middle or
superior third of the neck. The 213 patients with
In 1977, Kalnins et al. [45] retrospectively histologically positive lymph nodes were again
reported on the outcomes of 416 patients who divided into three groups based on whether the
had undergone radical neck dissection for squa- lymph node capsule (Fig. 6.20) was intact,
mous cell carcinoma of the oral cavity and tonsil- whether it was microscopically penetrated only
lar region. Three hundred forty determinate (Fig. 6.21), and whether soft tissue spread had
patients were followed for a minimum of 5 years. occurred (Fig. 6.22). Of the 160 patients with his-
Fig. 6.20 Metastatic
squamous cell
carcinoma in a cervical
lymph node. Well-
defined metastatic
squamous cell
carcinoma (arrow) is
noted within the
confines of the lymph
node and without
involvement of the
capsule. (Hematoxylin
& eosin, original
magnification ×40)
squamous cell
carcinoma within a
cervical lymph node
with a focus of
metastatic squamous cell
carcinoma
demonstrating minimal
erosion of the lymph
node capsule (arrow).
×40)
squamous cell
carcinoma spreading
from the lymph node
into the surrounding soft
tissues. Note the
disrupted node (long
arrow) and the spreading
tumor cells (short
arrow). (Hematoxylin &
eosin, original
tologically negative lymph nodes, disease- capsule was intact, the disease-specific survival
specific survival was 75% at 5 years. The for all patients was 33%. The disease-specific
disease-specific 5-year survival of patients with survival for microscopic penetration of the lymph
one positive lymph node, two positive lymph node capsule was 28%, and soft tissue spread led
nodes, and three or more lymph nodes was 49%, to a 5-year disease-specific survival rate of 11%.
30%, 13%, respectively. The overall 5-year Ghadjar et al. [46] retrospectively analyzed
disease-specific survival was 29% for all patients the neck dissection specimens of 133 patients
with positive lymph nodes. When the lymph node with squamous cell carcinoma of the head and
neck and preferentially studied those patients pleted follow-up for at least 9 months and com-
whose lymph nodes demonstrated extracapsular prised the analysis. There were 81 recurrences
extension (n = 98). Four of the 98 patients under- (25 nodal, 23 local, 3 distant, 4 nodal and local,
went selective neck dissection, 37 underwent 16 second primary, 10 unknown) and 50 deaths in
modified radical neck dissection, and 57 under- the elective neck dissection group and 146 recur-
went radical neck dissection. A total of 431 rences (108 nodal, 7 local, 3 distant, 8 nodal and
lymph nodes were examined including 231 local, 11 second primary, and 9 not known) and
lymph nodes with extracapsular extension and 79 deaths in the therapeutic neck dissection
200 lymph nodes without extracapsular exten- group. The difference between nodal recurrences
sion. A significant association between lymph in the two groups is noteworthy; 114 of 253
node size and the incidence of extracapsular patients (45.1%) in the therapeutic neck dissec-
extension was noted in the study. The mean diam- tion group, and 72 of 243 patients (29.6%) in the
eters for lymph nodes with and without extracap- elective neck dissection group. At 3 years postop-
sular extension were 11 and 9 mm, respectively eratively, elective neck dissection resulted in an
(p = 0.0004). Sixty-one patients demonstrated improved rate of overall survival (80%) com-
extracapsular extension in their positive lymph pared to the therapeutic neck dissection group
nodes that measured 10 mm or smaller while 22 (67.5%). At that time, patients in the elective
patients demonstrated extracapsular extension in neck dissection group had a higher rate of
their positive lymph nodes that measured 5 mm disease-free survival (69.5%) than the therapeu-
or smaller. Overall, 48% of positive lymph nodes tic neck dissection group (45.9%). The results of
smaller than 10 mm exhibited extracapsular this study reveal the survival benefits of elective
extension while 60% of positive lymph nodes neck dissection compared to watchful waiting
between 10 and 30 mm exhibited extracapsular followed by therapeutic neck dissection for
extension. lymph node recurrence in patients with early
The utility of performing an elective neck dis- stage, clinically node-negative oral squamous
section for clinically node-negative patients with cell carcinoma. The results show an absolute
oral squamous cell carcinoma is one of the most overall survival benefit of elective neck dissec-
contentious issues in oral/head and neck onco- tion of 12.5% points and a disease-free survival
logic surgery. Such patients may be thought to be benefit of 23.6% points. This fact translates to
acceptable candidates for exclusive surgical man- eight patients having to be treated with elective
agement of the primary cancer and watchful neck dissection to prevent one death, and four
waiting of the neck since approximately 70% of patients would need to be treated to prevent one
these patients will demonstrate a histologically recurrence.
negative neck if the neck is dissected electively Kuo et al. [48] addressed the issue of lymph
[47]. D’Cruz et al. [47] performed a prospective, node yield in oral cancer in recognition of prog-
randomized, controlled trial of 596 patients with nostic lymph node yield thresholds that have
T1 or T2 squamous cell carcinoma of the tongue, been identified and incorporated into treatment
floor of mouth, or buccal mucosa. Patients were guidelines for multiple human cancer sites
randomized to undergo either elective neck dis- including bladder, colorectal, and esophageal,
section (selective neck dissection I–III) or surgi- but not for oral cancer. There are no guidelines,
cal ablation of the primary cancer with monitoring and there is no consensus regarding the number
of the neck and therapeutic neck dissection only of lymph nodes removed, or lymph node yield,
when cervical metastases became apparent. The that indicates an acceptable neck dissection in
authors designed the study in part to determine if patients with cancer of the oral cavity. In order to
a survival difference exists between the elective address this issue, patients with oral cancer in the
neck dissection and therapeutic neck dissection National Cancer Database were accessed that
groups. Five hundred patients, 245 patients in the included 13,143 cases. Higher lymph node yields
elective neck dissection group, and 255 patients in neck dissection specimens were noted in
in the therapeutic neck dissection group com- males, young patients, and those procedures per-
formed in academic medical centers, in centers than 26 lymph nodes in the cN+ patients resulted
with higher case volumes, and certain geographic in significantly decreased survival rates. The
areas in the United States. Of the 6147 patients authors offered explanations for the observed
with known clinical lymph node status, 71.1% survival benefit associated with more extensive
underwent neck dissection and 79% had clini- lymph node dissections. In patients with cN0 dis-
cally negative necks. The rate of neck dissection ease, higher lymph node yields were associated
was 63.9% in the cN0 group and 98.3% in the with a greater likelihood of removing and identi-
cN+ group. Of the patients who underwent neck fying one positive lymph node. Removing more
dissection, the median lymph node yield was 21 lymph nodes increases the likelihood, therefore,
lymph nodes overall, 20 lymph nodes in the cN0 of a complete cancer surgery while also serving a
patients, and 25 lymph nodes in the cN+ patients. role regarding the possibility of adjuvant therapy
Multivariate analysis was performed on 3097 delivery, both advantages resulting in more favor-
cN0 patients who underwent neck dissection in able outcomes.
the cohort when controlling for patient age, sex,
insurance status, year of diagnosis, pT and pN
classification, tumor grade, surgical margin sta- 6.7 Human Papillomavirus
tus, radiation status, and chemotherapy status. A
threshold yield of 16 lymph nodes was validated Approximately 25% of human cancer is caused
in 2175 patients with N0 disease. For cN0 by infection, and until recently, the major interest
patients with fewer than 16 lymph nodes in head and neck squamous cell carcinoma has
removed, the frequency of identifying at least one been the involvement of Epstein–Barr virus in
positive lymph node was 16.3% whereas the like- nasopharyngeal cancer [49]. Recent attention has
lihood of identifying at least one positive lymph been drawn to the involvement of human papillo-
node in patients who had 16 or more lymph nodes mavirus (HPV) in head and neck squamous cell
removed was 27.2%. The identification of fewer carcinoma, particularly oropharyngeal cancer.
than 16 lymph nodes in the cN0 patients resulted Features distinguishing HPV-positive from HPV-
in significantly decreased survival rates. In cN+ negative oropharyngeal squamous cell carcinoma
patients, a threshold of 26 lymph nodes was vali- is the site predilection of tongue base and tonsil,
dated in 1903 patients. Identification of fewer basaloid histology (Fig. 6.23), younger age, high
Fig. 6.23
Histopathology of
HPV + base of tongue
squamous cell
carcinoma. Note the
basal cell-like (basaloid)
appearance of the tumor
cells. (Hematoxylin &
eosin, original
socioeconomic status, sexual behavior risk fac- Extrapolating this data to the United States popu-
tors, increasing incidence, and the distinct sur- lation suggests that in 2009–2010, there were
vival advantage of the HPV-positive cancers. approximately 15 million adult Americans with
Evidence also exists supporting the carcinogenic an oral HPV infection. Oral HPV infection seems
role of HPV in oral cavity sites, and sophisticated to be affected by numerous behavioral and social
assays indicate that approximately 3–10% of oral factors including tobacco use, oral and conven-
squamous cell carcinoma is HPV-related. The tional sexual practices, and immunosuppression,
number of cases is generally too small to establish but not alcohol consumption [3]. In fact, the odds
meaningful and durable conclusions regarding the of HPV infection increase significantly in a dose-
prognostic significance of HPV in oral squamous dependent fashion with increasing measure of
cell carcinoma. The detection of HPV in an oral current tobacco product use [3, 50] and are higher
cancer specimen is controversial, but most centers for women than men. Ever having performed oral
provide an initial screen with p16 immunohisto- sex, early oral and vaginal sexual debut, deep
chemistry (Fig. 6.24) that is relatively inexpensive tongue kissing, increasing lifetime oral and vagi-
to perform and is available in most pathology nal sex partners, and ever having performed vagi-
laboratories. The p16 immunohistochemistry nal and anal sex are all significantly associated
tests for high-risk HPV, including HPV 16 and 18 with oral HPV infection [3, 50, 51]. An observa-
that seem to be most often associated with HPV- tional study of oral HPV infection suggested that
related oral and oropharyngeal cancer. While the most oral infections with high-risk types are
presence of HPV in the nucleus of the cancer cell cleared within 1 year, with a median duration of
is demonstrated by p16 immunohistochemistry, infection of 6.9 months for any HPV, 6.3 months
the genetic expression of the oncoproteins E6 and for oncogenic HPV, and 7.3 months for HPV 16,
E7 must be demonstrated to establish a cause and in particular [52]. Factors typically associated
effect relationship between HPV and oral squa- with persistence of oral HPV infection include
mous cell carcinoma. At least theoretically sup- current tobacco users, age above 44 years, CD4
porting the relationship between HPV and some count below 500 cells/mm, and increased inci-
oral cancers is the fact that approximately 7% of dence secondary to sexual behaviors and expo-
Americans harbor HPV in their saliva [50]. sure frequency [53].
Fig. 6.24 3+ p16

immunohistochemistry
of a squamous cell
carcinoma of the tongue.
The 3+ grading of the
p16 indicates the
presence of high-risk
HPV in the cancer.
(Original magnification
×200)
HPV-associated oral and oropharyngeal squa- Surgical management of the mandible is an

mous cell carcinoma is a developing epidemic in important strategy to consider in patients with
the United States with distinctive demographic squamous cell carcinoma of the mandibular gin-
profiles, treatment responses, and inherent prog- giva (Fig. 6.25) as well as in many cases of squa-
noses. Additional data are required regarding mous cell carcinoma of the floor of mouth and
HPV-related oral squamous cell carcinoma to buccal mucosa, particularly when proximity of
possibly offer support of a cause and effect rela- the cancer to the mandible exists with periosteal
tionship as seems to have been established with fixation. Advanced cases of squamous cell carci-
oropharyngeal squamous cell carcinoma. noma of the tongue may also warrant mandibular
resection. The decision-making process for mar-
ginal mandibular resection vs. segmental resec-
6.8 Bone Invasion tion is an essential exercise for oral/head and
neck ablative surgeons and is based on known
The presence of mandibular invasion is a contro- outcomes according to the type of mandibular
versial issue in terms of whether it is an indepen- resection performed based on the type of invasion
dent prognostic factor in patients with oral noted radiographically. This decision-making
squamous cell carcinoma [54]. Prior editions of process is based on physical examination and
the staging manual of the AJCC have stated that radiographic assessment of the mandible. Precise
all forms of mandibular bone invasion were to be analysis of the extent of invasion of the mandible
designated as T4 tumors. The most recent edition is frequently difficult to determine clinically and
of this manual reserves T4 designations to those often difficult to ascertain with plain film exami-
cancers that invade beyond the cortex of the man- nation. Incipient bone invasion is typically mani-
dible (Table 6.1). Li et al. [54] performed a sys- fested by mandibular cortical erosion with
tematic review to determine if mandibular ultimate involvement of the medullary compo-
invasion, and cortical vs. medullary invasion spe- nent of the mandible. Negative radiological stud-
cifically, could be considered an independent ies may exist in the presence of microscopic bone
prognostic factor for patients with oral squamous invasion such that at least marginal resection of
cell carcinoma. Eighteen studies were included in the mandible should be considered under such
Li et al.’s meta-analysis with a total of 3746 circumstances. It is generally accepted that mar-
patients and 1444 patients with bone invasion. ginal mandibular resection should not be per-
When all types of mandibular invasion were conformed in patients with gross destruction of
sidered, the meta-analysis showed no statistical cortical bone or extension of cancer to the medul-
significance between disease-free survival and lary space of the mandible on radiographic evalu-
mandibular invasion (p = 0.43). Cortical invasion ation. Invasion of the mandibular canal or a large
alone had no effect on disease-free survival soft tissue disease on the medial or lateral corti-
(p > 0.05). When patients with medullary inva- ces of the mandible also represent contraindica-
sion were compared to patients with no bone tions to marginal resection. When tumor
invasion, adjusted data and unadjusted data indi- approximates the mandible, including with peri-
cated that medullary bone invasion of the man- osteal fixation yet without gross radiographic
dible possessed statistically and clinically erosion, marginal mandibular resection may rep-
relevant effects on prognosis. The authors con- resent an oncologically safe procedure [55].
cluded that mandibular invasion cannot be con- Petrovic et al. [55] retrospectively studied 1866
sidered a criterion for tumor staging, whereas patients with oral cavity cancer of whom 332
medullary invasion can be considered a criterion patients (18%) were considered suitable for mar-
for staging. Mandibular medullary invasion by ginal mandibular resection. Three hundred
oral squamous cell carcinoma, and not mandibu- twenty-six patients were included in the study.
lar invasion or mandibular cortical invasion, One hundred seven patients had gingival cancers,
could be an independent prognostic factor for 113 patients had floor of mouth cancers, 50
patients with oral squamous cell carcinoma. patients had buccal mucosal cancers, 34 patients
a b
Fig. 6.25 A stage IV (cT4N0M0) squamous cell carci- identifies bone erosion. Final pathology of the composite
noma of the left mandibular gingiva (a) that exhibited resection specimen demonstrated negative soft tissue mar-
bone erosion on preoperative panoramic radiograph (b). gins, infiltration of the medullary portion of the mandible
The patient underwent composite resection of his left by the cancer (arrow) (f) (Hematoxylin & eosin, original
mandibular gingiva, mandible and selective neck dissec- magnification ×100) with negative bone margins, and 1 of
tion (I–III) (c). Close examination of the medial aspect of 31 lymph nodes positive for metastatic squamous cell car-
the primary cancer identifies significant involvement of cinoma. A pathologic stage of pT4aN1 is offered
the medial gingiva (d). Specimen radiograph (e) similarly
Fig. 6.25 (continued) f
had retromolar cancers, and 22 patients had 6.9 Lymphovascular Invasion

tongue cancers. Two hundred seventy-seven
patients (85%) had no bone involvement and 49 Invasion of the rich lymphatic (Fig. 6.26) and
patients (15%) had bone involvement. Thirty-two vascular channels (Fig. 6.27) surrounding a squa-
(65.3%) of the 49 patients with bone involvement mous cell carcinoma of the oral cavity at least
demonstrated cortical bone erosion by the cancer theoretically represents an unfavorable prognos-
and 13 patients (26.5%) demonstrated medullary tic index. Involvement of the local lymphatics
bone erosion. In four patients, the type of bone and vasculature have long been considered histo-
involvement was not specified. Local recurrence- pathological features that promote locoregional
free survival in patients with and without bone recurrence or even distant metastases [56]. Adel
invasion was 62.8% and 76.2%, respectively et al. [56] assessed the outcomes of 571 consecu-
(p = 0.134). No patients demonstrating micro- tive patients who underwent primary surgical
scopic bone invasion required subsequent seg- treatment of squamous cell carcinoma of the oral
mental resection of the mandible. Eight patients cavity. In particular, the authors assessed the
displayed positive bone margins, and five of these independent influence of vascular invasion and
eight patients received adjuvant therapy; radia- lymphatic invasion on the development of locore-
tion therapy alone in four patients, and chemora- gional recurrence and distant metastases in these
diation therapy in one patient. No patients patients. The tumor subsites were tongue (n = 211
developed local/regional recurrence; however, patients), buccal mucosa (n = 209 patients), gin-
two of these patients died with distant metasta- giva (n = 83 patients), hard palate (n = 15
ses. The 5-year disease-specific survival was patients), lip (n = 18 patients), and floor of mouth
78.1% for the entire cohort in this study with a (n = 35 patients). Lymphatic invasion was noted
disease-specific survival of 79.7% for patients in 28 patients (4.9%) and vascular invasion was
without bone invasion compared to 66.0% for noted in 16 (2.8%) patients. There were signifi-
patients with bone invasion. cant associations between lymphatic invasion
Fig. 6.26 An embolus

of squamous cell
carcinoma located
within a lymphatic
channel (arrow).
×400)
Fig. 6.27 An embolus

of squamous cell
carcinoma (large arrow)
located within a small
vein (small arrows).
×200)
and T classification of the tumor (p = 0.009), (p < 0.001), extracapsular spread (p < 0.001),
nodal metastasis (p < 0.001), extracapsular perineural invasion (p < 0.001), depth of invasion
spread (p < 0.001), perineural invasion (p = 0.001), and pathologic differentiation
(p < 0.001), bone invasion (p = 0.004), depth of (p < 0.001), while no significant association was
invasion (p < 0.001), and pathologic differentia- observed with bone invasion (p = 0.327). These
tion (p = 0.002). There were significant associa- associations indicated that the histopathological
tions between vascular invasion and T findings of lymphatic and vascular invasion in the
classification (p = 0.025), nodal metastasis primary cancers were associated with positivity
for cervical metastases, extracapsular spread, nosis of oral squamous cell carcinoma will
perineural invasion, poor differentiation, and further elucidate the significance of these histo-
deeper tumor depth. The presence of lymphatic pathologic factors in the future.
and vascular invasion was not found to correlate
with the variables of local recurrence, regional
recurrence, or distant metastasis. Survival analy- References
sis indicated that the 5-year overall survival rates
for patients with and without lymphatic invasion 1. Wagner VP, Webber LP, Curra M, Klein IP, Meurer L,
Carrad VC, Martins MD. Bryne’s grading system pre-
were statistically significant at 49.3% and 70.3%, dicts poor disease-specific survival of oral squamous
respectively (p < 0.001). Overall survival of cell carcinoma: a comparative study among different
patients with vascular invasion were not signifi- histologic grading system. Oral Surg Oral Med Oral
cantly different (p = 0.511). In disease-specific Pathol Oral Radiol. 2017;123:688–96.
2. Johnson NW, Amarashinghe HK, Qualliotine JR,
survival analyses, the 5-year DSS rates for Fakhry C. Epidemiology. In: Bell RB, Fernandes RP,
patients with and without lymphatic invasion Andersen PE, editors. Oral, head and neck oncology
were statistically significant at 51.6% and 76%, and reconstructive surgery. St. Louis, MO: Elsevier;
respectively (p < 0.001). The difference in DSS 2018. p. 2–56.
3. Duncan LD, Winkler M, Carlson ER, Heidel RE,
according to vascular invasion was not signifi- Kang E, Webb DE. Prevalence of human papilloma-
cantly different between those patients with and virus in squamous cell carcinoma of the oral cavity
without vascular invasion in their specimens with correlation of p16 immunohistochemistry and
(p = 0.247). The disease-free survival rates for human papillomavirus polymerase chain reaction. J
Oral Maxillofac Surg. 2013;71:1367–75.
patients with and without lymphatic invasion 4. Vuity D, McMahon J, Takhiuddin S, Slinger C,
were 51.6% and 64.4%, respectively (p = 0.001). McLellan D, Wales C, MacIver C, Thomson E,
The difference in disease-free survival according McCaul J, Hislop S, Lamb C, Stalker E, Young D. Is
to vascular invasion was not statistically signifi- the 8th edition of the Union for International Cancer
Control staging of oral cancer good enough? Br J Oral
cant between those patients with and without vas- Maxillofac Surg. 2018;56:272–7.
cular invasion in their cancer specimens 5. Spiro RH, Huvos AG, Wong GY, Spiro JD, Gnecco
(p = 0.452). In summary, the authors found that CA, Strong EW. Predictive value of tumor thickness
lymphatic and vascular invasion did not impact in squamous carcinoma of the tongue and floor of
mouth. Am J Surg. 1986;152:345–50.
locoregional recurrence or distant metastases 6. Shim SJ, Cha J, Koom WS, Kim GE, Lee CG, Choi
after treatment. That said, vascular invasion was EC, Keum KC. Clinical outcomes for T1-2 N0-1 oral
not found to adversely affect patient survival, but tongue cancer patients underwent surgery with and
lymphatic invasion was associated with worse without postoperative radiotherapy. Radiat Oncol.
2010;5:1–7.
overall survival, disease-specific survival, and 7. Lydiatt WM, Patel SG, O’Sullivan B, Brandwein MS,
disease-free survival. Based on these results, the Ridge JA, Migliacci JC, Loomis AM, Shah JP. Head
authors indicated that the histopathologic demon- and neck cancers – major changes in the American
stration of either lymphatic or vascular invasion Joint Committee on Cancer eighth edition cancer
staging manual. CA Cancer J Clin. 2017;67:122–37.
might not be an indicator for adjuvant therapy in 8. Elston CW, Ellis IOI. Pathological prognostic factors
patients treated for squamous cell carcinoma of in breast cancer. I. The value of histological grade in
the oral cavity. breast cancer: experience from a large study with long-
term follow-up. Histopathology. 1991;19:403–10.
9. Gleason DF. Classification of prostatic carcinoma.
Cancer Chemother Rep. 1966;50:125–8.
6.10 Conclusion 10. Broders AC. Squamous cell epithelioma of the lip.
JAMA. 1920;74:656–64.
Oral squamous cell carcinoma is associated with 11. Barnes L, Eveson JW, Reichart P, Sidransky D. WHO
classification of tumors, vol. 9. Lyon: IARC; 2005.
numerous histopathologic factors that affect the 12. Muller S. Update from the 4th edition of the World
patient’s prognosis. As per the NCCN guidelines, Health Organization of head and neck tumours:
these factors dictate the administration of adju- tumours of the oral cavity and mobile tongue. Head
vant therapy. The molecular advances in the diag- Neck Pathol. 2017;11:33–40.
13. Bryne M, Koppang HS, Lilleng R, Stene T, Bang advanced head and neck cancer. New Engl J Med.
G, Dabelsteen E. New malignancy grading is a bet- 2004;350:1945–52.
ter prognostic indicator than Broders’ grading in 26. Magliocca KR. Surgical margins. The perspective

oral squamous cell carcinomas. J Oral Pathol Med. of pathology. Oral Maxillofac Surg Clin North Am.
1989;18:432–7. 2017;29:367–75.
14. Jacobsson PA, Eneroth CM, Killander D, Moberger 27. Weinstock YE, Allava I, Dierks EJ. Ptifalls in

G, Martenson B. Histologic classification and grad- determining head and neck surgical margins. Oral
ing of malignancy in carcinoma of the larynx (a pilot Maxillofac Surg Clin North Am. 2014;26:151–62.
study). Acta Radiol Ther Phys Biol. 1973;12:1–8. 28. Shapiro M, Salama A. Margin analysis. Squamous
15. Anneroth G, Batsakis J, Luna M. Review of the lit- cell carcinoma of the oral cavity. Oral Maxillofac
erature and a recommended system of malignancy Surg Clin North Am. 2017;29:259–67.
grading in oral squamous cell carcinoma. J Dent Res. 29. Dillon JK, Brown CB, McDonald TM, Ludwig DC,
1987;95:229–49. Clark PJ, Leroux BG, Futran ND. How does the
16. Bryne M, Koppang HS, Lilleng R, Kjaerheim
close surgical margin impact recurrence and survival
A. Malignancy grading of the deep invasive margins when treating oral squamous cell carcinoma? J Oral
of oral squamous cell carcinomas has high prognostic Maxillofac Surg. 2015;73:1182–8.
value. J Pathol. 1992;166:375–81. 30. Byers RM, Bland KI, Borlase B, Luna M. The prog-
17. Sawazaki-Calone I, Rangel ALCA, Bueno AG, Morais nostic and therapeutic value of frozen section deter-
CF, Nagai HM, Kunz RP, Souza RL, Rutkauskis L, minations in the surgical treatment of squamous
Salo T, Almangush A, Coletta RD. The prognostic cell carcinoma of the head and neck. Am J Surg.
value of histopathological grading systems in oral 1978;136:525–8.
squamous cell carcinomas. Oral Dis. 2015;21:755–61. 31. Johnson RE, Sigman JD, Funk GF, Robinson RA,
18. Brandwein-Gensler M, Teixeira MS, Lewis CM, Lee Hoffman HT. Quantification of surgical margin shrink-
B, Rolnitzky L, Hille JJ, Genden E, Urken M, Wang age in the oral cavity. Head Neck. 1997;19:281–6.
BY. Oral squamous cell carcinoma: histologic risk 32. Mistry RC, Qureshi SS, Kumaran C. Post-resection
assessment, but not margin status, is strongly predic- mucosal margin shrinkage in oral cancer: quantifica-
tive of local disease-free and overall survival. Am J tion and significance. J Surg Oncol. 2005;91:131–3.
Surg Pathol. 2005;29:167–78. 33. Cheng A, Cox D, Schmidt BL. Oral squamous cell
19. Almangush A, Coletta RD, Bello IO, Bitu C, Keski- carcinoma margin discrepancy after resection and
Santii H, Makinen LK, Kauppila JH, Pukkila M, pathologic processing. J Oral Maxillofac Surg.
Hagstrom J, Laranne J, Tommola S, Soini Y, Kosma 2008;66:523–9.
VM, Koivunen P, Kawalski LP, Nieminen P, Grenman 34. Ord RA, Aisner S. Accuracy of frozen sections in
R, Leivo I, Salo T. A simple novel prognostic assessing margins in oral cancer resection. J Oral
model for early stage oral tongue cancer. Int J Oral Maxillofac Surg. 1997;55:663–9.
Maxillofac Surg. 2015;44:143–50. 35. Ellis OG, David MC, Park DJ, Batstone MD. High-
20. Tarsitano A, Tardio ML, Marchetti C. Impact of peri- volume surgeons deliver larger surgical margins in oral
neural invasion as independent prognostic factor for cavity cancer. J Oral Maxillofac Surg. 2016;74:1466–72.
local and regional failure in oral squamous cell carci- 36. Edwards SP. Margin analysis – has free tissue trans-
noma. Oral Surg Oral Med Oral Pathol Oral Radiol. fer improved oncologic outcomes for oral squamous
2015;119:221–8. cell carcinoma? Oral Maxillofac Surg Clin North Am.
21. Chatzistefanou I, Lubek J, Markou K, Ord RA. The 2017;29:377–81.
role of perineural invasion in treatment decisions 37. Brennan JA, Mao L, Hruban RH, Boyle JO, Eby YJ,
for oral cancer patients: a review of the literature. J Koch WM, et al. Molecular assessment of histopatho-
Cranio-Maxillofac Surg. 2017;45:821–5. logical staging in squamous cell carcinoma of the
22. Liebig C, Ayala G, Wilks JA, Berger DH, Albo
head and neck. N Engl J Med. 1995;332:429–35.
D. Perineural invasion in cancer: a review of the lit- 38. Liu SA, Wang CC, Jiang RS, Wang WY, Lin

erature. Cancer. 2009;115:3379–91. JC. Genetic analysis of surgical margins in oral cavity
23. Cracchiolo JR, Xu B, Migliacci JC, Katabi N,
cancer. Br J Surg. 2018;105:e142–9.
Pfister DG, Lee NY, Patel SG, Ghossein RA, Wong 39. de Carvalho AC, Kowalski LP, Campos AHJF, Soares
RJ. Patterns of recurrence in oral tongue cancer with FA, Carvalho AL, Vettore AL. Clinical significance of
perineural invasion. Head Neck. 2018;40:1287–95. molecular alterations in histologically negative surgi-
24. National Comprehensive Cancer Network. NCCN
cal margins of head and neck cancer patients. Oral
clinical practice guidelines in oncology (NCCN Oncol. 2012;48:240–8.
guidelines). Head and neck cancers. Version 2.2018 – 40. Ferris RL, Xi L, Raja S, Hung JL, Wang J, Gooding
20 Jun 2018. www.nccn.org. Accessed 14 Jul 2018. WE, Kelly L, Ching J, Luketich JD, Godfrey
25. Bernier J, Domenge C, Ozsahin M, Matuszewska
TE. Molecular staging of cervical lymph nodes in
K, Lefebvre JL, Greiner RH, Giralt J, Maingon P, squamous cell carcinoma of the head and neck.
Rolland F, Bolla M, Cognetti F, Bourhis J, Kirkpatrick Cancer Res. 2005;65:2147–56.
A, van Glabbeke M. Postoperative irradiation with 41. Brockhoff HC, Kim RY, Braun TM, Skouteris C,

or without concomitant chemotherapy for locally Helman JI, Ward BB. Correlating the depth of inva-
sion at specific anatomic locations with the risk for 49. Shaw RJ, Brown J, Etti T, Risk JM. Prognostic factors
regional metastatic disease to lymph nodes in the in oral, oropharyngeal, and salivary gland cancer. In:
neck for oral squamous cell carcinoma. Head Neck. Brennan PA, Schliephake H, Ghali GE, Cascarini L,
2017;39:974–9. editors. Maxillofacial surgery. 3rd ed. St. Louis, MO:
42. de Matos LL, Manfro G, dos Santos RV, Stabenow Elsevier; 2017. p. 257–74.
E, de Mello ES, Alves VAF, Pinto FR, Kulesar MAV, 50. Gillison ML, Broutian T, Pickard RK, Tong ZY, Xiao
Brandao LG, Cernea CR. Tumor thickness as a pre- W, Kahle L, Graubard BI, Chaturvedi AK. Prevalence
dictive factor of lymph node metastasis and disease of oral HPV infection in the United States 2009–2010.
recurrence in T1N0 and T2N0 squamous cell carci- JAMA. 2012;307:693–703.
noma of the oral tongue. Oral Surg Oral Med Oral 51. Beachler DC, Weber KM, Margolick JB, Strickler
Pathol Oral Radiol. 2014;118:209–17. HD, Cranston RD, Burk RD, Wiley DJ, Minkoff
43. Weiss MH, Harrison LB, Isaacs RS. Use of decision H, Reddy S, Stammer EE, Gillison ML, D’Souza
analysis in planning a management strategy for the G. Risk factors for oral HPV infection among a high
stage N0 neck. Arch Otolaryngol Head Neck Surg. prevalence population of HIV-positive and at risk
1994;120:699–702. HIV-negative adults. Cancer Epidemiol Biomark
44. Masood MH, Farquhar DR, Vanleer JP, Patel SN, Prev. 2012;21:122–33.
Hackman TG. Depth of invasion on pathological 52. Kreimer AR, Pierce Campbell CM, Lin HY, Fulp

outcomes in clinical low-stage oral tongue cancer W, Papenfuss MR, Abrahamsen M, Hildesheim A,
patients. Oral Dis. 2018;24:1198–203. Villa LL, Salmeron JJ, Lazcano-Ponce E, Giulliano
45. Kalnins IK, Leonard AG, Sako K, Razack MS, Shedd AR. Incidence and clearance of oral human papil-
DP. Correlation between prognosis and degree of lomavirus infection in men: the HIM cohort study.
lymph node involvement in carcinoma of the oral cav- Lancet. 2013;382:877–87.
ity. Am J Surg. 1977;134:450–4. 53. Kreimer AR, Clifford GM, Boyle P, Franceschi

46. Ghadjar P, Schreier-Facklam H, Grater R, Evers C, S. Human papillomavirus types in head and neck squa-
Simcock M, Geretschlager A, Blumstein NM, Zbaren mous cell carcinomas worldwide: a systematic review.
P, Zimmer Y, Wilkens L, Aebersold DM. Quantitative Cancer Epidemiol Biomark Prev. 2005;14:467–75.
analysis of extracapsular extension of metastatic 54. Li C, Lin J, Men Y, Yang W, Mi F, Li L. Does med-
lymph nodes and its significance in radiotherapy plan- ullary versus cortical invasion of the mandible affect
ning in head and neck squamous cell carcinoma. Int J prognosis in patients with oral squamous cell carci-
Radiat Oncol Biol Phys. 2010;76:1127–32. noma? J Oral Maxillofac Surg. 2017;75:403–15.
47. D’Cruz AK, Vaish R, Kapre N, Dandekar M, Gupta 55. Petrovic I, Montero PH, Migilacci MC, Palmer FL,
S, Hawaldar R, Agarwal JP, Pantvaidya G, Chaukar Ganly I, Patel SG, Shah JP. Influence of bone inva-
D, Deshmukh A, Kane S, Arya S, Ghosh-Laskar S, sion on outcomes after marginal mandibulectomy
Chaturvedi P, Pai P, Nair S, Nair D, Badwe R. Elective in squamous cell carcinoma of the oral cavity. J
versus therapeutic neck dissection in node-negative Craniomaxillofac Surg. 2017;45:252–7.
oral cancer. N Engl J Med. 2015;373:521–9. 56. Adel M, Kao HK, Hsu CL, Huang JJ, Lee LY,

48. Kuo P, Mehra S, Sosa JA, Roman SA, Husain ZA, Huang Y, Browne T, Tsang NM, Chang YL, Chang
Burtness BA, Tate JP, Yarbrough WG, Judson KP. Evaluation of lymphatic and vascular invasion in
BL. Proposing prognostic thresholds for lymph node relation to clinicopathological factors and treatment
yield in clinically lymph node-negative and lymph outcome in oral cavity squamous cell carcinoma.
node-positive cancers of the oral cavity. Cancer. Medicine. 2015;94:e1510.
2016;122:3624–31.
Genetics of Oral Cancer
7
Anthony Morlandt and Hope Amm
Oral and oropharyngeal cancers affect as many as struggle with the patient who, after highly toxic
275,000 individuals worldwide, including and difficult treatments, recur either locally or in
53,000 in the United States in 2019 representing a distant fashion and as such suffer a shortened
about 11 cases per 100,000 per year. Survival for lifespan. Contemporary understanding of head
all stages has demonstrated a modest increase and neck malignancy has shifted in recent
over the past 40 years, from 50% to 65% disease- decades: from treatment consisting of radical sur-
specific survival over 5 years [1]. Over 90% of gical extirpation espoused by Halsted, Conley,
the cancers occurring in the oral cavity arise from and Martin, to an attempt to understand the
lining gingiva and mucosa and arise from the rap- molecular basis of tumors and develop novel
idly dividing squamous epithelial cells. therapies directed toward specific tumor types.
Classically, these oral squamous cell carcinomas The genetics of oral cancer is a very broad topic
(OSCC) were observed to occur mainly in men, and includes carcinogenesis via initiation, pro-
strike in the sixth and seventh decades of life, and motion, and progression, field cancerization, the
were caused by tobacco and alcohol consump- host immune response, and regulatory genes
tion. In the past decade, an increase in the num- associated with malignant transformation and
ber of cases affecting young people, in particular, metastasis.
white women less than 50 years old without iden- Carcinogenesis is theorized as a multistep
tifiable risk factors, has fueled curiosity regard- progression beginning with initiation, resulting
ing the genetic basis of these tumors [2]. There from environmental injury such as cigarette
are other cases which should respond to treat- smoke, oncogenic viruses such as HPV, or alco-
ment with surgery, radiotherapy, and/or chemo- hol consumption [3]. In other cases, regulatory
therapy, but do not. Head and neck oncologists genes including proto-oncogenes and tumor sup-
pressor genes contain inherited defects which ini-
tiate of carcinogenesis. Regardless of the cause
A. Morlandt (*)
Section of Oral Oncology, Department of Oral and of the defect, initiation represents a mutation
Maxillofacial Surgery, University of Alabama at which is self-perpetuating and irreversible, lead-
Birmingham, Birmingham, AL, USA ing to malignant transformation of normal cells.
e-mail: morlandt@uab.edu The promotion phase follows and is character-
H. Amm ized by unchecked cell growth, invasion, and
Department of Oral and Maxillofacial Surgery, metastasis. The role of chronic inflammation has
University of Alabama at Birmingham,
Birmingham, AL, USA been investigated in oral carcinogenesis and is
e-mail: hopeamm@uab.edu being investigated with respect to MMP

