Current diagnosis-and-therapy-for-head-and-neck-mal

CURRENT DIAGNOSIS AND THERAPY FOR HEAD AND NECK
MALIGNANCIES
CONSULTING EDITOR
NICHOLAS J. PETRELLI, MD, Medical Director, Helen F. Graham Cancer Center,
Newark, Delaware; and Professor of Surgery, Jefferson Medical College, Philadelphia,
Pennsylvania
GUEST EDITOR
WESLEY L. HICKS, JR, DDS, MD, FACS, Attending Surgeon, Department of Head and
Neck Surgery, Roswell Park Cancer Institute; Associate Professor of Otolaryngology,
Head and Neck Surgery, and Neurosurgery, School of Medicine and Biomedical
Sciences, State University of New York at Buffalo, Buffalo, New York; and Director,
Head and Neck Surgical Fellowship Program, Roswell Park Cancer Institute
CONTRIBUTORS
RONALD A. ALBERICO, MD, Associate Professor of Radiology, Assistant Clinical
Professor of Neurosurgery, School of Medicine and Biomedical Sciences, State
University of New York at Buffalo; Director of Neuroradiology/Head and Neck
Imaging, Department of Radiology, Roswell Park Cancer Institute, Buffalo, New York;
and Acting Director of Pediatric Neuroradiology, Buffalo Children’s Hospital, Buffalo,
New York
GARTH R. ANDERSON, PhD, Professor of Cellular and Molecular Biology, Department
of Cancer Genetics, Roswell Park Cancer Institute, Buffalo, New York
JIMMY J. BROWN, DDS, MD, FACS, Assistant Professor, Department of
Otolaryngology–Head and Neck Surgery, Charles R. Drew University of Medicine and
Science, Los Angeles, California
AMOS O. DARE, MD, Clinical Instructor, Department of Neurological Surgery, School of
Medicine and Biomedical Sciences, State University of New York at Buffalo, Buffalo,
New York
WADE DOUGLAS, MD, Fellow, Department of Head and Neck Surgery, Roswell Park
Cancer Institute, Buffalo, New York
KEVIN J. GIBBONS, MD, Program Director and Director of Skull Base Surgery,
Department of Neurological Surgery, School of Medicine and Biomedical Sciences,
State University of New York at Buffalo, Buffalo, New York
WILLIAM GIESE, MD, JD, Associate Professor, Department of Radiation Oncology,
Roswell Park Cancer Institute, Buffalo, New York
iii

RALPH W. GILBERT, MD, FRCSC, Associate Professor, Head and Neck Surgical
Oncology; Reconstructive Microsurgery, University Health Network; Princess
Margaret Hospital; and Department of Otolaryngology, University of Toronto,
Toronto, Canada
CHRISTINE G. GOURIN, MD, FACS, Assistant Professor, Department of
Otolaryngology–Head and Neck Surgery, Medical College of Georgia, Augusta,
Georgia
PATRICK J. GULLANE, MB, FRCSC, FACS, Otolaryngologist-in-Chief, University Health
Network; Wharton Chair in Head and Neck Surgery, Princess Margaret Hospital; and
Professor and Chairman, Department of Otolaryngology, University of Toronto,
Toronto, Canada
SYED HAMED S. HUSAIN, DO, Radiology Resident, School of Medicine and Biomedical
Sciences, State University of New York at Buffalo, Buffalo, New York
DOMINICK LAMONICA, MD, Director of Nuclear Medicine, Division of Diagnostic
Imaging, Roswell Park Cancer Institute; and Assistant Professor of Radiology and
Clinical Nuclear Medicine, School of Medicine and Biomedical Sciences, State
University of New York at Buffalo, Buffalo, New York
PABLO MOJICA-MANOSA, MD, Fellow, Department of Head and Neck Surgery,
JEFFREY N. MYERS, MD, PhD, Associate Professor of Head and Neck Surgery,
Department of Head and Neck Surgery, The University of Texas M.D. Anderson
Cancer Center, Houston, Texas
LARRY L. MYERS, MD, Department of Otolaryngology–Head and Neck Surgery,
University of Texas Southwestern Medical Center, Dallas, Texas
RYAN F. OSBORNE, MD, Director, Head and Neck Oncology, Cedars-Sinai Medical
Center; and Assistant Professor, Department of Otolarynology–Head and Neck
Surgery, Charles R. Drew University of Medicine and Science, Los Angeles, California
LANCE E. OXFORD, MD, Department of Otolaryngology–Head and Neck Surgery,
University of Texas Southwestern Medical Center, Dallas, Texas
CARSTEN E. PALME, MB BS, FRACS, Clinical Fellow, Oncologic Head and Neck
Surgery, Department of Otolaryngology, University of Toronto, Toronto, Canada
JAMES REIDY, DO, Fellow, Department of Head and Neck Surgery, Roswell Park Cancer
Institute, Buffalo, New York
NESTOR R. RIGUAL, MD, FACS, Associate Professor of Clinical Otolaryngology, School
of Medicine and Biomedical Sciences, State University of New York at Buffalo; and
Attending Surgeon, Section of Plastic and Reconstructive Surgery, Department of Head
and Neck Surgery, Roswell Park Cancer Institute, Buffalo, New York
JAMES K. SCHWARZ, MD, Assistant Professor, Department of Medicine, Roswell Park
Cancer Institute, Buffalo, New York
iv CONTRIBUTORS