https://doi.org/10.1007/978-3-030-30094-4_7
98 A. Morlandt and H. Amm
expression in periodontal disease (see below). Regulation of the immune response is compli-
Progression is the final stage of carcinogenesis cated and just beginning to be understood in can-
and follows a stepwise pattern. Knudson cer. Many molecules involved in immune
described the “two-hit hypothesis”, where initial regulation have shown genetic variation in
irreversible injury to one allele of a tumor sup- OSCC, including interleukins, chemokines, and
pressor gene does not result in phenotypic change programmed cell death ligand 1 (PD-L1). The
(tumor formation). Loss of heterozygosity role of interleukins in tumor promotion and pro-
describes this phenomenon, where the normally gression is a complicated story, with certain
functioning copy of the tumor suppressor gene interleukins having either tumor promoting or
inhibits cancer growth. The “second hit” by tumor suppressive effects depending on the tumor
mutagenic agents permanently disrupts the origin and tumor microenvironment. However, in
gene’s function, resulting in potential tumor cell OSCC, particular interleukin (IL) SNPs have
growth, leading to invasion and metastasis [4]. been associated with higher risk of OSCC or cer-
With the increase in affordability and accessi- tain stages of OSCC. Correlative studies have
bility of sequencing technologies, the ability to shown a strong association between susceptibil-
examine changes in DNA, RNA, and microRNA ity to OSCC and SNPs in the genes for IL-4,
(miRNA) has greatly expanded. Many of these IL-6, IL-8, IL-10, and TNFα [6]. In IL-10, a SNP
technologies have been used to study genomic in the promoter region was correlated with higher
variations and changes in gene expression in serum levels of IL-10 and increased risk of OSCC
head and neck cancer, including oral cancer, and in an Indian population [7]. Another study of
can serve as biomarkers of carcinogenesis. Many Indian populations showed that a polymorphism
mutations and single nucleotide polymorphisms in IL-18 was associated with OSCC risk com-
(SNPs) in genes associated with carcinogenesis pared to healthy controls [8]. This IL-18 SNP
have been suggested to play a role in susceptibil- was associated with other cancer types in differ-
ity and prognosis of OSCC. The Cancer Genome ent populations, but did not associate with
Network used a multiplatform approach to char- HNSCC in Iranian or Greek populations.
acterize the genomic alterations in 279 cases of Therefore, the relative risk associated with SNPs
HNSCC, including 172 cases of OSCC [5]. This may be dependent on ethnic background as well
study collected fresh, flash-frozen tumors from as cancer type. In a European population, IL-4
newly diagnosed HNSCC patients with either promoter polymorphisms were associated with
adjacent normal tissue or DNA from blood as increased OSCC risk and early-stage disease [9].
normal controls. Along with the collection of An IL-17A SNP was associated with late clinical
clinical data, they isolated DNA and RNA for stages and poor tumor differentiation [10]. A
whole exome, RNA, and miRNA sequencing; Taiwanese study found an IL-6 SNP (rs1800796)
methylation arrays; and DNA SNP arrays. The combined with poor dental care was significantly
study confirmed P53 and CDKN2A loss-of- associated with high-risk of OSCC [11]. In a
function mutations in many cases of smoking- study of stage I/II OSCC patients, low serum lev-
related HNSCC. HPV(−) tumors tended to have els of IL-6 correlated with longer disease-free
amplification of oncogenes (CCND1, MYC), survival [12]. In another study, high tumor stroma
growth factor receptor tyrosine kinases (EGFR, expression of S100A9 was shown to increase
FGFR1), and amplifications or mutations in IL-6 expression in OSCC cells in vitro and
PI3KCA. Mutations in FAT1 (23%) and NOTCH1 in vivo [13]. High S100A9 expression in early-
(19%) were also significantly prevalent in stage OSCC was correlated with shorter recur-
HNSCC. A subtype of OSCC with favorable rence-free survival. This highlights how
prognosis was identified with wild-type P53, but interleukins and upstream pathways, which
mutations in HRAS and CASP8. As the number regulate them, may play a role in the susceptibil-
of patients evaluated with multiomic platforms ity to OSCC.
increases, better definition of prognostic and pre- Another area of recent interest is the effect of
dictive biomarkers may be available. miRNA on gene expression and tumorigenesis.
7 Genetics of Oral Cancer 99
miRNAs are small noncoding RNAs that silence OSCC tumor tissue as well as the peripheral
gene expression by binding to complimentary blood of patients [19]. As immune checkpoint
regions on mRNAs and interrupting their expres- inhibitors continue to advance, HNSCC patients,
sion [14]. These miRNAs are known to regulate including OSCC, may benefit from these
many targets including interleukins. In a study of therapies.
esophageal squamous cell carcinoma, plasma Tumor microenvironment regulation has been
levels of IL-6 correlated with overall survival and implicated in the progression and invasion of
an increased incidence in recurrence [15]. They many human cancers [20]. Matrix metallopro-
discovered patients with a SNP in mir608, a teinases (MMP) are implicated in the degrading
miRNA, had higher plasma levels of IL-6, and the extracellular matrix that encapsulate tumors
analytical studies showed this variant was less allowing for invasion and promoting angiogene-
effective in regulating IL-6 expression. Another sis within tumors. In the promoter region of
study showed that miRNA let-7c was decreased MMP-1, a particular SNP (2G allele) has been
in OSCC patient samples, and that let-7c regu- associated with OSCC in Japanese, Taiwanese,
lates IL-8 in a tumor-suppressing manner [14]. Chinese, and Indian populations [21–25]. A
miRNAs are also being used to help develop oral MMP-2 promoter SNP correlates with OSCC in
diagnostics for OSCC. Yap et al. [16] used patient Thai, Taiwanese, and Chinese populations [21,
samples to identify a pattern of miRNA expres- 22, 26, 27]. Additional studies have identified
sion in OSCC. They showed increased miR-31 SNP in the promoters of MMP-3 and MMP-9
and miR-21 and decreased miR-99a, let-7c, miR- [28]. Additionally MMP-11 polymorphisms
125b, and miR-100 in OSCC samples. They used showed increased susceptibility for OSCC, and
this panel to differentiate between oral swirls patients with a certain MMP-11 allele (rs738792)
(sterile water used to wash the mouth) from had increased lymph node metastasis [29]. When
OSCC and control patients. They correctly iden- considering expression of MMP genes, an analy-
tified 100% of the OSCC patients and 67% of sis of microarray data from OSCC samples origi-
controls. More work is required for effective and nating from the tongue showed increased
noninvasive diagnostics, but further definition of expression of MMP-9 compared to normal tissue
OSCC genetics will allow refinement of ongoing [30]. In the future, detection of MMPs and related
attempts. SNPs may be valuable in determining the relative
Chemokines play an important role in regulat- risk for developing OSCC and the risk of OSCC
ing the immune system by attracting cells to par- progression.
ticular locations. SNPs in CCL4, a monocyte/ Genetics may also play a role in racial dispari-
macrophage attractant, were associated with ties observed in OSCC. Ancestry-related SNPs in
OSCC in a Taiwanese population. One particular DNA polymerase B (POLB) were detected in
SNP was associated with susceptibility to OSCC African American patients with HNSCC from
(increased presence in OSCC patients compared The Cancer Genome Network databases and
to healthy controls) but correlated with smaller were confirmed in additional patient databases
tumor size [17]. Another immune modulatory [31]. The POLB SNPs correlated with signifi-
molecule, programmed cell death protein 1 (PD- cantly increased expression on POLB and worse
1) is the target of a promising therapeutics overall survival and disease-free survival. As
designed to activate tumor-specific T-cells [18]. databases expand, include additional demo-
PD-1 binds to programmed death ligand-1 (PD- graphic information, and a wider variety of
L1) expressed by tumor cells, which suppresses patients; our ability to detect molecular differ-
the T-cell. Monoclonal antibodies bind to PD-1 to ences between types of tumors and markers of
inhibit the suppression and activate T-cell- prognosis will increase.
mediated cytotoxicity. Heineman et al. [18] Several studies have taken the approach of
showed HNSCC patients express high levels of identifying genetic signatures or candidate genes
PD-L1 relative to other tumor types. Another as prognostic indicators. In a study of 311
study showed increased expression of PD-L1 in patients, particular SNPs in cyclin D1 (rs9344)
and retinoblastoma (rs427686) were related to segment of cetuximab which, when excited by a
disease-free survival [32]. Another study com- source light at the appropriate wavelength, can
pared node-positive (LN+) versus node-negative be visualized using an optical scanner, delineat-
(LN-) patients and identified six immune-related ing the tumor. A significant advantage of this
genes (APOE, C3AR1, CD163, CXCR4, FN1, technology is the opportunity to assess surgical
TNFRSF9), which were increased in LN+ sam- margins in real time, both macro- and micro-
ples and correlated with significantly worse scopically, during ablative tumor surgery [39].
5-year node-free survival [33]. In node-positive The surgeon may thereby use the probe to see a
patients, a signature of 11 genes differentiated tumor’s actual margins in the operating room,
between patients with extracapsular spread and relying less on palpation and plain white light
those without [34]. A meta-analysis of SNPs in visualization, and the pathologist can use is for
case-controlled studies associated nine SNPs in real-time immunohistochemistry in the frozen
eight genes with increased risk of OSCC (IL-10, section room.
TGFb, HIF, COX2, XRCC3, CYP1A1, GSTM1, Non-antibody-based autofluorescence and tis-
MTHFR) by comparing OSCC patients to nor- sue reflectance involves an extrinsic light source
mal controls [35]. miRNA sequencing of Danish to excite fluorophores which are naturally occur-
patients and healthy controls identified three up- ring within the tissues and does not require an
regulated miRNAs and three down-regulated infusion with cetuximab or panitumumab.
miRNAs [36]. Three of these miRNAs (miR- Increased histone protein concentration, cross-
486-5p, miR-375, miR92b-3p) were detected in linked collagen, and hypervascular zones within
plasma and correlated with OSCC recurrence in a lesion (potentially resulting from angiogenesis)
Chinese patients with and without recurrence result in a difference between the excitation and
9–12 months post surgical treatment. Expression emission wavelengths, and may be visible to the
of five miRNAs was linked to OSCC compared observer. Many general dental practitioners use
to normal oral mucosa [37]. Evaluation of plasma these technologies in the dental office for screen-
samples showed miR-30a-5p and miR-796-5p ing purposes. A Cochrane review including 4002
were detectable in patients with OSCC compared patients with premalignant and malignant oral
to healthy controls. These studies emphasize the epithelial lesions suggested autofluorescence
diagnostic and prognostic possibilities that come light systems may aid the surgeon already plan-
with analyzed the molecular profile of OSCC. ning a biopsy by identifying the area of the lesion
Oral cancers are particularly difficult to diag- with the greatest metabolic activity but do not
nose for the general medical or dental practitio- possess the sensitivity or specificity to be used
ner. With no widely available screening assay for widespread population screening [40]. In the
using imaging (e.g., mammography, colonos- high-risk population of patients with previous
copy) or serum markers (e.g., PSA, CA 19-9), OSCC who had undergone surgery, radiotherapy,
astute clinical examination is necessary and or both, handheld light-based technology offered
relies on patient access to specialists in many no benefit to tumor or dysplasia detection over
cases. In addition to standard white light visual plain white light visualization by a trained practi-
examination and scalpel biopsy, novel detection tioner [41].
modalities such as fluorescence imaging using As precision medicine continues to gain in
antibody-based optical imaging are currently popularity, increasing the efficacy and safety of
being investigated [38]. Antibody-based optical treatments for patients is key to long-term out-
imaging offers a minimally invasive, antibody- comes. Currently, the only FDA-approved first-
specific approach to tumor detection and could line targeted therapy for OSCC is the epidermal
enhance traditional approaches to screening in growth factor receptor (EGFR) antagonist cetux-
the general practitioners’ office. Because EGFR imab [5, 42]. In 2016, the PDL-1 inhibitor pem-
is over-expressed in many tumors, especially brolizumab was approved for use in recurrent or
OSCCs, a fluorophore can be annealed to the Fc metastatic disease. The success of clinically
available biomarkers for tumor characterization 3. Farah CS, Jessri M, Currie S, Alnuaimi A, Yap T,
McCullough M. Aetiology of oral cavity cancer. In:
is key to understanding the therapeutic profile Kuriakose MA, editor. Contemporary oral oncol-
and treatment effect of immunotherapies. Chia ogy: biology, epidemiology, etiology, and prevention.
et al. [43] validated ex vivo models of HNSCC to New York, NY: Springer; 2017.
determine therapeutic response of tumors. They 4. Suresh A, Kuriakose M, Mohanta S, Siddapa
G. Carcinogenesis and field cancerization in oral
used patient-derived primary cultures to screen squamous cell carcinoma. In: Kuriakose MA, editor.
for sensitivity to several drugs. Next, patient- Contemporary oral oncology: biology, epidemiology,
derived xenografts (PDX) from the same patient etiology, and prevention. New York, NY: Springer;
were treated with drug; the cell culture indicated 2017.
5. Cancer Genome Atlas Network. Comprehensive
they were sensitive too. In each case shown, the genomic characterization of head and neck squamous
drug produced a significant decrease in tumor cell carcinomas. Nature. 2015;517:576. https://doi.
volume. For one OSCC patient, cells and PDX org/10.1038/nature14129.
models from both the primary site and metastatic 6. Serefoglou Z, Yapijakis C, Nkenke E, Vairaktaris
E. Genetic association of cytokine DNA poly-
site showed different genomic profiles and drug morphisms with head and neck cancer. Oral
sensitivities. The patient was treated with gefi- Oncol. 2008;44:1093–9. https://doi.org/10.1016/j.
tinib, an EGFR inhibitor, and showed significant oraloncology.2008.02.012.
regression within 6 weeks of treatment. The 7. Sharma U, Singhal P, Bandil K, Kumar A, Bose
S, Ahuja P, et al. Genetic variations of IL-10:
patient unfortunately progressed after 6 months Identification of novel variations and evaluation of the
of therapy, which was shown to a subpopulation impact of the SNPs/haplotype in the promoter region
of resistant tumor cells that expressed higher lev- with the progression of oral squamous cell carcinoma
els of YAP1. This study demonstrates the utility in indian population. Cytokine. 2018;103:99–108.
https://doi.org/10.1016/j.cyto.2017.09.016.
of patient-matched ex vivo studies to guide ther- 8. Singh PK, Ahmad MK, Kumar V, Hussain SR, Gupta
apy selection and identify treatment response R, Jain A, et al. Effects of interleukin-18 promoter
biomarkers. Many new therapies based on the (C607A and G137C) gene polymorphisms and their
expression of tumor markers or the presence of association with oral squamous cell carcinoma (OSCC)
in northern India. Tumour Biol. 2014;35:12275–84.
particular mutations are under development. https://doi.org/10.1007/s13277-014-2538-0.
Based on genomic information, the design of 9. Vairaktaris E, Yannopoulos A, Vassiliou S, Serefoglou
clinical trials is advancing. New basket and Z, Vylliotis A, Nkenke E, et al. Strong association
umbrella trials assign patients to treatment arms of interleukin-4 (−590 C/T) polymorphism with
increased risk for oral squamous cell carcinoma in
based on the presence of specific genetic altera- Europeans. Oral Surg Oral Med Oral Pathol Oral
tions rather than tumor type [44]. These trials aim Radiol Endod. 2007;104:796–802. https://doi.
to leverage the tumor genetic profiles to match org/10.1016/j.tripleo.2006.12.029.
actionable mutations with the proper drug, 10. Li N, Zhang C, Chen Z, Bai L, Nie M, Zhou B, et al.
Interleukin 17A and interleukin 17F polymorphisms
improve patient outcomes, and streamline clini- are associated with oral squamous cell carcinoma sus-
cal development. ceptibility in a Chinese population. J Oral Maxillofac
Surg. 2015;73:267–73. https://doi.org/10.1016/j.
joms.2014.09.009.
11. Chang JS, Lo HI, Wong TY, Huang CC, Lee WT,
References Tsai ST, Chen KC, Yen CJ, Wu YH, Hsueh WT,
Yang MW, Wu SY, Chang KY, Chang JY, Ou CY,
1. American Cancer Society Cancer Statistics Wang YH, Weng YL, Yang HC, Wang FT, Lin CL,
Center. 2019. https://cancerstatisticscenter.cancer. Huang JS, Hsiao JR. Investigating the association
org/?_ga=2.40982882.1193585586.1552078980- between oral hygiene and head and neck cancer. Oral
1333190596.1551989886#!/cancer-site/Oral%20 Oncol. 2013;49:1010–7. https://doi.org/10.1016/j.
cavity%20and%20pharynx. oraloncology.2013.07.004.
2. Tota JE, Anderson WF, Coffey C, Califano J, Cozen W, 12. Goda H, Okamoto M, Nakashiro KI, Hino S, Murase
Ferris R, St. John M, Cohen E, Chaturvedi A. Rising R, Hamakawa H. Prognostic impact of preoperative
incidence of oral tongue cancer among white men serum interleukin-6 levels in patients with early-stage
and women in the United States, 1973–2012. Oral oral squamous cell carcinoma, defined by sentinel
Oncol. 2017;67:146–52. https://doi.org/10.1016/j. node biopsy. Oncol Lett. 2017;14:7965–9. https://doi.
oraloncology.2017.02.019. org/10.3892/ol.2017.7183.
13. Fang WY, Chen YW, Hsiao JR, Liu CS, Kuo YZ, cell carcinoma. BMC Cancer. 2007;7:187. https://doi.
Wang YC, et al. Elevated S100A9 expression in tumor org/10.1186/1471-2407-7-187.
stroma functions as an early recurrence marker for 24. Chaudhary AK, Singh M, Bharti AC, Singh M, Shukla
early-stage oral cancer patients through increased S, Singh AK, et al. Synergistic effect of stromely-
tumor cell invasion, angiogenesis, macrophage sin-1 (matrix metalloproteinase-3) promoter (−1171
recruitment and interleukin-6 production. Oncotarget. 5A->6A) polymorphism in oral submucous fibrosis
2015;6:28401–24. https://doi.org/10.18632/ and head and neck lesions. BMC Cancer. 2010;10:369.
oncotarget.4951. https://doi.org/10.1186/1471-2407-10-369.
14. Peng CY, Wang TY, Lee SS, Hsieh PL, Liao YW, Tsai 25. Cao ZG, Li CZ. A single nucleotide polymorphism
LL, et al. Let-7c restores radiosensitivity and chemo- in the matrix metalloproteinase-1 promoter enhances
sensitivity and impairs stemness in oral cancer cells oral squamous cell carcinoma susceptibility in a
through inhibiting interleukin-8. J Oral Pathol Med. Chinese population. Oral Oncol. 2006;42:32. https://
2018;47:590–7. https://doi.org/10.1111/jop.12711. doi.org/10.1016/j.oraloncology.2004.08.015.
15. Yang PW, Huang PM, Yong LS, Chang YH, Wu CW, 26. Chaudhary AK, Pandya S, Mehrotra R, Singh M,

Hua KT, et al. Circulating interleukin-6 is associ- Singh M. Role of functional polymorphism of matrix
ated with prognosis and genetic polymorphisms metalloproteinase-2 (−1306 C/T and −168 G/T) and
of MIR608 in patients with esophageal squamous MMP-9 (−1562 C/T) promoter in oral submucous
cell carcinoma. Ann Surg Oncol. 2018;25:2449–56. fibrosis and head and neck squamous cell carcinoma
https://doi.org/10.1245/s10434-018-6532-4. in an Indian population. Biomarkers. 2011;16:577–
16. Yap T, Koo K, Cheng L, Vella LJ, Hill AF, Reynolds 86. https://doi.org/10.3109/1354750X.2011.609602.
E, et al. Predicting the presence of oral squamous 27. O-Charoenrat P, Khantapura P. The role of genetic
cell carcinoma using commonly dysregulated polymorphisms in the promoters of the matrix met-
microRNA in oral swirls. Cancer Prev Res (Phila). alloproteinase-2 and tissue inhibitor of metallo-
2018;11:491–502. https://doi.org/10.1158/1940- proteinase-2 genes in head and neck cancer. Oral
6207.CAPR-17-0409. Oncol. 2006;42:257–67. https://doi.org/10.1016/j.
17. Lien MY, Lin CW, Tsai HC, Chen YT, Tsai MH, Hua oraloncology.2005.07.008.
CH, et al. Impact of CCL4 gene polymorphisms and 28. Pereira AC, Dias do Carmo E, Dias da Silva MA,
environmental factors on oral cancer development and Blumer Rosa LE. Matrix metalloproteinase gene
clinical characteristics. Oncotarget. 2017;8:31424– polymorphisms and oral cancer. J Clin Exp Dent.
34. https://doi.org/10.18632/oncotarget.15615. 2012;4:e297–301. https://doi.org/10.4317/jced.50859.
18. Heineman TE, Widman A, Kuan EC, St John M. The 29. Lin CW, Yang SF, Chuang CY, Lin HP, Hsin

genetic landscape of programmed death ligand-1 (PD- CH. Association of matrix metalloproteinase-11
L1) alterations in head and neck cancer. Laryngosc polymorphisms with susceptibility and clinicopatho-
Investig Otolaryngol. 2017;2:99–103. https://doi. logic characteristics for oral squamous cell carci-
org/10.1002/lio2.79. noma. Head Neck. 2015;37:1425–31. https://doi.
19. Weber M, Wehrhan F, Baran C, Agaimy A, Buttner- org/10.1002/hed.23771.
Herold M, Preidl R, Neukam FW, Ries J. Pd-l1 expres- 30.
Thangaraj SV, Shyamsundar V, Krishnamurthy
sion in tumor tissue and peripheral blood of patients A, Ramani P, Ganesan K, Muthuswami M, et al.
with oral squamous cell carcinoma. Oncotarget. Molecular portrait of oral tongue squamous cell
2017;8:112584–97. https://doi.org/10.18632/ carcinoma shown by integrative meta-analysis of
oncotarget.22576. expression profiles with validations. PLoS One.
20. Kessenbrock K, Plaks V, Werb Z. Matrix metallo- 2016;11:e0156582. https://doi.org/10.1371/journal.
proteinases: regulators of the tumor microenviron- pone.0156582.
ment. Cell. 2010;141:52. https://doi.org/10.1016/j. 31. Ramakodi MP, Devarajan K, Blackman E, Gibbs D,
cell.2010.03.015. Luce D, Deloumeaux J, et al. Integrative genomic
21. Lin SC, Chung MY, Huang JW, Shieh TM, Liu
analysis identifies ancestry-related expression quan-
CJ, Chang KW. Correlation between functional titative trait loci on DNA polymerase beta and sup-
genotypes in the matrix metalloproteinases-1 pro- ports the association of genetic ancestry with survival
moter and risk of oral squamous cell carcinomas. disparities in head and neck squamous cell carcinoma.
J Oral Pathol Med. 2004a;33:323–6. https://doi. Cancer. 2017;123:849–60. https://doi.org/10.1002/
org/10.1111/j.1600-0714.2004.00214.x. cncr.30457.
22. Lin SC, Lo SS, Liu CJ, Chung MY, Huang JW, Chang 32.
Murali A, Varghese BT, Kumar RR, Kannan
KW. Functional genotype in matrix metalloprotein- S. Combination of genetic variants in cyclin D1 and
ases-2 promoter is a risk factor for oral carcinogen- retinoblastoma genes predict clinical outcome in oral
esis. J Oral Pathol Med. 2004b;33:405–9. https://doi. cancer patients. Tumour Biol. 2016;37:3609–17.
org/10.1111/j.1600-0714.2004.00231.x. https://doi.org/10.1007/s13277-015-4179-3.
23. Nishizawa R, Nagata M, Noman AA, Kitamura N, 33. Liu KYP, Lu XJD, Zhu Y, Yip S, Poh CF. Altered
Fujita H, Hoshina H, et al. The 2G allele of promoter immune-related gene expressions indicate oral cancer
region of matrix metalloproteinase-1 as an essential nodal disease. J Dent Res. 2018;97:709–16. https://
pre-condition for the early onset of oral squamous doi.org/10.1177/0022034518758045.
34. Wang W, Lim WK, Leong HS, Chong FT, Lim TK, the utility and limitations of repurposing an open-
Tan DS, et al. An eleven gene molecular signature for field optical imaging device for fluorescence-guided
extra-capsular spread in oral squamous cell carcinoma surgery in head and neck cancer patients. J Nucl
serves as a prognosticator of outcome in patients with- Med. 2017;58(2):246–51. https://doi.org/10.2967/
out nodal metastases. Oral Oncol. 2015;51:355–62. jnumed.115.171413.
https://doi.org/10.1016/j.oraloncology.2014.12.012. 40. Macey R, Walsh T, Brocklehurst P, Kerr AR,

35. Multani S, Saranath D. Genotypic distribution of sin- Liu JLY, Lingen M, Ogden G, Warnakulasuriya
gle nucleotide polymorphisms in oral cancer: global S, Scully C. Diagnostic tests for oral cancer and
scene. Tumour Biol. 2016;37:14501–12. https://doi. potentially malignant disorders in patients pre-
org/10.1007/s13277-016-5322-5. senting with clinically evident lesions. Cochrane
36. Yan Y, Wang X, Veno MT, Bakholdt V, Sorensen
Database Syst Rev. 2015;5:CD010276. https://doi.
JA, Krogdahl A, Sun Z, Gao S, Kjems J. Circulating org/10.1002/14651858.CD010276.pub2.
miRNAs as biomarkers for oral squamous cell 41. Sweeny L, Dean NR, Agnuson JS, Carroll WR,

carcinoma recurrence in operated patients. Clemons L, Rosenthal EL. Assessment of tis-
Oncotarget. 2017;8:8206. https://doi.org/10.18632/ sue autofluorescence and reflectance for oral
oncotarget.14143. cavity cancer screening. Otolaryngol Head
37. Pedersen NJ, Jensen DH, Lelkaitis G, Kiss K, Charabi Neck Surg. 2011;145(6):956–60. https://doi.
BW, Ullum H, Specht L, Schmidt AY, Nielsen FC, org/10.1177/0194599811416773.
von Buchwald C. MicroRNA-based classifiers for 42.
Polverini PJ, D’Silva NJ, Lei YL. Precision
diagnosis of oral cavity squamous cell carcinoma in therapy of head and neck squamous cell carci-
tissue and plasma. Oral Oncol. 2018;83:46. https:// noma. J Dent Res. 2018;97:614–21. https://doi.
doi.org/10.1016/j.oraloncology.2018.05.020. org/10.1177/0022034518769645.
38. Tipirneni KE, Rosenthal EL, Moore LS, Haskins AD, 43. Chia S, Low JL, Zhang X, Kwang XL, Chong FT,
Udayakumar N, Jani AH, Carroll WR, Morlandt AB, Sharma A, et al. Phenotype-driven precision oncology
Bogyo M, Rao J, Warram J. Fluorescence imaging as a guide for clinical decisions one patient at a time.
for cancer screening and surveillance. Mol Imaging Nat Commun. 2017;8:435. https://doi.org/10.1038/
Biol. 2017;19:645–55.. https://do.org/10.1007/ s41467-017-00451-5.
s11307-017-1050-5 44. Slosberg ED, Kang BP, Peguero J, Taylor M, Bauer
39. Moore LS, Rosenthal EL, Chung TK, de Boer E, TM, Berry DA, Braiteh F, Spira A, Meric-Bernstam F,
Patel N, Prince AC, Korb ML, Walsh EM, Young ES, Stein S, Piha-Paul SA, Salvado A. Signature program:
Stevens TM, Withrow KP, Morlandt AB, Richman JS, a platform of basket trials. Oncotarget. 2018;9:21383.
Carroll WR, Zinn KR, Warram JM. Characterizing https://doi.org/10.18632/oncotarget.25109.
Factors Affecting Response
and Survival in Radiotherapy
8
Michael Awadallah, Kurt Nisi, and Ketan J. Patel
8.1 Introduction pathways including a combination of those

mechanisms. These mechanisms include mitotic
Oral squamous cell carcinoma (OSCC) compro- cell death, apoptosis, immunogenic cell death,
mises a group of cancers with a prognosis that is and senescence via direct molecular bond break-
inferior to other head and neck cancer sites. age and/or generation of free radicals. Mitotic
Management of OSCC can comprise three differ- cell death is believed to be the primary pathway
ent modalities including surgery, radiation ther- in radiation-induced cell death. Electromagnetic
apy, and chemotherapy. These management radiation consists of electron and photon therapy
modalities can be used in combination or inde- with the latter being the most often used therapy
pendently. Although not typically used as a pri- for oral cancers. Other forms of radiation include
mary cure for oral squamous cell carcinomas, particle therapy, which consists of either protons,
radiation therapy remains the primary mainstay neutrons, and heavily charged ions such as
of management for oropharyngeal cancers in helium. Different radiation techniques can be
combination with chemotherapy. If surgery is administered postoperatively for the oral cancer
used primarily for curative intent, radiation patient. Some of these treatment techniques
could be used as an adjuvant for advance stage include brachytherapy, intraoperative radiation
disease, positive margins not amenable to re- therapy, and intensity-modulated radiation ther-
resection, multinodal disease, and other adverse apy. Intensity-modulated radiation therapy
features including perivascular or perineural (IMRT) using is the most commonly used radia-
disease. The concept behind administrating post- tion modality used in the adjunctive treatment of
operative radiation therapy is to eliminate any oral cancer. Different fractionation schedules or
residual microscopic tumor burden in the surgical the amount of radiation delivered to the tumor
field and prevent any potential recurrence. site exist; the standard fractionation schedule for
Radiation-induced cytotoxicity is via several oral cancer used in the United States consists of
1.8–2 Gy per fraction given once a day, 5 days a
M. Awadallah (*) · K. J. Patel week for 7–8 weeks totaling to a total of 35–39
Department of Oral and Maxillofacial Surgery, North treatments in total. Radiation treatment can be
Memorial Medical Center, Minneapolis, MN, USA used as a primary or an adjunct to surgery. Early
e-mail: Michael.Awadallah@northmemorial.com; stage tumors that are T1 and some select T2
Ketan.Patel@northmemorial.com
tumors can be treated with radiation alone with
K. Nisi comparable success rates to surgery with a con-
Minneapolis Radiation Oncology,
Minneapolis, MN, USA trol rate that ranges from 85% to 90%. Tumor cell
e-mail: Knisi@mropa.com survival after radiation is multivariable and is
https://doi.org/10.1007/978-3-030-30094-4_8
106 M. Awadallah et al.
contingent upon the type of radiation, frequency cancer is in the neck where occult spread to the
of radiation, amount of radiation, inherent radia- lymphatic basins can occur up to 30% of the
tion sensitivity of the tumor, and oxygen tension. time. The planning target volume (PTV) is the
Radiation-induced damage is most pronounced area to be irradiated to confirm treatment of the
during the M and G2 phase of the mitotic cycle, CTV. This is usually achieved by enlarging the
hence actively dividing tumors are more sensitive CTV by a 0.5–1 cm margin. The CTV and PTV
to radiation-induced death of the as compared to are considered to be comparable to surgical mar-
slowly dividing or senescent ones. An oxygen- gins in the radiation oncology realm (see
rich environment is radiosensitizer and will lead Fig. 8.4).
to more radiation-induced cell death as compared
to a hypoxic environment.
8.2 Contemporary Radiation

Therapy for Oral Cancer
By far the most common radiation delivery

mechanism for oral cancer is via intensity-
modulated and image-guided radiation therapy
(IMRT). This technique is a form of highly con-
formational radiation therapy in which high dose
volumes are adapted closely to the target area
while minimizing collateral damage to the adja-
cent healthy tissues. The basic principles in
achieving IMRT include outlining the target vol-
ume and margins using computed tomography
and image guidance programs, using multiple
beam directions to cross fire on targets, and cus- Fig. 8.1 IMRT treatment volumes of an anterior FOM
tomization of each of those radiation beams to with pathologically proven bilateral neck nodes in sagittal
deliver the conformed dose while modulating view
intensity. The pivotal step in this process is defin-
ing the volume and margins of the target. There
are three volumes that are measured between the
radiation oncologist and the medical physicist.
The gross tumor volume (GTV) is delineated
using multiple slices using CT, PET/CT, and/or
MRI in multiple planes, and this is where the
macroscopic disease is situated. Once the gross
tumor volume is outlined, the clinical target vol-
ume (CTV)is then estimated. The CTV is the
GTV margin in addition to the estimated micro-
scopic disease not visualized on imaging, usu-
ally a 0.5–1 cm extension beyond the GTV (see
Figs. 8.1, 8.2, and 8.3). The CTV is dependent
on aggressiveness of the disease, pattern of local
infiltration, pattern of regional spread, and pat-
Fig. 8.2 IMRT treatment volumes of an anterior FOM
tern of locoregional failure in the oral cavity. The with pathologically proven bilateral neck nodes in axial
most applicable situation of this concept in oral view
8 Factors Affecting Response and Survival in Radiotherapy 107
There are several different fractionation pro-

tocols used for management of residual or pri-
mary disease for oral cancer according to the
National Comprehensive Cancer Network
(NCCN) guidelines (see Fig. 8.5). Standard frac-
tionation treatment is modulated based on treat-
ment modality and high- vs. intermediate- vs.
low-risk tumors. For definitive treatment with
radiation alone, 66–70 Gy is delivered Monday
through Friday over 6–7 weeks for high-risk
tumors, 54–63 Gy for intermediate-risk tumors,
and 44–50 Gy for low-risk tumors. The standard
postoperative radiation fractionation schedule is
Monday through Friday at 2.0 Gy fractions per
day for a total dose of 60–66.6Gy of overall
treatment over 6 weeks for high-risk tumors,
54–63 Gy for intermediate- risk tumors, and
44–50 Gy for low-risk tumors. This treatment is
usually performed within 6 weeks after surgery
to allow for complete wound healing and preven-
tion of wound breakdown. The overall goal of
postoperative irradiation is to control locore-
gional persistence or recurrence at the primary
and regional sites. Altered fractionation sched-
Fig. 8.3 IMRT treatment volumes of an anterior FOM with ules are used to address the problem of acceler-
pathologically proven bilateral neck nodes in coronal view ated re-population; which is the concept of a
Fig. 8.4 Different dosage curves based on planned their respective IMRT treatment volumes
Total Days/ Time of

Gy/fractionation fractionations/day
dose week treatment
Definitive RT high risk 66-70 2 5 1 6-7 weeks
Definitive RT
81.6 1.2 5 2 7 weeks
hyperfractionation
Definitive RT
54-63 1.6-1.8 5 1 5-6 weeks
Intermediate risk
Definitive RT low risk 44-50 2 5 1 4-5 weeks
Adjunctive RT high 60-66
2 5 1 6-6.5 weeks
risk Gy
Adjunctive RT
54-63 2 5 1 5-6 weeks
intermediate risk
Adjunctive RT low risk 44-50 2 5 1 4-5 weeks
Fig. 8.5 IMRT schedules for primary therapy and post-operative therapy as per NCCN guidelines
tumor cell regenerative response to chemother- and locally advanced tumors with lymph node
apy, radiation, and surgery leading to increased metastasis. According to these studies, there was
cell division by residual tumor cells ultimately an 11% and 10% increase in 5-year overall sur-
leading to locoregional failure. This response is vival in the nonlocally advanced and locally
more accentuated in mucosal tissues that nor- advanced groups, respectively [1–3].
mally have a higher turnover rate at base line as Other radiation treatment modalities include
compared to other bodily tissues. Patients with brachytherapy, proton beam therapy, neutron beam
treatment delays either from increased time therapy, and stereotactic radiation therapy.
from surgery to radiation treatment or from pro- Brachytherapy involves the placement of a radio-
longed radiation treatment are good candidates active isotope via a carrier and placing it intersti-
for the altered fractionation schedule. tially into the central tumor bed to provide a high
Hyperfractionation is the delivery of multiple but localized dose of radiation. It can be used alone
small doses in a given day instead of one large or in conjuction with external beam radiation. It
dose. Accelerated fractionation delivers two or works via calculated radioactive decay of that spe-
more doses per day to allow for faster comple- cific isotope thereby emitting radiation to the
tion of the total radiation dose. Both hyper and tumor bed. Different isotopes can be used and
accelerated fractionation will lead to a greater include but are not limited to Iodine-125, Iridium
total radiation dose delivery over the course of 192, Palladium-103, Cesium 131, Gold 108, and
treatment as compared to conventional fraction- Ruthenium 106. The technique of the procedure
ation. Overall, the concept is to decrease the time itself involves the temporary placement of hollow
interval between radiation dosing to control for plastic catheters or tubes into the tumor bed, and
accelerated re-population. Hyperfractionation is the radioactive isotope is placed into the catheters.
usually used for oral cavity cancers as a defini- Fractionation schedule when using brachytherapy
tive therapy for high-risk tumors. The total frac- is 0.4–0.5 Gy per hour for several hours given
tionation dose is 81.6 Gy over 7 weeks with 5 days a week to a total dose of 60–70 Gy if given
1.2 Gy given twice a day according to the for definitive treatment. However, because of more
National Comprehensive Cancer Network advanced techniques in using conformational radi-
(NCCN ) guidelines. ation therapy, brachytherapy has fallen out of favor
Multiple retrospective studies have shown a by most if not all radiation oncologists.
statistically and a clinically survival benefit of Proton beam therapy (PBT) is a promising
postoperative radiation treatment for non-locally relatively new technique in particle radiation in
advanced tumors with lymph node metastasis which highly charged protons are delivered to the
tumor bed. The inherent physical characteristics 8.3 Indications

allow for constant dose delivery without attenua-
tion of energy then a rapid and steep decrease in Approximately 60–70% of oral cancer patients
energy beyond their target; allowing for less col- present with late-stage disease, stage 3 or 4, all of
lateral damage to healthy tissue and hence has an which will have at least one of the following char-
improved toxicity profile as compared to acteristics: frank bone invasion, extensive tumor
IMRT. However; this does not translate to better infiltration into the surround soft and hard tissues,
efficacy in repect to locoregional control at the nodal metastasis, and extracapsular extension of
primary or regional sites. The fractionation which all are indications for adjunct radiation
schedule for PBT is the same as that for IMRT. In therapy. Other indications for adjunct radiation
respect to oral cancer, there is no major advan- therapy include microscopic positive margins,
tage in using PBT over IMRT. The primary ben- lympho-vascular invasion, and perineural inva-
efits of PBT are in the reduction or even the sion. Primary radiation treatment is usually
elimination of catastrophic radiation sequela to reserved for those patients that cannot undergo
nearby critical structures like the brain, brain surgery secondary to severe comorbidities and/or
stem, and the orbital/periorbital tissues during a poor functional status. For tumors including
the treatment of sinonasal malignancies or other oropharynx, hypopharynx, nasopharynx, and lar-
malignancies that extend cephalad. The total ynx, radiation is usually used for primary man-
fractionation dose of PBT in the head and neck agement with addition of chemotherapy.
ranges from 66 to 83 Cobalt-gray equivalent
(CGE) depending on tumor site.
Stereotactic body radiation therapy (SBRT) is a 8.4 Prognostic Factors
type of external beam radiation that delivers greater in Radiation Therapy
doses of radiation to the tumor bed. Some of the
advantages of SBRT are a sharp dose fall off, good Tumor biological factors, surgical treatment ade-
local control rates, shorter overall treatment time as quacy, patient-specific factors, and type of radia-
compared to other radiation treatment modalities, tion treatment all play a role in affecting the
and acceptable toxicities. The fractionation regi- response to radiation treatment and the survival
mine can be either single dose SBRT or fraction- of the oral cancer patient. Tumor-specific factors
ated SBRT. In the single dose SBRT, a 13–18 Gy include the oxygen tension in the microenviron-
fraction is given at one time. While in the fraction- ment, radiosensitivity, tumor burden, epidermal
ated SBRT, a total of 36–48 is given over five to growth factor expression, nodal involvement,
eight fractions. At this time, the use of SBRT in the extranodal extension, and molecular composition
head and neck is limited to select patients that can- of the tumor [4]. Surgical treatment factors
not undergo surgery due to poor functional status mainly revolve around the presence of positive
or in recurrent tumors of the head and neck. margins after resection. Patient-specific variable
Neutron and heavy ion therapy are a form of include functional status and concurrent use of
high linear energy transfer. Heavy ion dose distri- tobacco during radiation. Radiation factors
bution is similar to that in proton therapy because include type of fractionation and timing to frac-
they deposit little energy until they reach the end of tionation [5].
their range. Most heavy particle therapy use either
carbon, helium, or neon as their charged particles.
However, neutron particles are uncharged, and 8.5 Oxygenation
their dose distribution is similar to that of tradi-
tional photon therapy but with at least a 20-fold As stated previously, the primary mechanism of
increase of energy deposition at the target site. The radiation tumor cell lysis is by direct killing via
implementation of these techniques are currently the formation of free radicals. This mechanism is
under investigation globally, but it will be some- further enhanced by the amount of oxygenation
time before light is shed on their efficacy. in the tumor environment; more oxygen leads to
more free radicals and hence tumor cell death. ers are then profiled upon degree of expression
Hence, oxygen is a potent modulator of radiosen- using a cluster analysis. Biological factors
sitivity. Multiple methods have been developed involved in modulating tumor response are those
to quantify oxygenation in a tumor bed environ- that affect proliferation, progression, cell cycle
ment with the initial being polyarographic elec- deregulation, apoptosis, and angiogenesis [10,
trodes. More contemporary techniques use 11]. The p53 tumor suppressor molecule regu-
exogenous and endogenous and exogenous lates the cell cycle and programmed cell death. In
hypoxia markers by immunohistochemistry. normal cells, the p53 tumor suppressor gene is
These techniques do not require any further inter- not expressed but following an insult to the cell
vention other than the initial incisional biopsy. the gene is translated to halt cell cycle division
Immunohistochemistry is used to detect the and promotes DNA repair mechanisms [12].
upregulation of certain genes that are modulated About 50% of head and neck SCC have muta-
by hypoxia. These genes are responsible for glu- tions on the p53 gene leading to the expression of
cose transport, pH regulation, and angiogenesis. a mutant protein that is nonfunctional/dysfunc-
The presence and degree of hypoxia in oral can- tional or it can lead to the absence of the protein
cer are associated with poorer outcomes after [13]. The presence of p53 derangements leads to
radiation treatment [6]. the obtunded ability of the cell to initiate or com-
plete apoptosis and to arrest the cell cycle. These
characteristics then lead to less susceptibility to
8.6 Tumor Volume radiation treatment and propels a self-propagating
cycle of cellular mutations and transformation
Tumor volume can be quantitively measured leading to further deregulation of the cell division
indirectly by integrated F-fluorodeoxy glucose cycle. This leads to tumor polyclonality and fur-
(FDG) uptake with positron emission tomogra- ther radio-resistance [4]. P53 mutations have a
phy (PET) and computed tomography in combi- strong correlation with locoregional failure and
nation with computer-aided software like RT primary radiation but not surgery [12]. The
image. This software will translate the amount of mutant type of p53 or the lack of p53 can lead to
FDG uptake to a semiquantitative standard a overexpression of Bcl2 [14]. This molecule
uptake value (SUV) [7]. Tumors with increased plays an important role in promoting cellular
SUV is a prognostic indicator and inversely asso- immortality by inhibiting apoptosis. This overex-
ciated with survival and poorer treatment pression has been associated with increased
response [8]. Multiple mechanisms have been put radio-resistance [15]. Detecting the molecule
worth as to explain this phenomenon and include ki-67 is an indirect immunohistochemical method
degree of hypoxia in a dense tumor, degree of measuring tumor growth rate. Ki-67 antibody is
proliferative potential, degree of metabolic activ- used to detect a certain cell nucleus antigen that
ity, tumor cell density, low apoptosis rate, or a becomes abundant during the G2 and M phase
combination of these elements. All of these ele- [16]. Several investigations have shown a corre-
ments are adverse factors in locoregional control lation between an increased ki-67 index and sur-
with radiation treatment [9]. vival, time to relapse, radio-resistance, and
regional spread to lymph nodes [17, 18].
Epidermal growth factor receptor is a trans-
8.7 Molecular Markers membrane glycoprotein of the tyrosine kinase
family. Activation of which leads to multiple
Molecular markers independently correlate with downstream signaling pathways that modulate
radiotherapy locoregional control and survival growth, differentiation, and apoptosis. The major-
[10]. Immunohistochemical staining is the ity if not all of oral squamous cell carcinomas
method of choice to test the expression and the have elevated expression of the EGFR glycopro-
degree of expression of these markers; the mark- tein. This is detected by immunohistochemistry. It
has been demonstrated that radiation induces alternative fractionation therapy was associated
auto-stimulation of the EGFR leading to down- with an 3.1% improvement of overall survival at
stream epidermal growth factor which in turn 5 years as compared to conventional fraction-
leads to an increase in proliferation, colony for- ation. This difference is even more pronounced
mation, and ultimately re-population leading to with the use of hyperfractionation where the
radio-resistance [4]. This characteristic of the absolute difference in overall survival at 5 years
EGFR has been associated with locoregional is 8.1% [27]. The Radiation Therapy Oncology
relapse in vitro and in vivo. However, it has not Group (RTOG) phase 3 randomized study 9003
been associated with distant metastasis [19]. comparing hyperfractionation and accelerated
fraction to conventional fractionation in head and
neck squamous cell carcinomas showed improved
8.8 Timing of Radiation Therapy locoregional control rate with the alternate sched-
and Fractionation Schedule ule but no improvement in overall survival [20].
A Cochrane review concluded that alternate frac-
Tumor cells that survive prior to the initiation of tionation schedule has more efficacy in control of
radiation therapy or during gaps in between frac- tumor burden at the primary site which translated
tionation therapy can repopulate the tumor col- to better overall survival [28].
ony given their potential to divide. Multiple head
and neck cancer studies have shown worse local
control rate if timing to initiation of radiation 8.9 Smoking and Malnutrition
therapy is prolonged [20–22]. A recent retrospec-
tive study looking at 15,064 patients showed It is well established that tobacco use is a signifi-
worse outcomes for those patients with prolonged cant risk factor for developing oral squamous cell
radiation treatment and prolonged postoperative carcinoma. One systematic review study esti-
interval between surgery and initiation of radia- mated that the population attributable risk for
tion therapy. In respect to postoperative interval developing oral squamous cell carcinoma from
time and time in radiation therapy, mortality smoking alone was 25%; being dose-dependent
increased after 5.7 and 7.9 weeks respectively with a life time cumulative use that is positively
[21]. Hence prolongation of the overall radiation correlated with developing oral cancer [29].
treatment time which includes time to initiation However, does this also translate to worse out-
of postoperative radiotherapy in addition to the comes in the efficacy of irradiation treatment? A
time needed for completion of radiation therapy study performed by Brown et al. on patients with
can have a negative impact on local control rate locally advanced stage III and stage IV head and
and survival of the patient [23–25]. For those neck cancer showed patients that continued to
patients who have prolonged overall radiation smoke throughout their course of postoperative
treatment time, then they might be candidates for radiation therapy had a lower response and over-
accelerated fractionation [26]. all survival rate than those who abstained from
The optimal type of fractionation has been the smoking during therapy [30]. A retrospective
topic of controversy for oral cancer and other chart review of a 101 patients with a mean fol-
head and neck cancers. The question is whether low-up time of 49 months performed by Chen
or not the conventional fractionation schedule is et al. showed patients that active smokers
less effective in local control than the alternate throughout radiation therapy had poorer disease-
fractionation schedules which includes acceler- free survival (42% vs. 65%), 5-year overall sur-
ated or hyperfractionation. Naturally, the next vival (23% vs. 55%), and locoregional control
question is which of the two types of the alternate (58% vs. 69%) as compared to smokers that
fractionation schedule is optimal for the oral can- stopped smoking prior to the initiation of radia-
cer patient. The MARCH meta-anaylsis trial that tion therapy [31]. Multiple theories of why smok-
included 34 trails with 11,969 patients found that ing during radiation therapy predisposes to a
worse prognosis and these include the continued (<25 kg/m2) had statistically significant decrease
potentiation and propagation of p53 mutations, in cancer-specific survival and overall survival as
continued hypoxia in the tumor environment and compared to those patients with higher pretreat-
less free radical production by radiation therapy ment BMI (>25 kg/m2) [33]. Multiple theories on
due to decreased oxygenation, potential upregu- why malnutrition is correlated with worse out-
lation of EGFR, and the repercussions of the comes after radiation therapy have been proposed
overall field cancerization effect that can lead and include association with increased tumor
metachronous or synchronous primaries. burden at the locoregional site with higher T and
Malnutrition and wasting are significant and N classifications, increased incidence of anemia,
prevalent problems for the oral cancer patient and worse functional status outcomes.
secondary to not only the disease sequela but also
due to comorbidities and social factors like
tobacco and/or alcohol abuse. This is com- 8.10 Pathologic Factors
pounded by the dysphagia, odynophagia, and
mucositis brought on during and after radiation In general, the same indications used to treat
therapy which at the very least negatively impacts oral cancer with postoperative radiation therapy
the patient’s quality of life (QOL). In addition, all are the ones that negatively affect the prognosis
patients will have some degree of decreased solid after radiation therapy. These same factors con-
and liquid oral intake after surgical ablation tribute to the advanced staging of oral cancers.
alone. Hence, patients planned for surgical and Even with improved survival rates due to earlier
adjunctive radiation treatment may benefit from detection and more robust treatment like IMRT,
alternative routes of supplemental nutrition that immunotherapy, altered fractionation, and other
include a nasogastric or a gastric feeding tube. treatment modalities, the overall 5-year sur-
According to the NCCN guidelines, intervention vival rates for OSCC ranges from 34% to 42%
with feed tube placement should be implemented and locoregional recurrence of 16–35% [34–
in patients with severe weight loss prior to treat- 36]. Cervical lymph node metastasis, extracap-
ment, patients with 5% weight loss or greater sular extension, and positive or close surgical
over one month, and patients woth 10% weight margins are the most critical prognostic factors
loss or greater over 6 months. Multiple studies in 5-year overall survival rate, disease-specific
have shown that malnutrition not affects QOL but survival, locoregional recurrence, and death
is associated with poorer treatment outcomes. An follow postoperative radiation therapy for
analysis study of the RTOG phase III prospective patients. The predominant site of failure is local
trial 90–93 done by Rabinovitch et al. was per- recurrence at the primary site followed by
formed to evaluate nutritional support with can- regional spread/ recurrence in the neck [37].
cer outcomes in patients with locally advanced The majority of locoregional failure will occur
head and neck cancer treated with definitive radi- in the first 2 years.
ation therapy. In their study, patients that were Local control rate is dependent on primary
malnourished to the point where pretreatment site, and clinical T stage. The 5-year local control
nutritional support was needed, had a worse rate of the oral tongue, the retromolar trigone, the
5 year locoregional control rate as compared to floor of the mouth, and other sites ranged from
patient who did not require any nutritional sup- 77% to 84%, 74–76%, 81–85%, 70–81%, respec-
port (29% vs. 57%, respectively) [32]. A retro- tively [38, 39]. Local control rate based on T
spective study by Pai et al. also showed similar stage at 5 years for T1–T2 and T3–T4 tumors
results when using radiation therapy for curative were 88–90% and 70–79%, respectively [38, 39].
intent on 1,562 patients by using their pretreat- Locoregional control rates were influenced by
ment body mass index (BMI) as an indirect mea- margin status, American Joint commission on
sure of their nutritional status. Their study Cancer clinical stage, and Extranodal Extension
showed patients with lower preradiotherapy BMI (ENE). The 5-year locoregional control rate for
negative margins and positive margins ranged dermal growth factor inhibitor like Cetuximab
from 77% to 79% and 47–52%, respectively [38, also showed improved locoregional control, dis-
39]. In respect to the AJCC clinical stage on pre- ease progression-free survival, and overall sur-
sentation, the locoregional control rate ranged vival for head and neck cancer patients as
from 87% to 100% for stage I, 74–88% for stage compared to radiation therapy alone [41].
II, 68% for stage III, and 62–72% for stage
IV. The 5-year locoregional control incremen-
tally decreases with each additional indication 8.12 Conclusion
for postoperative radiation, and this is more pro-
nounced after three or more indications [38, 39]. For locally advanced oral cancers surgery in
ENE leads to a drastic decrease in locoregional combination with postoperative radiation therapy
control and overall survival in patients with oral achieves optimal results in respect to locore-
squamous cell carcinoma treated with postopera- gional control, 5-year overall survival, and
tive radiation. The 5-year locoregional control is disease-free specific survival. IMRT with 3-D
61% as compared to 73% in patients without conformation has shown to provide an optimality
ENE. The 5-year overall survival in patients with in delivering the correct amount of radiation to
ENE was 21% as compared to 58% in patients the tumor bed and the surrounding environment,
without ENE. The 5-year cause-specific survival while minimizing unwanted damage and toxicity
was in patients with ENE was 38% as compared to the healthy surrounding tissue, especially spar-
to 81% in patients without ENE [39]. ing the major and minor salivary glands. The
goals of postoperative radiation therapy are pri-
marily two-fold at the primary and regional sites;
8.11 Additional Adjunctive to eradicate residual microscopic and macro-
Treatment scopic disease and to prevent recurrence. Ongoing
with Chemotherapy trials with escalation therapy and different frac-
tionation doses and frequency will have future
The RTOG 9501 and European Organization for therapy implications especially in patients with
Research and Treatment of Cancer (EORTC) high-risk features like nodal load and ENE.
22,301 randomized trials showed improvement
in locoregional control and disease-free survival
with concurrent administration of the chemo-
therapeutic agent cisplatin, especially for high- References
risk squamous cell carcinoma of the head and
1. Lavaf A, et al. Adjuvant radiotherapy improves over-
neck. A randomized trial by Bernier et al. placed all survival for patients with lymph node-positive
334 patients with stage 3 and 4 head and neck head and neck squamous cell carcinoma. Cancer.
cancer in two groups of a 167 patient each; one 2008;112(3):535–43.
2. Kao J, et al. Adjuvant radiotherapy and survival for
group received postoperative radiation therapy
patients with node-positive head and neck cancer: an
and the other group received postoperative con- analysis by primary site and nodal stage. Int J Radiat
current chemoradiation with Cisplastin. Oncol Biol Phys. 2008;71(2):362–70.
Progression-free survival was 47% for the con- 3. Sanguineti G, Forastiere AA. Determining the survival
benefit of adjuvant radiotherapy in patients with node-
current chemoradiation group as compared to
positive head and neck cancer. Nat Clin Pract Oncol.
36% for the radiation alone group. The overall 2008;5:694.
survival was 53% in the combined therapy group 4. Perri F, et al. Radioresistance in head and neck squa-
compared to the 40% in the radiation alone mous cell carcinoma: biological bases and therapeutic
implications. Head Neck. 2015;37(5):763–70.
group [40]. The overall incidence of relapse was
5. Silva P, et al. Clinical and biological factors affect-
also decreased for the combined therapy group ing response to radiotherapy in patients with head
as compared to the radiation alone group; 31% and neck cancer: a review. Clin Otolaryngol.
and 18%, respectively. Target therapy with epi- 2007;32(5):337–45.
6. Nordsmark M, Overgaard J. A confirmatory prog- 21. Ho AS, et al. Quantitative survival impact of compos-
nostic study on oxygenation status and loco-regional ite treatment delays in head and neck cancer. Cancer.
control in advanced head and neck squamous cell car- 2018;124(15):3154–62.
cinoma treated by radiation therapy. Radiother Oncol. 22. Overgaard J, et al. Five compared with six fractions
2000;57(1):39–43. per week of conventional radiotherapy of squamous-
7. Allal AS, et al. Standardized uptake value of 2-[18F] cell carcinoma of head and neck: DAHANCA
fluoro-2-deoxy-d-glucose in predicting outcome 6 and 7 randomised controlled trial. Lancet.
in head and neck carcinomas treated by radiother- 2003;362(9388):933–40.
apy with or without chemotherapy. J Clin Oncol. 23. González Ferreira JA, et al. Effect of radiother-

2002;20(5):1398–404. apy delay in overall treatment time on local con-
8. La TH, et al. Metabolic tumor volume predicts for trol and survival in head and neck cancer: review
recurrence and death in head-and-neck cancer. Int J of the literature. Rep Pract Oncol Radiother.
Radiat Oncol Biol Phys. 2009;74(5):1335–41. 2015;20(5):328–39.
9. Jeong J, et al. Estimate of the impact of FDG-avidity 24. Harris JP, et al. Association of survival with shorter
on the dose required for head and neck radiother- time to radiation therapy after surgery for US patients
apy local control. Radiother Oncol. 2014;111(3): with head and neck cancer. JAMA Otolaryngol Head
340–7. Neck Surg. 2018;144(4):349–59.
10. Buffa FM, et al. Molecular marker profiles predict 25. Graboyes EM, et al. Effect of time to initiation of
locoregional control of head and neck squamous cell postoperative radiation therapy on survival in sur-
carcinoma in a randomized trial of continuous hyper- gically managed head and neck cancer. Cancer.
fractionated accelerated radiotherapy. Clin Cancer 2017;123(24):4841–50.
Res. 2004;10(11):3745–54. 26. Sanguineti G, et al. Accelerated versus conven-

11. Begg AC. Predicting recurrence after radiother-
tional fractionated postoperative radiotherapy for
apy in head and neck cancer. Semin Radiat Oncol. advanced head and neck cancer: results of a multi-
2012;22(2):108–18. center phase III study. Int J Radiat Oncol Biol Phys.
12. Alsner J, Sorensen SB, Overgaard J. TP53 mutation 2005;61(3):762–71.
is related to poor prognosis after radiotherapy, but not 27. Lacas B, et al. Role of radiotherapy fractionation in
surgery, in squamous cell carcinoma of the head and head and neck cancers (MARCH): an updated meta-
neck. Radiother Oncol. 2001;59(2):179–85. analysis. Lancet Oncol. 2017;18(9):1221–37.
13. Somers KD, et al. Frequent p53 mutations in head and 28. Baujat B, et al. Hyperfractionated or accelerated

neck cancer. Cancer Res. 1992;52(21):5997–6000. radiotherapy for head and neck cancer. Cochrane
14. Csuka O, et al. Predictive value of p53, Bcl2 and bax Database Syst Rev. 2010;12:Cd002026.
in the radiotherapy of head and neck cancer. Pathol 29. Radoi L, Luce D. A review of risk factors for oral
Oncol Res. 1997;3(3):204–10. cavity cancer: the importance of a standardized
15. Haffty BG, Glazer PM. Molecular markers in clinical case definition. Community Dent Oral Epidemiol.
radiation oncology. Oncogene. 2003;22(37):5915–25. 2013;41(2):97–109, e78–91.
16. Sittel C, et al. Prognostic significance of Ki-67
30. Browman GP, et al. Influence of cigarette smoking on
(MIB1), PCNA and p53 in cancer of the oropharynx the efficacy of radiation therapy in head and neck can-
and oral cavity. Oral Oncol. 1999;35(6):583–9. cer. N Engl J Med. 1993;328(3):159–63.
17. Jacob R, et al. Heterogeneity of squamous cell car- 31. Chen AM, et al. Tobacco smoking during radiation
cinomas of the head and neck-analysis of tumor biotherapy for head-and-neck cancer is associated with
logic factors and proliferation rates. Laryngoscope. unfavorable outcome. Int J Radiat Oncol Biol Phys.
1996;106(9):1170–5. 2011;79(2):414–9.
18. Raybaud H, et al. Nuclear DNA content, an adjunct 32. Rabinovitch R, et al. Impact of nutrition support on
to p53 and Ki-67 as a marker of resistance to radia- treatment outcome in patients with locally advanced
tion therapy in oral cavity and pharyngeal squa- head and neck squamous cell cancer treated with
mous cell carcinoma. Int J Oral Maxillofac Surg. definitive radiotherapy: a secondary analysis of
2000;29(1):36–41. RTOG trial 90-03. Head Neck. 2006;28(4):287–96.
19. Ang KK, et al. Impact of epidermal growth factor 33. Pai PC, et al. Impact of pretreatment body mass

receptor expression on survival and pattern of relapse index on patients with head-and-neck cancer
in patients with advanced head and neck carcinoma. treated with radiation. Int J Radiat Oncol Biol Phys.
Cancer Res. 2002;62(24):7350–6. 2012;83(1):e93–e100.
20. Fu KK, et al. A Radiation Therapy Oncology Group 34. Amit M, et al. Improvement in survival of patients with
(RTOG) phase III randomized study to compare oral cavity squamous cell carcinoma: an international
hyperfractionation and two variants of accelerated collaborative study. Cancer. 2013;119(24):4242–8.
fractionation to standard fractionation radiotherapy 35. Zhang H, et al. Survival outcomes of patients with
for head and neck squamous cell carcinomas: first advanced oral cavity squamous cell carcinoma
report of RTOG 9003. Int J Radiat Oncol Biol Phys. treated with multimodal therapy: a multi-institutional
2000;48(1):7–16. analysis. J Otolaryngol Head Neck Surg. 2013;42:30.
36. Iyer NG, et al. Randomized trial comparing sur-

cell carcinoma of the oral cavity. Int J Radiat Oncol
gery and adjuvant radiotherapy versus concurrent Biol Phys. 1997;39(1):137–48.
chemoradiotherapy in patients with advanced, non- 39. Hinerman RW, et al. Postoperative irradiation for

metastatic squamous cell carcinoma of the head and squamous cell carcinoma of the oral cavity: 35-year
neck: 10-year update and subset analysis. Cancer. experience. Head Neck. 2004;26(11):984–94.
2015;121(10):1599–607. 40. Bernier J, et al. Postoperative irradiation with or without
37. Daly ME, et al. Intensity-modulated radiotherapy for concomitant chemotherapy for locally advanced head
oral cavity squamous cell carcinoma: patterns of fail- and neck cancer. N Engl J Med. 2004;350(19):1945–52.
ure and predictors of local control. Int J Rad Oncol 41. Ooishi M, et al. Patterns of failure after postop-

Biol Phys. 2011;80(5):1412–22. erative intensity-modulated radiotherapy for locally
38. Parsons JT, et al. An analysis of factors influencing advanced and recurrent head and neck cancer. Jpn J
the outcome of postoperative irradiation for squamous Clin Oncol. 2016;46(10):919–27.
Novel and Emerging
Chemotherapeutic Agents in Head
9
and Neck Cancer
Arielle Elkrief, Nicholas Makhoul,
and Nathaniel Bouganim
9.1 Introduction head and neck cancer patients with metastatic dis-
ease include cytotoxic chemotherapeutic agents or
Head and neck cancers are a heterogeneous group molecularly targeted agents.
of malignancies, and their management requires This chapter will describe the novel and
a multidisciplinary approach including input emerging chemotherapeutic agents in head and
from medical oncologists, radiation oncologists, neck cancer. The role for immunotherapy will be
surgeons, dentists, specialized nursing care, outlined in a later chapter.
speech and language pathologists, physiothera-
pists, nutritionists, as well as psychologists [1].
Overall survival (OS) is improved when patients 9.2 Systemic Therapy
are treated at high-volume centers [2]. for Locoregionally Advanced
Patients with early stage disease (stage I or II) Disease
are treated with surgical resection or definitive
radiation therapy (RT) to the primary site. Locoregionally advanced squamous head and
Locoregionally advanced disease (stage III or IV) neck cancer is associated with high rates of local
is treated with a combined modality approach such recurrence of up to 50% [3–5] and rates of distant
as surgery and RT with or without chemotherapy metastases between 4% and 26% [6–8].
given the increased risk of local recurrence and Chemotherapy has therefore been integrated into
distant metastasis in this patient population. the multimodality treatment plans in an effort to
Patients with metastatic disease require systemic improve the rates of both locoregional and distant
therapy as well as best supportive care. Patient recurrence, as well as to reduce patient morbidity
prognosis is often poor with median survival related to surgery and radiation using a functional
between 6 and 12 months. Therapeutic options for organ preservation approach. These approaches
can be classified into induction chemotherapy
A. Elkrief · N. Bouganim (*) (neoadjuvant chemotherapy), concurrent chemo-
Department of Oncology, McGil University, radiotherapy, and sequential chemoradiotherapy
Montreal, QC, Canada (combined induction chemotherapy followed by
e-mail: arielle.elkrief@mail.mcgill.ca; nathaniel. concurrent chemoradiotherapy). Prior to initia-
bouganim@mcgill.ca
tion of a multimodality treatment regimen, indi-
N. Makhoul vidual patient characteristics such as age,
Department of Oral and Maxillofacial Surgery, McGil
University, Montreal, QC, Canada comorbidities, performance status, and support
e-mail: nicholas.makhoul@mcgill.ca system should be assessed.

https://doi.org/10.1007/978-3-030-30094-4_9
118 A. Elkrief et al.
Various prospective studies have validated the 9.4 Treatment Regimens

role for chemotherapy in this patient population. for Recurrent Metastatic
Although there was no survival benefit for single- Disease in Previously
agent induction chemotherapy in comparison to Untreated Patients
surgery or RT alone, it was found that there was
an OS benefit in patients receiving cisplatin plus The median survival in patients with metastatic
fluorouracil [9]. Concurrent chemotherapy sig- head and neck cancer is poor and approaches
nificantly improved OS in comparison to surgery 6–12 months depending on disease and individ-
or RT alone [9]. In contrast, the benefit for ual patient factors, such as performance status,
sequential chemoradiotherapy is still unclear presence of comorbidities, and disease-related
[10–13] with suggested benefit for high-risk factors. Systemic treatment options are chosen
patients with bulky N2b, N3 nodal status, or based on whether the patient has already received
those with T3 or T4 disease [12]. systemic treatment as part of organ preservation
strategy or if they have already received a first-
line agent for presence of systemic or recurrent
9.3 Chemotherapy Regimens disease. The role of traditional cytotoxic agents,
for Locoregionally Advanced targeted molecular agents, and checkpoint inhibi-
Disease tor therapy in the treatment of metastatic or recur-
rent head and neck cancer will be discussed in
For induction chemotherapy, a three-drug detail herein. A small subgroup of patients with
combination of cisplatin, fluorouracil, plus a good performance status who recur may be can-
taxane is most commonly used and is the didates for “salvage” therapy with curative intent,
approach of choice [3, 14, 15]. Important tox- but most patients require a palliative approach
icities include myelosuppression, febrile neu- using the regimens discussed in this chapter. It is
tropenia, stomatitis, dysphagia, nausea, and important to note that best supportive care is also
anorexia [16]. an important component to the management plan
For concurrent therapy in patients with good in all of these patients.
performance status, high-dose bolus cisplatin In otherwise healthy patients with a good per-
(100 mg/m2 on days 1, 22, 43) can be adminis- formance status and those with advanced disease
trated concurrently with RT [17]. Given the fre- who are not appropriate candidates for curative
quent onset of both acute and late-onset adverse therapy, combinations of platinum-based chemo-
events, other dosing regimens are sometimes therapy with fluorouracil or a taxane is the pre-
used. The most commonly associated toxicities ferred approach [22–25]. The data supporting
included hematological toxicities, stomatitis, this recommendation are discussed in detail
dysphagia, as well as nausea and vomiting, neu- below.
rotoxicity, and nephrotoxicity [18]. Although Cisplatin (100 mg/m2 intravenous on day 1)
not as effective as cisplatin in the treatment of and fluorouracil (1000 mg/m2/day continuous
locally advanced squamous head and neck can- infusion over 4 days), and in comparison to
cer [19], weekly carboplatin (AUC of 1.5–2) is single-agent cisplatin or methotrexate, this dou-
an appropriate choice for patients with renal dis- blet regimen was associated with higher response
ease or poor performance status [20]. rates across all studies, albeit no survival benefit
Myelosuppression is an important limitation; was shown. For example, the EORTC Head and
however, there is less neurotoxicity associated Neck Cancer Cooperative Group conducted a
with this treatment. Carboplatin in combination randomized controlled trial three arms (1) cispla-
with fluorouracil given concurrently with RT is tin, methotrexate, bleomycin, and vincristine
another approach [21]. (CABO), (2) combination cisplatin and fluoro-
9 Novel and Emerging Chemotherapeutic Agents in Head and Neck Cancer 119
uracil (CF), and (3) cisplatin alone in previously methotrexate. Cetuximab (discussed later in this
untreated head and neck metastatic squamous chapter) can be added to these regimens, and as
cell cancer. Both CABO and CF were superior in seen in the EXTREME trial, when added to platin-
terms of overall response rates with no difference fluorouracil, confers an OS and PFS improvement
in progression-free survival or overall survival. when compared to cisplatin- fluorouracil alone
The Southwest Oncology Group (SWOG) per- [26]. Best supportive care is also an important
formed a randomized controlled trial of (1) CF, component to the management plan.
(2) carboplatin plus fluorouracil, and (3) single- The role of immune checkpoint inhibition
agent methotrexate. Once again, both the combi- with pembrolizumab in the first-line setting is
nation regimens (CF and carboplatin plus discussed in the final section of this chapter.
fluorouracil) were associated with improved
response rates in comparison to methotrexate
alone with similar median survival times across 9.5 pidermal Growth Factor
E
all three groups. There was however increased Receptor (EGFR)-Targeted
incidence of adverse events in the combination Therapy
treatment groups. Further adding to the data sup-
porting a doublet treatment regimen approach, a EGFR is a member of the ErbB/Her group of
study by Jacobs et al. randomized patients to ligand-activated receptor tyrosine kinases (RTKs)
receive either cisplatin alone, fluorouracil alone, [27]. Through ligand-binding and activation of
or their combination and once again found various downstream pathways, these receptors
improved overall response rates with no signifi- promote cancer cell proliferation, migration,
cant difference in survival outcomes. Toxicities angiogenesis, and tumor resistance to chemother-
were more important in the combination treat- apy [28–32]. EGFR expression occurs in over
ment arm, with alopecia and myelosuppression 90% of squamous head and neck cancers, and
being the most important. One study by Gibson overexpression is associated with decreased sur-
et al. failed to show any statistically significant vival, resistance to radiotherapy, locoregional
difference in response rate or survival between recurrence, and increased rate of distant metasta-
the single-agent and combination arms, and tox- ses [27, 33].
icities were similar in both groups.
We have seen above the data for combination
therapy for fluorouracil and cisplatin or carbopla- 9.5.1 M
onoclonal Antibodies (mAb)
tin; however, both cisplatin and carboplatin have Against EGFR
been combined with a taxane regimen, either
paclitaxel or docetaxel. No statistically signifi- Monoclonal antibodies targeting EFGR and used
cant benefit in response rate or overall survival in the treatment of locoregionally advanced squa-
exists with this regimen; however, common gas- mous head and neck cancer include cetuximab,
trointestinal adverse events and lack of need for panitumumab, zalatumumab, and nimotuzumab.
prolonged infusion time make the taxane regi- Their mechanism of action is through direct inhi-
men more convenient. bition of ligand-receptor binding [27].
Cisplatin is sometimes replaced for carbopla-
tin in the taxane combination for more frail indi- 9.5.1.1 Cetuximab
viduals as the side effect profile is more favorable Cetuximab is a highly specific, human-murine
with less ototoxicity, kidney failure, vomiting, chimeric immunoglobulin G (IgG) mAb target-
and neuropathy; however, this has not been vali- ing EGFR [27]. As demonstrated in this land-
dated in phase III trials. mark randomized controlled trial by Bonner
Single-agent therapy is reserved for patients et al., when administered at a dose of 400 mg/m2
with poor performance status and options include 1 week prior to RT followed by 250 mg/m2
single-agent taxane, cisplatin, carboplatin, or weekly during high-dose RT in patients with
locally advanced squamous head and neck can- 9.5.1.3 Zalutumumab

cer, cetuximab was associated with improved OS Zalutumumab, another fully humanized IgG anti-
(49.0 months compared to 29.3 months HR0.74; EGFR, works in a similar fashion as cetuximab
p = 0.03) and locoregional control (24.4 months and zalutumumab. Similar to panitumumab,
compared to 14.9 months (HE 0.68; p = 0.05) zalutumumab has failed to show any benefit in
[34]), in comparison to high-dose radiation ther- the treatment of patients with squamous cell head
apy alone. Progression-free survival was also and neck cancer in multiple phase III randomized
improved in the combination treatment arm. controlled trials [42–44].
This improvement in outcome was particularly
important in patients 65 years of age or less with 9.5.1.4 Nimotuzumab
good performance status, albeit the study was Nimotuzumab, another fully humanized IgG
not powered to detect differences in subgroups. anti-EGFR is now being compared to the admin-
In this study, there were no statistically signifi- istration of cisplatin in phase III trials in the man-
cant difference in the incidence of grade 3 or agement of locally and regionally advanced
higher adverse events; however, patients treated nasopharyngeal carcinoma when administered
with cetuximab may have higher incidence of during radiotherapy following preoperative
serious radiation dermatitis and another rare, but chemotherapy.
important side effect is the occurrence of
cetuximab- induced infusion reaction, particu-
larly in the first cycle. Interstitial lung disease 9.5.2 T
yrosine Kinase Inhibitors
was also an important side effect [35]. Current (TKI) Against EGFR
available data did not show any benefit to the use
of concurrent cetuximab plus cisplatin with RT The intracellular domain of EGFR has important
and is therefore not currently indicated in the tyrosine kinase activity. TKIs serve to inhibit the
treatment of locally advanced squamous head activation and subsequent phosphorylation of
and neck cancer [36]. In the metastatic setting, a EGFR [27]. In contrast to EGFR mAb, the small
randomized, phase III clinical trial in patients nature of these molecules allow for good GI
with recurrent or metastatic squamous head and absorption and therefore are prescribed orally in
neck cancer the addition of cetuximab was com- a daily fashion [27]. At the time of writing of this
pared with cisplatin/carboplatin plus fluoroura- text book, TKIs are under review in several ran-
cil. Chemotherapy plus cetuximab was associated domized, controlled trials, and none of the TKIs
with prolonged OS, PFS, and response rates. The have been approved in the treatment for squa-
main toxicities associated with the addition of mous head and neck cancer.
cetuximab in this trial were severe hypomagne-
semia, rash, and sepsis [26]. 9.5.2.1 Gefitinib
In a randomized phase III trial, the addition of
9.5.1.2 Panitumumab gefitinib to docetaxel did not improve survival for
Panitumumab is a fully humanized IgG anti- patients with recurrent or metastatic head and
EGFR mAb, and like cetuximab, it inhibits EGFR neck cancer [45] despite a phase II trial showing
ligand-dependant activation. Multiple prospec- an overall response rate of 10.6% [46].
tive and randomized studies have failed to show
an overall survival benefit in adding panitu- 9.5.2.2 Erlotinib
mumab to concurrent regimens in head and neck Erlotinib, the second most common TKI was
cancer, and its use is also associated with combined with cisplatin and compared to cispla-
increased toxicity such as grade 3 rash or muco- tin alone in a phase II trial in which the cisplatin
sal inflammation [37–40]. There was however a was given concurrently with definitive RT. In this
survival benefit p16-negative patients in the met- study, there was no improvement in the response
astatic setting [41]. rate or survival [47].
9.5.2.3 Lapatinib The VEGFR-directed therapies currently being

Lapatinib, a dual TKI, selectively inhibits the acti- studied in clinical models in squamous head and
vation of EGFR as well as HER-2 [27]. A phase II neck cancer include bevacizumab, sorafenib,
trial compared the addition of lapatinib to stan- sunitinib, and vandetanib. Other VEGF inhibitors
dard chemoradiotherapy and showed promising that are currently under investigation for head
results for the complete response rate in patients and neck cancers include pazopanib, axitinib,
with locally advanced squamous head and neck nilotinib, and linifanib [54–57].
cancer [48]. At this time, no benefit was shown in
survival. In the metastatic setting, no objective
response rate was observed in a phase II trial [49]. 9.6.1 M
onoclonal Antibodies (mAb)
Against the VEGFR
9.5.2.4 Afatinib
Afatinib is an irreversible TKI and, similarly to 9.6.1.1 Bevacizumab
lapatinib, binds to the Erb2 receptor to inhibit Bevacizumab is an antiangiogenic mAb against
EGFR [27]. Preliminary results from a phase II VEGFR. A phase II study in patients with locally
trial in the metastatic setting showed that there is advanced squamous head and neck cancer com-
significant disease activity for afatinib, and that it pared the addition of bevacizumab to concurrent
may be comparable to cetuximab [50]. It is cur- radiation therapy with cetuximab and peme-
rently being studied in the locally advanced trexed. The addition of bevacizumab increased
setting. toxicity without improvement in efficacy or clini-
cal outcomes [58]. Another phase II trial in squa-
9.5.2.5 Dacomitinib mous head and neck patients with locally
Dacomitinib is an irreversible tyrosine kinase advanced disease studying the addition of bevaci-
inhibitor for both EGFR and HER2. Two phase II zumab to concurrent intensity-modulated RT
clinical trials in recurrent or metastatic head and with cetuximab and cisplatin was associated with
neck cancer demonstrated important clinical favorable clinical outcomes with the most com-
activity with the most common grade 3 adverse mon grade 3 adverse events being lymphopenia,
event being diarrhea [51, 52]. Exploratory analy- mucositis, and dysphagia [59]. In the metastatic
ses suggest that certain subgroups of patients setting, a phase II trial showed an overall response
with specific biomarkers may have improved rate of 30% with the addition of bevacizumab to
responses to dacomitinib, but these findings need pemetrexed with frequent (15%) bleeding
to be validated in phase III randomized control adverse events [60].
trials before their implementation into clinical
practice.
9.6.2 T
yrosine Kinase Inhibitors
(TKIs) Against VEGFR
9.6 ascular Endothelial Growth
V
Factor Receptor (VEGFR)- 9.6.2.1 Sorafenib
Directed Therapies Sorafenib is a multiple kinase inhibitor targeting
VEGFR, RAF, and platelet-derived growth factor
Vascular endothelial growth factor is an impor- receptor (PDGFR) [27]. To date, evidence has
tant cytokine for tumor angiogenesis, which is been conflicting with two phase II trials in the
essential for tumor growth and metastatic dis- recurrent or metastatic setting showing little clin-
semination [27]. Overexpression of VEGFR in ical activity [61, 62] and a more recent phase II
patients with squamous head and neck cancer is trial showing an overall response rate of 55%
associated with worse OS [53], making the with the combination of sorafenib with paclitaxel
VEGFR pathway an appealing therapeutic target. and carboplatin [63].
9.6.2.2 Sunitinib phase II trial of patients with platinum-resis-

Sunitinib, a second multiple kinase inhibitor tar- tant recurrent or metastatic head and neck
geting VEGFR, PDGFR, RET, and c-kit was eval- squamous cell carcinoma. This study showed
uated as palliative monotherapy in patients with encouraging response rates of 35% with
metastatic head and neck cancer [27]. Outcomes improved PFS and OS in comparison with sim-
were poor with a significant amount of grade 3–5 ilar patient cohorts [72].
hemorrhage [64]. A second study was closed after
interim analysis due to only one out of the 19
patients in the study showing partial response [65]. 9.9 Immune Checkpoint
Inhibitors
9.6.2.3 Vandetanib
Vandetanib has activity against EGFR, VEGFR, The advent of immune checkpoint inhibitors has
and RET [27]. Currently its use has only been revolutionized the therapeutic landscape in many
shown to be feasible in the phase I setting [66] solid tumors and is now emerging as an important
with preclinical data showing it may overcome therapeutic option in the treatment of metastatic
resistance to EGFR as well as RT [67]. head and neck cancer.
At the time of writing of this text, the data on
the use of immune checkpoint inhibitors in the
9.7 P13K/AKT/mTOR Pathway first-line setting have been presented in abstract
Inhibitors form only and therefore should be interpreted
with caution until regulatory authorities approve
An important therapeutic hurdle to the use of these agents in this setting. Nevertheless, the pre-
EGFR and VEGFR inhibition is resistance to liminary results are promising and merit
these molecules, either primarily or by prolonged discussion.
use [68]. Prolonged treatment with EGFR can In an open-label, phase III, randomized con-
induce initiation of feedback loops thereby acti- trolled study (NCT02358031), patients were ran-
vating the P13/AKT pathway which promotes domly assigned to single-agent pembrolizumab,
protein synthesis, cell survival, and tumor growth a PD-L1 inhibitor, versus pembrolizumab plus
[27]. The mTOR pathway is another important flourouracil/platinum combination, versus cetux-
pathway promoting tumor growth through regu- imab and a fluorouracil/platinum combination.
lation of cell proliferation, cell motility, and pro- Patients were stratified based on PD-L1 score
tein synthesis and has shown to be stimulated in which was evaluated using the combined positive
57–81% of patients with squamous head and score (CPS). Single-agent pembrolizumab
neck cancer [27]. Temsirolimus is an mTOR improved OS in comparison to the cetuximab and
inhibitor that was studied in a phase II trial in fluorouracil/platinum combination in patients
patients with cetuximab-resistant metastatic with a high CPS score (above 20). Overall sur-
squamous head and neck cancer. This study vival in the pembrolizumab arm was 14.9 months
showed a nonstatistically significant improve- compared to 10.7 months. Strangely, this OS
ment in response rate. Two other studies evaluat- benefit did not translate in an improvement in
ing the combination of temsirolimus with PFS or response rate. As expected, toxicity was
erlotinib and everolimus (a second mTOR inhibi- less in the single-agent pembrolizumab arm. A
tor) with cetuximab and cisplatin were termi- similar benefit in OS was seen in the pembroli-
nated early due to toxicity [69–71]. zumab and fluorouracil/cisplatin arm
(13.0 months compared to 10.7 months), and
once again no significant differences were seen in
9.8 Palbociclib response rates or PFS.
In the second-line setting, pembrolizumab
Palbociclib, a selective cyclin-dependant was recently approved in the second-line setting
kinase (CDK) 4/6 inhibitor was evaluated in a for patients with metastatic head and neck squa-
mous carcinoma. The KEYNOTE-040 random- and neck squamous cell carcinoma is the lung.
ized controlled, phase III trial of patients who Around 30% of patients with metastatic head and
had failed standard platinum-based chemother- neck squamous cell carcinoma who undergo pul-
apy was randomized to either pembrolizumab or monary metastatecomy experience long-term
standard of care with either methotrexate, survival. Poor prognostic factors in this approach
docetaxel, or cetuximab [73]. Crossover was include male sex, oral cavity lesions, lymph node
allowed at progression. There was a small but involvement, and incomplete resection [76–78].
non-negligible improvement in overall survival
in the pembrolizumab group of 8.4 months ver-
sus 6.9 months, and this benefit was most impor- 9.11 Drug Resistance
tant in those with PDL-1 expression greater than
50%. There were less grade 3 or higher adverse Despite significant improvements in the survival
events in the chemotherapy arm, but as expected rates and organ preservation seen in the treatment
a higher incidence of grade 1–2 immune-related of head and neck cancer care, significant chal-
adverse events, hypothyroidism being the most lenges still exist as many patients experience
common. As a result of these studies and two drug resistance. Sensitivity to chemotherapeutic
other studies showing favorable response rate, agents is associated with tumor heterogeneity,
pembrolizumab was approved by the FDA for the which is a result of patient factors (ethnic differ-
treatment of recurrent or metastatic head and ences, age, weight, gender), and genetic differ-
neck squamous cell carcinoma at a dose of ences in clonal tumor cells [79]. Mechanisms of
200 mg intravenous every 3 weeks. resistance will be discussed in this section of this
In a phase III randomized controlled study in chapter.
the second- and later-line settings in patients with Firstly, decreased concentration of antineo-
recurrent or metastatic head and neck squamous plastic agent within the tumor cells is an impor-
cell carcinoma, nivolumab, a PD-1 inhibitor dem- tant mechanism of resistance and is thought to
onstrated an overall survival benefit (7.7 months occur through a ATB-Binding Cassette (ABC)-
versus 5.1 months), and this was most important mediated mechanism [80]. The ABC plays an
for patients with PDL-1 status more than 1% [74]. important role in the transportation of antineo-
It is important to note that crossover was not plastic treatments outside of the cell and also
allowed in this study. Based on the results of this transports nutrients within the tumor cells thus
study, the FDA approved nivolumab in this setting, allowing for drug resistance.
at a dose of 240 mg intravenous every 2 weeks. Secondly, head and neck squamous carcinoma
Darvalumab, another PD-1 inhibitor, has cells are able to perform DNA repair, mediated by
shown clinical activity in a phase II study of base-excision repair (BER). For example, poly-
patients with recurrent or metastatic and previ- morphisms in genes encoding BERs have been
ously treated squamous cell head and neck can- described, such as ERCC1 (C8092A), which
cer [75]. The role for ipilimumab, a CTLA-4 plays a role in mRNA stability and DNA repara-
inhibitor, is currently undergoing investigation tion capability, and ERCC1 expression may be
(NCT02369874). associated with improved chemoradiation sensi-
tivity perhaps clinical outcome as well [81].
Thirdly, there is an increased capability of
9.10 Oligometastatic Disease tumor dissemination through a variety of mecha-
nisms. For example, tumors that show an overex-
In carefully selected patients with oligometa- pression of p53 are resistant to both chemo and
static disease (limited metastatic disease) good radiation therapy, and this has been associated
performance status and who are good candidates with increased tumor progression and decreased
for aggressive management, it may be reasonable survival rates [82]. There may also be increased
to consider metastasectomy. As seen above, one chemoresistance through matrix metalloprotein-
of the most common sites of metastasis for head ase (MMP) through a Fas/FasL-mediated fash-
ion, as some studies have indicated that 2. Wuthrick EJ, Zhang Q, Machtay M, Rosenthal DI,
Nguyen-Tan PF, Fortin A, et al. Institutional clinical
polymorphisms in MMPs are independently trial accrual volume and survival of patients with head
associated with increased chemotherapy resis- and neck cancer. J Clin Oncol. 2015;33(2):156–64.
tance [83]. 3. Posner MR, Hershock DM, Blajman CR, Mickiewicz
Lastly, inactivation of antineoplastic drugs E, Winquist E, Gorbounova V, et al. Cisplatin and
fluorouracil alone or with docetaxel in head and neck
within the tumor cells can occur, also contribut- cancer. N Engl J Med. 2007;357(17):1705–15.
ing to resistance. This may be particularly impor- 4. Brockstein B, Haraf DJ, Rademaker AW, Kies MS,
tant for EGFR-mediated resistance, as EGF Stenson KM, Rosen F, et al. Patterns of failure, prog-
expression may be critical for maintaining tumor nostic factors and survival in locoregionally advanced
head and neck cancer treated with concomitant chemo-
cell proliferation, and thus perhaps resistance to
radiotherapy: a 9-year, 337-patient, multi-institutional
cetuximab [84]. However, it is still not been experience. Ann Oncol. 2004;15(8):1179–86.
determined which of the EGFR ligands predict 5. Bourhis J, Le Maitre A, Baujat B, Audry H, Pignon
response to anti-EGFR treatment in patients with JP. Meta-analysis of chemotherapy in head NCCG,
et al. Individual patients’ data meta-analyses in head
squamous cell carcinoma head and neck cancer.
and neck cancer. Curr Opin Oncol. 2007;19(3):188–94.
Resistance may also be due to autocrine growth 6. Alvi A, Johnson JT. Development of distant metasta-
factor production [85]. More data are required in sis after treatment of advanced-stage head and neck
order to fully elucidate the mechanisms of resis- cancer. Head Neck. 1997;19(6):500–5.
7. Bhatia R, Bahadur S. Distant metastasis in malig-
tance to single-agent cetuximab, but this may be
nancies of the head and neck. J Laryngol Otol.
related to the capability of EGF to inhibit epithe- 1987;101(9):925–8.
lial differentiation, and this may be in a cancer 8. Garavello W, Ciardo A, Spreafico R, Gaini RM. Risk
stem cell-related fashion [86, 87]. factors for distant metastases in head and neck squa-
These findings prompt the need for continued mous cell carcinoma. Arch Otolaryngol Head Neck
Surg. 2006;132(7):762–6.
search for biomarkers for resistance, with the 9. Pignon JP, le Maitre A, Maillard E, Bourhis J, Group
goal of a personalized approach when prescrib- M-NC. Meta-analysis of chemotherapy in head and
ing therapy for head and neck squamous cell car- neck cancer (MACH-NC): an update on 93 ran-
cinoma patients. domised trials and 17,346 patients. Radiother Oncol.
2009;92(1):4–14.
10. Haddad R, O'Neill A, Rabinowits G, Tishler R, Khuri
F, Adkins D, et al. Induction chemotherapy followed
9.12 Conclusions and Future by concurrent chemoradiotherapy (sequential chemo-
Directions radiotherapy) versus concurrent chemoradiotherapy
alone in locally advanced head and neck cancer
(PARADIGM): a randomised phase 3 trial. Lancet
The treatment of head and neck cancer continues Oncol. 2013;14(3):257–64.
to be challenging due to its heterogeneous nature 11. Hitt R, Grau JJ, Lopez-Pousa A, Berrocal A, Garcia-
as well as its increased incidence of resistance to Giron C, Irigoyen A, et al. A randomized phase III
trial comparing induction chemotherapy followed by
conventional chemoradiation as well as targeted chemoradiotherapy versus chemoradiotherapy alone
therapy. The distinct responses of HPV-positive as treatment of unresectable head and neck cancer.
and HPV-negative patients require further study. Ann Oncol. 2014;25(1):216–25.
Advancements in elucidation of cancer cell biol- 12. Cohen EE, Karrison TG, Kocherginsky M, Mueller J,
Egan R, Huang CH, et al. Phase III randomized trial
ogy have allowed for the development of several of induction chemotherapy in patients with N2 or N3
targeted therapies; however, more phase III trials locally advanced head and neck cancer. J Clin Oncol.
need to be undertaken in order to implement 2014;32(25):2735–43.
these targeted therapies in daily practice. 13. Ghi MG, Paccagnella A, Ferrari D, Foa P, Alterio D,
Codeca C, et al. Induction TPF followed by concomi-
tant treatment versus concomitant treatment alone
in locally advanced head and neck cancer. A phase
References II-III trial. Ann Oncol. 2017;28(9):2206–12.
14. Lorch JH, Goloubeva O, Haddad RI, Cullen K, Sarlis
1. Wheless SA, McKinney KA, Zanation AM. A pro- N, Tishler R, et al. Induction chemotherapy with cis-
spective study of the clinical impact of a multidisci- platin and fluorouracil alone or in combination with
plinary head and neck tumor board. Otolaryngol Head docetaxel in locally advanced squamous-cell can-
Neck Surg. 2010;143(5):650–4. cer of the head and neck: long-term results of the
TAX 324 randomised phase 3 trial. Lancet Oncol. cell carcinoma of the head and neck. J Clin Oncol.
2011;12(2):153–9. 1992;10(2):257–63.
15. Vermorken JB, Remenar E, van Herpen C, Gorlia T, 25. Gibson MK, Li Y, Murphy B, Hussain MH, DeConti
Mesia R, Degardin M, et al. Cisplatin, fluorouracil, RC, Ensley J, et al. Randomized phase III evaluation of
and docetaxel in unresectable head and neck cancer. cisplatin plus fluorouracil versus cisplatin plus pacli-
N Engl J Med. 2007;357(17):1695–704. taxel in advanced head and neck cancer (E1395): an
16. Blanchard P, Bourhis J, Lacas B, Posner MR,
intergroup trial of the Eastern Cooperative Oncology
Vermorken JB, Cruz Hernandez JJ, et al. Taxane- Group. J Clin Oncol. 2005;23(15):3562–7.
cisplatin-fluorouracil as induction chemotherapy 26. Rivera F, Garcia-Castano A, Vega N, Vega-Villegas
in locally advanced head and neck cancers: an indi- ME, Gutierrez-Sanz L. Cetuximab in metastatic or
vidual patient data meta-analysis of the meta-analysis recurrent head and neck cancer: the EXTREME trial.
of chemotherapy in head and neck cancer group. J Expert Rev Anticancer Ther. 2009;9(10):1421–8.
Clin Oncol. 2013;31(23):2854–60. 27. Dorsey K, Agulnik M. Promising new molecular

17. Adelstein DJ, Li Y, Adams GL, Wagner H Jr, Kish JA, targeted therapies in head and neck cancer. Drugs.
Ensley JF, et al. An intergroup phase III comparison 2013;73(4):315–25.
of standard radiation therapy and two schedules of 28. Hynes NE, Horsch K, Olayioye MA, Badache A. The
concurrent chemoradiotherapy in patients with unre- ErbB receptor tyrosine family as signal integrators.
sectable squamous cell head and neck cancer. J Clin Endocr Relat Cancer. 2001;8(3):151–9.
Oncol. 2003;21(1):92–8. 29. Kruger JS, Reddy KB. Distinct mechanisms medi-
18. Forastiere AA, Goepfert H, Maor M, Pajak TF, Weber ate the initial and sustained phases of cell migration
R, Morrison W, et al. Concurrent chemotherapy and in epidermal growth factor receptor-overexpressing
radiotherapy for organ preservation in advanced laryn- cells. Mol Cancer Res. 2003;1(11):801–9.
geal cancer. N Engl J Med. 2003;349(22):2091–8. 30. Masuda M, Toh S, Koike K, Kuratomi Y, Suzui M,
19. Fountzilas G, Ciuleanu E, Dafni U, Plataniotis G, Deguchi A, et al. The roles of JNK1 and Stat3 in
Kalogera-Fountzila A, Samantas E, et al. Concomitant the response of head and neck cancer cell lines to
radiochemotherapy vs radiotherapy alone in patients combined treatment with all-trans-retinoic acid and
with head and neck cancer: a Hellenic Cooperative 5-fluorouracil. Jpn J Cancer Res. 2002;93(3):329–39.
Oncology Group Phase III Study. Med Oncol. 31. Vivanco I, Sawyers CL. The phosphatidylinositol

2004;21(2):95–107. 3-Kinase AKT pathway in human cancer. Nat Rev
20. Posner MR, Norris CM, Wirth LJ, Shin DM, Cullen Cancer. 2002;2(7):489–501.
KJ, Winquist EW, et al. Sequential therapy for the 32.
Ellerbroek SM, Halbleib JM, Benavidez M,
locally advanced larynx and hypopharynx cancer sub- Warmka JK, Wattenberg EV, Stack MS, et al.
group in TAX 324: survival, surgery, and organ pres- Phosphatidylinositol 3-kinase activity in epidermal
ervation. Ann Oncol. 2009;20(5):921–7. growth factor-stimulated matrix metalloproteinase-9
21. Bourhis J, Sire C, Graff P, Gregoire V, Maingon P, production and cell surface association. Cancer Res.
Calais G, et al. Concomitant chemoradiotherapy 2001;61(5):1855–61.
versus acceleration of radiotherapy with or with- 33. Agulnik M. New approaches to EGFR inhibition for
out concomitant chemotherapy in locally advanced locally advanced or metastatic squamous cell carci-
head and neck carcinoma (GORTEC 99-02): an noma of the head and neck (SCCHN). Med Oncol.
open-label phase 3 randomised trial. Lancet Oncol. 2012;29(4):2481–91.
2012;13(2):145–53. 34. Bonner JA, Harari PM, Giralt J, Azarnia N, Shin DM,
22. Clavel M, Vermorken JB, Cognetti F, Cappelaere P, de Cohen RB, et al. Radiotherapy plus cetuximab for
Mulder PH, Schornagel JH, et al. Randomized com- squamous-cell carcinoma of the head and neck. N
parison of cisplatin, methotrexate, bleomycin and vin- Engl J Med. 2006;354(6):567–78.
cristine (CABO) versus cisplatin and 5-fluorouracil 35. Bonner JA, Harari PM, Giralt J, Cohen RB, Jones
(CF) versus cisplatin (C) in recurrent or metastatic CU, Sur RK, et al. Radiotherapy plus cetuximab
squamous cell carcinoma of the head and neck. A for locoregionally advanced head and neck cancer:
phase III study of the EORTC Head and Neck Cancer 5-year survival data from a phase 3 randomised trial,
Cooperative Group. Ann Oncol. 1994;5(6):521–6. and relation between cetuximab-induced rash and sur-
23. Forastiere AA, Metch B, Schuller DE, Ensley JF,
vival. Lancet Oncol. 2010;11(1):21–8.
Hutchins LF, Triozzi P, et al. Randomized compari- 36. Ang KK, Zhang Q, Rosenthal DI, Nguyen-Tan PF,
son of cisplatin plus fluorouracil and carboplatin Sherman EJ, Weber RS, et al. Randomized phase
plus fluorouracil versus methotrexate in advanced III trial of concurrent accelerated radiation plus cis-
squamous-cell carcinoma of the head and neck: a platin with or without cetuximab for stage III to IV
Southwest Oncology Group study. J Clin Oncol. head and neck carcinoma: RTOG 0522. J Clin Oncol.
1992;10(8):1245–51. 2014;32(27):2940–50.
24. Jacobs C, Lyman G, Velez-Garcia E, Sridhar KS,
37. Saloura V, Vokes EE. EGFR-based bioradiotherapy in
Knight W, Hochster H, et al. A phase III randomized SCCHN. Lancet Oncol. 2015;16(2):129–30.
study comparing cisplatin and fluorouracil as single 38. Mesia R, Henke M, Fortin A, Minn H, Yunes Ancona
agents and in combination for advanced squamous AC, Cmelak A, et al. Chemoradiotherapy with or with-
out panitumumab in patients with unresected, locally with chemoradiotherapy (CRT) in patients (pts) with
advanced squamous-cell carcinoma of the head and locally advanced, unresected squamous cell carci-
neck (CONCERT-1): a randomised, controlled, open- noma of the head and neck (SCCHN). J Clin Oncol.
label phase 2 trial. Lancet Oncol. 2015;16(2):208–20. 2010;28(15 Suppl):5505.
39. Siu LL, Waldron JN, Chen BE, Winquist E, Wright 49. Abidoye OO, Cohen EE, Wong SJ, Kozloff MF,
JR, Nabid A, et al. Effect of standard radiotherapy Nattam SR, Stenson KM, et al. A phase II study
with cisplatin vs accelerated radiotherapy with pani- of lapatinib (GW572016) in recurrent/metastatic
tumumab in locoregionally advanced squamous cell (R/M) squamous cell carcinoma of the head
head and neck carcinoma: a randomized clinical trial. and neck (SCCHN). J Clin Oncol. 2006;24(18
JAMA Oncol. 2017;3(2):220–6. Suppl):5568.
40. Giralt J, Trigo J, Nuyts S, Ozsahin M, Skladowski 50. Seiwert TY, Fayette J, Cupissol D, del Campo JM,
K, Hatoum G, et al. Panitumumab plus radiotherapy Clement PM, Hitt R, et al. A randomized, phase II
versus chemoradiotherapy in patients with unre- study of afatinib versus cetuximab in metastatic or
sected, locally advanced squamous-cell carcinoma recurrent squamous cell carcinoma of the head and
of the head and neck (CONCERT-2): a randomised, neck†. Ann Oncol. 2014;25(9):1813–20.
controlled, open-label phase 2 trial. Lancet Oncol. 51. Abdul Razak AR, Soulières D, Laurie SA, Hotte