IGOR SIROTKIN, MD, Radiology Resident, School of Medicine and Biomedical Sciences,
State University of New York at Buffalo, Buffalo, New York
DANIEL L. STOLER, PhD, Assistant Professor, Department of Experimental Pathology,
MAUREEN SULLIVAN, DDS, Chief, Department of Dentistry and Maxillofacial
Prosthetics, Roswell Park Cancer Institute, Buffalo, New York
DAVID J. TERRIS, MD, FACS, Porubsky Professor and Chairman, Department
of Otolaryngology–Head and Neck Surgery, Medical College of Georgia, Augusta,
Georgia
KEITH WILSON, MD, Associate Professor, ENT/Head and Neck Surgery, University
of Cincinnati, Cincinnati, Ohio
SAM M. WISEMAN, MD, FRCS(C), Assistant Professor of Surgery, University of British
Columbia School of Medicine; and Attending Surgeon, Department of Surgery, St.
Paul’s Hospital, Vancouver, British Columbia, Canada
ROBERT L. WITT, MD, Chief, Section of Otolaryngology, Department of Surgery,
Christiana Care Health System, Newark, Delaware; and Assistant Professor,
Department of Otolaryngology, Jefferson Medical College, Philadelphia, Pennsylvania
MAHER N. YOUNES, MD, Postdoctoral Fellow, Department of Head and Neck Surgery,
University of Texas M. D. Anderson Cancer Center, Houston, Texas
CONTRIBUTORS v

FORTHCOMING ISSUES
April 2004
Multidisciplinary Approach to
Anal Cancer
Morton S. Kahlenberg, MD, and
Charles R. Thomas, Jr, MD, Guest Editors
July 2004
Palliative Care
Lawrence D. Wagman, MD,
Guest Editor
October 2004
Adjuvant Therapy of Pancreatic
Adenocarcinoma
John P. Hoffman, MD, Guest Editor
RECENT ISSUES
October 2003
Intraoperative Radiotherapy
Hollis W. Merrick, MD, and
Charles R. Thomas, Jr, MD, Guest Editors
July 2003
Management of Peritoneal Surface
Malignancy
Paul H. Sugarbaker, MD, FACS, FRCS,
Guest Editor
April 2003
Emerging Perspectives in
Soft Tissue Sarcoma
Raphael E. Pollack, MD, PhD
Guest Editor
VISIT THESE RELATED WEB SITES
Access your subscription at:
http://www.TheClinics.com

Foreword
Current diagnosis and therapy for head
and neck malignancies
Consulting Editor
Approximately 1,334,100 new cancer cases were diagnosed in 2003. Since
1990, over 17 million new cancer cases have been diagnosed. According to
the American Cancer Society, these new cancer cases do not include carci-
noma in situ of any site except urinary bladder and do not include basal and
squamous skin cancers. In 2003, approximately 556,500 Americans died of
cancer, which is equivalent to more than 1500 people a day.
Cancers of the oral cavity and pharynx were diagnosed in an estimated
27,700 new cases in 2003. These incidence rates are more than twice as high
in men as in women and are greatest in men who are over age 50. Neverthe-
less, incidence rates for cancers of the oral cavity and pharynx continued to
decline in the 1990s in both African American and white males and females.
There were an estimated 7200 deaths in 2003 from oral cavity and pharyng-
eal cancer. The known risk factors for these cancers are cigarettes, cigars,
pipe smoking, and the use of smokeless tobacco. Excessive consumption
of alcohol is also a risk factor.
In this issue of the Surgical Oncology Clinics of North America, under the
direction of Wesley Hicks, Jr., DDS, MD, an outstanding array of authors
has been assembled to discuss many clinical and scientific issues regarding
cancers of the head and neck. Dr. Hicks is a member of the Department
of Head and Neck Surgery at the Roswell Park Cancer Institute in Buffalo,
New York, and an Associate Professor of Surgery at the State University of
New York at Buffalo.
Nicholas J. Petrelli, MD
1055-3207/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.soc.2003.12.010
Surg Oncol Clin N Am
13 (2004) xiii–xiv

The article by Wiseman, Stoler, and Anderson on the role of genomic
instability in the pathogenesis of squamous cell carcinoma of the head and
neck is especially interesting. These researchers are from the Departments of
Surgical Oncology, Experimental Pathology, and Cancer Genetics, respec-
tively. Cancer predisposition genes and the genetic heterogeneity of head
and neck tumors are discussed in detail.
On the clinical side, the article by Osborne and Brown from the Division
of Otolaryngology/Head and Neck Surgery at the University of California–
Los Angeles Medical Center deals with carcinoma of the oral pharynx with
an analysis of subsite treatment heterogeneity. This article provides a clear
discussion of the clinical treatment modalities in relationship to histopatho-
logic characteristics.
As I have stated in previous forewords for the Surgical Oncology Clinics
of North America, this issue is a must-read for trainees in the three major
disciplines of surgery, radiation oncology, and medical oncology. Trainees
in pathology and radiology should also make this issue a part of their educa-
tional matriculation.
I congratulate Dr. Hicks and his colleagues on an outstanding issue of the
Surgical Oncology Clinics of North America.
Nicholas J. Petrelli, MD
Consulting Editor
Helen F. Graham Cancer Center
4701 Ogletown-Stanton Road
Suite 1212
Newark, Delaware 19713, USA
Jeﬀerson Medical College
Philadelphia, Pennsylvania
xiv N.J. Petrelli / Surg Oncol Clin N Am 13 (2004) xiii–xiv

Preface
Current diagnosis and therapy for head
and neck malignancies
Guest Editor
Squamous cell carcinoma is the most common histologic malignancy of
the head and neck region. Despite this monotonous pathologic presentation,
there are a plethora of treatment options and clinical outcomes based on
both the site and stage of the primary tumor. The recent movement to com-
bined modality therapy has been driven by the clinical need to improve dis-
ease-free survival while minimizing functional and cosmetic morbidity. Here
much more work can and should be done. Advancement in treatment
and survival in head and neck surgery, in our opinion, requires further
translational research efforts melding clinical expertise with bench scientific
discovery.
Oncologic head and neck surgery is one of the most clinically challenging
and complex areas of surgical oncology. This issue of the Surgical Oncology
Clinics of North America presents what we believe is a rational organ-specific
approach to malignancies of this region. This issue is not intended to be an
exhaustive explanation regarding the armamentarium or clinical paradigms
for the treatment of head and neck cancer. We anticipate, however, that it
will serve as a solid foundation for those who wish to pursue a personal clin-
ical interest in head and neck surgical oncology.
We anticipate that this issue will give readers a general overview of head
and neck cancer and the common surgical/medical approaches to this dis-
ease. Each article is a self-contained clinical caveat with a complete explan-
ation of how specific subsites within the head and neck region can be
Wesley L. Hicks, Jr, MD
doi:10.1016/j.soc.2003.12.009
Surg Oncol Clin N Am
13 (2004) xv–xvi

evaluated and treated. After reviewing the articles in aggregate, my belief
was affirmed that an understanding of the clinical nuances pertinent to each
head and neck subsite must be mastered to obtain effective and improved
clinical outcomes.
I wish to express my sincere gratitude and thanks to the contributing
authors, whose diligent and exacting work made this issue possible.
Wesley L. Hicks Jr, DDS, MD, FACS
Department of Head and Neck Surgery
Roswell Park Cancer Institute
School of Medicine and Biomedical Sciences
State University of New York at Buffalo
Elm & Carlton Streets
Buffalo, NY 14263, USA
E-mail address: wesley.hicks@roswellpark.org
xvi W.L. Hicks / Surg Oncol Clin N Am 13 (2004) xv–xvi

The role of genomic instability
in the pathogenesis of squamous cell
carcinoma of the head and neck
Sam M. Wiseman, MD, FRCS(C)a,1
,
Daniel L. Stoler, PhDb
, Garth R. Anderson, PhDa,c,*
a
Department of Surgical Oncology, Roswell Park Cancer Institute,
Elm and Carlton Streets, Buffalo, NY 14263, USA
b
Department of Experimental Pathology, Roswell Park Cancer Institute,
c
Department of Cancer Genetics, Roswell Park Cancer Institute,
Human beings are composed of a highly complex community of cells, and
each cell type has its own role that is defined by the genetic instructions it
expresses. For cells to function normally, their genetic instructions must be
accurately transmitted from one generation to the next. The information
carried by the genetic code must be accurately replicated and efficiently
repaired to ensure the survival of cells, organisms, and species. In humans,
the importance of maintaining the integrity of their genetic blueprint can be
appreciated by the approximately 130 genes involved in DNA repair alone
[1]. When these cellular self-repair mechanisms break down, or when bathed
in an environment of genotoxic compounds, cells become genomically
unstable. Ultimately, this genomic destabilization, or instability, results in
the natural selection of a genomically heterogeneous cellular mass, or
cancer, that threatens the survival of the organism as a whole. The impor-
tance of genomic instability in tumorigenesis and tumor evolution may be
seen in the work of numerous investigators who have demonstrated that
cells must undergo multiple genetic alterations to become neoplastic. Loeb
[2,3] and Jackson and Loeb [4] determined that the normal rate of mutation
is insufficient to allow for the observed genetic change in neoplasms to take
1
Current address: Department of Surgery, St. Paul’s Hospital, 1081 Burrard Street,
Vancouver, British Columbia, Canada, V6Z 1Y6.
* Corresponding author.
E-mail address: garth.anderson@roswellpark.org (G.R. Anderson).
doi:10.1016/S1055-3207(03)00118-2
Surg Oncol Clin N Am 13 (2004) 1–11