2015;16(2):221–32. SJ, Singh S, Winquist E, et al. A phase II trial of
41. Vermorken JB, Stohlmacher-Williams J, Davidenko I, dacomitinib, an oral pan-human EGF receptor (HER)
Licitra L, Winquist E, Villanueva C, et al. Cisplatin inhibitor, as first-line treatment in recurrent and/or
and fluorouracil with or without panitumumab in metastatic squamous-cell carcinoma of the head and
patients with recurrent or metastatic squamous-cell neck†. Ann Oncol. 2013;24(3):761–9.
carcinoma of the head and neck (SPECTRUM): an 52. Kim HS, Kwon HJ, Jung I, Yun MR, Ahn M-J, Kang
open-label phase 3 randomised trial. Lancet Oncol. BW, et al. Phase II clinical and exploratory biomarker
2013;14(8):697–710. study of dacomitinib in patients with recurrent and/or
42. Bastholt L, Specht L, Jensen K, Brun E, Loft A, metastatic squamous cell carcinoma of head and neck.
Petersen J, et al. Phase I/II clinical and pharmacoki- Clin Cancer Res. 2015;21(3):544–52.
netic study evaluating a fully human monoclonal anti- 53. Kyzas PA, Stefanou D, Batistatou A, Agnantis

body against EGFr (HuMax-EGFr) in patients with NJ. Prognostic significance of VEGF immunohisto-
advanced squamous cell carcinoma of the head and chemical expression and tumor angiogenesis in head
neck. Radiother Oncol. 2007;85(1):24–8. and neck squamous cell carcinoma. J Cancer Res Clin
43. Machiels JP, Subramanian S, Ruzsa A, Repassy
Oncol. 2005;131(9):624–30.
G, Lifirenko I, Flygare A, et al. Zalutumumab 54. Brands RC, Knierim LM, De Donno F, Steinacker V,
plus best supportive care versus best supportive Hartmann S, Seher A, et al. Targeting VEGFR and
care alone in patients with recurrent or metastatic FGFR in head and neck squamous cell carcinoma
squamous-cell carcinoma of the head and neck in vitro. Oncol Rep. 2017;38(3):1877–85.
after failure of platinum- based chemotherapy: an 55. Swiecicki PL, Zhao L, Belile E, Sacco AG, Chepeha
open-label, randomised phase 3 trial. Lancet Oncol. DB, Dobrosotskaya I, et al. A phase II study evaluat-
2011;12(4):333–43. ing axitinib in patients with unresectable, recurrent or
44. Schick U, Gujral DM, Richards TM, Harrington KJ, metastatic head and neck cancer. Investig New Drugs.
Nutting CM. Zalutumumab in head and neck cancer. 2015;33(6):1248–56.
Expert Opin Biol Ther. 2012;12(1):119–25. 56. Hsu HW, de Necochea-Campion R, Williams V,

45. Argiris A, Ghebremichael M, Gilbert J, Lee JW,
Duerksen-Hughes PJ, Simental AA Jr, Ferris RL, et al.
Sachidanandam K, Kolesar JM, et al. Phase III ran- Linifanib (ABT-869), enhances cytotoxicity with poly
domized, placebo-controlled trial of docetaxel with or (ADP-ribose) polymerase inhibitor, veliparib (ABT-
without gefitinib in recurrent or metastatic head and 888), in head and neck carcinoma cells. Oral Oncol.
neck cancer: an eastern cooperative oncology group 2014;50(7):662–9.
trial. J Clin Oncol. 2013;31(11):1405–14. 57. Kramer B, Hock C, Birk R, Sauter A, Stuck BA,
46. Cohen EE, Rosen F, Stadler WM, Recant W, Stenson Hormann K, et al. Targeted therapies in HPV-
K, Huo D, et al. Phase II trial of ZD1839 in recurrent positive and -negative HNSCC – alteration of EGFR
or metastatic squamous cell carcinoma of the head and VEGFR-2 expression in vitro. Anticancer Res.
and neck. J Clin Oncol. 2003;21(10):1980–7. 2016;36(6):2799–807.
47. Martins RG, Parvathaneni U, Bauman JE, Sharma 58. Argiris A, Bauman JE, Ohr J, Gooding WE, Heron
AK, Raez LE, Papagikos MA, et al. Cisplatin and DE, Duvvuri U, et al. Phase II randomized trial
radiotherapy with or without erlotinib in locally of radiation therapy, cetuximab, and pemetrexed
advanced squamous cell carcinoma of the head and with or without bevacizumab in patients with
neck: a randomized phase II trial. J Clin Oncol. locally advanced head and neck cancer. Ann Oncol.
2013;31(11):1415–21. 2016;27(8):1594–600.
48. Harrington KJ, Berrier A, Robinson M, Remenar E, 59. Fury MG, Xiao H, Sherman EJ, Baxi S, Smith-

Housset M, Mendoza FH, et al. Phase II study of oral Marrone S, Schupak K, et al. Phase II trial of beva-
lapatinib, a dual-tyrosine kinase inhibitor, combined cizumab + cetuximab + cisplatin with concurrent
intensity-modulated radiation therapy for patients in combination with cetuximab and cisplatin as first-
with stage III/IVB head and neck squamous cell car- line therapy in recurrent and metastatic (R/M) head
cinoma. Head Neck. 2016;38(Suppl 1):E566–70. and neck squamous cell carcinoma (HNSCC). J Clin
60. Argiris A, Karamouzis MV, Gooding WE, Branstetter Oncol. 2012;30(15 Suppl):e16061.
BF, Zhong S, Raez LE, et al. Phase II trial of peme- 72. Adkins D, Oppelt PJ, Ley JC, Trinkaus K, Neupane
trexed and bevacizumab in patients with recurrent PC, Sacco AG. Multicenter phase II trial of palbo-
or metastatic head and neck cancer. J Clin Oncol. ciclib, a selective cyclin dependent kinase (CDK)
2011;29(9):1140–5. 4/6 inhibitor, and cetuximab in platinum-resistant
61. Williamson SK, Moon J, Huang CH, Guaglianone HPV unrelated (−) recurrent/metastatic head and
PP, LeBlanc M, Wolf GT, et al. Phase II evaluation neck squamous cell carcinoma (RM HNSCC). J Clin
of sorafenib in advanced and metastatic squamous Oncol. 2018;36:6008.
cell carcinoma of the head and neck: Southwest 73. Cohen EEW, Soulieres D, Le Tourneau C, Dinis

Oncology Group Study S0420. J Clin Oncol. J, Licitra L, Ahn MJ, et al. Pembrolizumab versus
2010;28(20):3330–5. methotrexate, docetaxel, or cetuximab for recurrent or
62. Elser C, Siu LL, Winquist E, Agulnik M, Pond GR, metastatic head-and-neck squamous cell carcinoma
Chin SF, et al. Phase II trial of sorafenib in patients (KEYNOTE-040): a randomised, open-label, phase 3
with recurrent or metastatic squamous cell carcinoma study. Lancet. 2019;393(10167):156–67.
of the head and neck or nasopharyngeal carcinoma. J 74. Ferris RL, Blumenschein G Jr, Fayette J, Guigay J,
Clin Oncol. 2007;25(24):3766–73. Colevas AD, Licitra L, et al. Nivolumab for recurrent
63. Blumenschein GR, Glisson BS, Lu C, Sabichi AL, squamous-cell carcinoma of the head and neck. N
Ginsberg LE, Bartos CI, et al. Final results of a phase Engl J Med. 2016;375(19):1856–67.
II study of sorafenib in combination with carboplatin 75. Zandberg DP, Algazi AP, Jimeno A, Good JS, Fayette
and paclitaxel in patients with metastatic or recurrent J, Bouganim N, et al. Durvalumab for recurrent or
squamous cell cancer of the head and neck (SCCHN). metastatic head and neck squamous cell carcinoma:
J Clin Oncol. 2012;30(15 Suppl):5592. Results from a single-arm, phase II study in patients
64. Machiels JP, Henry S, Zanetta S, Kaminsky MC,
with >/=25% tumour cell PD-L1 expression who have
Michoux N, Rommel D, et al. Phase II study of suni- progressed on platinum-based chemotherapy. Eur J
tinib in recurrent or metastatic squamous cell carci- Cancer. 2019;107:142–52.
noma of the head and neck: GORTEC 2006-01. J Clin 76. Shiono S, Kawamura M, Sato T, Okumura S,

Oncol. 2010;28(1):21–8. Nakajima J, Yoshino I, et al. Pulmonary metas-
65. Choong NW, Kozloff M, Taber D, Hu HS, Wade J tasectomy for pulmonary metastases of head and
3rd, Ivy P, et al. Phase II study of sunitinib malate neck squamous cell carcinomas. Ann Thorac Surg.
in head and neck squamous cell carcinoma. Investig 2009;88(3):856–60.
New Drugs. 2010;28(5):677–83. 77. Liu D, Labow DM, Dang N, Martini N, Bains M, Burt
66. Papadimitrakopoulou VA, Frank SJ, Cohen EW,
M, et al. Pulmonary metastasectomy for head and
Hirsch FR, Myers JN, Heymach JV, et al. Phase I study neck cancers. Ann Surg Oncol. 1999;6(6):572–8.
of vandetanib with radiation therapy with or without 78. Mochizuki T, Okumura S, Ishii G, Ishikawa Y,

cisplatin in locally advanced head and neck squamous Hayashi R, Kawabata K, et al. Surgical resection for
cell carcinoma. Head Neck. 2016;38(3):439–47. oral tongue cancer pulmonary metastases. Interact
67. Sano D, Matsumoto F, Valdecanas DR, Zhao M,
Cardiovasc Thorac Surg. 2010;11(1):56–9.
Molkentine DP, Takahashi Y, et al. Vandetanib restores 79.
Lopez-Verdin S, Lavalle-Carrasco J, Carreon-
head and neck squamous cell carcinoma cells’ sensi- Burciaga RG, Serafin-Higuera N, Molina-Frechero
tivity to cisplatin and radiation in vivo and in vitro. N, Gonzalez-Gonzalez R, et al. Molecular mark-
Clin Cancer Res. 2011;17(7):1815–27. ers of anticancer drug resistance in head and neck
68. Zibelman M, Mehra R. Overview of current treatment squamous cell carcinoma: a literature review. Cancers.
options and investigational targeted therapies for 2018;10(10):pii: E376.
locally advanced squamous cell carcinoma of the head 80. Becker M, Levy D. Modeling the transfer of

and neck. Am J Clin Oncol. 2016;39(4):396–406. drug resistance in solid tumors. Bull Math Biol.
69. Chawla A, Adkins D, Worden FP, Rao KA, Hu HS, 2017;79(10):2394–412.
Price KAR, et al. Effect of the addition of temsirolimus 81. Quintela-Fandino M, Hitt R, Medina PP, Gamarra S,
to cetuximab in cetuximab-resistant head and neck Manso L, Cortes-Funes H, et al. DNA-repair gene
cancers: results of the randomized PII MAESTRO polymorphisms predict favorable clinical outcome
study. J Clin Oncol. 2014;32(15 Suppl):6089. among patients with advanced squamous cell car-
70. Bauman JE, Arias-Pulido H, Lee SJ, Fekrazad MH, cinoma of the head and neck treated with cisplatin-
Ozawa H, Fertig E, et al. A phase II study of temsi- based induction chemotherapy. J Clin Oncol.
rolimus and erlotinib in patients with recurrent and/or 2006;24(26):4333–9.
metastatic, platinum-refractory head and neck squa- 82. Cutilli T, Leocata P, Dolo V, Altobelli E. Evaluation
mous cell carcinoma. Oral Oncol. 2013;49(5):461–7. of p53 protein as a prognostic factor for oral
71. Chung CH, Wang H, Tsottles N, Gourin CG, Agrawal cancer surgery. Br J Oral Maxillofac Surg.
N, Molinolo A, et al. A phase I study of everolimus 2013;51(8):922–7.
83. Blons H, Gad S, Zinzindohoue F, Maniere I,

of EGFR ligands on endocytic sorting of the receptor.
Beauregard J, Tregouet D, et al. Matrix metallo- Traffic. 2009;10(8):1115–27.
proteinase 3 polymorphism: a predictive factor of 86. Setubal Destro Rodrigues MF, Gammon L, Rahman
response to neoadjuvant chemotherapy in head and MM, Biddle A, Nunes FD, Mackenzie IC. Effects of
neck squamous cell carcinoma. Clin Cancer Res. cetuximab and erlotinib on the behaviour of cancer
2004;10(8):2594–9. stem cells in head and neck squamous cell carcinoma.
84. Ansell A, Jedlinski A, Johansson AC, Roberg
Oncotarget. 2018;9(17):13488–500.
K. Epidermal growth factor is a potential biomarker 87. Gemenetzidis E, Gammon L, Biddle A, Emich H,
for poor cetuximab response in tongue cancer cells. J Mackenzie IC. Invasive oral cancer stem cells dis-
Oral Pathol Med. 2016;45(1):9–16. play resistance to ionising radiation. Oncotarget.
85. Roepstorff K, Grandal MV, Henriksen L, Knudsen 2015;6(41):43964–77.
SL, Lerdrup M, Grovdal L, et al. Differential effects
Immunotherapy in Oral Cancer:
A Fourth Dimension of Cancer
10
Treatment
Marcus A. Couey, Rom S. Leidner,
Simon W. Young, and R. Bryan Bell
10.1 T
he Immune System in Head The central role of the immune system in both
and Neck Cancer the prevention and evolution of cancer has been
described by Schreiber et al. in the concept of
Although many components of the immune sys- immunoediting [2]. Immunoediting starts with
tem are involved in the antitumor immune the surveillance of the body for abnormal cells by
response, the most critical immune cell is the T the immune system. In some cases, the immune
cell. The Cancer Immunity Cycle, as described system is successful in targeting and eliminating
by Chen and Mellman, explains the steps required a tumor and preventing the occurrence of clini-
for the adaptive immune system to target tumors, cally detectable cancer. However, sometimes the
including uptake and presentation of antigens by immune system cannot eliminate a subset of the
dendritic cells (DC), priming and activation of T tumor cells, as those cells have acquired pheno-
cells in the lymph nodes, homing of T cells to the types that subvert immune targeting. After a
site of the tumor, and T cell recognition and period of equilibrium, where these resistant
destruction of tumor cells [1] (Fig. 10.1). Cancer tumor cells exist in quiescence, they may begin to
cell death then exposes more potential antigens to proliferate and spread. This is called immune
immune recognition, which can perpetuate the escape and is now recognized as one of the hall-
cycle. Each step in this sequence has biochemical marks of cancer [3].
stimulators and inhibitors, and appreciation of Advances in whole-exome sequencing have
these steps is critical in developing strategies to provided great insights into the molecular land-
overcome immune evasion by tumors. scape of head and neck cancer. Remarkable
efforts by several groups, including The Cancer
Genome Atlas (TCGA), have identified the most
M. A. Couey · R. S. Leidner · R. B. Bell (*) commonly altered molecular pathways involved
Earle A. Chiles Research Institute in the Robert in head and neck squamous cell carcinoma
W. Franz Cancer Center, Providence Cancer Institute,
Portland, OR, USA (HNSCC). These data have shown that HNSCCs
e-mail: Marcus.Couey@providence.org; Rom. have a moderate-to-high mutational burden and
Leidner@providence.org; Richard.Bell@providence. are heterogeneous tumors. The most common
org involved pathways are tumor-suppressor genes
S. W. Young such as p53, which are either inactivated or
Department of Oral and Maxillofacial Surgery, mutated. Successful restoration of functional
University of Texas Health Science Center at
Houston, Houston, TX, USA tumor-suppressor pathways in HNSCC through
e-mail: Simon.Young@uth.tmc.edu targeted therapy has so far remained elusive.

https://doi.org/10.1007/978-3-030-30094-4_10
130 M. A. Couey et al.
Trafficking of
T cells to tumors
(CTLs)
4
Priming and activation

(APCs & T cells) 3
blood Infiltration of T cells

vessel 5 into tumors
(CTLs, endothelial cells)
lymph node
tumor
Cancer antigen 2
presentation
(dendritic cells/APCs) 6
Recognition of
cancer cells by T cells
(CTLs, cancer cells)
1 7
Release of
cancer cell antigens Killing of cancer cells
(cancer cell death) (Immune and cancer cells)
Fig. 10.1 Therapies that might affect the cancer immunity cycle. (From: Chen DS., Mellman I. “Oncology meets
immunology: the cancer immunity cycle.” Immunity 39.1 (2013): 1–10)
Oncogenic pathways which have been identified employ to suppress the natural antitumor capa-
in a significant percentage of tumors, such as bilities of immune cells. Impaired antitumor
PI3K, are theoretically more targetable than responses in HNSCC may be caused by altera-
tumor suppressors. However, results of therapies tions in the generation, processing, and/or pre-
such as PI3K and mTOR inhibitors have been sentation of T cell epitopes derived from
mostly disappointing [4]. The heterogeneity of tumor-associated antigens (TAA) by human leu-
cells within a tumor, even those tumors that test kocyte antigen (HLA) class I and/or class II mol-
positive for targetable mutations, may play an ecule [6]. Other mechanisms for evading immune
important role in the difficulties encountered targeting by tumors include upregulation of
with targeted therapies [5]. immune-suppressive “checkpoint” ligands (e.g.,
While the relatively high mutational rate of PD-L1) [7], release of inhibitory cytokines by
HNSCC creates difficulty in applying targeted suppressive immune cells including regulatory T
therapies, it may be advantageous for the applica- cells (Treg), myeloid-derived suppressor cells
tion of immunotherapy through the presence of (MDSC) and tumor-associated macrophages
more mutations that may be recognized as non- (TAM) in the tumor microenvironment [8–10], or
self by the immune system. However, there are the secretion of immune-suppressive mediators
numerous escape mechanisms that tumors (e.g., TGF-beta) [11]. Knowledge of the specific
10 Immunotherapy in Oral Cancer: A Fourth Dimension of Cancer Treatment 131
Augmentation of tumor-
Immune recognition Removal of physical
specific T-cells.
of tumor and mobilization and stochastic barriers
Reversion of tumor
of anti-tumor effectors to immune rejection.
immune evasion.
Vaccines Co-inhibitory blockade Tumor vascular resistance

Pharmacologic killing Co-stimulatory activation Demoplastic stroma
-Chemotheraphy Activation of APCs Extreme hypoproliferation
-Targeted therapies
Innate immune recognition Hypoxic microenvironments
Radiotherapy
ACT (CARs, TCR xfer)
Fig. 10.2 Strategies for immune checkpoint combinations: depicted. (Modified from: Ai, M, Curran MA. “Immune
different classes of therapeutic approaches that have syner- checkpoint combinations from mouse to man.” Cancer
gistic potential for future combination immunotherapies are Immunology, Immunotherapy 64.7 (2015): 885–892)
mechanisms of immune suppression in HNSCC 10.2 A

New Standard of Care
has led to numerous approaches to overcome in Recurrent or Metastatic
immune suppression in the tumor microenviron- Head and Neck Cancer
ment (Fig. 10.2).
Treatment and prognosis of HNSCC vary 10.2.1 PD-1 Inhibitors
depending on anatomical location. For oropha-
ryngeal SCC, carcinogenesis is often driven by One of the ways that tumors can halt the
infection with high-risk strains of HPV, usually immune response is through signaling mole-
HPV-16. It has been shown that the molecular cules called checkpoints. These inhibitory cell
pathways and immunobiology of HPV-positive receptors block T cell activity and lead to T cell
tumors are distinct from HPV-negative disease apoptosis. This is normally a safeguard against
[12]. In addition, HPV-positive HNSCCs autoimmunity. However tumors can upregulate
respond better to standard therapies and most these inhibitory receptors, which promote eva-
immunotherapies [13]. The significance of sion of immune targeting. Monoclonal antibod-
HPV in oral cancer specifically remains uncer- ies that block the interaction between these
tain, and there is currently no standard method receptors and their ligands, known as check-
for identifying HPV as the etiologic agent in point inhibitors (CPI), have shown great prom-
oral cavity SCC that is both accepted and avail- ise for many types of solid tumors in recent
able universally [14]. This chapter is meant to years (Fig. 10.3).
focus on cancers arising in the oral cavity and The most well-studied checkpoint inhibitors
therefore will limit the discussion to HPV- in HNSCC are the antiprogrammed cell death
negative HNSCC. protein 1 (anti-PD-1) antibodies nivolumab and
Current therapies and those in development pembrolizumab. Through interaction with its
are focused on strategic targeting of the various ligands programmed death ligand 1 (PD-L1) and
steps in the Cancer Immunity Cycle. The great 2 (PD-L2), PD-1 acts as an immune rheostat that
promise of immunotherapy lies with the potential modulates the immune response within the tumor
for lasting, or durable, responses. This is referred [1]. The blockade of PD-1 has been shown to pre-
to as the “tail at the end of the survival curve” vent the inhibition of T cell activity in the tumor
(Fig. 10.5) and is perhaps the most exciting pros- microenvironment, thereby permitting cellular
pect of immunotherapy. cytotoxicity against tumor cells.
Combination
First in-human use Pembrolizumab lpilimumab
Coley first to use IL-2 approved for IL-2 approved of CD19 CAR T- and Nivolumab and Nivolumab Atezolizumab Atezolizumab
bacterial toxins as treatment in for treatment cells in acute approved for approved for Pembrolizumab approved for approved for
immune therapy in metastatic renal in metastatic lymphoblastic metastatic metastatic approved for metastatic metastatic
cancer cell cancer melanoma lymphoma melanoma melanoma metastatic RCC bladder cancer NSCLC
1890 1988 1992 1995 1998 2006 2010 2011 2014 2015 2016
First in-human Interferon First in- lpilimumab Pembrolizumab lpilimumab Nivolumab Pembrolizumab Pembrolizumab
ACT leading to approved for human use of approved and Nivolumab approved approved for and Nivolumab approved for 1st
response in cancer adjuvant modified TCR for use in approved for 2nd for adjuvant refractory approved for line in metastatic
treatment in in melanoma metastatic line in metastatic therapy in Hodgkin metastatic H&N NSCLC
melanoma melanoma NSCLC melanoma lymphoma cancer
- Cytokine therapy - Anti PD1/L1 therapy - Combination therapy
- Adoptive cell therapy - Anti CTLA4 therapy
Fig. 10.3 Timeline of breakthroughs in cancer immuno- J. Moslehi. “Cardiovascular Toxicities Associated with
therapy. (From: Wang, Daniel Y., Gosife Donald Okoye, Cancer Immunotherapies.” Current Cardiology Reports
Thomas G. Neilan, Douglas B. Johnson, and Javid 19:21. 2017)
In late 2016, the FDA granted approval of body, cetuximab. ORR was 16% with a median
nivolumab, following the results from the ran- duration of response of 8 months [19]. A survival
domized phase III trial, CheckMate141 [15]. This benefit with pembrolizumab in platinum-refrac-
study evaluated nivolumab versus investigator’s tory R/M disease was later confirmed in
choice (IC) chemotherapy in patients with recur- KEYNOTE-040. In this randomized, controlled
rent or metastatic (R/M) HNSCC that had pro- phase III trial, patients receiving pembrolizumab
gressed within 6 months of platinum-based had a median OS of 8.4 months, vs 6.9 months
chemotherapy. This was the first randomized, for IC chemotherapy. Patients with >50% of
controlled data showing a survival benefit in R/M tumor cells expressing PD-L1 experienced
HNSCC since the EXTREME regimen, consist- increased benefit with pembrolizumab (median
ing of the combination of platinum, 5-fluorouracil OS 11.6 months) [20].
and cetuximab [16]. The study was ended early Recently, results were announced from the
after meeting its primary endpoint of overall sur- first phase III trial of anti-PD-1 therapy in first-
vival (OS), and patients from the control arm were line R/M HNSCC. KEYNOTE-048 randomized
then allowed to crossover to receive nivolumab. A patients to one of three arms: pembrolizumab
recent update reported 2-year OS of 16.9% in monotherapy, pembrolizumab plus platinum and
nivolumab-treated patients, which was more than 5-FU, or cetuximab plus platinum and 5-FU
double the rate for patients that received standard (EXTREME regimen). This trial showed that in
therapy (6.0%) [17]. Significant OS benefit was PD-L1-expressing patients (~85% of patients in
seen regardless of HPV or PD-L1 status. this population), pembrolizumab monotherapy
In mid-2016, the FDA granted accelerated was superior to the EXTREME regimen, with
approval of pembrolizumab for platinum- 2-year OS of 30.2% vs 18.6%. Further, when
refractory R/M HNSCC based on results from looking at the total population (regardless of
the phase Ib KEYNOTE-012 trial [18]. This trial PD-L1 status), patients who received pembroli-
studied pembrolizumab in patients with R/M dis- zumab with cisplatin and 5-FU had 29% 2-year
ease whose tumors had progressed within survival compared with 18.7% with the
6 months of platinum-based cytotoxic chemo- EXTREME regimen. In June 2019, the FDA
therapy. The objective response rate (ORR) for approved the use of pembrolizumab with or with-
pembrolizumab was 18%, with 71% of responses out cisplatin/5-FU (determined by PD-L1 status)
lasting 12 months or more. The KEYNOTE-055 based on data from this trial [21].
trial studied pembrolizumab in R/M HNSCC Importantly, the anti-PD-1 agents are much
who had progressed on platinum and the anti- better tolerated than cytotoxic chemotherapy,
epidermal growth factor receptor (EGFR) anti- with grade 3–5 drug-related adverse events
occurring in 13–14% of patients receiving anti- 10.3 Additional Checkpoint

PD-1 versus 36% with standard of care in phase Inhibitors
III, randomized trials. The most common adverse
events associated with these agents are fatigue, 10.3.1 PD-L1
nausea, rash, decreased appetite, and pruritus.
Hypothyroidism occurs in 7.7–13% of patients, Blockade of checkpoints can be achieved through
vs. about 1% with standard treatment. targeting either the receptor (e.g., PD-1), or the
Pneumonitis is the most severe adverse event ligand. Of the two ligands for PD-1, PD-L1 is
associated with anti-PD-1 and can be life- generally thought to play a more prominent role
threatening if not recognized early and treated. in immunosuppression than PD-L2 [22]. In addi-
Numerous physical and social quality of life tion, PD-L1’s binding is not exclusive to the
(QOL) measures were assessed in CheckMate141 PD-1 receptor; it can also block the immune-
which showed advantages of PD-1 inhibitors stimulating ligand B7–1, causing immune sup-
across the board compared with standard of care pression by a separate mechanism (Fig. 10.4)
chemotherapy. While pembrolizumab and [23]. Furthermore, targeting PD-L1 specifically
nivolumab have been the most highly studied may reduce the risk of pneumonitis compared
anti-PD-1 therapies, there are other anti-PD-1 with anti-PD-1 therapy, as pneumonitis is thought
agents currently in development. Spartalizumab to be at least partially mediated by PD-L2 [24].
(PDR001) and cemiplimab (REGN2810) are Therefore, the blockade of PD-L1 is expected to
being tested alone and in combination with other confer a different treatment response and adverse-
immunotherapies in multiple solid tumor types effect profile than anti-PD-1 therapy. Several
including HNSCC. anti-PD-L1 antibodies are currently being stud-
Cancer immunotherapy
with mAb to PD-1
T cell
PD-L1 T cell
B7-1
TCR
APC
PD-1 PD-L2 MΦ
RGMb
PD-L1
PD-L1
PD-L2
PD-L2
PD-1
Treg
immune cell
MHC-Ag Ag
PD-L1
PD-L2
Cancer Cancer immunotherapy
PD-L1
with mAb to PD-L1 or PD-L2
PD-L2
Fig. 10.4 Human cancer immunotherapy with anti-PD-1 the PD-1 and PD-L1 pathway.” Trends in molecular medi-
and anti-PD-L1/L2 antibodies. (From: Ohaegbulam, KC., cine 21.1 (2015): 24–33)
et al. “Human cancer immunotherapy with antibodies to
ied in HNSCC, including durvalumab, avelumab, Cytotoxic T-lymphocyte-Associated Protein 4

and atezolizumab. (CTLA-4). CTLA-4 is a receptor expressed on
Recently, the results were announced from the the surface of T cells that binds with B7 ligands
phase II HAWK trial of durvalumab in platinum- on antigen-presenting cells (APC), causing T
refractory R/M HNSCC with high PD-L1 expres- cell anergy and apoptosis. While PD-1 inhibits
sion, defined as greater than 25% tumor cell effector cells within the tumor microenviron-
expression (NCT02207530). ORR was 16%, and ment (TME), CTLA-4 inhibits T cell activation
55% of responses were ongoing at data cutoff. within the periphery, largely in the lymph nodes
Overall survival at 1 year was 33.6%. There was [26]. Blockade of CTLA-4 prevents immune-
a striking difference in response rate based on suppressive signaling, while also freeing the B7
HPV status, with an ORR of 29.4% in HPV- ligands to bind the costimulatory receptor
positive disease and 10.8% in HPV-negative dis- CD28. Additionally, antibodies targeting
ease. The incidence of grade 3 or higher adverse CTLA-4 cause Treg depletion through antibody-
events was relatively low at 8%. One patient dis- dependent cell-mediated cytotoxicity (ADCC),
continued therapy because of a treatment-related thereby reducing immunosuppressive influences
adverse event, and there were no treatment- in the TME [27].
related deaths [25]. There are a large number of Ipilimumab, a fully humanized IgG1 anti-
clinical trials studying durvalumab in combina- CTLA-4 antibody, was the first checkpoint inhib-
tion with other immune-modulating agents, as itor to demonstrate improved survival in
well as chemotherapy and radiation, which will melanoma and subsequently became the first
be discussed later in the chapter. drug in this class to gain FDA approval [28].
Ipilimumab has since shown activity in numerous
solid tumor types; however, immune-related
10.3.2 CTLA-4 adverse events are more common and more
severe than with the anti-PD-1 agents.
In addition to PD-1/PD-L1, there are numerous Tremelimumab is fully humanized IgG2 anti-
other immune receptors that act as checkpoints CTLA- 4 antibody, which is being studied in
through inhibition of immune cells (Fig. 10.5). many clinical trials across numerous cancer types
The first immune checkpoint to be identified including HNSCC, mostly in combination with
and targeted for cancer immunotherapy was durvalumab (anti-PD-L1) (Table 10.1).
Fig. 10.5 T cell

costimulatory and CD28 PD-L1
coinhibitory receptors. ICOS CTLA-4
Shown are the families 4-1BB T-cell PD-1
of T cell costimulatory T-cell OX40 BTLA T-cell
and immune checkpoint Co-Stimulatory TIM-3 Co-Inhibitory
GITR
receptors as well as Receptors Receptors
CD27 VISTA
those which affect
TWEAKR LAG-3
dendritic cells
responsible for T cell HVEM TIGIT
activation. (From: Ai, M, TIM-1
Curran MA. “Immune
checkpoint combinations
from mouse to man.”
Cancer Immunology, CD40
Immunotherapy 64.7 Dendritic Cell
(2015): 885–892) Co-Stimulatory APC
Receptors
4-1BB
Table 10.1 Clinical trials: other checkpoint inhibitors

Phase Tumor type Combination Comparison Trial Expected completion date
CTLA-4
Ipilimumab III R/M HNSCC Nivolumab Ipi placebo NCT02823574 08/2020
III R/M HNSCC Nrvolumab Plat/5FU/Cetux NCT02741570 08/2020
Tremelimumab II R/M HNSCC Durvalumab Mono vs combo NCT02319044 09/2018
III R/M HNSCC Durvalumab SOC chemo NCT02369874 11/2018
III R/M HNSCC Durvalumab SOC chemo NCT02551159 12/2018
TIM-3
TSR-022 I Advanced solid tumors Anti-PDl NCT02817633 06/2020
LY3321367 I Advanced LY3300054 NCT03099109 06/2020
MBG453 I/II Advanced malignancies PDR001 NCT02608268 03/2019
LAG-3
Relatlimab I Advanced solid tumors Nivolumab NCT02966548 07/2020
I/IIa Advanced solid tumors Nivolumab NCT01968109 12/2020
TSR-033 I Advanced solid tumors Anti-PDl NCT03250832 05/2021
REGN3767 I Advanced malignancies REGN2810 NCT03005782 03/2022
LAG525 Ι/II Advanced malignancies PDR001 NCT02460224 08/2019
TIGIT
BMS-986207 I/IIa Advanced solid tumors Nivolumab NCT02913313 12/2022
OMP-313M32 I LA/M solid tumors Nivolumab NCT03119428 10/2019
MTIG7192A I LA/M tumors Atezolizumab NCT02794571 09/2019
LA locally advanced, R recurrent, M metastatic, Ipi ipilimumab, plat platinum, cetux cetuximab, mono monotherapy,
SOC standard of care
10.3.3 TIM-3 10.3.4 LAG-3
T cell Immunoglobulin and Mucin Domain 3 The immune checkpoint Lymphocyte Activation
(TIM-3) is another immune checkpoint expressed Gene-3 (LAG-3) has been shown to suppress
on the surface of T cells. High TIM-3 is a marker responses of CD8+ cytotoxic T cells and NK cells
of T cell exhaustion, similar to PD-1, and has and to promote the suppressive influence of
been shown in preclinical and clinical studies to Tregs. LAG-3 is co-expressed with PD-1 on dys-
be upregulated in cases of progressive disease functional or exhausted T cells, and anti-
after anti-PD-1 therapy [29]. Preclinical models LAG-3 in preclinical studies has demonstrated
have also shown increased cytokine production synergy with anti-PD-1/PD-L1 to improve anti-
and activity of cytotoxic T cells with blockade of tumor immune responses [31]. There are at least
TIM-3 and PD-1 pathways in combination com- four monoclonal antibodies being evaluated in
pared with PD-1 pathway blockade alone [30]. phase I–II clinical trials for advanced solid
Therefore, there is sound rationale for studying tumors including HNSCC (Table 10.1).
TIM-3 in combination with therapies targeting
the PD-1/PD-L1 pathway. Additionally, blockade
of TIM-3 has been shown in preclinical models to 10.3.5 TIGIT
promote immune responses and reduce suppres-
sive forces via multiple targets aside from CD8+ T cell Immunoglobulin and ITIM Doman
T cells, including CD4+ T cells, natural killer (TIGIT) is another immune checkpoint that
(NK) cells, Tregs, MDSCs, and DCs. At least dampens the immune response through interac-
three monoclonal antibodies are currently in tions with multiple cell types, including effector
phase I-II trials for advanced solid malignancies T and NK cells, DC cells, and suppressive Tregs.
(Table 10.1). The combined blockade of TIGIT and PD-L1
synergistically promotes CD8+ T cell effector line therapy compared to chemotherapy

function within tumors [32]. There are at least (NCT02823574, NCT02741570). This combina-
three anti-TIGIT antibodies currently being eval- tion has been approved for the treatment of
uated in phase I–II clinical trials for advanced patients with advanced melanoma and advanced
solid tumors (Table 10.1). renal cell carcinoma. Combination
nivolumab + ipilimumab has also shown signifi-
cant promise in the treatment of patients with
10.4 Combinations of Checkpoint advanced non-small-cell lung cancer [33]. Data
Inhibitors from these trials are not yet mature and will be
presented at future meetings.
There is abundant preclinical evidence support- Tremelimumab (anti-CTLA-4) is now being
ing a combinatorial approach to cancer immuno- tested in numerous clinical trials for HNSCC in
therapy. For example, CTLA-4 acts relatively combination with durvalumab (anti-PD-L1). The
early in the cancer immunity cycle during T cell CONDOR trial was a phase II, randomized trial of
priming and activation, while PD-1 comes into durvalumab and tremelimumab alone and in com-
play later in the cycle by modulating immune bination in patients with platinum-refractory R/M
effector cell function within tumors. Additionally, HNSCC who had low PD-L1 levels. Median over-
blockade of one immune checkpoint can lead to all survival was 7.6 months in the combination
the increased expression of other checkpoints by arm, 6.0 months for durvalumab alone, and
tumor cells leading to immune escape [29]. 5.5 months for tremelimumab alone [34]. EAGLE
Therefore, it has been hypothesized that adding (NCT02369874) and KESTREL (NCT02551159)
one or more additional CPI to PD-1/PD-L1 ther- are randomized phase III trials studying this com-
apy may improve response rates, particularly in bination in platinum-refractory disease and as
PD-L1-negative patients, and prevent resistance first-line treatment of R/M HNSCC, respectively.
to single-agent therapy (Fig. 10.6). There is a large body of preclinical evidence for
Two large studies are currently testing the antitumor activity of other combinations of
nivolumab (anti-PD-1) alone or in combination C. In fact, many of the newer CPI are being devel-
with ipilimumab (anti-CTLA-4) in R/M oped as combinations from the start, most com-
HNSCC. CheckMate714 is a double-blinded, monly in combination with PD-1/PD-L1 pathway
randomized, phase II study in both the platinum- antagonists. For example, all the current anti-
refractory and first-line settings, and TIM-3, anti-TIGIT, and anti-LAG-3 trials evaluat-
CheckMate651 is an open-label, randomized ing tolerability and efficacy in advanced solid
phase III study of the same combination as first- tumors are testing these agents in combination with
100
Chemotherapy
Targeted or Vaccine therapy
80
Percent survival
Checkpoint Blockade
60 Checkpoint Combinations
40
20
0
Time
Fig. 10.6 Strategies for immune checkpoint combina- approaches. (Modified from: Ai, M, Curran MA. “Immune
tions: Shown is the current goal of the field of immuno- checkpoint combinations from mouse to man.” Cancer
therapy to increase the percentage of patients experiencing Immunology, Immunotherapy 64.7 (2015): 885–892)
durable, complete responses through combination therapy
anti-PD-1 or anti-PD-L1 antibodies. While clinical shown effector T cell -mediated antitumor
trials for these other combinations are still in early immune responses and Treg depletion with anti-
phases for the most part, there are early reports of bodies from subclasses capable of ADCC, as well
success such as the phase I/IIa trial of relatlimab as synergy with both CTLA-4 and PD-1 blocking
(anti-LAG-3) with nivolumab (anti-PD-1) in agents. There are currently two antibodies in
advanced solid tumors. In an expansion cohort of clinical trials in advanced solid tumors
melanoma patients who were refractory to PD-1 (Table 10.2). JTX-2011, a humanized IgG1
therapy alone, 11.5% had objective responses and monoclonal ICOS agonist antibody, is the fur-
37.7% had stable disease with the combination. thest along in development. It was found to be
Patients with ≥1% LAG-3 expression had an ORR well tolerated in phase I trials and is currently
of 18%, and patients who also had prior exposure undergoing phase II testing in several tumor
to CTLA-4 had an ORR of 24% [35]. This trial types including HNSCC [39].
includes head and neck patients as well, and study
completion is expected in late 2019.
10.5.2 Tumor Necrosis Factor (TNF)
Receptor Superfamily
10.5 Costimulatory Agonists
The TNF superfamily of receptors (TNFRs) are
In addition to T cell receptor (TCR) recognition involved in immune cell activation, proliferation,
of MHC-presented antigens, the activation of T and survival. There are several members of this
cells requires specific costimulatory signals group that are being targeted with agonist pharma-
(Fig. 10.5). TCR ligation without the second ceuticals in cancer immunotherapy clinical trials,
costimulation signal leads to T cell anergy and including OX40, 4-1BB, CD27, glucocorticoid-
immune tolerance [36]. The development of ago- induced TNFR-related protein (GITR), and CD40.
nist antibodies that activate costimulatory recep- OX40 is transiently expressed on CD4+ and
tors has added a new dimension to cancer CD8+ T cells and Tregs following TCR ligation.
immunotherapy. While the activity of checkpoint OX40 is also expressed on NK cells, NKT cells,
inhibitors has been described as “releasing the and neutrophils, and its ligand OX40L is tran-
breaks” on the immune system, costimulatory siently expressed on APCs and some T cells.
agonists have been described as “stepping on the OX40 appears to be important for T cell survival
gas.” Preclinical studies have shown that costim- and expansion, and for differentiation of T cells,
ulatory agonists have synergistic activity with skewing toward an effector phenotype [40].
CPI, and there are currently many such agents OX40 has shown synergistic activity with check-
being tested in clinical trials alone and in combi- point blockade in preclinical studies [41].
nation with other immunotherapies. Furthermore, OX40 is highly expressed in tumor-
infiltrating lymphocytes (TILs) in HNSCC, par-
ticularly CD4+ T cells, providing strong rationale
10.5.1 CD28 Superfamily for testing in clinical trials in this disease [42,
43]. There are six OX40-targeting antibodies in
Inducible T cell costimulator (ICOS) is a member pharmaceutical pipelines that are currently in
of the CD28 superfamily that promotes CD4+ T clinical trials for advanced solid tumors, includ-
cell growth, differentiation and effector function, ing HNSCC (Table 10.2). Interestingly, one such
as well as survival and memory of both CD4+ and agent was recently tested in a phase Ib neoadju-
CD8+ T cells [37]. ICOS is only expressed at low vant trial prior to surgery: MEDI6469 was admin-
levels on naïve T cells but is rapidly upregulated istered intravenously at various time intervals
upon TCR ligation [38]. It is also highly expressed prior to definitive surgical resection in 17 patients
on Tregs, and ICOS signaling can therefore con- with stage II–IVA HNSCC [44]. Fifty percent of
tribute to immune suppression, contrary to its the patients treated experienced an increase in the
action on effector T cells. Preclinical studies have percentage of tumor-reactive CD8+ T cells in the
Table 10.2 Clinical trials: costimulatory agonists