place. Genomic instability is now seen as an essential enabling component
that allows tumors to evolve [5].
With a yearly global incidence of 500,000 cases, head and neck squa-
mous cell carcinoma (HNSCC) ranks as the sixth most prevalent cancer
worldwide [6]. The American Cancer Society estimates that there will be
28,900 new cases of oral cavity and pharynx cancers diagnosed and 8900 new
cases of larynx cancer diagnosed in 2002 [7]. Despite advancements in medi-
cine, surgery, and radiation therapy, the long-term survival of individuals
diagnosed with HNSCC has not increased significantly over the past 20 years,
with the 5-year survival rate remaining at 52% in the United States [8]. It is
estimated there will be 11,100 deaths from cancer at these head and neck sites
in 2002 [8]. Mortality from this disease correlates with tumor size and the
presence of local nodal or distant metastatic disease [9–11]. This article
provides an overview of the clinical and experimental evidence, and
implications, of genomic instability as a major force driving HNSCC tumori-
genesis and evolution.
Many patients with head and neck cancer are cancer predisposed
HNSCC arises from a complex interaction between the host (genetic
factors) and the environment. Tobacco exposure has long been recognized
as increasing the risk of developing HNSCC between 2- and 20-fold [12].
More than 50 years ago, Slaughter et al [13] recognized the ‘‘field can-
cerization’’ that occurs in patients with HNSCC as a consequence of pro-
longed carcinogenic exposure of the upper aerodigestive tract. It is this
‘‘field cancerization’’ that is believed to be responsible for the 2% yearly
incidence of second primary tumors that develop in this patient population.
Furthermore, the concurrent consumption of alcohol with tobacco may
have a multiplicative effect on the risk of developing HNSCC [14–16]. In
addition to tobacco and alcohol exposure, other environmental factors that
are currently believed to play a role in HNSCC development include viruses,
radiation exposure, and certain nutritional deficiencies [17–19].
Epidemiologic and experimental evidence suggests that, because of an
inherent inability to maintain their genomic integrity in the presence of specific
environmental stressors, certain individuals demonstrate a predisposition to
developing head and neck tumors. The occurrence of this malignancy, even in
the absence of environmental carcinogen exposure, supports this concept. The
current authors recently described a cohort of 40 nonsmoking, nondrinking
patients with HNSCC treated at Roswell Park Cancer Institute. These
patients tended to be elderly (median age, 60 y), female, and white. In addition,
they had oral cavity primary tumors and were predisposed to second primary
tumor development. Despite a lack of exposure to tobacco and alcohol, 10
patients (25% of study population) eventually developed a second primary
tumor. The occurrence of second primary tumors in this patient population
suggests a possible genetic pre11disposition of these individuals to HNSCC
2 S.M. Wiseman et al / Surg Oncol Clin N Am 13 (2004) 1–11

development [20]. Information concerning whether nonsmoking patients with
HNSCC have a prognosis different from smoking patients with HNSCC is
limited. Koch and McQuone [21] described a cohort of 46 nonsmokers
(individuals who never used tobacco on a regular basis) who developed
HNSCC, and compared them to a large cohort of smokers who developed
HNSCC. Of the 46 nonsmokers in this study, 37 patients (84%) were also
nondrinkers. The overall length of survival of the patients in this study did not
vary significantly with either smoking history or drinking history.
Family members of patients with HNSCC also are predisposed to upper
aerodigestive tract tumor development. In a study performed in The
Netherlands, Copper et al [22] found that first-degree relatives of patients
with HNSCC had an increased risk of developing upper aerodigestive tract
tumors (relative risk [RR], 3.5), with an especially high risk amongst siblings
(RR, 14.6). Foulkes et al [23] performed a similar study in southern Brazil
and found first-degree relatives of patients with HNSCC to have a relative
risk of 3.5 times the general population for developing HNSCC. In this
study, siblings had a relative risk of 8.6 for developing the disease. In a study
performed in the United States, however, Goldstein et al [24] found only
a slight increase in the relative risk of first-degree relatives of patients with
HNSCC for developing these tumors themselves (RR, pharynx, 1.7; RR,
oral, 1.2). In a study examining 26 individuals with multiple primary upper
aerodigestive tract tumors, Foulkes et al [25] demonstrated that the relative
risk of developing HNSCC was significantly higher in relatives of individuals
who developed multiple versus single primary tumors (RR, 7.89 versus 3.53,
respectively). Only the study by Goldstein et al [24] collected smoking
details for relatives, and matching was performed according to racial group.
Thus, epidemiologic studies suggest the families of patients with HNSCC
are themselves genetically predisposed to developing HNSCC, although the
magnitude of this predisposition is limited.
There are several rare ‘‘cancer predisposition’’ syndromes that arise from
genes that maintain genomic integrity. These syndromes have a constellation
of associated malignancies, including HNSCC. These cancer predisposition
syndromes include the genomic instability syndromes, Werner’s syndrome,
Bloom syndrome, Fanconi’s anemia, and ataxia telangectasia. Patients who
have Bloom syndrome or Werner’s syndrome have deficient RecQ-like
DNA helicases [26,27]. Werner’s syndrome is an autosomal recessive dis-
order that has been linked to the WRN locus on chromosome 8p. Homo-
zygous individuals age prematurely and are cancer predisposed [26]. Bloom
syndrome is an autosomal recessive disorder that arises from an alteration
of the Bloom syndrome gene, which encodes a DNA helicase that interacts
with topoisomerase III and plays an important role in DNA repair. Affected
individuals have impaired fertility, immunodeficiency, dwarfism, and are
cancer predisposed [27]. Genomic instability arises in these individuals from
impaired DNA repair, and consequently, these individuals are vulnerable to
DNA damage and subsequent cancer development [26,27].
3S.M. Wiseman et al / Surg Oncol Clin N Am 13 (2004) 1–11

Fanconi’s anemia is a rare genetic disorder in which affected individuals
are predisposed to developing squamous cell carcinoma (SCC) of the gingiva,
tongue, and mandible. This disorder is also characterized by developmental
abnormalities and a propensity to develop hematologic disorders and bone
marrow failure. It is currently believed that the protein defects that arise in
affected individuals may result in loss of regulation of DNA repair [28].
Ataxia telangiectasia is a rare genetic disorder that arises from a defect in the
ataxia telangiectasia gene (ATM gene). The ATM gene functions in survey-
ing for DNA damage and is responsible for activating DNA repair and
apoptosis genes. Individuals homozygous for the defective gene develop
progressive neuromuscular degeneration, an unsteady gait, and have facial
or conjunctival telangiectasias, and are predisposed to cancer development.
Heterozygous individuals also have a cancer predisposition [29].
Hereditary nonpolyposis colorectal cancer is associated with a high fre-
quency of microsatellite instability. Mismatches of nucleotides may occur
during DNA replication when DNA polymerase inserts the wrong bases into
newly synthesized DNA. Normally, these DNA mismatches are repaired by
mismatch repair enzymes, and individuals with germ-line mutations in the
genes that encode these mismatch repair enzymes have impaired DNA
replication fidelity. The integrity of the mismatch repair enzyme systems is
measured by microsatellite instability, or a measure of the integrity of short,
tandemly repeated DNA sequences (microsatellites) distributed throughout
the human genome [30]. Individuals diagnosed with hereditary nonpolyposis
colorectal cancer have been reported to be at increased risk of developing
HNSCC [31].
Further evidence suggesting patients with HNSCC are cancer predisposed
may be appreciated on experimental studies applying mutagen sensitivity
testing to the HNSCC patient population. Hsu et al [32] developed the
bleomycin mutagen sensitivity test as a method of assessing cellular DNA
repair capacity. This test is performed by exposing lymphocytes in culture to
the drug bleomycin and quantifying the number of chromosome breaks per
lymphocyte in culture. This test has been successfully able to identify
individuals at high risk for developing head and neck tumors [33,34]. The
highest rates of chromosomal breakage are observed in individuals with
a family history of HNSCC and in those with multiple primary tumors
[33,34]. The DNA repair capacity after exposure to the carcinogen
benzo(a)pyrene diol epoxide also is impaired in patients with HNSCC [35].
In addition, investigators have found that certain polymorphisms of the
DNA repair gene XRCC1 are associated with an increased risk for HNSCC
development [36].
Genomic instability and head and neck cancer
Genomic instability may be broadly considered either chromosomal or
intrachromosomal. Intrachromosomal genomic instability, the form most

frequently observed in sporadic disease, consists of amplifications, deletions,
inversions, and oligobase or point mutations. These alterations can be readily
assessed by comparative genomic hybridization (CGH) and microsatellite
instability measurements. Independent measures of intrachromosomal
instability can be assessed by cytogenetic techniques and CGH on ordered
Bac clone arrays. Chromosomal instability, changes in ploidy status, and
chromosomal translocations can be visualized by cytogenetic analytic
techniques. Although not all chromosomes are affected in each tumor, all
autosomal chromosome arms have been described as being affected in head
and neck tumors [17,18,37]. Although clustering aberrations have been
described, consistent cyogenetic abnormalities common to all HNSCCs have
not been demonstrated [17,18,37]. CGH analyses performed on HNSCCs
have demonstrated multiple chromosomal aberrations found in these tumors
[38–40]. Using CGH methodology, Hashimoto et al [40] were able to
demonstrate a correlation between specific chromosomal aberrations and
pathologic tumor stage in 32 patients with HNSCC.
The literature has been conflicting regarding the role of microsatellite
instability in head and neck tumor progression. Piccinin et al [41]
demonstrated low rates of microsatellite instability in patients who had
HNSCC with a single primary or multiple primary tumors. Both groups had
similarly low rates of microsatellite instability. Using makers for 11
chromosomal loci, El-Naggar et al [42] examined microsatellite instability
in peripheral blood, dysplastic tissue, and SCCs. Although microsatellite
instability was absent in the blood, levels were twice as high in the cancers as
in the dysplastic tissue (30% versus 15%, respectively). This finding led these
authors to conclude that microsatellite instability played an important role in
HNSCC progression. Field et al [43], however, were unable to demonstrate
any correlation between microsatellite instability and clinicopathologic
features (tumor site, tumor grade, nodal metastasis, disease stage, history of
prior treatment, or alcohol consumption) in patients. Thus, whereas
measurements of microsatellite instability in HNSCC suggest genomic
instability is driving tumor progression, they have been of limited clinical
applicability.
The study of the fraction of chromosomal arms on which allele loss is
observed, or the fractional allelic loss rate (FAL), has demonstrated how
measurements of genomic instability in patients with HNSCC may be of
clinical value. By studying 80 HNSCCs, Field et al [44] were able to dem-
onstrate a positive correlation between FAL and tumor grade, neck nodal
status, and overall patient prognosis.
Cancer predisposition genes
Cancer predisposition genes may play an important role in the increased
genetic susceptibility observed in patients who have HNSCC. Important