Expected
Phase Tumor type Combination Trial completion date
ICOS
GSK3359609 I Advanced solid Pembrolizumab NCT02723955 05/2020
tumors
JTX-2011 Ι/II Advanced Ipilimumab or nivolumab or NCT02904226 12/2022
malignancies pembrolizumab
OX40
MEDI0562 I Advanced solid None NCT02318394 Completed,
tumors pending
publication
I Advanced solid Durvalumab or tremelimumab NCT02705482 12/2019
tumors
Ib HNSCC or None NCT03336606 12/2024
melanoma
(neoadjuvant)
PF-4518600 I LA/M cancers Utomilumab NCT02315066 01/2021
Ib/II Advanced solid Avelumab +/− utomilumab NCT02554812 05/2020
tumors
MOXR0916 Ib LA/M solid tumors Atezolizumab NCT02410512 08/2018
GSK3174998 I Advanced solid Pembrolizumab NCT02528357 01/2020
tumors
BMS-986178 I/IIa Advanced solid Nivolumab and/or ipilimumab NCT02737475 10/2021
tumors
INCAGN1949 I A/M solid tumors None NCT02923349 02/2019
4-1BB
Urelumab III A/M solid tumors/ Nivolumab NCT02253992 09/2019
NHL
I Malignant tumors Nivolumab NCT02534506 06/2019
I Metastases in Nivolumab + SBRT NCT03431948 02 2020
advanced solid
tumors
Utomilumab I LA/M cancers PF-4518600 NCT02315066 01/2021
Ib/II Advanced solid Avelumab +/− PF-4518600 NCT02554812 05 2020
tumors
GITR
TRX518 I Melanoma and None NCT01239134 12/2018
other solid tumors
I Advanced solid Pembrolizumab or nivolumab or NCT02628574 09/2020
tumors gemcitabine
GWN323 I Advanced solid PDR001 NCT02740270 12/2019
tumors
MK-4166 I Advanced solid Pembrolizumab NCT02132754 10/2019
tumors
MK-1248 I Advanced solid Pembrolizumab NCT02553499 10/2018
tumors
MEDI1873 I Advanced solid None NCT02583165 01/2021
tumors
INCAGN01876 I/II Advanced None NCT02697591 08/2018
malignancies
BMS-986156 I/IIa Advanced solid Nivolumab NCT02598960 05/2020
tumors
CD27
Varlilumab Ι/Π Advanced Nivolumab NCT02335918 01/2020
refractory solid
tumors
Table 10.2 (continued)
Expected
Phase Tumor type Combination Trial completion date
CD40
SEA-CD40 I Advanced Pembrolizumab NCT02376699 09/2022
malignancies
CDX-1140 I Advanced solid None NCT03329950 12/2020
tumors
Selicrelumab Ib LA/M solid tumors Atezolizumab NCT02304393 10/2019
Ib Advanced solid Emactuzumab NCT02760797 Completed,
tumors pending
publication
APX005M I Solid tumors None NCT02482168 12/2018
ABBV-927 I Advanced solid ABBV-181 NCT02988960 11/2019
tumors
tumor after anti-OX40 treatment. Early-phase cacy in several cancer models [46]. There are cur-
clinical trials are underway with the goal of rently seven GITR agonist monoclonal antibodies
understanding how best to incorporate OX40 into in phase I–II clinical trials for patients with
combination trials with CPI and conventional advanced solid tumors (Table 10.2).
therapies. CD27 is a costimulatory receptor that is con-
4-1BB is can be found on many cell types, stitutively expressed on lymphoid cells, includ-
including cells of hematopoietic and neuronal ing T cell s, B-cells, and NK cells and is
origins. Like OX40, 4-1BB is transiently upregulated on CD4+ and CD8+ after activation.
expressed on CD4+ and CD8+ T cells following Binding of CD27 with its ligand CD70 on acti-
activation. Binding to 4-1BB ligand induces cell vated APCs promotes clonal expansion of T cells,
proliferation and survival, promotes effector in addition to effector and memory T cell differ-
functions, and stimulates memory cell differenti- entiation and survival. Contrary to the above
ation. Preclinical studies have shown potent anti- TNFRs, CD27 also stimulates T cell priming.
tumor responses predominantly mediated by Preclinical studies of CD27 agonists have shown
cytotoxic CD8+ T cells. 4-1BB agonists have also efficacy in several tumor models [47]. Varlilumab,
demonstrated enhancement of NK-mediated a humanized IgG1 monoclonal antibody agonist
ADCC [45], making these agents attractive for to CD27, is currently being tested in clinical tri-
combinations with targeted therapies and immu- als for a variety of cancers, including one study
notherapies capable of tumor killing or Treg with a phase II cohort in combination with
depletion through ADCC. Two 4-1BB agonists nivolumab in HNSCC (Table 10.2).
are currently in clinical trials in solid tumors CD40 is a member of the TNF-receptor super-
(Table 10.2). family expressed on multiple cell types including
GITR, similarly to OX40 and 41-BB, is tran- dendritic cells, B cells, monocytes, and some
siently expressed on CD4+ and CD8+ T cells fol- tumor types including HNSCC [48]. Rather than
lowing TCR ligation. Like these other TNFRs, directly activating T cells, CD40 agonists have
GITR appears to be less important for T cell been shown to activate dendritic cells to induce T
priming, and more involved with promoting cell responses. These agents can also activate
effector cell functions. GITR is also expressed on macrophages to mount a T cell independent anti-
DCs, monocytes, granulocytes, and NK cells and tumor response and can induce ADCC and
is constitutively expressed on Tregs. Preclinical complement- dependent cytotoxicity (CDC)
studies have shown that GITR ligation can over- through interaction with NK cells [49]. Phase I
come Treg-mediated immunosuppression, and studies have shown favorable toxicity profiles
agonists have shown impressive antitumor effi- and therapeutic promise in targeting CD40 [50].
There is a strong rationale for combining CD40 daily injections into the neck for 10 days prior to
agonists with other immunotherapies, including surgery, and injections five times a month for up
CPI. Currently, there are at least five of these to a year after surgery.
agents in active clinical trials, alone and in com- Building on this work, another group is inves-
bination with other immunotherapies tigating in HNSCC, a blend of cytokines termed
(Table 10.2). IRX-2, which primarily consists of IL-2, IL-1β,
IFN-γ, and TNF-α. The formulation is adminis-
tered perilymphatically and is combined with
10.6 Cytokines systemic low-dose cyclophosphamide for Treg
depletion, along with indomethacin and zinc to
Cytokine therapy with IFN-α or IL-2 has been uti- inhibit immunosuppressive elements within the
lized in the treatment of cancer for over 30 years. systemic circulation and TME. In 2011, nonran-
IFN-α use has greatly diminished due to marginal domized phase II data were presented from 27
effectiveness and concerns for acute toxicities patients with Stage II-IVa HNSCC receiving
[51]. High-dose IL-2 (HD-IL2) is also associated perilymphatic IRX-2. The regimen was found to
with severe, although generally reversible, acute be well-tolerated, with no grade 4 or higher tox-
toxicities including hypotension, renal failure, and icities. 3-year OS and DFS rates were 69% and
thrombocytopenia. However, HD-IL2 is capable 62%, respectively; median DFS and OS were not
of producing durable responses in a minority of reached after follow-up of at least 3 years [54]. A
patients with metastatic melanoma and renal cell larger, phase II trial in HNSCC is currently
carcinoma (RCC). Patients must be selected for underway (NCT02609386) and a phase Ib trial in
good functional status and organ function, and the combination with durvalumab (anti-PD-1L) and
therapy must be administered in specialized cen- tremelimumab (anti-CTLA-4) is planned to com-
ters with hospitalization required for the duration mence in October 2018 (NCT03381183).
of treatment [52]. The antitumor activity of intra- Another cytokine mixture-dubbed Multikine
venous HD-IL2, however, has not been demon- has been tested in HNSCC. This formulation
strated in cancers outside of melanoma and RCC. contains 14 interleukins, interferons (IFN), che-
The well-established immune-stimulating mokines, and colony-stimulating factors. In
effects of various cytokines have sustained inter- 2005, published data from a phase II clinical trial
est in researching clinical applications for can- in T2-3, N0-2, M0 HNSCC showed an ORR of
cers outside of melanoma and RCC, including 41% in 17 patients treated with neoadjuvant peri-
HNSCC. In 2002, results were published from a lymphatic Multikine injections [55]. A large
randomized phase III trial of perioperatively phase III trial of Multikine in HNSCC completed
administered perilymphatic IL-2 in Stage II-IVb enrollment in late 2016, and results are pending
oral or oropharyngeal SCC. This therapy (NCT01265849).
appeared to be safe and efficacious, significantly IL-15 is cytokine of particular interest cur-
improving disease-free survival and OS. The rently in cancer immunotherapy. IL-15 shares
5-year survival rate in the perilymphatic IL-2 the immunostimulatory characteristics of IL-2;
group was 73%, compared with 55% in the con- however, IL-15 does not promote Treg expan-
trol group, and disease-free survival (DFS) rates sion or activation-induced cell death (AICD) of
were 64% and 51%, respectively [53]. Despite effector T cells, which are characteristics of
the impressive improvement in outcomes from IL-2 [56]. Two variants of IL-15, modified to
this randomized, phase III study, perilymphatic improve the pharmacokinetic and pharmacody-
IL-2 has not received much attention in the field. namic properties when administered intrave-
One possible reason for this includes the inten- nously, are currently in clinical development.
sive dosing regimen used in the study, including Recombinant human IL-15 (rhIL-15) was found
to be well-tolerated and produced substantial associated molecule, receptor signaling leads to

increases in circulating NK and CD8+ T cells release cytokines and interferons, which initiate
[57]. Additional phase I trials are combining an immune response. Each TLR subtype recog-
CPI with rhIL-15 or hetIL-15, another modified nizes a specific type of PAMP or DAMP, for
IL-15, for metastatic or refractory cancer example, LPS for TLR 4 or double-stranded viral
(NCT03388632, NCT02452268). RNA for TLR3. While the subtypes TLR can be
expressed on a variety of cell types, they are all
expressed on dendritic cells [60]. Currently, com-
10.7 Targeting the Innate pounds of clinical interest are agonists of TLR3,
Immune System 7, 8, or 9, which are present on endosomes, and
TLR4, which is present on the cell surface.
Traditionally, the immune system has been TLR7 and 8 are closely related structurally
divided into “innate” and “adaptive” compo- and functionally, recognizing single-stranded
nents. The term “innate” refers to sensors that do RNA from viruses or bacteria. Activation of these
not require rearrangement of genes. Receptors TLRs induces production of cytokines and type I
present on innate immune cells, such as dendritic interferons [61]. Imiquimod, a topical TLR7 ago-
cells and macrophages, are highly conserved nist, was FDA-approved in 1997 for the treat-
between individuals and species. “Adaptive” ment of genital warts after clinical trials showed
immune components require rearrangement of efficacy against this virus-induced pathology.
genes, leading to great diversity in receptors but Imiquimod later demonstrated activity against
are therefore specific to individuals. Adaptive actinic keratosis, a premalignancy, as well as
immune components include T cell receptors, basal cell carcinoma, and now has FDA approval
B-cell receptors, and antibodies. for both of these conditions [51]. Intratumoral
The innate and adaptive immune systems are injection of TLR7 and TLR7/8 agonists are
not separate systems but are in fact intimately undergoing early clinical investigation in multi-
related. Innate immune cells are involved in the ple tumor types including HNSCC (Table 10.3).
activation of the adaptive immune system through The phase II Active8 trial randomized patients
cytokine signaling and antigen presentation [58]. with R/M HNSCC to receive SOC platinum/5FU/
Therefore, the innate immune system is capable cetuximab with or without the TLR8 agonist,
of targeting threats directly through “innate” motolimod. Although motolimod failed to
pathways and also of “priming” adaptive immune improve PFS or OS in the intent-to-treat popula-
cells including T cells. Appreciation of this relation, HPV+ patients and those with injection site
tionship has made targeting the innate immune reactions experienced significant benefit [62].
system attractive for cancer immunotherapy, as These results suggest that motolimod may benefit
well as autoimmune diseases. certain subgroups of HNSCC patients, based on
HPV status or other biomarkers. A phase Ib trial
of motolimod combined with cetuximab in the
10.7.1 TLR Agonists neoadjuvant setting showed evidence of immune
response in the peripheral blood and resected
Toll-like receptors (TLR) are components of tumor specimens [63]. The study protocol was
innate immunity that recognize foreign mole- amended in 2016 to add nivolumab to the combi-
cules (pathogen-associated molecular patterns nation of cetuximab and motolimod, and those
[PAMPs], e.g., lipopolysaccharide [LPS]) or results are pending [64].
products of damaged tissues (danger-associated TLR9 is a subset of TLR that recognizes CpG-
molecular patterns [DAMPs], e.g., HMGB1) rich DNA, a PAMP. Activation of TLR9 leads to
[59]. After recognition of a foreign or danger- TNF and type I IFN production, which in turn
Table 10.3 Clinical trials: innate immune activators

Target Drug Phase Tumor type Combination Trial Completion date
TLR7 LHC165 I Advanced PDR001 NCT03301896 08/2019
malignancies
TLR7/8 MEDI9197 I Solid tumors Durvalumab and/or NCT02556463 08/2020
palliative RT
NKTR-262 I Advanced or NKTR-214 (modified NCT03435640 12/2019
metastatic solid IL-2)
tumors
TLR8 Motolimod I R/M/ Cyclophosphamide NCT02050635 05/2021
persistent;progressive
solid tumors
Ib Resectable HNSCC Cetuximab and NCT02124850 Unknown
(neoadjuvant) nivolumab
TLR9 SD-101 I/II R/M HNSCC Pembrolizumab NCT02521870 02/2020
IMO-2125 I Solid tumors None NCT03052205 04/2020
MGN1703 I Advanced solid Ipilimumab NCT02668770 05/2020
malignancies
STING ADU-S100 I Advanced or Ipilimumab NCT02675439 12/2020
metastatic solid
tumors
I Advanced or PDR001 NCT03172936 05/2019
metastatic solid
tumors
MK-1454 I Advanced or Pembrolizumab NCT03010176 10/2021
metastatic solid
tumors
R recurrent, M metastatic
can activate T cells. Preliminary data are avail- mice show a markedly defective CD8+ T cell
able from a phase I/II study of the TLR9 agonist, priming [66]. Preclinical studies have also shown
SD-101, with pembrolizumab in R/M that activation of the STING pathway can
HNSCC. Out of 16 PD-1-naïve patients currently increase effector T cell tumor infiltration [67].
enrolled, ORR in 10 evaluable patients is 40% ADU-S100 is a cyclic dinucleotide (CDN) that
(four PR, one SD, five PD). Final results of the was discovered to activate all known human
study after data maturation are anticipated, but STING alleles and is currently undergoing phase
these early results are promising [65]. Other I clinical evaluation in advanced solid tumors in
TLR9 agonists in phase I studies in solid tumors combination with ipilimumab or PDR001 (anti-
include IMO-2125 and MGN1703 (Table 10.3). PD-1). Another CDN STING agonist, known as
IMO-2125 is also in phase III development in MK-1454, is also in phase I trials in advanced/
combination with CPI in melanoma. metastatic solid tumors, alone or in combination
with pembrolizumab (Table 10.3).
STING agonists are local therapy—they are
10.7.2 STING Agonists injected into the tumor and have no systemic
effect. One group is developing a novel interven-
The STING (stimulator of interferon genes) path- tion using a biomaterial containing CDN ligands
way has recently been recognized as a critical that is called STINGblade, which is implanted
component of the antitumor response. STING is locally into the resection site at the time of sur-
an endoplasmic reticulum protein which binds to gery and is extremely effective at preventing local
cytosolic (tumor) DNA, causing activation den- recurrence following total or subtotal surgical
dritic cells. Experimental studies in STING −/− resection [68]. Using several different models of
Table 10.4 Targeting of tumor-associated antigens

Class Advantages Concerns Examples
Tissue differentiation • Shared antigens • Expression on normal tissues MART-1
antigens • “Off the shelf” treatments can • Potential for on-target, off-tumor gp100
be developed toxicity CEA
CD19
Tumor germline (“tumor- • Shared antigens • Potential for on-target off-tumor ΝY-ESOl
testis”) antigens • “Off the shelf” treatments can toxicity MAGE-A3
be developed • May be expressed in a low
• Potentially tumor-specific frequency of cancers
Normal proteins • Shared antigens • On-target, off-tumor toxicity hTERT
overexpressed by cancer • “Off the shelf” treatments can EGFR
cells be developed mesothelin
Viral proteins • Shared antigens • Low frequency of virus-associated HPV
• “Off the shelf” treatments can cancers EBV
be developed MCC
• Tumor-specific, thus minimal risk
of on-target off-tumor toxicity
Tumor-specific mutated • Tumor specific, thus minimal • Currently requires surgical resection Mum-1
antigens risk of on-target off-tumor for next-generation sequencing B-catenin
toxicity • Most immunogenic mutations CDK4
• Shared driver hot-spot identified so far are patient-specific ERBB2IP
mutations can potentially be • Extended time to develop
targeted personalized treatment targeting
mutations
Modified from: Ilyas S, Yang JC. Landscape of Tumor Antigens in T cell Immunotherapy. Journal of immunology
(Baltimore, Md: 1950). 2015;195(11):5117–5122. doi:https://doi.org/10.4049/jimmunol.1501657)
HNSCC, they showed that antitumor activity was specific mutated antigens or “neoantigens.”
host-STING and CD8-dependent, indicating that Alternative targets include tumor-associated anti-
adaptive immune responses are required for con- gens (TAA) including tissue differentiation anti-
trol of disease and improved survival. Subsequent gens, antigens that are overexpressed on tumors
work demonstrated that a novel approach to ana- compared with normal tissue (e.g., EGFR), and
lyzing cytokine response using tumor explants cancer germline antigens that are not normally
treated ex vivo identified tumors with variable expressed on somatic cells but are aberrantly
immune responses to STING ligands, which expressed in tumor cells [70]. Each type of target
could enable personalization of the immunother- antigen caries potential advantages and disadvan-
apy-containing biomaterial to induce tumor cure. tages for vaccination, outlined in Table 10.4.
Numerous types of cancer vaccines have been
tested in preclinical models and in clinical trials.
10.8 Vaccines These include peptide vaccines, tumor lysates,
DNA or RNA vaccines, and cellular vaccines
Just as vaccines can train the immune system to including dendritic cells that have been exposed
recognize and destroy pathogens, thereby pre- to antigen and danger signals. Sipuleucel-T
venting infection, vaccination can also initiate (Provenge), an autogenous cellular vaccine tar-
antitumor immune responses [69]. To induce an geting prostatic acid phosphatase, was approved
effective immune response and avoid targeting of by the FDA for the treatment of metastatic
self-antigens, vaccination would ideally utilize prostate cancer in 2010. While Sipuleucel-T was
antigens that are expressed only on tumor cells shown to increase OS in a randomized phase III
and not on normal cells, i.e., tumor-specific anti- trial, there was no increase in PFS [71].
gens (TSA) including viral proteins and tumor-
Sipuleucel-T remains the only FDA-approved and is currently in phase I clinical trials in mul-
therapeutic vaccine for the treatment of cancer. tiple tumor types, with plans to expand to oral
The ability to identify neoantigens through cancer [72].
next-generation sequencing (NGS) techniques Therapeutic vaccines have been tested clini-
is appealing, in that neoantigens would theoretically in HNSCC, mostly showing modest effi-
cally avoid self-reactivity seen with tissue dif- cacy [73–76]. While vaccination can facilitate
ferentiation antigens or overexpressed antigens. antigen presentation, it does not address deficien-
However, this approach is based on patient- cies in T cell activation or suppressive forces
specific antigens and relies on complex mathe- within tumors. Furthermore, targeting specific
matical modeling to predict binding to the antigens may lead to immune editing and shed-
patient’s MHC types. Short-lived peptides are ding of that antigen, unless the mutant target is a
another potential source of tumor-associated true “driver” of oncogenesis. Therefore, vaccina-
antigens that have not been tolerized, due to the tion as a monotherapy is unlikely to be successful
usual rapid degradation in autophagosomes. By in most settings and should be approached as an
“freezing” these vesicles in vitro, preventing additional tool in the immunotherapy arsenal
fusion with lysosomes and subsequent destruc- rather than a standalone therapy. Future direc-
tion of the antigens, a source of tumor-associ- tions for cancer vaccines will undoubtedly
ated antigens can be obtained which otherwise involve other therapies such as costimulatory
would have been “thrown away” by normal cell agonists or checkpoint blockade to capitalize on
metabolism. This technique has been used to antigen presentation or reduce immune-
create a vaccine known as DRibbles, which has suppressive influences that may prevent immune
been shown to contain shared antigens capable response despite recognition of a tumor [77].
of cross-recognition between different tumors Ongoing trials in HNSCC are listed in Table 10.5.
Table 10.5 Clinical trials: therapeutic vaccines in HNSCC (including HPV-negative)

Estimated
completion
Category Target Product Phase Combination Trial date
Tissue CEA CEA(6D)/ I None NCT02999646 Completed,
differentiation TRICOM results pending
antigen
Tissue CEA GI-6207 I None NCT00924092 Completed,
differentiation results pending
antigen
Tissue MUC-l - I/II Tadalafil NCT02544880 04/2021
differentiation
antigen
Cancer germline MAGE-A3 Biropepimut-S II cyc, GM-CSF, NCT02873819 12/2020
antigen poly ICLC
Tumor lysate Tumor- derived Allovax II None NCT02624999 12/2018
antigens
Tumor lysate Tumor- derived MVX-ONCO-1 II None NCT02999646 06/2020
(irradiated) antigens
Tumor-associated CEA, MUC-l, NANT vaccine I/II chemo, RT, NCT03109764 01/2019
antigens Ras, Brachyury CPI, cytokines
Neoantigen— Patient-specific Vaccibody I/IIa None NCT03548467 03/2023
vaccine/antibody neoantigens VB10.NEO
hybrid
cyc cyclophosphamide, CPI checkpoint inhibitors
Table 10.6 Clinical trials: adoptive cell therapy in HNSCC (including HPV-negative)
Method Product Phase Combination Trial Estimated completion date
TIL LN-145 II Lymphodepletion/IL-2 NCT03083873 10/2018
TCR-engineered IMA201 I Lymphodepletion/IL-2 NCT03247309 12/2019
CAR-T T4 I None NCT01818323 6/2019
study of intratumorally injected pan-ErbB CAR

10.9 Adoptive Cell Therapy T cell s showed a disease control rate
(DCR = CR + PR + SD) of 69%, despite rapidly
Adoptive cell therapy (ACT) is an immunothera- progressing tumors at trial enrollment [82].
peutic approach that involves the extraction of T
cells from a patient, expansion of a T cell popula-
tion ex vivo, and infusion of the T cells back into 10.10 Oncolytic Viruses
the patient, usually after chemotherapeutic lym-
phocyte depletion and followed by administra- It has been shown that certain viruses can preferen-
tion of cytokines such as IL-2 [78]. Several tially replicate within malignant cells while pre-
methods of ACT are in development, including serving normal cells. Some viruses naturally
extracting tumor-infiltrating lymphocytes (TIL), possess this capability (myxoma virus, Newcastle
which are expected to include some T cells that disease virus (NDV), reovirus), while other types
have specificity for TAA. Alternatively, T cells of viruses (herpes simplex, vaccinia, adenovirus)
may be extracted from the peripheral blood and can be genetically modified to specifically infect
either (1) selected for tumor reactivity ex vivo, malignant cells [83]. In addition to direct tumor cell
(2) exposed to dendritic cells loaded with specific killing, oncolytic viruses can induce immunogenic
TAA, (3) transduced with specific T cell recep- cell death (ICD), essentially acting as a cancer vac-
tors (TCR) with affinity for known TAAs, or (4) cine [84]. As with many new cancer therapies in
transduced with a chimeric antigen receptor development, oncolytic viruses will likely find the
(CAR) that has the antigen-recognizing domain most utility when combined with immunotherapies
of an antibody and the signaling domain of a such as CPI, and this is reflected by the design of
TCR [60]. These approaches essentially skip the many current clinical trials (Table 10.7).
majority of the cancer immunity cycle, including In 2015, the first FDA approval of an onco-
antigen recognition, T cell activation and prolif- lytic virus was granted to a genetically engi-
eration, by directly introducing cancer-fighting T neered, GM-CSF-transduced Herpes Simplex
cells into the patient. ACT is therefore termed Virus (HSV-1), Talimogene Laherparepvec or
“passive immunity,” as it circumvents reliance on T-VEC, for use in locally advanced or nonresect-
the patient’s immune response. able melanoma. In HNSCC, T-VEC was tested in
ACT has proven to be a potent strategy in a small phase I/II study in combination with
melanoma, capable of inducing complete chemoradiotherapy in advanced HNSCC, show-
responses, and investigation of this method has ing an OS of 70.5% at median follow-up of
broadened to include cancers of epithelial origin 29 months. Patients all received post-therapy
[79]. There are currently three ACT clinical tri- neck dissection, which was not standard of care
als that include patients with HPV-negative for many of these patients; however, a pathologic
HNSCC (Table 10.6). Early results from the CR rate of 94% in neck dissection specimens was
LN-145 TIL trial showed an ORR of 38% in a promising finding [85]. T-VEC is being tested
eight patients evaluable [80, 81]. Additionally, in many other types of cancers and is currently in
preliminary data from a phase I dose-escalation a phase Ib/III study in HNSCC in combination
Table 10.7 Clinical trials: oncolytic viruses in HNSCC and solid tumors
Estimated completion
Virus Product Phase Tumor type Combination Trial date
Herpes T-VEC Ib/III R/M HNSCC Pembrolizumab NCT02626000 08/2020
virus
Measles MV-NIS I R/M HNSCC None NCT01846091 12/2018
Vaccinia Pexa-Vec I A/M solid tumors Ipilimumab NCT02977156 11/2019
Vaccinia p53MVA I Refractory solid Pembolizumab NCT02432963 02/2019
tumors
Adenovirus Enadenotucirev I A/M epithelial Nivolumab NCT02636036 03/2019
tumors
A advanced, R recurrent, M metastatic
with pembrolizumab [86]. HF10, a spontane- with locoregional disease, but stated that there were
ously occurring mutant HSV-1 virus, was shown too few patients to power a statistical analysis for
to be well-tolerated in a phase I trial in HNSCC patients with distal metastases alone [93]. From
[87], and in a phase II trial in advanced mela- review of the company’s webpage, it appears that
noma, HF10 combined with nivolumab showed further development is currently focused on
an ORR of 41% with a 16% rate of CR [88]. myeloma, breast, and pancreatic cancer [94].
Another example is Cavatak™, a coxsackie-
virus developed by Viralytics, which seeks out
and attaches itself to a protein that is highly 10.11 T
he Role of Conventional
expressed on the surface of many cancer cells, Therapies in Activating
intercellular adhesion molecule-1 (ICAM-1). the Immune System
Since ICAM-1 is expressed in HNSCC [89], a
phase 1 clinical trial studying Cavatak with pem- While the efficacy of traditional cancer treat-
brolizumab has been designed and is currently in ments, including chemotherapy, radiation, and
its final stages of preparation before opening for targeted therapies, has historically been
recruitment. GL-ONC1, an attenuated vaccinia ascribed to direct cytotoxicity or inhibition of
virus, was well-tolerated with concurrent chemo- cellular activities, there is increasing apprecia-
radiotherapy in HNSCC [90] and is in phase II tion for the immune-stimulating effects of
development in recurrent ovarian cancer these treatments. As described above, the
(NCT02759588). Oncorine (H101), an E1B- Cancer Immunity Cycle begins with the release
deleted adenovirus, was approved in China in of cancer cell antigens. This is achieved by a
2005 for use in HNSCC after a phase III study process known as Immunogenic Cell Death
showed an ORR of 78% when combined with (ICD), in which the killing of tumor cells can
cisplatin and 5-FU [91, 92]. The use of H101 so elicit an antitumor immune response. In addi-
far remains limited to China. tion to promoting recognition of tumor anti-
A trial called REO 018 trial was initially gens through ICD, many chemotherapeutic
designed as a randomized phase III study of agents have been shown to modulate immuno-
Reolysin, a reovirus, in combination with carbopla- suppressive influences, e.g., through depletion
tin and paclitaxel in platinum-refractory of Tregs or MDSCs [95].
HNSCC. The study was reformatted after interim Many chemotherapeutics have been assessed
analysis found differential responses in patients for the various components of ICD. While cis-
with locoregional disease versus patients with meta- platin has previously been thought to be incapa-
static disease alone. The company claimed a statisti- ble of inducing bona fide ICD on its own, recent
cally significant increase in PFS and OS in patients work on HNSCC models indicate that cisplatin
Table 10.8 Immunotherapy with chemotherapy in HNSCC

Chemo agent Phase Settiug Immunotherapy Timing of immunotherapy Trial
Docetaxel + cisplatin + 5FU I PULA Durvalumab Induction (before chemo/rad) NCT02997332
Evofosfamide I LA or M Ipilimumab Concurrent (second line) NCT03098160
Docetaxel Ι/II R/M Pembrolizumab Concurrent (second line) NCT02718820
Platinum + 5FU III R/M Pembrolizumab Concurrent (first line) NCT02358031
Table 10.9 Immunotherapy with radiation in HNSCC

Type of radiation Phase Setting Immunotherapy Timing in relation to RT Trial
With surgery
Neoadjuvant SBRT I/II Curative Nivolumab Neoadjuvant + adjuvant NCT03247712
IMRT II LA Pembrolizumab Neoadjuvant + adjuvant NCT02296684
Without surgery
IMRT II LA Pembrolizumab Concurrent NCT02707588
"High dose" II M Pembrolizumab Concurrent NCT03085719
Re-irradiation II R Pembrolizumab During and after NCT02289209
Re-irradiation I/II R Nivolumab Before, during and after NCT03317327
SBRT II M Nivolumab During and after NCT02684253
Proton SBRT Observational R/M Nivolumab Before, during and after NCT03539198
can effectively induce ICD [96]. This strength- [100]. Nonetheless, there is abundant evidence
ens the rationale for combining SOC platinum for the synergistic effect of radiation and immu-
agents with immunotherapy. KEYNOTE-048 notherapy [101]. There are numerous studies
was the first randomized phase III trial of anti- combining immunotherapy and radiation in
PD-1 therapy plus chemotherapy in HNSCC. As HNSCC (Table 10.9), including a phase I study
mentioned previously in section 10.2, the posi- of neoadjuvant nivolumab combined with hypo-
tive results from this trial led to approval of pem- fractionated stereotactic body radiotherapy
brolizumab alone or in combination with (SBRT) prior to surgical resection in the defini-
platinum chemotherapy in first line R/M disease. tive setting (NCT03247712).
Additional trials of cytotoxic chemotherapy Additionally, there are mechanistic rationales
combined with immunotherapy in HNSCC are for combining targeted therapies with immuno-
listed in Table 10.8. therapy. The antitumor activity of cetuximab in
Radiation may also play a synergistic role in HNSCC is now appreciated to be primarily
combination with immunotherapy. In addition ADCC, as opposed to direct cytotoxicity. By
to induction of ICD, radiation therapy has been combining cetuximab with CPI or other immuno-
shown to recruit T cells to the irradiated tumor therapies, immune-suppressive forces within the
and increase susceptibility of tumor cells to TME could potentially be counteracted leading
cytotoxic effector cells [98]. Radiation also to increased efficacy over either agent alone
upregulates PD-L1 expression on tumor cells, [102]. An interim safety analysis of a phase II
which may limit the immunogenicity of radia- trial of pembrolizumab and cetuximab in R/M
tion alone, but offers a therapeutic opportunity HNSCC showed good tolerability with no DLTs
for combination with PD-1 inhibitors [99]. In [103]. There are several other efficacy studies
HNSCC, immunotherapy appears to increase underway which combine cetuximab with CPI
the antitumor response from radiation, rather with or without chemotherapy and radiation.
than facilitate distant abscopal responses Lenvatinib is a multiple kinase inhibitor
through an autovaccination effect of radiation approved for differentiated thyroid cancer and
advanced renal cell carcinoma. In addition to its great potential for understanding the effects of
effects on tumor angiogenesis and proliferation, immunotherapy in human cancer. Results from a
lenvatinib has been shown to decrease suppres- phase II study of neoadjuvant nivolumab in
sive TAM populations within tumors and to resectable HNSCC showed good tolerability and
increase numbers of effector CD8+ cells [104]. tumor reductions within 1 month in nearly half of
Interim analysis of a phase Ib/II clinical trial of evaluable patients [106]. Many other neoadjuvant
lenvatinib with pembrolizumab in metastatic immunotherapy trials in HNSCC are underway,
HNSCC showed an ORR of 41%, although the with a focus on analysis of immune effects within
rate of grade 3–4 adverse events was 73% [105]. the surgical specimens. The histological and clin-
Interim analysis of a phase I/II trial of pembroli- ical comparison of tumors before and after immu-
zumab in combination with vorinostat, a histone notherapy may provide much insight into the
deacetylase inhibitor, showed an ORR of 36% effects in vivo, including the potential to identify
with DCR of 56%. Another study of Acalabrutinib, biomarkers for response and further therapeutic
a Bruton’s tyrosine kinase inhibitor, in combina- targets.
tion with pembrolizumab in advanced HNSCC, Due to the large number of clinical trials com-
is underway. Trials of targeted therapies with bining chemotherapy, radiation, and/or targeted
immunotherapy in HNSCC are listed in therapies with immunotherapy in HNSCC in a
Table 10.10. variety of settings (neoadjuvant, adjuvant, recur-
There are a number of “preoperative window,” rent/metastatic), much will be learned about the
neoadjuvant immunotherapy studies underway in safety and efficacy of combinations. Studies not
HNSCC (Table 10.11). In addition to testing already listed in previous sections are presented
immunotherapies in earlier stages of disease with in Table 10.12. Moving forward, a major focus of
curative intent, studies such as these provide preclinical research and clinical trials moving
Table 10.10 Immunotherapy with targeted therapy in HNSCC

Targeted therapy Phase Setting Immunotherapy Timing of immunotherapy Trial
Acalabrutinib II LAa or R/M Pembrolizumab Concurrent NCT02454179
Vorinostat Ι/II LAa or M Pembrolizumab Concurrent NCT02538510
Cetuximab I/II R/M Nivolumab Concurrent NCT03370276
II R/M Pembrolizumab Concurrent NCT03082534
Lenvatinib I/II M Pembrolizumab Concurrent NCT02501096
Not amenable to surgery
a
Table 10.11 Preoperative “window of opportunity” immunotherapy trials

Expected
Imtnuaotherapy Phase Endpoint Trial completion date
Nivolumab II Response; indicators of NCT03021993 03/2020
immune response in tissue/
blood
Nivolumab +/− ipilimumab II Response, recurrence NCT02919683 04/2024
Nivolumab +/− ipilimumab or I/II Response, recurrence NCT02488759 12/2019
relatlimab or daratumumab
Cemiplimab (RENG2810) II Response, recurrence NCT03565783 01/2020
Durvalumab II Indicators of immune NCT02827838 01/2019
response in tissue/blood
Ipilimumab (intratumoral) I Indicators of immune NCT02812524 07/2019
MEDI0562 (anti-OX40) Ib Indicators of immune NCT03336606 12/2024
Table 10.12 Immunotherapy with chemotherapy/targeted therapy and radiation in HNSCC

Type of chemotherapy Phase Setting Immunotherapy Timing of immunotherapy Trial
With surgery
Carboplatin + nab-paclitaxel II LA Durvalumab Neoadjuvant and adjuvant NCT03174275
Cisplatin II LA Pembrolizumab Neoadjuvant and adjuvant NCT02641093
Without surgery
Cisplatin or cetuximab I LAa Nivolumab Before, during, after CRT NCT02764593
I LAa Durvalumab During radiation NCT03509012
Cetuximab III PULA Avelumab Before, during, after CRT NCT02999087
Ib LAa Ipilimumab During and after CRT NCT01860430
Cisplatin III LAa Pembrolizumab Before, during, after CRT NCT03040999
LA locally advanced, PULA previously untreated locally advanced, R recurrent, M metastatic
a
Not amenable to surgery
will be to determine the optimal doses and timing 115], and checkpoint inhibitors that target NK
of standard therapies to promote responses to dif- cells (e.g., anti-KIR, anti-NKG2A) [116, 117].
ferent types of immunotherapy. Moving forward in this new era of cancer
immunotherapy will require continuing integra-
tion between the clinic and laboratory. Not only
10.12 Conclusions will preclinical science remain critical in devel-
oping new approaches in patient care, but labo-
Immunotherapy is rapidly changing the standard ratory evaluation of pathologic tumor responses,
of care in oncology. The appearance of a tail at immune cell infiltrates, and circulating immune
the end of the survival curve with checkpoint components will allow full-circle analysis and
inhibition in advanced cancers provides a graphic understanding of the physiologic effects of
representation of the durable responses that can experimental treatments. The increasing num-
be achieved with this new group of therapies. ber of neoadjuvant trials will facilitate this route
This is the great promise of cancer immunother- of scientific discovery by providing postimmu-
apy, that is, the possibility of achieving lasting notherapy tissue samples for comparison with
responses or even cures. pretreatment biopsies. Cutting-edge technolo-
As evidence of the enthusiasm around immu- gies for specimen analysis, such as NGS allow-
notherapy, the number of new products in devel- ing whole- exome, RNA and T cell receptor
opment and early-phase clinical trials has sequencing, as well as advanced imaging tech-
skyrocketed in the last decade. In 2017, it was niques for multiplex immunohistochemistry,
estimated that there were 800–1000 cancer immu- will allow for greater understanding of the
notherapy trials in the US involving over 100,000 in vivo effects of various immunotherapies on
patients [107, 108]. The same year, a report from tumor biology. This work, along with clinical
the Pharmaceutical Research and Manufacturers outcome correlations, will be critical in identi-
of America found that there were 248 immuno- fying predictive biomarkers and prognostic
oncology agents in clinical trials, which only indicators and will provide evidence for future
included “the most recognized classes of immu- directions in cancer research.
notherapy” [109]. In addition to the therapies
described in this chapter, there are many other
immunotherapies and immune adjuncts in devel- References
opment, including but not limited to agents that
target tumor metabolism (e.g., IDO-1 and the 1. Chen DS, Mellman I. Oncology meets immunology:
the cancer-immunity cycle. Immunity. 2013;39:1–10.
adenosine pathway) [110–112], therapies to 2. Schreiber RD, Ikeda H, Dunn GP, Bruce AT, Old
deplete or inhibit Tregs, MDSC, or TAM (e.g., LJ. Cancer immunoediting: from immunosurveillance
anti-CCR4, PDE-5 inhibitors, anti-CSF1R) [113– to tumor escape. Nat Immunol. 2002;3:991–8.
3. Hanahan D, Weinberg R. Hallmarks of cancer: the 14. Lingen MW. Low etiologic fraction for high-risk
next generation. Cell. 2011;144:646–74. human papillomavirus in oral cavity squamous cell
4. Isaacsson Velho PH, Castro J, Gilberto CCH. Targeting carcinomas. Oral Oncol. 2013;49(1):1–8. https://doi.
the PI3K pathway in head and neck squamous cell car- org/10.1016/j.oraloncology.2012.07.002
cinoma. Am Soc Clin Oncol Meet. 2015;2015:123. 15. Nivolumab for SCCHN. U.S. Food and Drug
5. Cai Y, Dodhia S, Su GH. Dysregulations in the PI3K Administration. https://www.fda.gov/Drugs/
pathway and targeted therapies for head and neck squa- InformationOnDrugs/ApprovedDrugs/ucm528920.
mous cell carcinoma. Oncotarget. 2017;8:22203–17. htm.
6. Tran E, Robbins PF, Lu YC, Prickett TD, Gartner JJ, 16. Vermorken JB, Mesia R, Rivera F, et al. Platinum-
Jia L, Pasetto A, Zheng Z, Ray S, Groh EM, Kriley IR, based chemotherapy plus cetuximab in head and neck
Rosenberg SA. T-cell transfer therapy targeting mutant cancer. N Engl J Med. 2008;359:1116–27.
KRAS in cancer. N Engl J Med. 2016;375(23):2255– 17. Ferris RL, Blumenschein Jr GR, Fayette J, et al.
62. https://doi.org/10.1056/NEJMoa1609279. Nivolumab (Nivo) vs investigator’s choice (IC) in
PubMed PMID: 27959684; PMCID: PMC5178827. recurrent or metastatic (R/M) squamous cell carci-
7. Tumeh PC, Harview CL, Yearley JH, Shintaku IP, noma of the head and neck (SCCHN): 2-yr outcomes
Taylor EJ, Robert L, Chmielowski B, Spasic M, Henry in the overall population and PD-L1 subgroups of
G, Ciobanu V, West AN, Carmona M, Kivork C, CheckMate 141. Oral presentation at: 2018 American
Seja E, Cherry G, Gutierrez AJ, Grogan TR, Mateus Association for Cancer Research annual meeting,
C, Tomasic G, Glaspy JA, Emerson RO, Robins H, 14–18 Apr 2018, Chicago, IL.
Pierce RH, Elashoff DA, Robert C, Ribas A. PD-1 18. Pembrolizumab (KEYTRUDA). U.S. Food and
blockade induces responses by inhibiting adaptive Drug Administration. https://www.fda.gov/Drugs/
immune resistance. Nature. 2014;515(7528):568–71. InformationOnDrugs/ApprovedDrugs/ucm515627.
https://doi.org/10.1038/nature13954. PubMed PMID: htm. Accessed 25 Nov 2017.
25428505; PMCID: PMC4246418. 19. Bauml J, Seiwert TY, Pfister DG, et al. Pembrolizumab
8. Spranger S, Spaapen RM, Zha Y, Williams J, Meng for platinum- and cetuximab-refractory head and neck
Y, Ha TT, Gajewski TF. Up-regulation of PD-L1, cancer: results from a single-arm, phase II study. J
IDO, and T(regs) in the melanoma tumor microen- Clin Oncol. 2017;35:1542–9.
vironment is driven by CD8(+) T cells. Sci Transl 20. Soulieres D, et al. Updated survival results of
Med. 2013;5(200):200ra116. https://doi.org/10.1126/ the KEYNOTE-040 study of pembrolizumab vs
scitranslmed.3006504. PubMed PMID: 23986400; standard-of-care chemotherapy for recurrent or
PMCID: PMC4136707. metastatic head and neck squamous cell carcinoma.
9. Jie HB, Gildener-Leapman N, Li J, Srivastava RM, Abstract CT115. Presented at: American Association
Gibson SP, Whiteside TL, Ferris RL. Intratumoral for Cancer Research Annual Meeting; 14–18 Apr,
regulatory T cells upregulate immunosuppressive mol- Chicago; 2018.
ecules in head and neck cancer patients. Br J Cancer. 21. Burtness B, Harrington KJ, Greil R, Soulières
2013;109(10):2629–35. https://doi.org/10.1038/ D, Tahara M, De Castro JG, et al. LBA8_
bjc.2013.645. PubMed PMID: 24169351; PMCID: PRKEYNOTE-048: Phase III study of first-line pem-
PMC3833228. brolizumab (P) for recurrent/metastatic head and neck
10. Gordon SR, Maute RL, Dulken BW, Hutter G, George squamous cell carcinoma (R/M HNSCC). Ann Oncol.
BM, McCracken MN, Gupta R, Tsai JM, Sinha R, 2018;29(suppl_8):mdy424.045.
Corey D, Ring AM, Connolly AJ, Weissman IL. PD-1 22. Alsaab HO, Sau S, Alzhrani R, et al. PD-1 and PD-L1
expression by tumour-associated macrophages inhib- checkpoint signaling inhibition for cancer immu-
its phagocytosis and tumour immunity. Nature. notherapy: mechanism, combinations, and clinical
2017;545(7655):495–9. https://doi.org/10.1038/ outcome. Front Pharmacol. 2017;8:561. https://doi.
nature22396. PubMed PMID: 28514441; PMCID: org/10.3389/fphar.2017.00561.
PMC5931375. 23. Butte MJ, Keir ME, Phamduy TB, Sharpe AH,
11. Mehal WZ, Sheikh SZ, Gorelik L, Flavell RA. TGF- Freeman GJ. Programmed death-1 ligand 1 inter-
beta signaling regulates CD8+ T cell responses to acts specifically with the B7-1 costimulatory mol-
high- and low-affinity TCR interactions. Int Immunol. ecule to inhibit T cell responses. Immunity. 2007;27:
2005;17(5):531–8. https://doi.org/10.1093/intimm/ 111–22.
dxh233. 24. Xiao Y, Yu S, Zhu B, et al. RGMb is a novel bind-
12. Puram SV, Rocco JW. Molecular aspects of head and ing partner for PD-L2 and its engagement with
neck cancer therapy. Hematol Oncol Clin North Am. PD-L2 promotes respiratory tolerance. J Exp Med.
2015;29:971–92. 2014;211:943–59.
13. Fakhry C, Westra WH, Li S, Cmelak A, Ridge JA, 25. European Society for Medical Oncology. ESMO
Pinto H, Forastiere A, Gillison ML. Improved survival 2017: durvalumab shows promising clinical benefit
of patients with human papillomavirus–positive head in recurrent/metastatic head and neck squamous cell
and neck squamous cell carcinoma in a prospective carcinoma (11 Sep 2017). Accessed 24 Dec 2017.
clinical trial. J Natl Cancer Inst. 2008;100(4):261–9. http://www.esmo.org/Conferences/Past-Conferences/
https://doi.org/10.1093/jnci/djn011. ESMO-2017-Congress/News-Articles/.
26. Buchbinder EI, Desai A. CTLA-4 and PD-1 pathways: 37. Chen L, Flies DB. Molecular mechanisms of T cell
similarities, differences, and implications of their inhi- co-stimulation and co-inhibition. Nat Rev Immunol.
bition. Am J Clin Oncol. 2016;39(1):98–106. https:// 2013;13:227.
doi.org/10.1097/COC.0000000000000239. 38. Acton QA. Cytokines. In: Acton QA, editor.
27. Du X, et al. A reappraisal of CTLA-4 checkpoint Intercellular signaling peptides and proteins –
blockade in cancer immunotherapy. Cell Res. advances in research and application. Atlanta, GA:
2018;28(4):416. Scholarly Editions; 2013. p. 278.
28. Hodi FS, O'Day SJ, McDermott DF, Weber RW, 39. Burris HA, Callahan MK, et al. Phase 1 safety of
Sosman JA, Haanen JB, Gonzalez R, Robert C, ICOS agonist antibody JTX-2011 alone and with
Schadendorf D, Hassel JC, Akerley W, van den nivolumab (nivo) in advanced solid tumors; predicted
Eertwegh AJ, Lutzky J, Lorigan P, Vaubel JM, Linette vs observed pharmacokinetics (PK) in ICONIC. J Clin
GP, Hogg D, Ottensmeier CH, Lebbe C, Peschel C, Oncol. 2017;35(Suppl):abstr 3033.
Quirt I, Clark JI, Wolchok JD, Weber JS, Tian J, Yellin 40. Sturgill ER, Redmond WL. TNFR agonists: a review
MJ, Nichol GM, Hoos A, Urba WJ. Improved survival of current biologics targeting OX40, 4-1BB, CD27,
with ipilimumab in patients with metastatic mela- and GITR. AJHO. 2017;13(11):4–15.
noma. N Engl J Med. 2010;363(8):711–23. https:// 41. Guo Z, Wang X, Cheng D, Xia Z, Luan M, Zhang
doi.org/10.1056/NEJMoa1003466. PubMed PMID: S. PD-1 blockade and OX40 triggering synergistically
20525992; PMCID: PMC3549297. protects against tumor growth in a murine model of
29. Koyama S, Akbay EA, Li YY, et al. Adaptive resis- ovarian cancer. PLoS One. 2014;9:e89350.
tance to therapeutic PD-1 blockade is associated with 42. Montler R, Bell RB, Thalhofer C, Leidner R, Feng
upregulation of alternative immune checkpoints. Nat Z, Fox BA, Cheng AC, Bui TG, Tucker C, Hoen H,
Commun. 2016;7:10501. Weinberg A. OX40, PD-1 and CTLA-4 are selectively
30. Economopoulou P, Kotsantis I, Psyrri A. The prom- expressed on tumor-infiltrating T cells in head and
ise of immunotherapy in head and neck squamous cell neck cancer. Clin Transl Immunol. 2016;5(4):e70.
carcinoma: combinatorial immunotherapy approaches. https://doi.org/10.1038/cti.2016.16. PubMed PMID:
ESMO Open. 2017;1:e000122. 27195113; PMCID: PMC4855266.
31. Goldberg MV, Drake CG. LAG-3 in cancer 43. Bell RB, Leidner RS, Crittenden MR, et al. OX40
immunotherapy. Curr Top Microbiol Immunol. signaling in head and neck squamous cell carcinoma:
2011;344:269–78. overcoming immunosuppression in the tumor micro-
32. Pauken KE, Wherry EJ. TIGIT and CD226: tipping environment. Oral Oncol. 2016;52:1–10.
the balance between costimulatory and coinhibitory 44. Bell RB, Duhen R, Ballesteros-Merino C, Duhen T,
molecules to augment the cancer immunotherapy tool- Bifulco C, Piening BD, Bernard B, Pucilowska J,
kit. Cancer Cell. 2014;26(6):785–7. Tamakawa R, Redmond WL, Koguchi Y, Montler R,
33. Argiris A, Gillison M, Ferris RL, Harrington K, Morris G, Fisher B, Shuster M, Cheng A, Patel A,
Sanchez TK, Baudelet C, Geese WJ, Shaw J, Urba WJ, Fox BA, Curti BD, Leidner RS, Weinberg
Haddad R. A randomized, open-label, phase 3 study AD. Neoadjvuant anti-OX40 (MEDI6469) prior to
of nivolumab in combination with ipilimumab vs surgery in head and neck squamous cell carcinoma. J
extreme regimen (cetuximab + cisplatin/carboplatin Clin Oncol. 2018;36(Suppl):abstr 6011.
+ fluorouracil) as first-line therapy in patients with 45. Kohrt HE, Houot R, Weiskopf K, et al. Stimulation
recurrent or metastatic squamous cell carcinoma of natural killer cells with a CD137-specific antibody
of the head and neck-CheckMate 651. Ann Oncol. enhances trastuzumab efficacy in xenotransplant mod-
2016;27(Suppl_6):1016TiP. https://doi.org/10.1093/ els of breast cancer. J Clin Invest. 2012;122:1066–75.
annonc/mdw376.68. 46. Knee DA, Hewes B, Brogdon JL. Rationale for
34. Siu L, et al. A randomized, open-label, multicenter, anti-GITR cancer immunotherapy. Eur J Cancer.
global phase 2 study of durvalumab (D), tremelim- 2016;67:1–10.
umab (T), or D plus T, in patients with PD-L1 low/ 47. van de Ven K, Borst J. Targeting the T-cell co-
negative recurrent or metastatic head and neck squa- stimulatory CD27/CD70 pathway in cancer immu-
mous cell carcinoma: CONDOR. Int J Rad Oncol Biol notherapy: rationale and potential. Immunotherapy.
Phys. 2017;100(5):1307. https://doi.org/10.1016/j. 2015;7:655–67.
ijrobp.2017.12.021. 48. Cao W, Cavacini L, Tillman K, Posner M. CD40 func-
35. Ascierto PA, Bono P, Bhatia S, et al. Efficacy of BMS- tion in squamous cell cancer of the head and neck.
986016, a monoclonal antibody that targets lympho- Oral Oncology. 2005;41:462–9.
cyte activation gene-3 (LAG-3), in combination with 49. Vonderheide RH, Glennie MJ. Agonistic CD40
nivolumab in pts with melanoma. In: Proceedings antibodies and cancer therapy. Clin Cancer Res.
from the 2017 ESMO congress, 8–12 Sep 2017, 2013;19(5):1035–43.
Madrid, Spain. Abstract LBA18. 50. Vonderheide R, Dutcher J, Gribben J, et al. Phase I
36. Harding FA, McArthur JG, Gross A, Raulet DH, study of recombinant human CD40 ligand in cancer
Allison JP. CD28-mediated signaling co-stimulates patients. J Clin Oncol. 2001;19(13):3280–7.
murine T cells and prevents induction of anergy in 51. Hauschild A. Adjuvant interferon alfa for melanoma.
T-cell clones. Nature. 1992;356:607–9. Curr Oncol. 2010;17(3):6–7.
52. Payne R, Glenn L, Hoen H, et al. Durable responses ment naïve patients with recurrent or metastatic head
and reversible toxicity of high-dose interleukin-2 treat- and neck squamous cell carcinoma (HNSCC). Cancer
ment of melanoma and renal cancer in a Community Res. 2018;78(13 Suppl):CT098.
Hospital Biotherapy Program. J Immunother Cancer. 66. Woo S-R, Fuertes MB, Corrales L, et al. STING-
2014;2:13. https://doi.org/10.1186/2051-1426-2-13. dependent cytosolic DNA sensing mediates
53. De Stefani A, et al. Improved survival with peri- innate immune recognition of immunogenic
lymphatic interleukin 2 in patients with resectable tumors. Immunity. 2014;41(5):830–42. https://doi.
squamous cell carcinoma of the oral cavity and oro- org/10.1016/j.immuni.2014.10.017.
pharynx. Cancer. 2002;95(1):90–7. 67. Demaria O, De Gassart A, Coso S, et al. STING acti-
54. Wolf GT, Fee WE, Dolan RW, et al. Novel neoad- vation of tumor endothelial cells initiates spontane-
juvant immunotherapy regimen safety and survival ous and therapeutic antitumor immunity. Proc Natl
in head and neck squamous cell cancer. Head Neck. Acad Sci U S A. 2015;112(50):15408–13. https://doi.
2011;33(12):1666–74. https://doi.org/10.1002/ org/10.1073/pnas.1512832112.
hed.21660. 68. Baird JR, Bell RB, Troesch V, Friedman D, Bambina
55. Tímár J, et al. Neoadjuvant immunotherapy of oral S, Kramer G, Blair T, Medler T, Wu Y, Sun Z, de
squamous cell carcinoma modulates intratumoral Gruijl T, van de Ven R, Leidner RS, Crittenden MR,
CD4/CD8 ratio and tumor microenvironment: a Gough MJ. Evaluation of explant responses to STING
multicenter phase II clinical trial. J Clin Oncol. ligands: personalized immunosurgical therapy for
2005;23(15):3421–32. head and neck squamous cell carcinoma. Cancer Res.
56. Romano E, Margolin K. T cell modulatory cytokines. 2018;78(21):6308–19.
In: Butterfield LH, Kaufman HL, Marincola FM, edi- 69. Church SE, et al. Multiple vaccinations: friend or foe.
tors. Cancer immunotherapy principles and practice. Cancer J. 2011;17(5):379.
New York, NY: Demos LLC; 2017. 70. Ilyas S, Yang JC. Landscape of tumor antigens in T cell
57. Miller J, Morishima C, Conlon K, et al. A first-in- immunotherapy. J Immunol. 2015;195(11):5117–22.
human phase i study of subcutaneous outpatient recom- 71. Kantoff PW, et al. Sipuleucel-T immunotherapy for
binant human IL15 (rhIL15) in adults with advanced castration-resistant prostate cancer. N Engl J Med.
solid tumors. Clin Cancer Res. 2018;24(7):1525–35. 2010;363(5):411–22.
58. O’Neil LAJ. The innate immune system. In: Paul WE, 72. Page DB, et al. Glimpse into the future: harness-
editor. Fundamental immunology. 7th ed. Philadelphia, ing autophagy to promote anti-tumor immunity with
PA: Lippincott Williams & Wilkins; 2013. the DRibbles vaccine. J Immunother Cancer. 2016;
59. Land WG. The role of damage-associated molecu- 4(1):25.
lar patterns (DAMPs) in human diseases: part II: 73. Ferris RL, et al. Phase I dendritic cell p53 peptide
DAMPs as diagnostics, prognostics and therapeutics vaccine for head and neck cancer. Clin Cancer Res.
in clinical medicine. Sultan Qaboos Univ Med J. 2014;2014:2617.
2015;15(2):e157–70. 74. Voskens CJ, Sewell D, Hertzano R, et al. Induction
60. Gajewski TF. Manipulating innate immune path- of MAGE-A3 and HPV-16 immunity by Trojan vac-
ways as cancer immunotherapy. In: Butterfield LH, cines in patients with head and neck carcinoma. Head
Kaufman HL, Marincola FM, editors. Cancer immu- Neck. 2012;34(12):1734–46. https://doi.org/10.1002/
notherapy principles and practice. New York, NY: hed.22004.
Demos LLC; 2017. 75. Yoshitake Y, et al. Phase II clinical trial of multiple
61. Cervantes JL, et al. TLR8: the forgotten relative revin- peptide vaccination for advanced head and neck can-
dicated. Cell Mol Immunol. 2012;9:434. cer patients revealed induction of immune responses
62. Ferris RL, Saba NF, Gitlitz BJ, et al. Effect of adding and improved OS. Clin Cancer Res. 2014;2014:0202.
motolimod to standard combination chemotherapy and 76. Chindavijak S, Har-Noy M, Lausoontornsiri W. Effect
cetuximab treatment of patients with squamous cell car- of therapeutic vaccine on CTLA4 and tumor debulk-
cinoma of the head and neck: the Active8 Randomized ing response in recurrent and metastatic HNSCC. Clin
Clinical Trial. JAMA Oncol. 2018;4:1583–8. https:// Cancer Res. 2018;36(5 Suppl):115.
doi.org/10.1001/jamaoncol.2018.1888. 77. Yu G, et al. Combinational immunotherapy with Allo-
63. Ferris RL, et al. A phase 1b study of neoadjuvant DRibble vaccines and anti-OX40 co-stimulation leads
immune biomarker modulation with cetuximab and to generation of cross-reactive effector T cells and
motolimod in head and neck cancer (HNC). Int J tumor regression. Sci Rep. 2016;6:37558.
Radiat Oncol Biol Phys. 2016;94(4):867–8. 78. Redeker A, Arens R. Improving adoptive T cell ther-
64. Garlapow M. Motolimod plus cetuximab proves apy: the particular role of T cell costimulation, cyto-
efficacious in head and neck cancer. Target Oncol. kines, and post-transfer vaccination. Front Immunol.
2016; https://www.targetedonc.com/news/motolimod- 2016;7:345.
plus-cetuximab-efficacious-head-and-neck-cancer. 79. Rosenberg SA, Restifo NP. Adoptive cell transfer
Accessed 19 Aug 2018. as personalized immunotherapy for human cancer.
65. Cohen E, et al. Abstract CT098: phase Ib/II, open Science. 2015;348(6230):62–8.
label, multicenter study of intratumoral SD-101 in 80. Iovance biotherapeutics announces preliminary phase
combination with pembrolizumab in anti-PD-1 treat- 2 data for TIL Treatment in head and neck and cer-
vical cancers. 2018. http://ir.iovance.com/phoenix. adenovirus (H101) combined with cisplatin-based