cancer predisposition genes that have been found to be mutated in patients
with HNSCC include p16 and TP53 tumor-suppressor genes.
The p16 tumor-suppressor gene is located on chromosome 9p21 and
encodes a 16kDa protein that binds to the cyclin-dependent kinases (cdk) 4
and 6 and prevents them from complexing with cyclin D1. This inhibition of
D1-cdk 4/6 complex activity does not allow for retinoblastoma phosphory-
lation, and thus blocks the cell cycle G1/S transition. Somatic homozygous
deletions and mutations of the 9p21 region are commonly observed in patients
with sporadic HNSCC [45]. Recently, Yu et al [46] performed molecular
analyses of a family with a high incidence of HNSCC and melanoma, and
identified a germ-line p16 tumor-suppressor gene mutation. These results,
along with this group’s earlier description of a separate, unrelated family
harboring a p16 germ-line mutation and exhibiting a HNSCC predisposition,
suggest a familial HNSCC syndrome may exist [46].
Li-Fraumeni syndrome arises as a consequence of germ-line mutations in
the TP53 gene, which is located on chromosome 17. The affected individual
develops a sarcoma before age 45, has a first-degree relative with cancer
before age 45, and has another first- or second-degree relative with a history
of sarcoma at any age or a cancer diagnosed before age 45. Patients who
have Li-Fraumeni syndrome are at increased risk of developing laryngeal
cancer [47]. The TP53 gene is a well-known tumor-suppressor gene and
represents one of the most common sites for genetic abnormalities to be
found in human tumors [48]. The TP53 gene is frequently mutated in several
human cancers, including lung cancer, colon cancer, and breast cancer [49–
51]. Multiple investigators have demonstrated that 40% or more of sporadic
HNSCC contain TP53 mutations [52–54]. The TP53 gene product, the P53
protein, is believed to have a dual role in protecting the cell from cancer
development. The P53 protein causes cell cycle arrest, allowing damaged
DNA to be repaired; alternatively, it helps prevent DNA damage from being
passed on to the next cell generation by causing damaged cells to undergo
apoptosis before division. Therefore, cells that lack or only produce abnormal
P53 protein are more susceptible to malignant transformation [48].
In sporadic head and neck tumors, TP53 mutation is currently believed
to represent a relatively late step in head and neck tumor evolution. The
incidence of TP53 mutation in premalignant head and neck lesions has been
reported to be much lower (19%) than in invasive lesions (43%) [54]. Shin
et al [55] reported P53 immunohistologic expression by the adjacent normal
epithelium in 6 of 31 (19%) normal epithelium specimens adjacent to
squamous cell carcinomas, in 7 of 24 (29%) hyperplastic lesions, in 12 of 26
(46%) dysplastic lesions, and in 28 of 48 (58%) HNSCCs. In a group of 232
patients with HNSCC, Koch et al [56] described a much higher incidence of
TP53 mutations in smokers (44%) than in nonsmokers (18%). Field et al
[57] correlated tumor TP53 status and the patient’s history of smoking and
alcohol consumption, also suggesting both substances may be linked to
aberrant P53 expression in HNSCC.