zhtml?c=254507&p=irol-newsArticle&ID=2328242. chemotherapy in treating squamous cell cancer
81. Leidner R, et al. A phase 2, multicenter study to evalu- of head and neck or esophagus. Chin J Cancer.
ate the efficacy and safety of autologous tumor infil- 2004;23(12):1666–70.
trating lymphocytes (LN-145) for the treatment of 92. Garber K. China approves world’s first oncolytic
patients with recurrent and/or metastatic squamous virus therapy for cancer treatment. J Natl Cancer
cell carcinoma of the head and neck (HNSCC). J Clin Inst. 2006;98(5):298–300. https://doi.org/10.1093/
Oncol. 2018;36:TPS6096. jnci/djj111.
82. Papa S, et al. A phase I trial of T4 CAR T-cell immuno- 93. Oncolytics Biotech Inc. Announces additional data
therapy in head and neck squamous cancer (HNSCC). from REO 018 randomized study of REOLYSIN in
J Clin Oncol. 2018;36:3046. head and neck cancers. Calgary Oncol Biotech. 2014;
83. Pikor L, Bell JC, Kaufman HL. Oncolytic viruses. https://www.oncolyticsbiotech.com/press-releases/
In: Butterfield LH, Kaufman HL, Marincola FM, edi- detail/308/oncolytics-biotech-inc-announces-addi-
tors. Cancer immunotherapy principles and practice. tional-data-from. Accessed 19 Aug 2018.
New York, NY: Demos LLC; 2017. 94. Oncolytics Biotech Inc. Website studies current
84. Yaghchi CA, Zhang Z, Alusi G, Lemoine NR, Wang development plan. https://www.oncolyticsbiotech.
Y. Vaccinia virus, a promising new therapeutic agent for com/clinical-trials/studies. Accessed 19 Aug 2018.
pancreatic cancer. Immunotherapy. 2015;7(12):1249– 95. Kroemer G, et al. Immunogenic cell death in cancer
58. https://doi.org/10.2217/imt.15.90. therapy. Annu Rev Immunol. 2013;31:51–72. https://
85. Harrington KJ, Hingorani M, Tanay MA, Hickey doi.org/10.1146/annurev-immunol-032712-100008.
J, Bhide SA, Clarke PM, Renouf LC, et al. Phase I/ 96. Tran L, et al. Cisplatin alters antitumor immunity
II study of oncolytic HSVGM-CSF in combination and synergizes with PD-1/PD-L1 inhibition in head
with radiotherapy and cisplatin in untreated stage and neck squamous cell carcinoma. Cancer Immunol
III/IV squamous cell cancer of the head and neck. Res. 2017;5(12):1141–51.
Clin Cancer Res. 2010;16(15):4005. https://doi. 97. The ASCO Post. KEYNOTE-048: pembrolizumab
org/10.1158/1078-0432.CCR-10-0196. monotherapy in head and neck squamous cell carci-
86. Harrington KJ, et al. Safety and preliminary efficacy noma. 2018. http://www.ascopost.com/News/59121.
of talimogene laherparepvec (T-VEC) in combination Accessed 8 Aug 2018.
(combo) with pembrobrolizumab (Pembro) in patients 98. Van Limbergen EJ, De Ruysscher DK, Olivo
(pts) with recurrent or metastatic squamous cell car- Pimentel V, et al. Combining radiotherapy with
cinoma of the head and neck (R/M HNSCC): a mul- immunotherapy: the past, the present and the future.
ticenter, phase 1b study (MASTERKEY-232). J Clin Br J Radiol. 2017;90(1076):20170157. https://doi.
Oncol. 2018;36(Suppl):abstr 6036. org/10.1259/bjr.20170157.
87. Gildener-Leapman N, et al. A phase I trial of 99. Illidge T, et al. Radiation therapy induces an adap-
intratumoral administration of HF10 in patients tive upregulation of PD-L1 on tumor cells which
with refractory superficial cancer: immune may limit the efficacy of the anti-tumor immune
correlates of virus injection. J Clin Oncol. response but can be circumvented by anti-PD-L1.
2013;31(15 Suppl):3099. https://doi.org/10.1200/ Int J Rad Oncol. 2014;90(1):S776.
jco.2013.31.15_suppl.3099. 100. McBride SM, et al. A phase II randomized trial
88. Andtbacka, Robert HI, et al. Final results of a phase of nivolumab with stereotactic body radiotherapy
II multicenter trial of HF10, a replication-competent (SBRT) versus nivolumab alone in metastatic (M1)
HSV-1 oncolytic virus, and ipilimumab combination head and neck squamous cell carcinoma (HNSCC).
treatment in patients with stage IIIB-IV unresectable J Clin Oncol. 2018;36(Suppl):abstr 6009.
or metastatic melanoma. Abstract 166420. Presented 101. Gong J, et al. Radiation therapy and PD-1/PD-L1
at: American Society of Clinical Oncology Annual blockade: the clinical development of an evolv-
Meeting, 20 May 2017, Chicago, IL. ing anticancer combination. J Immunother Cancer.
89. Usami Y, Ishida K, Sato S, Kishino M, Kiryu 2018;6(1):46.
M, Ogawa Y, Okura M, Fukuda Y, Toyosawa 102. Ferris RL, et al. Rationale for combination of

S. Intercellular adhesion molecule-1 (ICAM-1) therapeutic antibodies targeting tumor cells and
expression correlates with oral cancer progres- immune checkpoint receptors: harnessing innate
sion and induces macrophage/cancer cell adhe- and adaptive immunity through IgG1 isotype
sion. Int J Cancer. 2013;133(3):568–78. https://doi. immune effector stimulation. Cancer Treat Rev.
org/10.1002/ijc.28066. 2018;63:48–60.
90. Mell LK, et al. Phase I trial of intravenous attenu- 103. Sacco AG et al. An open-label, non-randomized,
ated vaccinia virus (GL-ONC1) with concur- multi-arm, phase II trial evaluating pembrolizumab
rent chemoradiotherapy (CRT) for locoregionally combined with cetuximab in patients with recurrent/
advanced head and neck carcinoma. J Clin Oncol. metastatic (R/M) head and neck squamous cell car-
2015;33:–6026. cinoma (HNSCC): results of the interim safety anal-
91. Xia Z-J, et al. Phase III randomized clinical trial ysis. ASCO 2018 meeting abstract. http://abstracts.
of intratumoral injection of E1B gene-deleted asco.org/214/AbstView_214_225517.html
104. Kato Y, et al. Abstract A92: effects of lenvatinib on 112. Antonioli L, et al. Anti-CD73 in cancer immunother-
tumor-associated macrophages enhance antitumor apy: awakening new opportunities. Trends Cancer.
activity of PD-1 signal inhibitors. Mol Cancer Ther. 2016;2(2):95–109.
2015;14(12 Suppl 2):A92. 113. Vela M, et al. Chemokine receptor-specific antibod-
105. Taylor MH, et al. A phase 1b/2 trial of lenvatinib ies in cancer immunotherapy: achievements and
plus pembrolizumab in patients with squamous challenges. Front Immunol. 2015;6:12.
cell carcinoma of the head and neck. J Clin Oncol. 114. Weed DT, Vella JL, Reis IM, et al. Tadalafil reduces
2018;36:6016. myeloid-derived suppressor cells and regulatory
106. Ferris RL, et al. LBA46. An open-label, multicohort, T cells and promotes tumor immunity in patients
phase 1/2 study in patients with virus-associated with head and neck squamous cell carcinoma.
cancers (CheckMate 358): safety and efficacy of Clin Cancer Res. 2015;21(1):39–48. https://doi.
neoadjuvant nivolumab in squamous cell carci- org/10.1158/1078-0432.CCR-14-1711.
noma of the head and neck (SCCHN). Ann Oncol. 115. Gomez-Roca CA, Cassier PA, Italiano A, Cannarile
28(Suppl_5):mdx440.041. https://doi.org/10.1093/ M, Ries C, Brillouet A, Mueller C, Jegg AM,
annonc/mdx440.041. Meneses-Lorente G, Baehner M, et al. Phase I study
107. Financial Times. Calling time on the immunotherapy of RG7155, a novel anti-CSF1R antibody, in patients
gold rush. Financial Times. 2017. with advanced/metastatic solid tumors. J Clin Oncol.
108. Kolata G. A cancer conundrum: too many drug tri- 2015;33:3005.
als, too few patients. The New York Times. https:// 116. Beldi-Ferchiou A, Caillat-Zucman S. Control of NK
www.nytimes.com/2017/08/12/health/cancer-drug- cell activation by immune checkpoint molecules. Int
trials-encounter-a-problem-too-few-patients.html. J Mol Sci. 2017;18:2129.
Accessed 5 Sep 2017. 117. Cohen RB, Salas S, Even C, Kotecki N, Jimeno
109. Medicines in development for immuno-oncology A, Soulié AM, Tirouvanziam-Martin A, Zerbib R,
2017 Report. PhRMA.org. 2017. https://www. André P, Boyer-Chammard A, Fayette J. Abstract
phrma.org/medicines-in-development-immuno- 5666: safety of the first-in-class anti-NKG2A
oncology. Accessed 20 Aug 2018. monoclonal antibody monalizumab in combination
110. Jochems C, et al. The IDO1 selective inhibitor with cetuximab: a phase Ib/II study in recurrent or
epacadostat enhances dendritic cell immunogenic- metastatic squamous cell carcinoma of the head
ity and lytic ability of tumor antigen-specific T cells. and neck (R/M SCCHN). Cancer Res. 2017;77(13
Oncotarget. 2016;7(25):37762. Suppl):5666. https://doi.org/10.1158/1538-7445.
111. Ma SR, et al. Blockade of adenosine A2A receptor AM2017-5666.
enhances CD8(+) T cells response and decreases
regulatory T cells in head and neck squamous cell
carcinoma. Mol Cancer. 2017;16(1):99.
Recurrent Oral Cancer and Salvage
Options
11
Joshua E. Lubek and Michael Nagai
11.1 Introduction with aggressive surveillance and new advances in

therapy [1–5].
Oral cancer is the sixth most common cancer to Recurrence rates for oral squamous cell can-
occur worldwide, with squamous cell cancer cer are reported to occur in up to 30% of patients
accounting for greater than 90% of these cases depending on initial disease stage presentation
histologically. Considered to be a global health with most recurrences developing either locally
problem, there will be more than 300,000 new or regionally within 3 years of initial therapy.
cases of oral squamous cell cancer (OSCC) diag- Successful salvage therapy for recurrent disease
nosed each year and will be responsible for over will depend on numerous factors to include initial
140,000 deaths per year. Once thought to be pre- tumor staging, location of recurrence, and initial
dominantly a disease of older males with known treatment utilized (i.e., surgery, primary radio-
risk factors such as tobacco use, alcohol use, or therapy, or multimodality therapy as initial treat-
betel nut habit (high rates of buccal squamous ment strategy) [6–9].
cell carcinoma in certain countries such as India,
China, or Taiwan), recent epidemiologic data
would suggest that patients without risk factors 11.2 R
ecurrent Disease of the Oral
such as younger patients (<45 years of age, Cavity: General
females and nonsmokers) are developing these Considerations
cancers. Despite good prognosis for early-stage
disease, overall survival and disease relapse for Recurrent OSCC is devastating for the patient
late-stage oral cavity disease remains poor, even both from an emotional and physical standpoint.
Very often a patient has just recovered from the
healing phases of primary therapy to now have to
deal with the psychological impact of facing yet
J. E. Lubek (*)
Oral-Head and Neck Surgery/Microvascular further treatment. This fear along with the previ-
Reconstructive Surgery, Department of Oral and ous resultant physical sequelae such as trismus,
Maxillofacial Surgery, University of Maryland, decreased salivary function, dysarthria, dyspha-
Baltimore, MD, USA gia, or disfigurement will pose significant com-
e-mail: jlubek@umaryland.edu
plexities to quality of life considerations that will
M. Nagai factor into patient treatment decisions.
Oral-Head and Neck Surgery/Microvascular Surgery,
University of Maryland, Baltimore, MD, USA Meta-analysis and pooled prospective data for
e-mail: mnagai@umaryland.edu upper aerodigestive squamous cell carcinoma

https://doi.org/10.1007/978-3-030-30094-4_11
156 J. E. Lubek and M. Nagai
suggest that for stage I and II recurrences, the was increased risk for decreased survival. The
2-year recurrence-free survival is around 70% authors also found that recurrence after age
with 60–85% achieving or exceeding presurgical greater than 60 years was associated with
quality of life (QOL), with a modest complica- increased survival [14].
tion rate of 6%. For stage III disease, the 2-year Regional failures are those that occur within
recurrence-free survival drops to 30% with the the previously treated or untreated draining lym-
rates of significant complications also reaching phatic cervical system. In general, the first eche-
the 30th percentile and only 40% of patients lon lymphatics for oral squamous cell cancer are
achieving or exceeding their baseline QOL. For considered to be levels I–III of the anterior neck
stage IV disease recurrences, the outcomes were (submental, submandibular, and supraomohyoid
considerably worse with less than 25% of patients jugular chain lymphatics). Drainage patterns can
living disease-free for greater than 2 years and be altered in cases of previous neck treatment
more than half of patients dying within 9 months (surgical lymphadenectomy or radiotherapy)
[10]. Ultimately final treatment decisions will be resulting in failure to unusual locations such as
based upon long-term prognosis and patient mor- parotid or axillary lymphatics.
bidity as the guide to help stratify those patients Recurrence within the local site or neck will
who will benefit from salvage therapy with a depend on numerous factors to include margin
curative intent versus discussions for a palliative status, perineural invasion, tumor depth of inva-
approach. sion (DOI), tumor size and location within the
The classic definition of a locally recurrent oral cavity, pre-existing neck nodal disease, and
head and neck cancer is any lesion that is located extranodal tumor extension (ENE). Initial treat-
within a 2 cm distance from the index tumor or ment strategies employed will also factor into
that which occurs within 3 years of the index recurrence rates, salvage options, and overall
tumor [11]. This certainly is true for most recur- patient outcomes. Most series would demonstrate
rent OSCC; however, recent data would indicate that single modality therapy, upfront surgery,
that aggressive active surveillance is required for negative surgical margins, and the pathologically
much longer duration, as late recurrence and sec- node negative (N0) neck not only do better in
ond primary cancers both within the oral cavity terms of initial prognosis but also will have
and other sites within the head and neck can improved overall survival (OS) and disease-free
occur even in patients initially treated for very survival (DFS) when undergoing salvage therapy
early stage disease. In a series of 112 Stage I oral in those cases that do recur [15, 16]. Benefits of
squamous cell carcinomas, there was a 19.6% the elective neck dissection in terms of improved
incidence of late local failure (>36 months) survival and decreased neck failure were initially
despite 75% and 86% of patients having negative suggested by studies such as Kligerman et al.
margins (≥5 mm and ≥3 mm, respectively). The [17], with confirmatory results in a recent large
authors also noted a 11.6% rate of second prima- prospective trial by D’Cruz et al. [18] Patients in
ries within the head and neck [12]. these studies with OSCC who underwent elective
Time to recurrence is considered an important neck dissection in the N0 neck with DOI ≥ 3 mm
factor both in the prognosis of disease and in the not only OS and DFS, but recurrence within the
ability to successfully salvage a patient. Various dissected neck tended to be smaller burden of
studies suggest that recurrent disease within a disease. Failure in the dissected neck does bring
6-month interval is a significantly poor prognos- with a worse prognosis, and rates of successful
tic marker. Although this statement holds true, salvage are more controversial.
one could ask if this is truly recurrent disease ver- Margin status following surgical resection has
sus recidivistic occult disease [13]. In a large been considered to be a risk factor for recurrence
series of over 1600 patients with OSCC, Liu et al. within OSCC. In a meta-analysis by Anderson
reported a local recurrence rate of 31%, and that et al. Recurrence rates were pooled to give a 21%
recurrence within 18 months of initial surgery absolute risk reduction in local recurrence with
11 Recurrent Oral Cancer and Salvage Options 157
margins clear by more than 5 mm [19]. This has One should have a low threshold for rebiopsy of
been challenged with possibility of margins persistent or new oral leukoplakia, erythroplakia,
greater than 3 mm as adequate. Further contro- or ulcerations. Surveillance imaging can be very
versy exists with the finding of dysplasia at the helpful especially in situations of altered anat-
surgical margin and risk of recurrent OSCC. Most omy (flap reconstruction), trismus, or radiation.
authors would agree that both carcinoma-in-situ Computerized tomography (CT) imaging and
and severe dysplasia are increased risk factors for magnetic resonance imaging (MRI) are generally
recurrence; however, the presence of mild or first-line surveillance. Positron emission tomog-
moderate dysplasia is of more questionable sig- raphy (PET) is useful for detection of early recur-
nificance [20, 21]. In a series by Pu et al. of over rence prior to any anatomic changes with other
500 patients who were surgically treated for imaging modalities but can generate a false-
OSCC, positive or dysplastic margins were iden- positive finding if performed too early after sur-
tified in 20% of patients [22]. The authors con- gery or radiation due to inflammation [26]. A
cluded on multivariate analysis there was no 2016 meta-analysis of prospective studies involv-
significant difference in recurrence-free survival ing PET/CT for detection of recurrent head and
(RFS) or disease-free survival (DFS) in patients neck cancers concluded that PET/CT imaging
with mild dysplasia and negative margins; how- 3 months post treatment was beneficial in the
ever, patients with mild dysplasia who did not detection of residual or recurrent disease.
undergo re-excision demonstrated significantly Sensitivity and specificity of residual and recur-
worse DFS and RFS. Another series by rent regional and distant disease were 72.3%,
Gokavarapu et al. analyzed 425 patients with 88.3% and 84.6%, 94.9%, respectively [27]. Lin
OSCC resected with negative surgical margins et al. retrospectively reviewed 111 patients with
and revealed that patients with mild and moder- advanced resected OSCC who developed recur-
ate dysplasia at the margin had lower rates of sur- rences following adjuvant therapy and surveil-
vival with moderate dysplasia being an lance PET/CT imaging [28]. The authors
independent risk factor for survival [23]. As the concluded that scheduled periodic PET/CT sur-
data would suggest, dysplasia at the margin likely veillance is a valuable tool for early detection of
represents field cancerization change within the recurrent lesions in asymptomatic OSCC patients
oral cavity and would seem logical that there is as the presence of clinical symptoms and a short
increased risk of recurrence. time to positive PET/CT findings were adverse
Recurrence that occurs distantly is associated prognostic factors for clinical outcome in patients
with a poor long-term prognosis. Considerations with advanced OSCC.
for treatment should be based upon patient life
expectancy, disease burden, and treatment with
the specific purpose of addressing quality of life 11.3 Surgical Salvage
issues. Patients faced with both local recurrence
and simultaneous distant metastatic disease Prior to deciding on a surgical option, one must
should be considered for definitive treatment for consider the previous treatment rendered, time to
local source control to improve quality of life and recurrence, location of the recurrence, ease of
to minimize systemic palliative therapies and surgical access, ability to achieve the surgical
tumor burden. Clinical trials should also be con- objective (i.e., negative surgical margin), need for
sidered in these situations as well [24–26]. reconstruction, and patient comorbidities.
Surveillance for recurrent OSCC requires For patients who recur locally within the oral
both a thorough clinical exam to include indirect cavity, complete surgical resection should be
mirror exam and fiberoptic/rigid endoscopy espe- considered as a first-line therapy. With active sur-
cially in patients with trismus and periodic imag- veillance, these patients can be identified early
ing. Very often new symptoms such as pain or and surgery offers a good chance of cure, espe-
swelling should alert the clinician to a recurrence. cially in low-volume disease. In a recent
p ublished series, Ord et al. reported on a series of received adjuvant radiotherapy, there was no
112 patients with T1 oral tongue carcinoma overall or DFS benefit following recurrence [31].
patients followed long-term [12]. Of the 19.6% Matsuura et al. [16] compared a cohort of 46
of patients who recurred locally and of those 50% patients with recurrent OSCC who underwent
(n = 11/22) were salvaged with a second surgery salvage versus 199 patients without recurrence
requiring no further treatment. The other 50% following primary therapy. Independent risk fac-
continued to develop multiple failures but were tors for poor prognosis included positive lymph
able to be successfully salvaged with surgery node metastases and positive surgical margins.
provided they did not recur regionally. Patients who developed a recurrence and were
Surveillance for local disease becomes more successfully salvaged had similar overall survival
problematic in situations that make clinical exam as compared to those who never developed a
more difficult such as in cases of pre-existing flap recurrence (54.7% vs. 70.7%, overall survival
reconstruction, trismus, adjuvant radiotherapy, or p = 0.158 respectively). The authors also note
disease occurring within the maxilla. that patients who continued to develop recur-
Regional recurrence within the neck can be rences after salvage surgery had similar overall
divided into two categories to include recurrence survival for those who received palliative
within the previously untreated neck and those treatment.
that have received treatment. This is further com- It cannot be overstated that the surgical and
plicated as the previously treated neck can be fur- reconstructive undertaking for these patients can
ther subdivided into patients who were treated be daunting. They often require many adjunct
with definitive surgery alone, definitive radio- services during the course of treatment, and ade-
therapy, or multimodality therapy. Patients who quate supportive care is a must. It is therefore
have only undergone previous surgery generally recommended that these complex cases be com-
have better rates of successful salvage as the pleted at centers that regularly manage this type
option for combined modality therapy still exists of disease and have the available resources as
(i.e., surgery +/− chemoradiotherapy) as opposed well as experienced surgical and reconstructive
to those treated initially with radiotherapy. teams. It is important to note that advances in
In a series by Wong et al. patients who under- reconstructive techniques and standardization
went salvage surgery were found to have signifi- among perioperative protocols at many centers
cant improved recurrent survival time when has increased the number of candidates available
compared to radiotherapy/chemoradiotherapy. to undergo salvage surgery and decreased the
Some may argue that these were patients who incidence of some complications. In the 1960s
were more likely to have unresectable recurrent and 1970s, Gilbert and Kagan found that only
disease thus biasing the statistical analysis [29]. 18.3% of patients with recurrent carcinoma of the
In a 2018 series from Memorial Sloan Kettering, oral cavity or oropharynx were considered eligi-
190 recurrent OSCC patients (all were initially ble for salvage surgery due to limitations related
N0) were evaluated [30]. The authors reported a to reconstructive methods [32]. With the advance-
15% recurrence rate within the neck and that sig- ment of microvascular techniques and availabil-
nificantly poorer outcomes in DFS as compared ity of reliable donor site options this percentage
to those who did not recur in the neck (32% vs. has dramatically increased, and free flaps are
74% DFS, respectively). Patients that were able now becoming the standard of care in many sal-
to receive combined surgery and adjuvant radio- vage operations allowing patients options never
therapy +/− chemotherapy had better overall sur- before seen with functional outcomes that are
vival as compared to those who received either exceptional. Dense, fibrotic scar tissue from prior
surgery or radiotherapy alone. A smaller series of surgery, radiation or a combination of both can
24 OSCC recurrences by Skelenica et al. noted make conventional local tissue flaps and nonvas-
that salvage surgery significantly improved over- cularized grafts unreliable for reconstruction of
all survival; however, in the 50% of patients who these defects. Bringing nonirradiated tissue with
its native vascular supply to the site of recurrence patients with early T stage tumors, longer disease-
can lessen the burden of tissue healing and reduce free intervals, and favorable comorbidity profile
local wound complications. Although it may will have a high likelihood of success.
increase surgical time and is associated with
some additional morbidity, the risks of aggres-
sive surgery are often outweighed by the 11.4 Radiation Therapy
improved quality of life, reported by many for Salvage
patients, following salvage surgery. Over the last
decade, the utility, reliability, and function of free Adjuvant radiotherapy should be utilized in a
tissue transfer has had a significant impact on the standard fashion per guidelines for those
functional outcomes for those patients undergo- patients considered radiation-naïve who recur
ing recurrent cancer care [33]. In a recent study locoregionally during standard surveillance.
by Smirk and Kyzas, 25 of 29 patients underwent Patients who meet criteria include advanced T
salvage surgery with free flap reconstruction. The stage disease, perineural/lymphovascular inva-
authors reported that surgical complications were sion, close margins, and nodal disease/
relatively uncommon with only one total flap ENE. Recurrent disease within the neck should
failure [34]. Both this current study and a series be treated initially with surgery if possible, irre-
by Kostrzewa et al. found that over 50% of spective of previous neck dissection or suspi-
patients were gastrostomy-dependent after sal- cion of extranodal disease, taking into
vage surgery, highlighting the high risk of long- consideration patient morbidity. This approach
term dysphagia despite successful salvage will help reduce tumor burden and increase the
surgery and reconstruction [35]. This impact will effectiveness of adjuvant radiotherapy. Local
only continue to grow as the population ages and recurrence that has been resected to a negative
the treatment of cancers improve; patients will be surgical margin without adverse features should
living longer with the desire to surgically treat continue with close surveillance only [36].
recurrences when feasible. Patients who present with early recurrence or
Surgical salvage can also include palliative even recidivistic OSCC before planned postop-
options such as tumor debulking, ablation with- erative radiotherapy (PORT) present with a diffi-
out reconstruction, and the creation of an elective cult challenge. The question as to why these
surgical airway for prolonged comfort in an patients have developed disease so early within
attempt to improve quality of life. Noncurative the primary treatment window is paused for con-
surgery, which reduces the primary tumor bulk cern. Perhaps surgical wounds have not fully
can reduce pain and bleeding, improve swallow- healed, or patients are recovering in rehabilita-
ing and nutrition while enhancing airway patency. tion facilities and are too weak to start adjuvant
Surgical resection with application of a simple therapy, or maybe the patient is delaying therapy
reconstruction bar or coverage with a pedicled for personal reasons. Regardless this early recur-
flap such as a pectoralis or supraclavicular may rence whether within the local site, neck or as
reduce operating time, blood loss and provide dermal or distant metastases should alert the cli-
sufficient reconstruction of certain defects. In nician as to the aggressiveness of the tumor. A
select settings, using newer endovascular tech- recent series by Hosni et al. analyzed a cohort of
niques, embolization and vessel stenting may OSCC patients of which 88 patients (15% of
offer symptom control for bleeding related to patients in the series) developed early recurrence
major vascular erosion.[24, 25]. Following recur- while awaiting the start of PORT. The authors
rences, surgical salvage should be explored as the divided the patients into two cohorts; 70 patients
first-line option for treatment, especially if the in the salvage group (locoregional disease only)
goal is cure. Despite increased morbidity and and 18 patients in the palliative group (had devel-
lower success rates compared with surgery in the oped locoregional and distant disease) [37].
setting of primary disease, surgery among those Radiation dosages were adjusted accordingly to
the specific groups. The authors concluded that induced adverse events outweigh its therapeutic
significant risk factors for early recurrence use and alternative treatments should be sought.
included the oral tongue subsite and microscopic Ionizing radiation can have a significant impact
positive margin. The 3-year OS salvage and pal- on the native tissues of the head and neck, and the
liative rates were 71% and 41%, respectively. risks of serious complications such as carotid ste-
Disease-free survival following salvage PORT nosis or rupture, neurosensory loss, speech defi-
was 36% with extranodal disease and volume of cits, dysphagia, and osteoradionecrosis must be
gross disease being associated with poor DFS carefully assessed against any potential benefits,
following salvage PORT. This study certainly especially in terms of overall survival. In a large
highlights the importance of “total treatment multicenter review of reirradiation for recurrent
package time” (85–100 days) with longer head and neck cancers, Caudell et al. showed a
surgery- PORT interval times greater than 22.1% risk of grade ≥3 complications, and that
6 weeks increasing the odds risk for local failure late toxicities continue to increase up to 5 years
[38]. following reirradiation, plateauing at 48–66%
For patients who may be candidates for sur- [41]. Although there is no general consensus
gery but have recurrent tumors that are not ame- regarding the total dose, fractionation scheme, or
nable to salvage due to anatomic considerations, mode of delivery, most centers will reirradiate
or for patients who are medically unfit for sur- qualifying patients to doses >50 Gy and will
gery, then reirradiation with or without concur- often limit the field to the site of recurrence [42,
rent chemotherapy may offer a survival benefit or 43]. Another consideration is potential retreat-
a palliative alternative [39]. Patient selection is ment with proton beam therapy. The unique
critical, and when choosing reirradiation as a physical properties of this type of radiation allow
modality, then the risk of acute and late toxicity for higher doses to be delivered to tumors, while
must be appreciated. Also important is the assess- minimizing the dose to previously irradiated nor-
ment of tumor characteristics, prognosis, and mal tissues. While the initial evaluations have
how willing the patient is to undergo a potentially been promising [44, 45], further prospective tri-
toxic regime as treatment-related fatalities from als with longer follow-up times are needed to
bleeding, tissue necrosis, and infection can be assess the efficacy, tolerability, and cost effec-
substantial. Historically, radiation was thought to tiveness of this treatment modality.
be a single use treatment, but with the advance- Another potential use of reirradiation is in the
ments in modern conformational radiotherapy palliative setting. Although data are limited, pallia-
techniques such as IMRT and volumetric- tive radiation could afford patients not candidates
modulated arch therapy, the therapeutic ratio of for aggressive reirradiation; in specific situations,
reirradiation has changed such that a select group treatment options geared toward improving symp-
of people may benefit from a second round of toms. An effective palliative approach that mini-
treatment, and it is considered safe to do so [40]. mizes the burden of longer protracted courses of
Those that may be candidates for reirradiation treatment is considered in this approach [26].
include patients who have undergone surgical Often, lower doses of radiation with increased
salvage with their tumors showing high-risk fea- fractions are delivered in an attempt to reduce
tures on final pathology, patients who have surgi- acute toxicity with the hope that late complications
cally unresectable disease, or patients whose with increased fraction size will be irrelevant [46,
performance status limits their operative choices. 47]. Alleviating pain, dysphagia, bleeding, and
It is generally considered that tumors which have impending airway compromise should be the
recurred at the site of previous radiation to a full goals of treatment and can have a dramatic impact
treatment dose (>50 Gy) within 6 months of ther- on a patient’s outlook, ability to function, and live
apy have radioresistant disease and would be in a comfortable setting.
unlikely to benefit from additional dosing. In Patients who have undergone treatment of
these patients, the potential risks of radiation- head and neck cancer pose a difficult therapeutic
challenge. Surgical clearance of the recurrent or needed [48]. Treatment options include single-
second primary disease will remain the preferred agent therapy and combination regimens using
treatment modality of choice, but when disease either conventional cytotoxic chemotherapy and/
and patient specific factors are critically evalu- or molecularly targeted agents combined with
ated, then the addition of reirradiation can best supportive care. Checkpoint inhibitor immu-
become an important aspect of achieving the notherapy is an option for patients with progres-
greatest survival outcomes. Patients undergoing sive disease after they have failed initial
salvage surgery with curative intent should be platinum-based chemotherapy. Currently, con-
considered for repeat radiotherapy when high- current chemoradiotherapy over chemotherapy
risk pathologic features are encountered. These or radiation alone is favorable and has demon-
include positive margins, perineural invasion, strated survival benefits in multi-institutional
lymphovascular invasion, and extranodal exten- group settings [49]. There are a multitude of
sion. Radiation alone or the addition of radiation therapeutic drug classes available and include
to supplement treatment has well-recognized conventional cytotoxic platinum containing com-
benefits. This, however, must be balanced against pounds (cisplatin and carboplatin), taxanes
the risks associated with the radiation, and each (docetaxel, paclitaxel, and nabpaclitaxel), metho-
treatment plan will need to be tailored to each trexate, fluorouracil, and the monoclonal EGFR
individual patient and their overall therapeutic receptor inhibitor cetuximab. Varying combina-
objectives. Eckardt et al. compared the survival tions of these medications have been used with
of patients treated with surgery with that of and without reirradiation, and with the exception
patients treated with multimodality treatment of combining cetuximab with a platinum-based
over a 20-year period and found a 31.0% rate of compound as shown in the EXTREME trial [50],
salvage with surgery as compared with a 15.4% no specific treatment algorithm has shown any
rate with surgery and radiotherapy, with a salvage significant survival benefit over another [51].
rate of 0% for patients treated with radiotherapy When choosing a systemic therapy, prognostic
or supportive therapy alone [7]. indicators, such as ECOG performance status,
histologic differentiation of the tumor, and prior
response to any therapeutic agents, along with
11.5 Systemic Therapy patient-specific co-morbidities need to be taken
and Immunotherapy into consideration to help guide prescribing deci-
sions and avoid excessive toxicities.
Systemic therapy to treat head and neck cancer Immunotherapy has been one of the most dis-
has been well established and has recently seen cussed and potentially exciting treatments for
promising advancements with the development recurrent advanced stage OSCC in recent years,
of targeted immunotherapy which has begun to and the antitumor defense of these medications is
play a significant role in the management of those widely recognized. Although it is still considered
affected by advanced-stage recurrent disease. a second-line therapy, the mechanisms through
Salvage therapy options for patients with unre- which these agents work are believed to be the
sectable, previously irradiated cancers of the key toward recognizing improved host recogni-
head and neck are limited, and ultimately most tion and destruction of cancer, ultimately result-
patients who recur will require some form of pal- ing in effective and sustainable long-term survival
liative systemic therapy. At present, chemother- outcomes. There has been a paradigm shift in the
apy is the treatment most often used in this way cancer is viewed, from a biologic and host
population; however, the optimal regimen for interaction standpoint to the way treatment strat-
these patients has not been clearly defined in ran- egies are being implemented. In 2016, based on
domized clinical trials. Unfortunately, median the results of the phase III CHECKMATE 141
survival with chemotherapy alone has approxi- trial and the phase I KEYNOTE 012 trial, the
mated 6 months, and alternative strategies are FDA approved the PD-1 checkpoint inhibitors
nivolumab (Opdivo, Bristol-Myers Squibb) and results of trials become available, the stratifica-
pembrolizumab (Keytruda, Merck), respectively, tion of particular tumors and or patients into
for the treatments of patients with recurrent or specific treatment groups will hopefully result
metastatic head and neck squamous cell carci- in a survival benefit and loss of morbidity.
noma that had progressed after treatment with While surgical salvage will continue to be the
chemotherapy [52, 53]. Ferris et al. reported foundation upon which additional treatment
long-term follow-up data of nivolumab versus choices can be applied, promising novel thera-
chemotherapy and noted an overall 2-year sur- pies are on the horizon and undergoing evalua-
vival of 16.9% in the immunotherapy PD-L1 tion in large-scale prospective clinical trials,
group versus a 6% overall survival in the chemo- which may alter the scope of the current day
therapy cohort. The nivolumab group also approach to cancers of the oral cavity. The
reported over 50% less side effects and signifi- oncologic team now has multiple tools at their
cant toxicities as compared to the chemotherapy disposal to help patients throughout the course
arm. These immune checkpoint inhibitors may of their treatment and all options should be
represent a ground-breaking advance for treat- explored when delivering care. Treatment
ment of solid tumors including head and neck plans need to be developed in a multidisci-
cancers that exhibit poor responses to chemo- plinary setting and reflect more than just sur-
and radiotherapy [54]. Currently, numerous vival outcomes. Patient input and preferences
ongoing clinical trials are investigating various should be a large component of the decision to
immunotherapy agents—alone and in combina- apply any therapy in the recurrent setting.
tion—for management of patients with head and Quality of life needs to be at the core of the
neck cancer. Encouraging and pertinent clinical decision-making process, and aggressive tumor
results have been achieved in head and neck can- surveillance with close patient recall needs to
cers, but many challenges remain for the clinical be performed to identify those patients that can
impact of immunotherapy to be improved: pre- be identified with local recurrence and poten-
dictive biomarkers are needed for patient selec- tially be successfully salvaged.
tion, and associations of several immunotherapies
or with conventional drugs need to be tested [35].
At this time, only 10–20% of patients will be References
responders to immune therapy, and this must be
weighed against the high financial costs of these 1. National Cancer Institute. Surveillance, epidemiology
drugs on the healthcare system. As further insight and end results program. cancer stat facts: oral cavity
into the tumor-host microenvironment is gained and pharynx cancer. https://seer.cancer.gov/statfacts/
html/oralcav.html.
and more specific targeted therapies are devel- 2. Patel SC, Carpenter WR, Tyree S, Couch ME,
oped, there hopes to be a substantial improve- Weissler M, Hackman T, et al. Increasing incidence
ment in long-term survival while at the same time of oral tongue squamous cell carcinoma in young
reducing adverse events. white women, age 18 to 44 years. J Clin Oncol.
2011;29:1488–94.
3. Farquhar DR, Tanner AM, Masood MM, et al. Oral
tongue carcinoma among young patients an analysis
11.6 Summary of risk factors and survival. Oral Oncol. 2018;84:7–11.
4. Hussein AA, Helder MN, de Visscher JG, et al. Global
incidence of oral and oropharynx cancer in patients
For patients with recurrent oral cancer, the younger than 45 years versus older patients: a system-
prognosis remains poor. It is clear that a better atic review. Eur J Cancer. 2017;82:115–27.
understanding of the molecular and biological 5. Haas I, Hauser U, Ganzer U. The dilemma of fol-
interactions between the host and various low-up in head and neck cancer patients. Eur Arch
Otorhinolaryngol. 2001;58:177–83.
tumor types is needed and will facilitate the 6. Ho AS, Kraus DH, Ganly I, et al. Decision making
development of more precisely targeted agents. in the management of recurrent head and neck cancer.
As further biomarkers are discovered and Head Neck. 2014;36:144–51.
7. Eckardt A, Barith EL, Kokemueller H, et al. Recurrent cancer. Oral Surg Oral Med Oral Pathol Oral Radiol.
carcinoma of the head and neck: treatment strategies 2017;123:330–7.
and survival analysis in a 20-year period. Oral Oncol. 24. Cocks H, Ah-See K, Capel M, et al. Palliative and sup-
2004;40:427–32. portive care in head and neck cancer: United Kingdom
8. Liao CT, Chang JT, Wang HM, et al. Salvage therapy national multidisciplinary guidelines. J Laryngol Otol.
in relapsed squamous cell carcinoma of the oral cav- 2016;130:198–207.
ity: how and when? Cancer. 2008;112:94. 25. Kim J, Kim S, Albergotti W, et al. Selection of ideal
9. Ord RA, Kolokythas A, Reynolds MA. Surgical sal- candidates for surgical salvage of head and neck
vage for local and regional recurrence in oral cancer. J squamous cell carcinoma: effects of the Charlson
Oral Maxillofac Surg. 2006;64:1409–14. age comorbidity index and oncologic characteris-
10. Goodwin WJ Jr. Salvage surgery for patients with tics on a 1-year survival and hospital course. JAMA
recurrent squamous cell carcinoma of the upper Otolaryngol Head Neck Surg. 2015;141:1059–65.
aerodigestive tract: when do the ends justify the 26. Lok BH, Jiang G, Gutiontov S, et al. Palliative head
means? Laryngoscope. 2000;110:1–18. and neck radiotherapy with the RTOG 8502 regi-
11. Warren S, Gates DC. Multiple malignant primary men for incurable primary or metastatic cancers. Oral
tumors: a survey of the literature. Am J Cancer. Oncol. 2015;51:957–62.
1932;16:1358–414. 27. Cheung PK, Chin RY, Eslick GD. Detecting residual/
12. Ord RA, Isaiah A, Dyalram D, Lubek JE. Is long term recurrent head neck squamous cell carcinomas using
follow up mandatory for stage I oral tongue cancer? PET or PET/CT systematic review and meta analysis.
J Oral Maxillofac Surg. 2018;76:2676–83. https://doi. Otolaryngol Head Neck Surg. 2016;154:421–32.
org/10.1016/j.joms.2018.06.169. 28. Lin HC, Kang CJ, Huang SF, et al. Clinical impact
13. Schwartz GJ, Mehta RH, Wenig BL, et al. Salvage of PET/CT imaging after adjuvant therapy in patients
treatment for recurrent squamous cell carcinoma of with oral cavity squamous cell carcinoma. Eur J Nucl
the oral cavity. Head Neck. 2000;22:34–41. Med Mol Imaging. 2017;44:1702–11.
14. Liu SA, Wong YK, Lin JC, et al. Impact of recurrence 29. Wong LY, Wei WI, Lam LK, Yuen AP. Salvage of
interval on survival of oral cavity squamous cell car- recurrent head and neck squamous cell carcinoma after
cinoma patients after local relapse. Otolaryngol Head primary curative surgery. Head Neck. 2003;25:953–9.
Neck Surg. 2007;136:112–8. 30. Mizrachi A, Migliacci JC, Montero PE, et al. Neck
15. Brandwein-Gensler M, Teixeira MS, Lewis CM, et al. recurrence in clinically node negative oral cancer: 27
Oral squamous cell carcinoma: histologic risk assess- year experience at a single institution. Oral Oncol.
ment, but not margin status, is strongly predictive of 2018;78:94–101.
local disease-free and overall survival. Am J Surg 31. Skelenica S, Gardiner S, Dierks EJ, et al. Survival
Pathol. 2005;29:167. analysis and risk factors for recurrence in oral squa-
16. Matsuura D, Valim TD, Kulcsar MAV, et al. Risk fac- mous cell carcinoma does surgical salvage affect out-
tors for salvage surgery failure in oral cavity squamous come. J Oral Maxillofac Surg. 2010;68:1270–5.
cell carcinoma. Laryngoscope. 2018;128:1113–9. 32. Gilbert H, Kagan AR. Recurrence patterns in squa-
17. Kligerman J, Lima RA, Soares JR, et al. mous cell carcinoma of the oral cavity, pharynx and
Supraomohyoid neck dissection in the treatment of larynx. J Surg Oncol. 1974;6:357–80.
T1/T2 squamous cell carcinoma of the oral cavity. Am 33. Knoetgen J, Choudry U, Finical SJ, et al. Head and
J Surg. 1994;168:391. neck reconstruction with a second free flap following
18. D’Cruz AK, Vaish R, Kapre N, et al. Elective versus resection of a recurrent malignance. Ann Plast Surg.
therapeutic neck dissection in node negative oral can- 2005;55:378–83.
cer. N Engl J Med. 2015;373:521. 34. Smirk R, Kyzas P. Outcome of salvage procedures
19. Anderson CR, Sisson K, Moncrieff M. A meta-analysis for recurrent oral and oropharyngeal cancer. Br J Oral
of margin size and local recurrence in oral squamous Maxillofac Surg. 2018;56:847–53.
cell carcinoma. Oral Oncol. 2015;51:464–9. 35. Kostrzewa JP, Lancaster WP, Iseli TA, et al. Outcomes
20. Shah A. Postoperative pathologic assessment of surgi- of salvage surgery with free flap reconstruc-
cal margins in oral cancer a contemporary review. J tion for recurrent oral and oropharyngeal cancer.
Oral Maxillofac Pathol. 2018;22:78–85. Laryngoscope. 2010;120:267–72.
21. Sopka DM, Li T, Lang MN, et al. Dysplasia at the mar- 36. Rosenthal DI, Mohamed ASR, Garden AS, et al. Final
gin? Investigating the case for subsequent therapy in report of a prospective randomized trial to evaluate the
‘low-risk’ squamous cell carcinoma of the oral tongue. dose response relationship for postoperative radiation
Oral Oncol. 2013;49(11):1083–7. therapy and pathologic risk groups in patients with
22. Pu Y, Wang Y, Huang X, et al. The influence of mild head and neck cancer. Int J Radiat Oncol Biol Phys.
dysplasia at the surgical margin on the prognosis of 2017;98:1002–11.
oral squamous cell carcinoma. Int J Oral Maxillofac 37. Hosni A, Huang SH, Chiu K, Xu W, Su J, Bayley A,
Surg. 2016;45:1372–7. Bratman SV, Cho J, Giuliani M, Kim J, O’Sullivan B,
23. Gokavarapu S, Parvataneni N, Pavagada S, et al. Ringash J, Waldron J, Spreafico A, Yu E, de Almeida
Mild to moderate dysplasia at surgical margin is JR, Monteiro E, Chepeha DB, Irish JC, Goldstein DP,
significant indicator of survival in patients with oral Hope A. Predictors of early recurrence prior to planned
post-operative radiation therapy for oral cavity squa- an optimal fractionation scheme. Head Neck.
mous cell carcinoma and outcomes following salvage 2008;30:1586–91.
intensified radiation therapy. Int J Radiat Oncol Biol 47. Corry J, Peters LJ, Costa ID, et al. The ‘QUAD
Phys. 2019, 103:363–73. https://doi.org/10.1016/j. SHOT’—a phase II study of palliative radiotherapy
ijrobp.2018.09.013. for incurable head and neck cancer. Radiother Oncol.
38. Rosenthal DI, Liu L, Lee JH, et al. Importance of the 2005;77:137–42.
treatment package time in surgery and postoperative 48. Jacob C, Layman G, Velez-Garcia E, et al. A phase III
radiation therapy for squamous carcinoma of the head randomized study comparing cisplatin and fluoroura-
and neck. Head Neck. 2002;24:115–26. cil as single agents and in combination for advanced
39. McDonald MW, Lawson J, Garg MK, et al. ACR squamous cell carcinoma of the head and neck. J Clin
appropriateness criteria retreatment of recurrent head Oncol. 1992;10:257–63.
and neck cancer after prior definitive radiation expert 49. Spencer SA, Harris J, Wheeler RH, et al. Final report
panel on radiation oncology head and neck cancer. Int of RTOG 9610, a multi-institutional trial of reirra-
J Radiat Oncol Biol Phys. 2011;80:1292. diation and chemotherapy for unresectable recurrent
40. Sulman EP, Schwartz DL, Le TT, et al. IMRT reir- squamous cell carcinoma of the head and neck. Head
radiation of head and neck cancer – disease con- Neck. 2008;30:281–8.
trol and morbidity outcomes. Int J Radiat Oncol. 50. Vermorkin JB, Mesia R, Rivera F, et al. Platinum-
2009;73:399–409. based chemotherapy plus cetuximab in head and neck
41. Caudell JJ, Ward MC, Riaz N, et al. Volume, dose and cancer. N Engl J Med. 2008;359:1116–27.
fractionation considerations for IMRT based reirradia- 51. Forastiere AA, Metch B, Schuller DE, et al.
tion in head and neck cancer: a multi-institution analy- Randomized comparison of cisplatin plus fluorouracil
sis. Int J Radiat Oncol Biol Phys. 2018;100:606–17. and carboplatin plus fluorouracil versus methotrexate
42. Merlotti A, Mazzola R, Alterio D, et al. What is the in advanced squamous cell carcinoma of the head and
role of post operative re-irradiation in recurrent and neck: a southwest oncology group study. J Clin Oncol.
second primary squamous cell cancer of the head and 1992;10:1245.
neck? A literature review according to PICO criteria. 52. Ferris RL, Blumenschein G, Fayette J, et al. Nivolumab
Crit Rev Oncol Hematol. 2017;111:20. vs investigators choice in recurrent metastatic squa-
43. Popovtzer A, Gluck I, Chepeha DB, et al. The pattern mous cell carcinoma of the head and neck: 2 year long
of failure after reirradiation of recurrent squamous cell term survival update of CheckMate 141 with analyses
head and neck cancer: implications for defining the by PD-L1 expression. Oral Oncol. 2018;81:45–51.
targets. Int J Radiat Oncol Biol Phys. 2009;74:1342–7. 53. Hsu C, Lee S-H, Ejadi S, et al. Antitumor activity and
44. Phan J, Sio TT, Nguyen TP, et al. Reirradiation of head safety of pembrolizumab in patients with PD-L1 posi-
and neck cancers with proton therapy: outcomes and tive nasopharyngeal carcinoma: interim results from a
analysis. Int J Radiat Oncol bio Phys. 2016;96:30–41. phase 1b study. Ann Oncol. 2015;26:93–102.
45. Romesser PB, Cahlon O, Scher ED, et al. Proton 54. Outh-Gauer S, Alt M, Le Tourneau C, et al.
beam reirradiation for recurrent head and neck cancer: Immunotherapy in head and neck cancers: a new chal-
multi-institutional report on feasibility and early out- lenge for immunologists, pathologists and clinicians.
comes. Int J Radiat Oncol Bio Phys. 2016;95:386–95. Cancer Treat Rev. 2018;65:54–64.
46. Chen AM, Vaughn A, Narayan S, et al. Palliative
radiation therapy for head and neck cancer: toward
Oral Reconstruction
12
J. Collin, B. Turner, and R. Fernandes
12.1 General Considerations commonly split or full thickness skin, but also
mucosal grafts. The fourth rung includes local
The general concept behind reconstructive surgery advancement or rotation flaps supplied by a ran-
is to assess what is missing and what is available to dom vascular pattern. The fifth includes regional
replace it, bearing in mind that where possible pedicled flaps. These flaps provide a greater area
“like should replace like.” For any given defect, of soft tissue coverage than is possible with local
the reconstruction options range from simple to flaps in the mouth, without requiring the facilities
complex, along a spectrum frequently referred to and expertise for microvascular reconstruction.
as the reconstructive ladder [1]. The reconstructive They are frequently thought of as less susceptible
matrix is an elaboration of this linear concept that to failure than microvascular flaps, but it should
considers technical requirements and the potential be pointed out that this is not necessarily the case.
risks to the patient on additional axes [2]. For example, pectoralis major myocutaneous
At the bottom of the reconstructive ladder, flaps are associated with a partial necrosis rate of
some very small defects, especially of the tongue, up to 29% [3, 4]. Furthermore, while pedicled
can be left to heal by secondary intention. This is flaps are versatile, they are sometimes esthetically
rarely appropriate for malignant lesions, as even inferior to free flaps. The final rung is free tissue
the theoretical lesion of zero volume would still transfer. Free flaps excel the previously mentioned
require enough margin to preclude this approach. options with respect to versatility, size, and some-
The second rung is primary closure. This approach times cosmetic and functional outcomes. Free
is appropriate for small defects that can be closed flaps are available to close any size of defect in the
without significant functional deficit. It should be oral cavity, and their freedom from a regional
borne in mind that some small defects, particu- pedicle allows them to be placed in any location.
larly of the tongue, will have better functional Their disadvantages include the need for costly
results with primary closure than with a flap of resources and expertise, increased operative time,
excessive bulk for the defect. Similarly, a small lip donor site morbidity, and a low, but nonzero rate
defect can be closed primarily in a highly esthetic of flap loss.
manner. The third rung is occupied by grafts,
12.2 Reconstruction of the Lips

J. Collin · B. Turner · R. Fernandes (*)
University of Florida College of Medicine
The goals of lip reconstruction should be both
Jacksonville, Jacksonville, FL, USA
e-mail: John.Collin@jax.ufl.edu; Benjamin.Turner@ functional and esthetic. Oral competence is
jax.ufl.edu; Rui.Fernandes@jax.ufl.edu important during mastication and the oral stage
https://doi.org/10.1007/978-3-030-30094-4_12
166 J. Collin et al.
of deglutition, particularly with fluids, but also 12.3 Reconstruction

when at rest to prevent drooling. Continuity of of the Vermillion
the orbicularis oris muscle and adequate peri-
oral sensation are required to fulfill this func- Distortion of the vermillion and the white roll is
tion. The size of the oral stoma also has a readily apparent; therefore careful reconstruction
bearing on function. All of these factors con- is important for cosmesis. Small defects that do
tribute to the esthetics of the lower face. not involve the underlying orbicularis muscle
Careful reconstruction of the anatomic land- may heal satisfactorily by secondary intention.
marks of the lip, such as the white roll, vermil- However, the process is slow (25 days on aver-
ion border, and Cupid’s bow will facilitate age) and can result in contracture [5, 6].
cosmetic reconstruction. Alternatively, primary closure following small
The unique structure and appearance of the vertically oriented fusiform excisions will give
lips, involving transition from skin, keratinized excellent results. Where redundant ‘dog-ear’ tis-
dry, and nonkeratinized wet mucosa, means sue is likely to occur, a V-Y island of mucosa can
that using existing lip tissue for reconstruction be advanced from the labial mucosa into the ver-
gives better results than when distant tissue is million or laterally from adjacent vermillion [7].
employed. Preservation of contiguous, inner- Larger superficial defects of the upper or
vated orbicularis oris muscle, and hence func- lower vermillion are best managed by resection
tion is also more likely. Local flaps including of the entire vermillion (lip shave) and advance-
rotational, advancement, and cross-lip flaps ment of a flap of labial mucosa. This affords a
have become mainstays of reconstruction for close cosmetic match, and sensation is usually
larger deformities not amenable to direct or regained; however, atrophy and contracture can
sliding lip closure. be apparent. The vermillion of the opposing lip
can also be transferred on a single or double ped-
icle and can include underlying muscle pedicled
12.2.1 A Timeline of Lip on the labial vessels [8]. This technique requires
Reconstruction Techniques a second procedure for division and inset, argu-
ably for little benefit compared with mucosal
1000 BC Shushruta First mention of lip repair advancement. Vermillion reconstruction with
1597 AD Tagliacozzi Upper and lower lip repair tongue [9], buccal mucosal, or myomucosal flaps
using forearm flap can give acceptable results. Grafting anal verge
1768 Louie Wedge excision described
mucosa has also been described [10].
1838 Sabattini Full thickness switch from
lower to upper
1845 Dieffenbach Cheek advancement for upper
lip repair 12.4 Lower Lip Reconstruction
1857 Von Bruns Cheek advancement for lower
lip repair 12.4.1 Defects Less Than 50%
1872 Estlander Upper to lower switch at
commissure
1898 Abbe Lip switch for bilateral cleft Depending on the laxity of the lower lip (mainly
lip a function of age), most defects of less than one
1909 Lexer Tongue flaps for lip repair third to half the width of the lower lip can be
1954 Schuchardt Sliding inferiorly based closed primarily with good cosmetic and func-
cheek flaps
tional outcomes. The excision is extended inferi-
!969 Bakamjian Deltopectoral flap to lower lip
defect orly to complete a “v” or shield shape. The point
1974 Karapandzic Advancement along of the ‘v can be curved laterally to follow the
nasolabial fold for lower lip labiomental groove. In wider resections, this
defects groove can be followed on one or both sides of
12 Oral Reconstruction 167
the base of the resection to aid in tissue mobiliza-

tion. Some surgeons utilize a ‘w’-shaped inci-
sion, but this gives a less cosmetic outcome.
12.4.2 Defects Greater Than 50%
Lower lip defects of this proportion are unlikely

to be able to be closed primarily without causing
unacceptable microstomia.
12.4.2.1 Karapandzic Flap

Unilateral or bilateral transdermal curvilinear inci-
sions following the nasolabial crease(s) allow
mobilization and approximation of the remaining Fig. 12.1 Webster–Bernard Burrow flap markings prior
lower lip [11]. The incision extends through der- to lower lip resection
mis only, leaving the neurovascular supply to the
lower lip elements intact. This improves upon
Gillie’s fan flap and other advancement techniques
that employ full thickness incisions that disrupt the
neurovasculature. Reconstruction of defects up to
80% of the lower lip have been described [12],
with good preservation of function and cosmesis.
A degree of micostomia and blunting of the com-
missures is associated, however.
12.4.2.2 Lip Switch Flaps

A reverse Abbe flap can be employed whereby a
segment of upper lip with the same vertical
dimension, but around 50% of the width of the
lower lip defect is pedicled on the labial artery.
Cosmetically, harvest should only be from lateral
to the philtrum. In theory bilateral flaps can be
harvested, but the two resultant pedicles make
oral intake challenging until the flaps are inset.
Fig. 12.2 Webster–Bernard-von burrow flap following
closure
12.4.3 Subtotal Defects
12.4.3.1 Cheek Advancement Flaps close the lower lip. Webster recommended modi-
Bilateral horizontal cheek advancement flaps fication of this technique for lower lip defects
where described by Bernard (1852) and von using only partial thickness incisions and placing
Burrow (1853) for reconstruction of large lip the triangles to be excised within the nasolabial
defects [13]. Tissue is advanced from the cheek and labiomental creases [14] (Figs. 12.1 and
by extending incisions laterally from the com- 12.2). This concept was further refined by
missure and excising four triangles in cases Pirgousis and Fernandes for lower lip reconstruc-
affecting the upper lip, and three triangles to tion [15] (Figs. 12.3 and 12.4).
of deviation or distortion of the philtrum and

nasal base. Defects up to around one quarter to
one third of upper lip width may be satisfactorily
closed, particularly if laterally situated. The phil-
tral region is less forgiving again, and while up to
around 50% can be closed primarily, there is a
tendency for flattening and upwards retraction of
the vermillion here.
12.4.4.2 Abbe Flap

Sabattini first reported the use of a 2-stage ped-
icled “lip switch” flap in 1838 [18], although
the labial artery-based flap was popularized by
Abbe in 1898 for reconstruction of bilateral
cleft lip [19]. The lower lip flap is designed
Fig. 12.3 Fernandes flap incisions following lower lip with width half that of the defect (thus the
resection
transverse discrepancy is equalized between
upper and lower lips), height equal to the defect,
and the pedicle lateral. The central lower lip is
the preferred donor site, as it is hair-bearing in
males and leaves the least visible scar. The
white roll should be marked prior to incision
and potential obscuration due to bleeding,
edema, and pallor. The flap is raised including
skin, muscle, and mucosa, but with preserva-
tion of the lateral vermilion incorporating the
labial vessels. The flap is rotated and inset, tak-
ing care to reapproximate the orbicularis oris
muscle and align the white roll. After
2–3 weeks, the pedicle is divided and the flap
inset. For defects of the commissure, Estlander
described a similar flap where the commissure
Fig. 12.4 Fernandes flap following closure is the rotation point, and therefore no secondary
insetting is required.
12.4.3.2 Gate Flap 12.4.4.3 Perialar Crescenteric

Bilateral Fujimori “Gate” flaps are effectively Advancement
nasolabial flaps that can be combined with advance- Initially described by Webster [20], a perialar
ment of the lower labial mucosa to reconstruct the incision can be made unilaterally or bilaterally to
entire lower lip [16]. A similar approach has been recruit lateral tissue for closure of the upper lip
employed to reconstruct the upper lip too [17]. defects (Figs. 12.5, 12.6, and 12.7). This tech-
nique can also be combined with an Abbe flap to
reconstruct the philtrum.
12.4.4 Upper Lip
12.4.4.4 Reverse Karapandzic
12.4.4.1 Primary Closure Incisions following the melolabial groove upward
The upper lip is less forgiving when considering to join an upper lip defect can be used to advance
primary closure of defects. It exhibits less tissue lateral tissue in a similar fashion as the
laxity, and asymmetry is more noticeable because Karapandzic flap for closure of lower lip defects.
Fig. 12.7 Late postoperative appearance of perialar cre-

senteric advancement
Fig. 12.5 Perialar cresenteric flap markings for and

upper lip defect
Fig. 12.8 Upper lip defect
mucosa advanced from intraorally to recreate the

vermilion. The Gillies commissuroplasty
involves excision of a triangular segment of skin
lateral to the rounded commissure, to a point
comparable with the normal side. A vermilion
Fig. 12.6 Perialar cresenteric advancement after closure flap from the opposing lip is then lifted and
rotated into this, and a mucosal flap advanced to
form the vermilion of the donor site.
12.4.5 Commissuroplasty
Many of the local flaps described lead to blunting 12.4.6 Total Lip Defects
of the commissure. The simplest correction is to
make a horizontal full thickness incision through A combination of the techniques already
the blunted commissure, extending laterally to described can be employed to reconstruct total
correspond with the position of the contralateral defects, although more commonly free tissue
normal commissure. Epithelium superior and transfer is used. By far the most frequently
inferior to the incision is excised and labial used option is the radial-free flap (Figs. 12.8,
Fig. 12.9 Radial flap

prior to disconnection
12.5 Buccal Mucosa and Cheek
The function of the cheek chiefly is to bound the

oral cavity and retain food between the occlusal
surfaces of the teeth during mastication.
However, morbidity secondary to resection of
the cheek is not limited to this function. For
example, a large resection closed primarily or
left to heal by secondary intention can lead to
trismus, particularly if postoperative radiation
therapy is indicated. Small defects can be closed
primarily, but larger defects limited to the inner
aspect will likely benefit from resurfacing. This
can be accomplished with split thickness skin
Fig. 12.10 Radial flap inset into upper lip defect
grafts or buccal fad pad [28], as well as regional
flaps as described in the section on reconstruc-
12.9, and 12.10), giving soft, pliable tissue and tion of the floor of the mouth (Fig. 12.11). Free
the option to include palmaris longus tendon tissue transfer for a mucosa-only defect will gen-
[21]. The palmaris longus tendon can be uti- erally be in the form of a radial forearm flap due
lized as a sling between the commissures, or to to its pliability and lack of bulk. Full thickness
the malar periosteum to improve form and oral defects require consideration of reconstruction
competence. The flexor carpi radialis tendon or of both the intraoral and extraoral aspects of the
a nonvascularized fascia lata graft can be used cheek. While hybrid solutions can be consid-
in a similar manner. Coaptation of the lateral ered, for example, a regional or free flap for the
antebrachial cutaneous nerve to the mental lateral aspect and a split thickness skin graft for
nerve has been described to restore sensation the mucosal aspect, these defects are most com-
to the reconstruction [22]. Other free flaps that monly reconstructed with a folded or dual pad-
have been described for lip reconstruction dled flap. The intervening surface can be
include the gracilis for lower lip reconstruction de-epithelialized to lie behind the oral sphincter,
[23–25] and temporal scalp for upper lip whether this is preserved or re- established.
defects [26, 27]. However, if a significant lip resection is required,
12.6 Floor of Mouth
The floor of the mouth can be defined as the

mucosal and muscular sling extending from the
ventral surface of the tongue, bounded anteriorly
and laterally by the lingual aspect of the mandib-
ular gingiva, and posteriorly by the retromolar
trigone. The mucosal layer acts as the inferior
limit of the oral cavity and provides a reservoir
for food during mastication. The muscles of the
floor of the mouth are instrumental in deglutition.
The mylohyoid, the geniohyoid, and the anterior
belly of the digastric muscle draw the aerodiges-
tive tract anteriorly and superiorly during swal-
lowing, anchoring the tongue and increasing the
diameter of the fauces. The musculature of the
floor of the mouth, particularly the genioglossus,
Fig. 12.11 Buccal mucosa reconstruction with submen-
tal island pedicled flap
also acts as a passive sling at rest, to prevent pos-
terior displacement of the tongue and consequent
airway compromise. Finally, the range of motion
the intervening tissue can instead be incorpo- of the tongue depends on adequate separation
rated in the lip repair, keeping in mind that the from the mandible, both laterally and medially,
greater the hiatus in the oral sphincter, the greater which is essential for speech and swallowing.
will be the detriment to oral competence. Both Reconstruction of the floor of the mouth
radial forearm [29] and anterolateral thigh [30] should aim to restore both these functions. The
donor sites can be used for bipaddled flaps. The tongue must be resuspended anteriorly, and suf-
first is more pliable and thinner; the second can ficient tissue must be preserved or transferred to
be made larger and confers less donor site mor- ensure adequate tongue mobility and prevent
bidity, as it can generally be closed primarily. ankyloglossia. One commonly employed option
Full thickness resections of the cheek often for reconstruction of pure floor of mouth defects
require sacrifice of the buccal or marginal man- is the submental artery island flap. This flap was
dibular branches of the facial nerve, with conse- first described by Martin et al. in 1993 [33]. It is
quent paralysis of the oral aspects of facial large, reliable, and easily raised with minimal
expression. Although facial reanimation is a donor site morbidity (and even improvement in
topic of its own, the broad strokes bear mention- cosmetic soft tissue profile). The pedicle is up to
ing here. The simplest procedures, and often the 8 cm in length, and cutaneous dimensions up to
most appropriate in the oncological setting, are 7 × 18 cm can be harvested, sufficient to recon-
the static procedures, in which the oral commis- struct most pure floor of mouth defects. It confers
sure is suspended from the deep temporal fascia less donor morbidity than a radial forearm-free
by a strip of autogenous tissue, usually fascia flap, which is the main alternative among the free
lata, or of a prosthetic material, whether biologic flaps. The main proviso regarding the use of this
(acellular dermal graft) or synthetic (PTFE, flap is the question of oncological safety due to
polypropylene) [31]. Dynamic procedures, compromised nodal dissection of level 1. In one
which permit modulation of facial expression, small series, four of nine patients undergoing SIF
involve either the restoration of nerve continuity suffered local or regional recurrence thought to
or the interposition of innervated muscle, com- be attributable to incomplete nodal harvest [34].
monly temporalis transposition or a gracilis neu- On the other hand, Howard et al. found that in 50
romuscular free flap [32]. patients undergoing SIF, all with clinically nega-
tive level I nodes, none experienced recurrence should always be considered to improve proprio-
attributable to the flap [35]. It is generally ception of the reconstructed tongue and facilitate
accepted that the flap is contraindicated in necks cortical plasticity.
with clinically positive nodes, particularly in
level I. A history of radiation is a relative contra-
indication, though good outcomes have been 12.7.1 Primary Closure
described in this setting [36, 37]. Finally, since
the submental vessels arise from the facial ves- Small defects of the free oral tongue can often be
sels, the flap cannot be performed in patients who closed primarily. As defect size increases, the
have undergone a neck dissection with sacrifice option of healing by secondary intention should
of the facial vessels. be considered as this can result in a more natural
In patients in whom SIF is contraindicated, tongue morphology, while dehiscence is often the
and those with an unfavorable defect, alternate natural course for primary closure anyway.
regional flaps may be appropriate, particularly if
this permits the use of nonirradiated tissue. The
infrahyoid island flap [38] is of a similar size and 12.7.2 Pedicled Flap Reconstruction
character to the SIF and would often lie in the
same radiation field. Its chief advantage over the Where free tissue transfer is precluded, tongue
SIF is that it obviates the question of oncological defects can be reconstructed with a variety of
safety in level I. The supraclavicular and pectora- pedicled flaps. Prior to development of free tissue
lis major flaps are robust options and would gen- transfer, the pectoralis major and deltopectoral
erally lie outside any previously operated or flap were commonly used, with the facial artery
radiated field. However, these flaps depend on a myomucosal flap an option for smaller volume
broad pedicle, which often introduces excess tis- defects. There has recently been a renaissance in
sue into the defect or the tunnel from donor site to the use of other regional flaps such as submental
defect. Many such patients will therefore be best island, supraclavicular island, trapezius island,
served with a radial forearm-free flap, which pro- and infrahyoid flaps.
vides a thin, supple reconstruction and negligible
pedicle bulk.
12.7.3 Free Flap Reconstruction
12.7 Tongue Reconstruction Free flap reconstruction is advocated to restore

form, prevent contracture, and reduce tethering in
Tongue function depends on complex interplay glossectomy defects one quarter of the tongue
between sensory and motor components of both size or greater [40]. The flap should be designed
voluntary and involuntary nervous systems. to recreate premorbid morphology in all three
Restoration of tongue function is important for dimensions as far as possible.
mastication, deglutition and articulation, and as a The radial forearm-free flap is the workhorse
result, quality of life. Evidence suggests that of tongue reconstruction for good reason
tongue resection has a significant effect on qual- (Fig. 12.12). It provides thin, pliable, soft tissue,
ity of life compared with other oropharyngeal and large-caliber vessels of consistent anatomy
structures, proportional to the size of resection. A for microvascular anastomosis. Harvest can be
reconstruction that recreates the biomechanics of simultaneous with resection and presents lim-
the healthy tongue leads to better function and ited morbidity. Neural coaptation of the lateral
even cortical adaptation to the neotongue [39]. In antebrachial cutaneous nerve to the lingual
terms of speech, the ability for the anterior tongue nerve affords a sensate flap with some evidence
to contact the palate is particularly important. that this helps with function and resists flap
Speech therapy following tongue reconstruction atrophy [41].
preserved, hyolaryngeal suspension should also

be considered to reduce the risk of persistent
aspiration [43].
12.8 Mandibular Reconstruction
The mandible provides support for the lower

third of the face and attachment for the muscula-
ture of the tongue, floor of mouth, and hyoid.
Reconstruction is often necessary to restore both
form and function following ablation. Additional
consideration should be made for facilitation of
dental rehabilitation when considering mandibu-
lar reconstruction, in the form of fixed or remov-
able prostheses.
Fig. 12.12 Radial free flap to reconstruct hemiglossec-
tomy defect
12.8.1 Classification of Defects
Brown et al. developed a classification system for

oncological mandibular defects that is helpful in
guiding reconstructive options [44]. Based on
four mandibular ‘corners’ of the angles and
canine regions, defects increase in size and com-
plexity from class I to class IV and are subclassi-
fied by whether the condyles are included too.
The classification indicates the average length
of bone required for reconstruction, and it can be
appreciated that morbidity will increase with
class if the defect is not reconstructed, particu-
larly with respect to mandibular continuity.
Fig. 12.13 ALT flap reconstruction following total
Mandibular reconstruction may aim to restore
glossectomy
only the hard tissue defect (autogenous bone or
reconstruction bar alone) or only soft tissue,
Defects greater than three quarters of the whether intra or extraoral, or both hard and soft
tongue may benefit from reconstruction with tis- tissue defects. In all cases, meticulous intraoral
sue of greater substance such as anterolateral closure should be insured to reduce the risk of
thigh (ALT) or rectus abdominis-free flaps. ALT infection and fistulation.
harvest results in a more easily hidden scar, and
potential sensory loss is less troublesome. Total
glossectomy reconstruction is particularly chal- 12.8.2 Soft Tissue Reconstruction
lenging. Recreation of three-dimensional mor-
phology is critical, with emphasis on height and a Recruitment of intraoral soft tissue can permit
protuberant tip, to restore speech and swallowing primary closure, and often the loss of alveolar
function [42] (Fig. 12.13). Bulky flaps, around height associated with bone resection will facili-
one third oversized to the defect are recom- tate this. The buccal fat pad can be mobilized to
mended to achieve this aim. When the larynx is assist with closure of small posterolateral soft tis-
sue defects, as can an inferiorly pedicled facial

artery myomucosal flap.
Larger soft tissue defects associated with
mandibular resection can be reconstructed with
regional flaps such as the submental island,
supraclavicular island, pectoralis major, or latis-
simus dorsi flap. Alternatively, free tissue transfer
can be performed, most commonly radial or ALT
flaps.
12.8.3 Hard Tissue Reconstruction
In some cases, restoration of mandibular continu-

ity following segmental mandibulectomy may
not be deemed necessary, particularly in elderly, Fig. 12.14 Postoperative 3D CT of fibular flap recon-
struction for Brown Class I mandibulectomy
edentulous patients with class 1 defects.
Restoration of mandibular continuity can be
achieved by a load-bearing reconstruction bar
alone. However, even with good soft tissue clo-
sure, there is risk of hardware exposure, particu-
larly if radiotherapy is to be administered. A
nonvascularized bone graft, usually from the iliac
crest, can be inserted with the plate, but again,
considering radiotherapy is likely to be indicated
within 6 weeks of surgery, this choice is precari-
ous. Segmental defects are therefore preferably
reconstructed with osseous free flaps to permit
radiotherapy in the short term and facilitate future
dental rehabilitation. This process has been aided
greatly in recent years by the use of virtual surgi-
cal planning.
The fibular-free flap is commonly used to
reconstruct mandibular defects for a number of
reasons. The bone stock allows insertion of
implants for dental rehabilitation, osteotomizing Fig. 12.15 Postoperative 3D CT of fibular free flap
the bone permits restoration of mandibular form, reconstruction for Brown class IV mandibulectomy defect
and in class IV defects, it may be the only option
to provide sufficient length of bone (Figs. 12.14
and 12.15). Additionally, the flap can include fas- for smaller defects or as a free flap based on the
ciocutaneous and/or muscle paddles to provide DCIA system with or without muscle and or
for soft tissue repair. A double-barreled bone skin. The skin paddle is bulky and restricted in
arrangement can be used to increase mandibular terms of its placement compared with the fibular
height if needed. flap, which also has a longer pedicle. The mor-
The iliac crest arguably contributes the best phology of the ilium lends itself well to recon-
bone stock in terms of both volume and quality. struction of Brown Class I or II defects
It can be harvested as a nonvascularized graft (Figs. 12.16 and 12.17).
Fig. 12.16 3D CT of DCIA reconstruction of mandible
Fig. 12.18 Postoperative 3D CT of scapula flap recon-

struction for Brown Class II mandibulectomy defect
12.9 Palatal and Maxillary Defects
The function of the palate is to separate the nasal

passage and maxillary sinuses from the mouth
and to provide a superior limit to the oral cavity.
Insufficiency of this structure has predictable
effects: nasal speech, difficulty propelling the
food bolus posteriorly during deglutition, ten-
dency for food to become trapped in the nasal
cavity or maxillary sinus, and nasal regurgitation.
Defects with greater vertical extent may also
involve the orbital floor or contents. Soft palate
Fig. 12.17 3D CT of DCIA reconstruction of mandible. defects result in velopharyngeal insufficiency,
Inferior view
chiefly characterized by nasal speech, but also
disposing to retrograde nasal aspiration, espe-
cially of fluids.
The lateral border of the scapula is the next Maxillectomy defects have been classified by
most commonly used free flap in mandibular a number of authors, but arguably the most useful
reconstruction (Fig. 12.18). It can be osteoto- in terms of guiding reconstructive options is that
mized, and receive implants, although not as proposed by Brown et al. [45]. The first level of
reliably as fibula or DCIA reconstructions.
classification is the vertical: Class I (low maxil-
Simultaneous harvest is also challenging. lectomy) defects do not cause an oronasal fistula;
class II defects do, but do not involve the orbit;

class III defects involve the orbital adnexa with
orbital retention; class IV defects involve orbital
enucleation or exenteration; class V describes
orbitomaxillary defects without oronasal fistula;
and class VI describes nasomaxillary defects.
Class II–IV defects are further classified
according to their horizontal extent: Horizontal
class ‘a’ does not involve the alveolus, ‘b’ is a
laterally located defect less than the hemipalate,
‘c’ is also less than half the palate, but located
anteriorly, and ‘d’ is greater than half the palate.
Simple mucosal defects overlying the hard
palate can be left to close by secondary intention.
Fig. 12.19 DCIA flap inset to reconstruct Brown and
Although the process is lengthy, pain and remu- Shaw class III defect
cosalization can be aided by an acrylic cover
plate retained by bone screws or dental cribs.
Most oncologic resections for lesions overlying
the hard palate will require maxillectomy to
obtain adequate margins. It is sometimes possible
to preserve the continuity of the soft palate, thus
simplifying reconstruction. Small class I defects
can sometimes be closed by advancement of the
buccal mucosa, since the decreased alveolar
height postresection prevents excess tension.
Regional pedicled flap options include buccal fat
pad, facial artery myomucosal flaps, temporalis,
and temporoparietal flaps. Free flaps can be sub-
divided into those providing hard tissue (fibula, Fig. 12.20 Brown and Shaw class IIb maxillectomy
deep circumflex iliac artery (Fig. 12.19), scapula, defect
lateral arm, and osteocutaneous radial forearm)
and those providing soft tissue alone (radial fore-
arm (Figs. 12.20 and 12.21), rectus abdominis,
free latissimus dorsi, and anterolateral thigh).
Osseus flaps afford the potential for osseointe-
grated implant placement for dental rehabilita-
tion (Figs. 12.22 and 12.23).
Many maxillary defects are adequately
addressed with an obturator appliance, which fills
the defect, re-establishes the separation between
oral and nasal cavities, and can incorporate pros-
thetic dentition (Figs. 12.24 and 12.25). The
potential advantages of an obturator are reduced
operative time and morbidity due to less extensive
surgery, greater ease of cavity surveillance for
recurrence, and earlier restoration of dental func-
tion. On the other hand, obturators require serial Fig. 12.21 Radial-free flap to reconstruct Brown and
revision as the cavity matures, and they can be dif- Shaw Class IIb defect
Fig. 12.22 Panoramic radiograph showing implants

placed within fibula flap reconstruction of Brown and
Shaw Class II maxillectomy Fig. 12.25 Same patient as Fig. 12.24 with obturator in
situ
[45]. Moreno et al. found that the horizontal (pala-

tal) extent of the defect was more significant than
the vertical in weighting toward flap reconstruc-
tion [46]. When an obturator is contemplated,
maxillary teeth should be preserved during resec-
tion if possible, to allow for retention and stability
of the obturator. When this is not feasible, obtura-
tor retention can be achieved with a two-part
Fig. 12.23 Panoramic radiograph showing implants design or osseointegrated implants.
placed within scapula flap reconstruction of Brown and
Shaw Class IV maxillectomy
References
1. Ray E. Head and neck reconstructive surgery. In:
Cancer treatment and research. New York, NY:
Springer; 2018. p. 123–43. http://www.ncbi.nlm.nih.
gov/pubmed/29435840. Accessed 3 Sep 2018.
2. Erba P, Ogawa R, Vyas R, Orgill DP. The reconstruc-
tive matrix: a new paradigm in reconstructive plastic
surgery. Plast Reconstr Surg. 2010;126(2):492–8.
3. El-Marakby HH. The reliability of pectoralis major
myocutaneous flap in head and neck reconstruction. J
Egypt Natl Canc Inst. 2006;18(1):41–50. http://www.
ncbi.nlm.nih.gov/pubmed/1723785. Accessed 31 Aug
2018.
4. Shah JP, Haribhakti V, Loree TR, Sutaria
P. Complications of the pectoralis major myocutane-
ous flap in head and neck reconstruction. Am J Surg.
1990;160(4):352–5.
Fig. 12.24 Brown and Shaw class IIc maxillectomy 5. Zitelli JA. Wound healing by secondary inten-
defect tion. A cosmetic appraisal. J Am Acad Dermatol.
1983;9(3):407–15.
6. Leonard AL, Hanke CW. Second intention healing for
ficult to use in an inflamed field during radiation intermediate and large postsurgical defects of the lip. J
therapy. It is therefore not always clear whether a Am Acad Dermatol. 2007;57(5):832–5.
given defect would be better served with a free flap 7. Jin X, Teng L, Zhang C, Xu J, Lu J, Zhang B, et al.
Reconstruction of partial-thickness vermilion defects
or an obturator. Brown et al. suggest that as a gen-
with a mucosal V–Y advancement flap based on the
eral rule, the larger the defect, the greater the orbicularis oris muscle. J Plast Reconstr Aesth Surg.
potential advantages of free flap reconstruction 2011;64(4):472–6.
8. Wagner DJD, Newman DMH. Bipedicled axial www.ncbi.nlm.nih.gov/pubmed/17979111. Accessed