Genetic heterogeneity of head and neck tumors
Despite considerable histologic homogeneity, HNSCCs do not all exhibit
uniform biologic behavior. Tumors from different head and neck sites display
a wide range of biologicbehaviors, despite often being separatedby only a very
small anatomic distance, and they are not treated in a uniform manner.
Recently, investigators have provided evidence that suggests, at the genetic
level, HNSCCs are actually a heterogeneous group of disease entities. Takes
et al [58] studied expression of several proteins (P53, retinoblastoma (Rb),
cyclin D1, myc, bcl-2, epidermal growth factor receptor (EGFR), neuro-
glioblastoma derived oncogene (neu), E-cadherin, epithelial cellular adhesion
molecule (Ep-CAM), desmoplakin 1, and nonmetastatic protein 23 [nm23])
in 33 laryngeal cancers, 31 pharyngeal cancers, and 36 oral cancers. These
authors found that cyclin D1 had a very high level of expression in the pharynx
cancer (P = 0.0004) and EGFR had a very low level of expression in the
larynx cancer (P0.0001). Rodrigo et al [59] examined 38 laryngeal cancers,
29 oropharyngeal cancers, and 37 hypopharyngeal cancers for the following:
amplification of oncogenes at the 11q13 region (CCND1, FGF3, FGF4,
EMS1), oncogenes MYC and ERBB1, for integration of the human papilloma
virus types 6b and 16, loss of heterozygosity at P53 and NAT2, and cellular
DNA content. This group found that FGF3 and FGF4 had a significantly
higher degree of amplification in the hypopharyngeal tumors (P = 0.006 and
P = 0.0002, respectively). Aneuploid tumors were found in a significantly
lower proportion of larynx tumors than in other sites (P = 0.03). These
observed differences in genes and gene expression, at different head and neck
disease sites, provide early evidence suggesting that HNSCC may actually
represent a genetically heterogeneous group of diseases. Furthermore, there
may be different pathways of tumorigenesis and tumor progression that are
responsible for the observed clinical differences in biologic behavior of disease
originating from different sites. The genetic heterogeneity of head and neck
tumors is not surprising when the genomic instability exhibited by these
tumors is considered. Not only are these tumors genetically heterogeneous
from one tissue site to the next, however, but they may also exhibit
considerable intratumoral heterogeneity. Jacob et al [60] evaluated five tumor
regions in 12 patients who underwent surgery for oropharyngeal carcinoma.
Specimens were each evaluated by immunohistochemical assessment for
proliferation markers (Ki67 and proliferating cell nuclear antigen), for
quantitative DNA content, and morphologic tumor-front grading. Their
results demonstrated a considerable variation of proliferation and differen-
tiation both intratumorally (within the same tumor) and extratumorally
(between different tumors). Consistent with the genomic instability exhibited
by head and neck malignancies, these results suggest that HNSCCs are
genetically heterogeneous tumors. As has been suggested, perhaps head and
neck tumors should be studied in a manner similar to how they are viewed and
treated clinically, as separate and distinct disease entities [59].

Summary
Measurements of genomic instability, or identification of genes respon-
sible for instability, may potentially be used as molecular markers to predict
disease course and response to therapy. Other possible applications include
use of genomic instability measurements, or genes, as tools to screen for
primary or recurrent disease. Methodologies for detection of genetic
mutations in saliva, blood, and sputum have already been described
[61,62]. Brennan et al [63] have described a molecular technique for
analyzing histopathologically negative margins and lymph nodes for the
presence of p53 gene mutation. This study showed that a positive molecular
margin significantly predicted disease recurrence.
The recognition that HNSCC is a genetically heterogeneous disease repre-
sents a major step toward developing an understanding of its underlying
genetic basis. To develop an insight into this genetically heterogeneous
disease, investigators must not only focus their efforts on specific head and
neck disease sites. Laser-capture microdissection represents a powerful tool
for isolating very specific cell populations from tumors [64]. Leethanakul et al
[65] performed laser-capture microdissection on oral cavity SCC to construct
stage-specific cDNA libraries. Sequencing of 96 clones from each of the six
libraries constructed suggested the existence of 132 novel genes, which may
play a role in the pathogenesis of HNSCC.
The current literature suggests that many individuals diagnosed with
HNSCC are genetically predisposed to developing malignancy because of
some inherent deficiency of their capacity to maintain their genome in the
presence of environmental stressors. Head and neck cancers are highly het-
erogeneous tumors and exhibit a wide variety of forms of genomic
instability. Thus, genomic instability may be viewed as a fundamental force
driving head and neck tumorigenesis and evolution. Future study of the
specific genetic mechanisms that underlie genomic instability in the HNSCC
patient population is needed. It is only through study of this fundamental
force that drives the development of these tumors that clinicians may gain
the insight required to develop new diagnostic and therapeutic modalities to
benefit the HNSCC patient population as a whole.
References
[1] Wood RD, Mitchell M, Sgouros J, Lindahl T. Human DNA repair genes. Science 2001;
291(5507):1284–9.
[2] Loeb LA. A mutator phenotype in cancer. Cancer Res 2001;61(8):3230–9.
[3] Loeb LA. Mutator phenotype may be required for multistage carcinogenesis. Cancer Res
1991;51(12):3075–9.
[4] Jackson AL, Loeb LA. The mutation rate and cancer. Genetics 1998;148(4):1483–90.
[5] Anderson GR, Stoler DL, Brenner BM. Cancer: the evolved consequence of a destabilized
genome. Bioessays 2001;23(11):1037–46.