cross-lip flap for correction of major vermilion 3 Sep 2018.
deficiency after cleft lip repair. 2017. https://doi. 24. Sacak B, Gurunluoglu R. The innervated graci-
org/101597/1545-1569_1994_031_0148_baclff_23. lis muscle for microsurgical functional lip recon-
co_2. struction: review of the literature. Ann Plast Surg.
9. Ceran C, Demirseren ME, Sarici M, Durgun M, Tekin 2015;74(2):204–9. http://content.wkhealth.com/
F. Tongue flap as a reconstructive option in intraoral linkback/openurl?sid=WKPTLP:landingpage
defects. J Craniofac Surg. 2013;24(3):972–4. &an=00000637-201502000-00015. Accessed 3 Sep
10. Levi B, Ricci JA, Donelan MB. Vermillion reconstruc- 2018.
tion with anal verge transitional epithelium. J Burn 25. Gurunluoglu R, Glasgow M, Williams SA,
Care Res. 2017;39(2):1. Gurunluoglu A, Antrobus J, Eusterman V. Functional
11. Karapandzic M. Reconstruction of lip defects by local reconstruction of total lower lip defects using inner-
arterial flaps. Br J Plast Surg. 1974;27(1):93–7. http:// vated gracilis flap in the setting of high-energy bal-
www.ncbi.nlm.nih.gov/pubmed/4593704. Accessed listic injury to the lower face: preliminary report.
31 Aug 2018. J Plast Reconstr Aesthet Surg. 2012;65(10):1335–
12. Sood A, Paik A, Lee E. Lower lip reconstruction: 42. http://linkinghub.elsevier.com/retrieve/pii/
karapandzic flap. Eplasty 2013;13:ic17. http://www. S1748681512002471. Accessed 3 Sep 2018.
ncbi.nlm.nih.gov/pubmed/23409208. Accessed 3 Sep 26. Lyons GB, Milroy BC, Lendvay PG, Teston
2018. LM. Upper lip reconstruction: use of the free superfi-
13. Bernard C. Cancer de la levre inferieure opere par un cial temporal artery hair-bearing flap. Br J Plast Surg.
procede nouveau. Bull Mem Soc Chir. 1853;3:357– 1989;42(3):333–6. http://www.ncbi.nlm.nih.gov/
64. https://ci.nii.ac.jp/naid/10011662354/#cit. pubmed/2758210. Accessed 3 Sep 2018.
14. Webster RC, Coffey RJ, Kelleher RE. Total and 27. Chang K-P, Lai C-S, Tsai C-C, Lin T-M, Lin
partial reconstruction of the lower lip with inner- S-D. Total upper lip reconstruction with a free tem-
vated muscle-bearing flaps. Plast Reconstr Surg. poral scalp flap: long-term follow-up. Head Neck.
1960;25(3):360–71. 2003;25(7):602–5. http://www.ncbi.nlm.nih.gov/
15. Pirgousis P, Fernandes R. Reconstruction of subtotal pubmed/12808664. Accessed 3 Sep 2018.
defects of the lower lip: a review of current techniques 28. Dean A, Alamillos F, García-López A, Sánchez J,
and a proposed modification. J Oral Maxillofac Surg. Peñalba M. The buccal fat pad flap in oral recon-
2011;69(1):295–9. struction. Head Neck. 2001;23(5):383–8. http://www.
16. Fujimori R. Gate for the total reconstruction of the ncbi.nlm.nih.gov/pubmed/11295812. Accessed 3 Sep
lower lip. Br J Plast Surg. 1980;33(3):340–5. http:// 2018.
www.ncbi.nlm.nih.gov/pubmed/7426811. Accessed 29. Liu F, Wang L, Pang S, Kan Q. Reconstruction of
31 Aug 2018. full-thickness buccal defects with folded radial fore-
17. Aytekin A, Ay A, Aytekin O. Total upper lip recon- arm flaps: a retrospective clinical study. Medicine
struction with bilateral fujimori gate flaps. Plast (Baltimore). 2017;96(32):e7344. http://insights.
Reconstr Surg. 2003;111(2):797–800. ovid.com/crossref?an=00005792-201708110-00002.
18. Agostini T. The Sabattini-Abbé flap: a historical note. Accessed 3 Sep 2018.
Plast Reconstr Surg. 2009;123(2):767. https://insights. 30. Liu Z-M, Wu D, Liu X-K, Liu W-W, Li H, Li Q, et al.
ovid.com/crossref?an=00006534-200902000-00051. Reconstruction of through-and-through cheek defects
Accessed 3 Sep 2018 with folded free anterolateral thigh flaps. J Oral
19. Abbe R. A new plastic operation for the relief of Maxillofac Surg. 2013;71(5):960–4. http://linking-
deformity due to double harelip. Plast Reconstr Surg. hub.elsevier.com/retrieve/pii/S027823911201573X.
1968;42(5):481–3. http://www.ncbi.nlm.nih.gov/ Accessed 3 Sep 2018.
pubmed/4880838. Accessed 31 Aug 2018 31. Liu YM, Sherris DA. Static procedures for the man-
20. Webster JP. Crescentic peri-alar cheek excision for agement of the midface and lower face. Facial Plast
upper lip flap advancement with a short history of upper Surg. 2008;24(2):211–5. http://www.thieme-connect.
lip repair. Plast Reconstr Surg. 1955;16(6):434–64. de/DOI/DOI?10.1055/s-2008-1075836. Accessed 3
21. Fernandes R, Clemow J. Outcomes of total or near- Sep 2018.
total lip reconstruction with microvascular tissue 32. Coyle M, Godden A, Brennan PA, Cascarini L,
transfer. J Oral Maxillofac Surg. 2012;70(12):2899– Coombes D, Kerawala C, et al. Dynamic reanimation
906. http://linkinghub.elsevier.com/retrieve/pii/ for facial palsy: an overview. Br J Oral Maxillofac
S027823911200359X. Accessed 3 Sep 2018. Surg. 2013;51(8):679–83. https://www.sciencedirect.
22. Sakai S, Soeda S, Endo T, Ishii M, Uchiumi E. A com/science/article/pii/S0266435613000107?via%3
compound radial artery forearm flap for the recon- Dihub. Accessed 3 Sep 2018.
struction of lip and chin defect. Br J Plast Surg. 33. Martin D, Pascal JF, Baudet J, Mondie JM, Farhat
1989;42(3):337–8. JB, Athoum A, et al. The submental island flap:
23. Cordova A, D’Arpa S, Moschella F. Gracilis free a new donor site. Anatomy and clinical applica-
muscle transfer for morpho-functional reconstruction tions as a free or pedicled flap. Plast Reconstr Surg.
of the lower lip. Head Neck. 2008;30(5):684–9. http:// 1993;92(5):867–73.
34. Cariati P, Cabello Serrano A, Marin Fernandez tion of the oral tongue. Ann Otol Rhinol Laryngol.
AB, Perez de Perceval Tara M, Juliá MA, Ildefonso 2005;114(9):681–7.
Martinez Lara M. Is submental flap safe for the onco- 40. Brown L, Rieger JM, Harris J, Seikaly H. A longitudi-
logical reconstruction of the oral cavity? J Stomatol nal study of functional outcomes after surgical resec-
oral Maxillofac Surg. 2018. https://linkinghub.else- tion and microvascular reconstruction for oral cancer:
vier.com/retrieve/pii/S2468785518301307. Accessed tongue mobility and swallowing function. J Oral
29 Aug 2018. Maxillofac Surg. 2010;68(11):2690–700.
35. Howard BE, Nagel TH, Donald CB, Hinni ML, Hayden 41. Biglioli F, Liviero F, Frigerio A, Rezzonico A,
RE. Oncologic safety of the submental flap for recon- Brusati R. Function of the sensate free forearm flap
struction in oral cavity malignancies. Otolaryngol after partial glossectomy. J Cranio-Maxillofac Surg.
Head Neck Surg. 2014;150(4):558–62. http://jour- 2006;34(6):332–9.
nals.sagepub.com/doi/10.1177/0194599814520687. 42. Kimata Y, Sakuraba M, Hishinuma S, Ebihara S,
Accessed 29 Aug 2018. Hayashi R, Asakage T, et al. Analysis of the relations
36. Wu Y, Tang P, Qi Y, Xu Z, He Y. Evaluation for sub- between the shape of the reconstructed tongue and
mental island flap. Zhonghua Kou Qiang Yi Xue Za postoperative functions after subtotal or total glossec-
Zhi. 2002;37(6):418–20. http://www.ncbi.nlm.nih. tomy. Laryngoscope. 2003;113(5):905–9.
gov/pubmed/12641952. Accessed 31 Aug 2018. 43. Weber RS, Ohlms L, Bowman J, Jacob R, Goepfert
37. Taghinia AH, Movassaghi K, Wang AX, Pribaz H. Functional results after total or near total glossec-
JJ. Reconstruction of the upper aerodigestive tract tomy with laryngeal preservation. Arch Otolaryngol
with the submental artery flap. Plast Reconstr Surg. Neck Surg. 1991;117(5):512–5.
2009;123(2):562–70. https://insights.ovid.com/cros 44. Brown JS, Barry C, Ho M, Shaw R. A new clas-
sref?an=00006534-200902000-00016. Accessed 31 sification for mandibular defects after oncologi-
Aug 2018. cal resection. Lancet Oncol. 2016;17(1):e23–30.
38. Deganello A, Manciocco V, Dolivet G, Leemans CR, h t t p : / / l i n k i n g h u b. e l s ev i e r. c o m / r e t r i eve / p i i /
Spriano G. Infrahyoid fascio-myocutaneous flap as an S1470204515003101. Accessed 31 Aug 2018.
alternative to free radial forearm flap in head and neck 45. Brown JS, Rogers SN, McNally DN, Boyle M. A
reconstruction. Head Neck. 2007;29(3):285–91. http:// modified classification for the maxillectomy defect.
www.ncbi.nlm.nih.gov/pubmed/17230563. Accessed Head Neck. 2000;22(1):17–26.
3 Sep 2018. 46. Moreno MA, Skoracki RJ, Hanna EY, Hanasono
39. Mosier K, Liu W-C, Behin B, Lee C, Baredes MM. Microvascular free flap reconstruction versus
S. Cortical adaptation following partial glossectomy palatal obturation for maxillectomy defects. Head
with primary closure: implications for reconstruc- Neck. 2009;32(7):860–8.
Predicting Quality of Life (QoL)
of Oral Cancer
13
Roderick Y. Kim, Jason N. Burkes,
and Fayette C. Williams
13.1 Introduction 1. Allows the surgeon to discuss candidly the

limitation of ablative and reconstructive
Surgical and adjuvant therapy for treatment of options while aligning patient expectations
oral cancer is ultimately designed to treat, pre- with the surgeon.
serve, and restore patients back to a functional 2. Determines the main goal of therapy since QoL
life. Unfortunately, quality of life (QoL) has not is subjective and should be individualized.
been a primary focus in the advancement of care 3. Ultimately permits the patient to make the
for the oral cancer patient, although it often has a best-informed treatment decisions.
significant impact on the patient. QoL studies
have attempted to address, in a standardized man- Therefore, before even any treatment is ren-
ner, the dilemma of improving the patient’s sub- dered, the patient’s expectations are established.
jective effect, and in objective manner, the This in turn improves the overall treatment out-
unintended consequences of clinical treatment come by cultivating the patient expectations and
decisions. motivation. Through these evidence-based prac-
The best way to deal with a problem is to pre- tices, surgeons are better prepared to predict
vent it from ever happening in the first place. As quality of life issues in oral cancer patients to
surgeons, this involves carefully considering all advance the overall care of the patient.
aspects of perioperative care as well as the long-
term implications of the recommended treatment.
Predicting the QoL life in oral cancer patients 13.2 Quality of Life (QoL)
allows multiple comportments to occur:
In the treatment of oral cancer, much focus in the
past has been in regards to understanding the
pathophysiology, surgical, and medical manage-
R. Y. Kim (*) · F. C. Williams ment. However, it is no surprise treating cancers
Division of Maxillofacial Oncology and that arise in the oral cavity has consequences
Reconstructive Surgery, John Peter Smith Health beyond the locoregional control and survival of
Network, Fort Worth, TX, USA the patient. These treatment implications include
e-mail: Rkim01@jpshealth.org
esthetics, social cues, speech, breathing, eating,
J. N. Burkes and affection. Rogers in 2007 noted that, although
Department of Oral and Maxillofacial Surgery, Walter
Reed Nation Military Medical Center, Bethesda, we have improved care with multimodality
MD, USA methods through tumor board consensus, little

https://doi.org/10.1007/978-3-030-30094-4_13
182 R. Y. Kim et al.
debate occurs in terms of what effect the clini- bridging of the gap between expectations and
cians have in health-related quality of life out- reality, as Morton et al. noted, “a larger gap
comes [1]. As we perform surgical interventions, between perceived reality and one’s expectation,
along with adjuvant therapies, it is easy to wit- [leads to] a poor QoL [6].”A striking example of
ness the significant impact the treatments have on reality versus expectation is of McNeil in the
the physical, mental, emotional, and psychoso- “Fireman” study [7], where patients expressed
cial well-being of the patient. Therefore, it is that they would rather die than have a total laryn-
important for clinicians to be familiar in evidence- gectomy [5] due to preconceived perception of
based QoL measurements to allow objective the postoperative self. The issue may in part be
assessments beyond the treatment efficacy and due to the fact during the consultation in clinic,
disease outcome. commonly the surgery and treatment were dis-
Most common QoL parameters for head and cussed and seldomly the expectations.
neck cancer include the University of Washington In the literature, QoL scores are frequently
QoL (UW-QoL) and the European Organization obtained at pre-, during, and post-therapy. Having
for Research and Treatment of Cancer QoL objective data to properly counsel, the patient
(EORTC-QoL) [2–4]. These are considered allows for a candid conversation, alleviates
health-related quality of life parameters patient fears regarding treatment, and establishes
(HRQOL). UW-QoL, currently in version 4, is patient expectations. Through open communica-
originally described by Hassan SJ and Weymuller tions based on evidence-based practices, the
EA published in 1993 in the journal of Head and expectations of the patients and the surgeon are
neck [3]. Its benefits include a short self- better aligned. This allows identification of gaps
administered questionnaire that requires no input in patient’s needs, which can be augmented by a
from the provider, guided toward head and neck multidisciplinary team consisting of a social
cancer, and multifactorial questions that allow work, nutrition, support group, psychosocial
identification of subtle changes. A short 12-item counseling, physical therapy, occupational ther-
questionnaire domains, which takes about apy, and speech/language pathology. These per-
10 minutes to complete, the score is scaled from sonnel are all essential to address the functional
0 being the worst to 100 being the best, and a and QoL changes that occur after oral cancer sur-
composite score can be calculated by averaging gery. The ultimate goal in treating patients with
each domain score. Patients can also rank the oral cancer is to improve their outcome in all
importance of the domain in their quality of life. aspects of care, in which QoL is clearly a large
Currently the domains include pain, appearance, factor.
activity, recreation, swallowing, chewing, speech,
shoulder, taste, saliva, mood, and anxiety. In
some recent studies, UW-QoL was noted to be 13.3 Glossectomy
the most frequently used HRQOL parameter [2]. and Reconstruction
Thus, it is considered an important tool for
assessing the progression of disease and effec- Tongue is the most common subsite in the oral
tiveness of treatment, perhaps most suited for cavity for squamous cell carcinoma. Tongue can-
patients undergoing surgery [5]. cer has implications in multiple parameters of
The EORTC-QoL, specific to the cancer type, QoL and function since speech, swallowing, oral
consists of 35 questions. A composite score is hygiene, taste, affection, and airway are all at risk
calculated by sum of individual scores, with 1 for dysfunction. Ganziano alerted in 2002 that
being the best and 4 being the worst. The domains dysphagia after surgical resection of tongue can-
include pain, swallowing, sense, speech, social cer can lead to malnutrition, dehydration, weight
eating, social contact, and sexuality [2, 4]. This loss, functional loss, fear of eating, and drinking
questionnaire focuses on the symptoms and side and thus lead to depression with decreased QoL
effects of treatment. These QoL parameters allow [8]. Although traditional teaching is that 50% of
13 Predicting Quality of Life (QoL) of Oral Cancer 183
the tongue can be resected without affecting ated with better functional results [17].
speech and swallowing, there are multiple studies Interestingly the bulk, size, and type of flap uti-
that describe resection of large tongue cancers lized is poorly understood for QoL with no spe-
with negative effects on QoL [9–11]. This was cific amount or size criteria to obtain maximal
highlighted by Brown et al. in 2006, who showed results [14]. Yang et al. in 2016 measured the
speech and swallowing were worse when the QoL difference between Pectoralis Major
tongue resection was greater than 50% [11]. Pedicled Flap (PMPF) and the Anterolateral
These tumors are more likely to require adjuvant Thigh (ALT) free flap and brought to the atten-
radiotherapy, which is known to decrease QoL tion how the current measure of success for
[10, 12]. Interestingly, Zhang et al. have noted reconstruction was based only the survival of the
that as long as the tongue is replaced, it seems to flap, rather than the patient’s QoL [18]. They also
decrease functional outcomes compared to pre- noted that although PMPF as a locoregional flap
operative function [10]. Not surprisingly, there provided ample reliable soft tissue, but it was
are multiple data that show without replacement limited by pliability, restriction in arc of rotation,
the functional outcome is worse [13, 14]. Perhaps and poor esthetics. Recently with the popularity
the correct question we should ponder is, what is of free tissue transfers, investigators are attempt-
the QoL difference that is reasonable for the ing to answer if there is superiority between the
patient? Not only is it important to make sure the kind of free tissue transfers. Yuan et al. compared
patient has realistic expectations of their opera- the functional difference between the ALT-free
tive course, whether this is better or worse than flap and radial forearm-free flap (RFFF) for
the expectations of the surgeon, but perhaps it is reconstruction of tongue defects [17]. They used
equally important that we shed light on QoL UW-QoL and EORTC-QoL to discuss differ-
changes that will occur in some capacity with the ences between RFFF and the ALT-free flap. The
surgery. results showed the QoL and oral function
One aspect that we as surgeons have an inte- improved from 6 to 12 months postoperatively
gral effect on is the method of reconstruction. for both RFFF and the ALT-free flap reconstruc-
This includes healing by secondary intention, tions. The authors noted the ALT-free flap
autologous/allogenic graft, locoregional tissue required longer operative time and was better
transfer, and free tissue transfer. Typically, the suited for larger tongue defects requiring tissue
decision making is first formulated by the size bulk while most partial tongue defects were bet-
and location of tumor—floor of the mouth ter reconstructed with RFFF.
(FOM), lateral, ventral, and base of tongue. The Overall, the literature rather seems to suggest
tumor size determines if a hemi, partial, subtotal, that significant implications for QoL lies within
or total glossectomy is warranted. Before the radiotherapy, cancer stage, and socioeconomic
popularity of vascularized free flaps, McConnel status rather than the type of tongue reconstruc-
and colleagues in 1987 found reconstructed tion [10, 12, 18]. Unfortunately, the long-term
tongue defects with split-thickness skin graft QoL data are still lacking, and further investiga-
(STSG) had the best outcomes for speech and tions, prospective in nature, could provide better
swallowing after they noted tongue mobility to insight concerning the superiority of particular
be the most significant factor in the determination tongue reconstructive methods.
of postoperative speech results [15]. Since then,
there are increasing evidence for differences in
QoL with differing reconstructive options [12, 13.4 Maxillectomy
14, 16]. Locoregional flaps have good postopera-
tive QoL outcomes, minimize operating room Another location of oral cancer that affects a
time, morbidity, and postoperative care but are multitude of QoL domains is the maxilla. Maxilla
usually limited to T1–T2 lesions [13]. affects chewing, swallowing, speech, and facial
Vascularized free tissue transfer has been associ- esthetics. There are classifications from Brown
and Shaw, as well as Okay for prosthetically flaps provide a definitive reconstruction with good
derived classification of maxillectomy defects results in small defects, it is associated with
[19, 20]. These defect classifications, which the increased hospital stay and morbidity. They reiter-
details are beyond the scope of this chapter, have ate the point that larger defects are more difficult
been utilized in numerous studies on to fit obturators and associated with problems of
QoL. Additionally, obturators have been classi- leakage and hypernasal speech. Yet, their main
fied by the obturator functioning scale (OFS), outcome showed that obturator restores QoL
which has domains of ability to eat, speak, satis- almost completely, while free flaps do not.
faction with lip position, cosmetic effect, and Furthermore, people with large obturators cope
lubrication of mouth with saliva [21]. and ultimately assimilates to the prosthesis, but
A controversial topic concerning the maxillec- free flap patients do not always get used to their
tomy is the reconstruction of the defect with obtu- reconstruction, leading to poorer QoL. They con-
rator reconstruction versus locoregional/ free flap cluded that obturator, if available, shortens ther-
reconstruction. Conventional teaching cautioned apy time, restores function, and is still one of the
the flap reconstruction, as surveillance of tumors most promising ways of improving QoL.
was difficult and only large tumors would be Therefore, in terms of the literature, similar to
noted as a recurrence. Having the convenience reconstructive option for tongue defects, there is
and assurance of open, visible surgical site yet to be comprehensive data that favor one
allowed easy surveillance and outweighed obtura- method of reconstruction for maxillectomy
tor issues of poor fit, expenses, and frequent visits defects. It is important to note that oral rehabilita-
with the prosthodontist. For those proponents of tion following ablation plays a key role in pro-
obturators, Signetmartin et al. in 2015 studied the moting QoL and self-esteem even if chewing is
understandability of speech after maxillectomy not improved [1]. The surgeon and the patient,
obturator placement. They noted the obturator is based on the patient’s goals, expectations, and
considered one of the most efficient rehab tools, changes of QoL, need to establish a coordinated
immediately enhancing intraoral function and treatment plan.
decreases cosmetic deformity by supplying miss-
ing teeth and soft tissue support [22]. Particularly
they showed 2B defects or lower had better obtu- 13.5 Dental Implants
rator function and QoL, but postoperative radia-
tion caused overall lower QoL. In contrast, in The surgical management of oral, head, and neck
patients who have poor manual dexterity and can- malignancies frequently results in deterioration of
not clean the crusting obturators, the obturator oral function, speech, and swallowing.
may be nonfunctional [23] and could lead to a Rehabilitation of the orofacial form aspires not
poor QoL. Although the overall QoL noted in the only to restore oral function but also to improve
literature for obturators is good, if the patient had speech, swallowing, and facial appearance. The
postoperative radiation, it was associated with a use of vascularized free flaps for reconstruction
poorer QoL due to hyposalivation, lack of social significantly diminishes disfigurement of these
eating, and poor understandability. A comparative patients and provides a foundation for restoring
study by Genden et al., evaluating the functional oral function. Unfortunately, the resulting neo-oral
outcomes of hard palate defects reconstruction structures can be suboptimal [26] and often unfa-
with RFFF and obturator, showed equivalent suc- vorable for prosthodontic rehabilitation. To facili-
cess in diet, mastication, articulation, appearance, tate oral rehabilitation in this situation, the
speech, and taste. The RFFF, however, had placement of endosseous implants for support of a
improved satisfaction scores and social scores fixed prosthesis has been found to optimize func-
[24]. In contrast, Brandao et al. performed a study tion, esthetics, and quality of life (QoL) [27–31].
specifically addressing obturator versus free tis- Dholam et al. found most head and neck cancer
sue transfer [25]. Their data showed although free patients reported improved QoL in function, pain,
in addition to psychological and social disabilities tional surgical procedures, risk minimization
with any prosthetic rehabilitation [32]. Nonetheless associated with implant placement in irradiated
the ability of oral function is significantly worse in tissues, as well as restoring oral function sooner to
patients without implant- supported overdenture. assist in rehabilitation of speech and swallowing.
Ablative techniques often further complicate satis- There are risks involved with PARS dental-
factory retention and stability of oral prosthesis implant placement, though these have been found
necessitating the application of implants for reten- to be low incidence when compared to placement
tion. The McGill consensus statement concerning of dental implants in irradiated tissues. These
overdentures for the general population in 2002 risks, while marginal include poor anatomical
concluded there was overwhelming evidence for positioning of the implant, possible delay in pos-
two-implant overdentures as the first choice of tablative chemoradiation therapy, possibility of
treatment in edentulous mandibles [33]. In support recurrent disease, and risk of postradiation treat-
of the McGill statement, the York census statement ment complications [29, 38].
concluded in 2009 the QoL, and patient satisfac- Radiation worsens treatment outcomes in
tion was greater in mandibular implant-retained regard to oral function, pain, and jaw opening.
overdentures when compared to dentures without However, it’s no longer considered a contraindi-
retention [34]. Intriguingly, the number of implants cation for placement of dental implants. The
installed does not appear to influence the QoL or placement of implants in irradiated bone second-
denture satisfaction in patients [27, 29, 31]. arily during PR surveillance has an 84.3% suc-
The placement of dental implants in the head cess rate [39]. Claudy et al. found in a systematic
and neck cancer patient can be achieved during review and meta-analysis approximately 34%
the primary ablative/reconstructive surgery higher risk in implant failure if the implants were
(PARS) or secondarily during post-radiotherapy placed within 12 months of completion of radio-
(PR) surveillance. Since the majority of osseoin- therapy [40]. The time delay for PR dental-
tegration occurs within 6 weeks following implant implant placement can require patients to wait
placement [35], some authors advocate implant 18 months or more after initial treatment to obtain
placement at the end of the ablative/reconstruc- an implant-retained prosthesis [26, 29, 31, 38,
tion procedure when postoperative radiotherapy is 41]. Site placement has been found to influence
indicated [29, 36, 37]. This is mainly due to evi- PR implant placement with higher implant fail-
dence showing primary placement of implants for ure rates associated with irradiated maxilla [42].
oral rehabilitation to be optimal when placed into Interestingly, Curi et al. found higher PR implant
vascularized bone at the time of the PARS as a failure rates among older females as well as con-
two-stage implant technique [26, 29, 31, 36]. To ventional conformal radiotherapy modality when
minimize complications during radiotherapy, the compared to intensity-modulated radiotherapy
implants are covered by soft tissue at time of pri- (IMRT) [38]. An additional risk factor for
mary placement. The second stage of implant increased P-RS implant failure is total radiation
exposure with abutment placement is completed therapy dosages greater than 50 Gy.
6 months after completing radiation therapy [28].
This allows the patient to regain oral function
with an implant-retained prosthesis in as little as 13.6 The Patient
10 months after the initial ablative procedure.
Patients who underwent immediate implant place- Management of the patient after treatment has
ment during PARS followed by postoperative traditionally concentrated on survival, locore-
radiotherapy were found to have an equal chance gional disease control, and function [14], but
of one-year implant survival after prosthesis social well-being, psychological [43], and voca-
placement as noncancer/nonradiated patients [29, tional restoration [44] are emerging as critical
37]. Some of the main advantages of PARS factors in treatment success. QoL is a crucial fac-
implant placement include avoidance of addi- tor in treatment outcomes and is subjectively
determined by the patient. Valdez and Brennan in cancer patients resulting in inadequate treat-
explained QoL as “an abstract, subjective, and ment [46, 50, 58].
multidimensional conceptualization of a patient’s In addition to the psychological morbidities
perception of self” [2]. It subjectively improves and psychosocial challenges of HNC patients,
with time following acute oncological treatment there are many physical and financial conse-
as the patient learns to manage both physical and quences that impact the patient’s quality of life.
somatic dysfunctions, notwithstanding continu- As previously mentioned, the most notable of
ing dysfunction [45]. Nevertheless, establishing a these physical effects involve the esthetics of the
QoL closest to the predisease state is vital to the face, speech, voice, and swallowing. These phys-
overall well-being of the patient. ical toxicities, notably radiation as mentioned
Psychological morbidity is a significant con- throughout this chapter, frequently contribute to
tributor to the overall health and QoL of the development of malnutrition and loss of muscle
patient [46]. Head and neck cancer (HNC) mass. Protein calorie malnutrition, low body
patients are more disposed to psychological dis- mass index, and weight loss leads to poor QoL,
tresses (PD) when compared to other types of reduced survival, and impedes all cancer thera-
cancer patients [43, 47]. PD is an undifferenti- pies (surgery, chemotherapy, radiotherapy) [51,
ated amalgamation of symptoms varying from 59]. All cancer patients should be screened rou-
anxiety, depression, functional disabilities, and tinely for malnutrition to optimize nutritional sta-
cognitive challenges to behavioral deficits [48, tus, treatment modalities, and improve their QoL.
49]. The situation can lead to significant impair-
ment of daily living and social functions.
Predictive factors are influenced by the com- 13.7 Conclusion
bined psychosocial aspects when patients are not
married and live alone [50–52]. Additional con- There is no doubt oral cancer and surgical inter-
tributor to the PD is financial hardship obtained vention has significant impact on the QoL of
from treatment cost, medical debt, and reduced patients. The QoL of the patient will evolve from
income [53, 54]. The definitive course of PD is initial diagnosis, during treatment, and post-
largely undetermined when left untreated, but treatment surveillance. Patient-reported QoL out-
suggestive of a natural continuum to major comes have the potential to provide more
depression disorder [49]. individualized treatment and care, as the decisions
The prevalence of major depression disorder are determined by the QoL issues most important
(MDD) among HNC patients vary from 3.7% to to the patient. QoL is an important tool for evalua-
20% [55]. Following cancer diagnosis and acute tion of patient as well as influencing the treatment
treatment, the incidence of MDD in HNC patients outcome. It should be a critical part of surgeon’s
within the first year is 15–50% [43, 56] with peak care for the patient, as much as the scalpel.
symptoms occurring within 2–3 months follow-
ing diagnosis [43] and a suicide rate more than
four times the rate of the general population [57]. References
Depression is a well-documented contributor to
reduced treatment compliance and poor progno- 1. Rogers SN, Ahad SA, Murphy AP. A structured review
and theme analysis of papers published on ‘quality of
sis. It is an independent determinant of QoL in life’ in head and neck cancer: 2000–2005. Oral Oncol.
HNC patients associated with poorer quality of 2007;43(9):843–68.
life, weaken immune system, increased hospital 2 . Valdez JA, Brennan MT. Impact of oral can-
length of stay, and impaired abilities to perform cer on quality of life. Dent Clin N Am.
2018;62(1):143–54.
activities of daily living (ADLs) [43, 49]. 3. Hassan SJ, Weymuller EA. Assessment of quality
Unfortunately, physicians have a marked ten- of life in head and neck cancer patients. Head Neck.
dency to underestimate the severity of depression 1993;15(6):485–96.
4. Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, 17. Yuan Y, Zhang P, He W, Li W. Comparison of oral
Cull A, Duez NJ, Filiberti A, Flechtner H, Fleishman function: free anterolateral thigh perforator flaps ver-
SB, de Haes JC, Kaasa S. The European Organization sus vascularized free forearm flap for reconstruction
for Research and Treatment of Cancer QLQ-C30: in patients undergoing glossectomy. J Oral Maxillofac
a quality-of-life instrument for use in interna- Surg. 2016;74(7):1500–e1.
tional clinical trials in oncology. J Natl Cancer Inst. 18. Yang Y, Li F, Li W. Factors that affect the qual-

1993;85(5):365–76. ity of life of patients with oral cancer who have
5. Parikh SL, Chen AY. Quality of life in head and had their defects reconstructed immediately after
neck cancer. In: Eisele DW, Smith RV, editors. excision of the tumour. Br J Oral Maxillofac Surg.
Complications in head and neck surgery. 2nd ed. 2016;54(4):410–4.
Philadephia, PA: Mosby; 2009. p. 215–8. 19. Brown JS, Shaw RJ. Reconstruction of the maxilla
6. Morton RP, Davies DM, BAKER J, Baker GA, Stell and midface: introducing a new classification. Lancet
PM. Quality of life in treated head and neck cancer Oncol. 2010;11(10):1001–8.
patients: a preliminary report. Clin Otolaryngol Allied 20.
Okay DJ, Genden E, Buchbinder D, Urken
Sci. 1984;9(3):181–5. M. Prosthodontic guidelines for surgical reconstruc-
7. McNeil BJ, Weichselbaum R, Pauker SG. Speech tion of the maxilla: a classification system of defects.
and survival: tradeoffs between quality and quan- J Prosthet Dent. 2001;86(4):352–63.
tity of life in laryngeal cancer. N Engl J Med. 21. Kornblith AB, Zlotolow IM, Gooen J, Huryn JM,
1981;305(17):982–7. Lerner T, Strong EW, Shah JP, Spiro RH, Holland
8. Gaziano JE. Evaluation and management of oropha- JC. Quality of life of maxillectomy patients using an
ryngeal dysphagia in head and neck cancer. Cancer obturator prosthesis. Head Neck. 1996;18(4):323–34.
Control. 2002;9(5):400–9. 22. Seignemartin CP, Miranda ME, Luz JGC, Teixeira
9. Lazarus CL, Husaini H, Anand SM, Jacobson AS, RG. Understandability of speech predicts qual-
Mojica JK, Buchbinder D, Urken ML. Tongue ity of life among maxillectomy patients restored
strength as a predictor of functional outcomes and with obturator prosthesis. J Oral Maxillofac Surg.
quality of life after tongue cancer surgery. Ann Otol 2015;73(10):2040–8.
Rhinol Laryngol. 2013;122(6):386–97. 23. Rosenthal E, Couch M, Farwell GD, Wax MK. Current
10. Zhang L, Huang Z, Wu H, Chen W, Huang Z. Effect concepts in microvascular reconstruction. Otolaryngol
of swallowing training on dysphagia and depression Head Neck Surg. 2007;136(4):519–24.
in postoperative tongue cancer patients. Eur J Oncol 24.
Genden EM, Wallace DI, Okay D, Urken
Nurs. 2014;18(6):626–9. ML. Reconstruction of the hard palate using the radial
11. Brown JS, Rogers SN, Lowe D. A comparison of forearm free flap: indications and outcomes. Head
tongue and soft palate squamous cell carcinoma Neck. 2004;26(9):808–14.
treated by primary surgery in terms of survival and 25. Brandão TB, Vechiato Filho AJ, de Souza Batista VE,
quality of life outcomes. Int J Oral Maxillofac Surg. de Oliveira MCQ, Santos-Silva AR. Obturator pros-
2006;35(3):208–14. theses versus free tissue transfers: a systematic review
12. Yang ZH, Chen WL, Huang HZ, Pan CB, Li JS. Quality of the optimal approach to improving the quality of
of life of patients with tongue cancer 1 year after sur- life for patients with maxillary defects. J Prosthet
gery. J Oral Maxillofac Surg. 2010;68(9):2164–8. Dent. 2016;115(2):247–53.
13. Boyapati RP, Shah KC, Flood V, Stassen LF. Quality 26. Urken ML, Buchbinder D, Weinberg H, Vickery C,
of life outcome measures using UW-QOL question- Sheiner A, Parker R, Schaefer J, Som P, Shapiro A,
naire v4 in early oral cancer/squamous cell cancer Lawson W, Biller HF. Functional evaluation fol-
resections of the tongue and floor of mouth with lowing microvascular oromandibular reconstruc-
reconstruction solely using local methods. Br J Oral tion of the oral cancer patient: a comparative study
Maxillofac Surg. 2013;51(6):502–7. of reconstructed and nonreconstructed patients.
14. Breeze J, Rennie A, Dawson D, Tipper J, Rehman Laryngoscope. 1991;101(9):935-950.
KU, Grew N, Pigadas N. Patient-reported quality of 27. Kumar VV, Jacob PC, Ebenezer S, Kuriakose MA,
life outcomes following treatment for oral cancer. Int Kekatpure V, Baliarsing AS, Al-Nawas B, Wagner
J Oral Maxillofac Surg. 2018;47(3):296–301. W. Implant supported dental rehabilitation following
15.
McConnel FM, Pauloski BR, Logemann JA, segmental mandibular reconstruction-quality of life
Rademaker AW, Colangelo L, Shedd D, Carroll W, outcomes of a prospective randomized trial. J Cranio-
Lewin J, Johnson J. Functional results of primary Maxillofac Surg. 2016;44(7):800–10.
closure vs flaps in oropharyngeal reconstruction: a 28. Hayter JP, Cawood JI. Oral rehabilitation with end-
prospective study of speech and swallowing. Arch osteal implants and free flaps. Int J Oral Maxillofac
Otolaryngol Head Neck Surg. 1998;124(6):625–30. Surg. 1996;25(1):3–12.
16. Villaret AB, Cappiello J, Piazza C, Pedruzzi B,
29. Schoen PJ, Raghoebar GM, Bouma J, Reintsema

Nicolai P. Quality of life in patients treated for cancer H, Burlage FR, Roodenburg JLN, Vissink
of the oral cavity requiring reconstruction: a prospec- A. Prosthodontic rehabilitation of oral function in
tive study. Acta Otorhinolaryngol. 2008;28(3):120. head–neck cancer patients with dental implants placed
simultaneously during ablative tumour surgery: an 41. Marker P, Siemssen SJ, Bastholt L. Osseointegrated
assessment of treatment outcomes and quality of life. implants for prosthetic rehabilitation after treat-
Int J Oral Maxillofac Surg. 2008;37(1):8–16. ment of cancer of the oral cavity. Acta Oncol.
30. Karayazgan-Saracoglu B, Atay A, Korkmaz C, Gunay 1997;36(1):37–40.
Y. Quality of life assessment of implant-retained 42. Colella G, Cannavale R, Pentenero M, Gandolfo

overdentures and fixed metal-acrylic resin prostheses S. Oral implants in radiated patients: a sys-
in patients with marginal mandibulectomy. J Prosthet tematic review. Int J Oral Maxillofac Implants.
Dent. 2017;118(4):551–60. 2007;22(4):616.
31. Korfage A, Raghoebar GM, Slater JJH, Roodenburg 43. Lydiatt WM, Moran J, Burke WJ. A review of depres-
JL, Witjes MJ, Vissink A, Reintsema H. Overdentures sion in the head and neck cancer patient. Clin Adv
on primary mandibular implants in patients with oral Hematol Oncol. 2009;7(6):397–403.
cancer: a follow-up study over 14 years. Br J Oral 44. Shah JP, Patel SG, Singh B. Head and neck surgery
Maxillofac Surg. 2014;52(9):798–805. and oncology. Amsterdam: Elsevier Health Sciences;
32. Dholam KP, Dugad JA, Sadashiva KM. Impact of oral 2012.
rehabilitation on patients with head and neck can- 45. Morton RP. Studies in the quality of life of head and
cer: a study using the Liverpool Oral Rehabilitation neck cancer patients: results of a two-year longitudinal
Questionnaire and the Oral Health Impact Profile-14. study and a comparative cross-sectional cross-cultural
J Prosthet Dent. 2017;117(4):559–62. survey. Laryngoscope. 2003;113(7):1091–103.
33. Feine JS, Carlsson GE, Awad MA, Chehade A,
46. Singer S. Quality of life issues. In: Vermorken JB,
Duncan WJ, Gizani S, Head T, Lund JP, MacEntee M, Budach V, Leemans CR, Machiels JP, Nicolai P,
Mericske-Stern R, Mojon P. The McGill consensus O'Sullivan B, editors. Critical issues in head and neck
statement on overdentures. Mandibular two-implant oncology: key concepts from the fifth THNO meeting.
overdentures as first choice standard of care for New York, NY: Springer; 2016. p. 275–80.
edentulous patients. Int J Oral Maxillofac Implants. 47. Singer S, Krauß O, Keszte J, Siegl G, Papsdorf

2002;17(4):601. K, Severi E, Hauss J, Briest S, Dietz A, Brähler E,
34. Thomason JM, Feine J, Exley C, Moynihan P, Müller Kortmann RD. Predictors of emotional distress in
F, Naert I, Ellis JS, Barclay C, Butterworth C, Scott B, patients with head and neck cancer. Head Neck.
Lynch C. Mandibular two implant-supported overden- 2012;34(2):180–7.
tures as the first choice standard of care for edentulous 48. Neilson K, Pollard A, Boonzaier A, Corry J, Castle
patients—the York Consensus Statement. Br Dent J. D, Smith D, Trauer T, Couper J. A longitudinal study
2009;207(4):185. of distress (depression and anxiety) up to 18 months
35. Schoen PJ, Reintsema H, Raghoebar GM, Vissink A, after radiotherapy for head and neck cancer. Psycho-
Roodenburg JLN. The use of implant retained man- Oncology. 2013;22(8):1843–8.
dibular prostheses in the oral rehabilitation of head 49.
Drapeau A, Marchand A, Beaulieu-Prévost
and neck cancer patients. A review and rationale for D. Epidemiology of psychological distress. In:
treatment planning. Oral Oncol. 2004;40(9):862–71. Mental illnesses-understanding, prediction and con-
36.
Sclaroff A, Haughey B, Gay WD, Paniello trol. London: InTech; 2012.
R. Immediate mandibular reconstruction and place- 50. Kugaya A, Akechi T, Okuyama T, Nakano T, Mikami
ment of dental implants: at the time of ablative surgery. I, Okamura H, Uchitomi Y. Prevalence, predictive fac-
Oral Surg Oral Med Oral Pathol. 1994;78(6):711–7. tors, and screening for psychologic distress in patients
37. Schepers RH, Slagter AP, Kaanders JHAM, Van den with newly diagnosed head and neck cancer. Cancer.
Hoogen FJA, Merkx MAW. Effect of postoperative 2000;88(12):2817–23.
radiotherapy on the functional result of implants 51. Carrera PM, Kantarjian HM, Blinder VS. The finan-
placed during ablative surgery for oral cancer. Int J cial burden and distress of patients with cancer:
Oral Maxillofac Surg. 2006;35(9):803–8. understanding and stepping-up action on the fi nancial
38. Curi MM, Condezo AFB, Ribeiro K, Cardoso
toxicity of cancer treatment. CA Cancer J Clin.
CL. Long-term success of dental implants in patients 2018;68(2):153–65.
with head and neck cancer after radiation therapy. Int 52. de Oliveira RL, dos Santos RF, de Carvalho SHG,
J Oral Maxillofac Surg. 2018;47(6):783–8. Agripino GG, Canto MMN, de Vasconcelos Carvalho
39. Nobrega AS, Santiago JF Jr, de Faria Almeida DA, M, Marinho SA, Gallottini M, de Santana Sarmento
dos Santos DM, Pellizzer EP, Goiato MC. Irradiated DJ. Prospective evaluation of quality of life in patients
patients and survival rate of dental implants: a sys- with head and neck cancer. Oral Surg Oral Med Oral
tematic review and meta-analysis. J Prosthet Dent. Pathol Oral Radiol. 2017;123(3):350–7.
2016;116(6):858–66. 53. Tarsitano A, Pizzigallo A, Ballone E, Marchetti

40. Claudy MP, Miguens SAQ Jr, Celeste RK, Camara C. Health-related quality of life as a survival predictor
Parente R, Hernandez PAG, da Silva AN Jr. Time for patients with oral cancer: is quality of life associ-
interval after radiotherapy and dental implant fail- ated with long-term overall survival? Oral Surg Oral
ure: systematic review of observational studies Med Oral Pathol Oral Radiol. 2012;114(6):756–63.
and meta-analysis. Clin Implant Dent Relat Res. 54. Yabroff KR, Dowling EC, Guy GP Jr, Banegas MP,
2015;17(2):402–11. Davidoff A, Han X, Virgo KS, McNeel TS, Chawla
N, Blanch-Hartigan D, Kent EE. Financial hardship 57. Zeller JL. High suicide risk found for patients

associated with cancer in the United States: findings with head and neck cancer. JAMA. 2006;296(14):
from a population-based sample of adult cancer survi- 1716–7.
vors. J Clin Oncol. 2016;34(3):259. 58. Passik SD, Dugan W, McDonald MV, Rosenfeld B,
55. Ringash J, Bernstein LJ, Devins G, Dunphy C,
Theobald DE, Edgerton S. Oncologists’ recognition
Giuliani M, Martino R, McEwen S. Head and neck of depression in their patients with cancer. J Clin
cancer survivorship: learning the needs, meeting the Oncol. 1998;16(4):1594–600.
needs. Semin Radiat Oncol. 2018;28(1):64–74. 59. Cereda E, Cappello S, Colombo S, Klersy C, Imarisio
56. Sehlen S, Lenk M, Herschbach P, Aydemir U,
I, Turri A, Caraccia M, Borioli V, Monaco T, Benazzo
Dellian M, Schymura B, Hollenhorst H, Dühmke M, Pedrazzoli P. Nutritional counseling with or with-
E. Depressive symptoms during and after radio- out systematic use of oral nutritional supplements in
therapy for head and neck cancer. Head Neck. head and neck cancer patients undergoing radiother-
2003;25(12):1004–18. apy. Radiother Oncol. 2018;126(1):81–8.

Deepak Kademani - Improving Outcomes in Oral Cancer - A Clinical and Translational Update-Springer International Publishing (2020)

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Deepak Kademani - Improving Outcomes in Oral Cancer - A Clinical and Translational Update-Springer International Publishing (2020)

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

Deepak Kademani - Improving Outcomes in Oral Cancer - A Clinical and Translational Update-Springer International Publishing (2020)

Uploaded by

Copyright:

Available Formats

Improving Outcomes

ISBN 978-3-030-30093-7 ISBN 978-3-030-30094-4 (eBook)

© Springer Nature Switzerland AG 2020

1 Oral Epithelial Dysplasia���������������������������������������������������������������� 1

1.1 Introduction in the oral cavity compared to normal mucosa in

leukoplakia or keratosis), nonhomogenous (clini- reported to undergo malignant transformation

Mild Moderate Severe

Low Suspicion High Excision with Excision and

6 monthly review 6 montly review Severe Malignancy

Fig. 1.1 Authors’ algorithm for management of oral epithelial dysplasia

1.11.3 Medical Treatments after surgical excision of OEDs. In cohort study

2.1 Introduction allowing for the identification of biomarkers that

A variant of the amino acid l-cysteine, N-acetyl-­ 2.2.2 Topical Bleomycin

2.2.3 Polyphenols 2.2.3.2 Curcumin

3.1 Introduction be diagnosed on average at a statistically signifi-

© Springer Nature Switzerland AG 2020 23

p­ remalignant disease in this population [15]. The a

3.2.2 Lugol’s Iodine

B. A clinically evident, seemingly

No further action is Lesion Lesion persists Definitive Use cytologic adjunct to

Adjuncts used to evaluate

Positive result: Higher Negative result: Lower

Biopsy and histopathological

Positive diagnosis Negative diagnosis

3.8 Conclusion cal recommendations regarding screening for oral

4.1 Introduction 4.2 Changes to the American

© Springer Nature Switzerland AG 2020 33

4.3 Depth of Invasion exophytic lesions, thickness may be greater than

Justine Moe, Andrew Baker, and Brent Ward

© Springer Nature Switzerland AG 2020 45

defects can be reconstructed using an obturator

5.5.3 Oral Tongue

The oral tongue is the most common site of

5.5.4 Buccal Mucosa

function of tumor aggressiveness rather than

with frozen sections have been shown to be

limited N+ OCSCC, with adjuvant radiotherapy

5.10 Sentinel Node Biopsy

has not yet been validated. Furthermore, the

5.11 Dental Rehabilitation

Dental rehabilitation has become an integral

5.13 Conclusion nccn.org/professionals/physician_gls/default.aspx.

6.1  merican Joint Committee

Table 6.1 (continued) 6.2 Grade

Fig. 6.3 Broders grade

Fig. 6.4 Broders grade

Fig. 6.5 Broders grade

Fig. 6.6 Broders grade

6.4 Status of the Surgical removing oral squamous cell carcinoma. To this

Fig. 6.19 The tumor

Fig. 6.24 3+ p16

HPV-associated oral and oropharyngeal squa- Surgical management of the mandible is an

had retromolar cancers, and 22 patients had 6.9 Lymphovascular Invasion

Fig. 6.26 An embolus

Fig. 6.27 An embolus

© Springer Nature Switzerland AG 2020 97

8.1 Introduction pathways including a combination of those

1 Oral Epithelial Dysplasia�� 1

1.1 Introduction in the oral cavity compared to normal mucosa in

1.11.3 Medical Treatments after surgical excision of OEDs. In cohort study

2.1 Introduction allowing for the identification of biomarkers that

A variant of the amino acid l-cysteine, N-acetyl- 2.2.2 Topical Bleomycin

2.2.3 Polyphenols 2.2.3.2 Curcumin

3.1 Introduction be diagnosed on average at a statistically signifi-

p remalignant disease in this population [15]. The a

3.2.2 Lugol’s Iodine

3.8 Conclusion cal recommendations regarding screening for oral

4.1 Introduction 4.2 Changes to the American

4.3 Depth of Invasion exophytic lesions, thickness may be greater than

5.5.3 Oral Tongue

5.5.4 Buccal Mucosa

5.10 Sentinel Node Biopsy

5.11 Dental Rehabilitation

5.13 Conclusion nccn.org/professionals/physician_gls/default.aspx.

6.1 merican Joint Committee

Table 6.1 (continued) 6.2 Grade

6.4 Status of the Surgical removing oral squamous cell carcinoma. To this

had retromolar cancers, and 22 patients had 6.9 Lymphovascular Invasion

8.1 Introduction pathways including a combination of those

8.2 Contemporary Radiation

tumor bed. The inherent physical characteristics 8.3 Indications

Various prospective studies have validated the 9.4 Treatment Regimens

9.5.2.3 Lapatinib The VEGFR-directed therapies currently being

9.6.2.2 Sunitinib phase II trial of patients with platinum-resis-

occurring in 13–14% of patients receiving anti- 10.3 Additional Checkpoint

10.3.3 TIM-3 10.3.4 LAG-3

to be well-tolerated and produced substantial associated molecule, receptor signaling leads to

11.1 Introduction with aggressive surveillance and new advances in

12.2 Reconstruction of the Lips

of deglutition, particularly with fluids, but also 12.3 Reconstruction

12.4.2 Defects Greater Than 50%

12.4.2.1 Karapandzic Flap

12.4.2.2 Lip Switch Flaps

12.4.4.2 Abbe Flap

12.4.3.2 Gate Flap 12.4.4.3 Perialar Crescenteric

12.5 Buccal Mucosa and Cheek

12.6 Floor of Mouth

12.7 Tongue Reconstruction Free flap reconstruction is advocated to restore

12.8 Mandibular Reconstruction

12.8.3 Hard Tissue Reconstruction

12.9 Palatal and Maxillary Defects