[6] Parkin DM, Laara E, Muir CS. Estimates of the worldwide frequency of sixteen major
cancers in 1980. Int J Cancer 1988;41(2):184–97.
[7] American Cancer Society. Cancer facts and figures 2002. American Cancer Society:
Atlanta, GA; 2002.
[8] Swango PA. Cancers of the oral cavity and pharynx in the United States: an epidemiologic
overview. J Public Health Dent 1996;56(6):309–18.
[9] Quon H, Liu FF, Cummings BJ. Potential molecular prognostic markers in head and neck
squamous cell carcinomas. Head Neck 2001;23(2):147–59.
[10] Overgaard J, Hansrn HS, Overgaard M, et al. A randomized double-blinded phase III
study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic
larynx and pharynx carcinoma. Results of the Danish Head and Neck Cancer study
(DAHANCA) protocol 5–85. Radiother Oncol 1998;46:135–46.
[11] Van den Bogaert W, Van der Schueren E, Horiot JC, et al. The EORTC randomized trial
on three fractions per day and misonidazole in advanced head and neck cancer: prognostic
factors. Radiother Oncol 1995;35:100–6.
[12] Maier H, Dietz A, Gewelke U, Heller WD, Weidauer H. Tobacco and alcohol and the risk
of head and neck cancer. Clin Investig 1992;70(3–4):320–7.
[13] Slaughter DP, Southwick HW, Smejkal W. ‘‘Field cancerization’’ in oral stratified epi-
thelium. Cancer 1953;6(5):963–8.
[14] Blot WJ, McLaughlin JK, Winn DM, et al. Smoking and drinking in relation to oral
and pharyngeal cancer. Cancer Res 1988;48(11):3282–7.
[15] Elwood JM, Pearson JC, Skippen DH, Jackson SM. Alcohol, smoking, social and occu-
pational factors in the aetiology of cancer of the oral cavity, pharynx and larynx. Int
J Cancer 1984;34(5):603–12.
[16] Boffetta P, Mashberg A, Winkelmann R, Garfinkel L. Carcinogenic effect of tobacco
smoking and alcohol drinking on anatomic sites of the oral cavity and oropharynx. Int
J Cancer 1992;52(4):530–3.
[17] Scholes AG, Field JK. Genomic instability in head and neck cancer. Curr Top Pathol 1996;
90:201–22.
[18] Field JK. Genomic instability in squamous cell carcinoma of the head and neck. Anticancer
Res 1996;16(4C):2421–31.
[19] Friedlander PL. Genomic instability in head and neck cancer patients. Head Neck 2001;
23(8):683–91.
[20] Wiseman SM, Swede H, Stoler DL, Anderson GR, Rigual NR, Hicks WL, Douglas WG,
et al. Squamous cell carcinoma of the head and neck in nonsmokers and nondrinkers: an
analysis of clinicopathologic characteristics and treatment outcomes. Ann Surg Onc 2003;
10(5):551–7.
[21] Koch WM, McQuone S. Clinical and molecular aspects of squamous cell carcinoma of
the head and neck in the nonsmoker and nondrinker. Curr Opin Oncol 1997;9(3):257–61.
[22] Copper M, Jovanic A, Nauta J. Role of genetic factors in the aetiology of squamous cell
carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg 1995;121:157–60.
[23] Foulkes WD, Brunet JS, Kowalski LP, Narod SA, Franco EL. Family history of cancer is
a risk factor for squamous cell carcinoma of the head and neck in Brazil: a case-control
study. Int J Cancer 1995;63(6):769–73.
[24] Goldstein AM, Blot WJ, Greenberg RS, et al. Familial risk in oral and pharyngeal cancer.
Eur J Cancer B Oral Oncol 1994;30B(5):319–22.
[25] Foulkes WD, Brunet JS, Sieh W, Black MJ, Shenouda G, Narod SA. Familial risks of
squamous cell carcinoma of the head and neck: retrospective case-control study. BMJ
1996;313(7059):716–21.
[26] Lebel M. Werner syndrome: genetic and molecular basis of a premature aging disorder.
Cell Mol Life Sci 2001;58(7):857–67.
[27] Llerena JC Jr, Murer-Orlando M. Bloom syndrome and ataxia telangiectasia. Semin
Hematol 1991;28(2):95–103.

[28] Yamashita T, Nakahata T. Current knowledge on the pathophysiology of Fanconi anemia:
from genes to phenotypes. Int J Hematol 2001;74(1):33–41.
[29] Meyn MS. Ataxia-telangiectasia and cellular responses to DNA damage. Cancer Res 1995;
55(24):5991–6001.
[30] Lawes DA, SenGupta SB, Boulos PB. Pathogenesis and clinical management of hereditary
non-polyposis colorectal cancer. Br J Surg 2002;89(11):1357–69.
[31] Lynch HT, Kriegler M, Christiansen TA, Smyrk T, Lynch JF, Watson P. Laryngeal
carcinoma in a Lynch syndrome II kindred. Cancer 1988;62(5):1007–13.
[32] Hsu TC, Johnston DA, Cherry LM, et al. Sensitivity to genotoxic effects of bleomycin in
humans: possible relationship to environmental carcinogenesis. Int J Cancer 1989;15;43(3):
403–9.
[33] Cloos J, Spitz MR, Schantz SP, et al. Genetic susceptibility to head and neck squamous cell
carcinoma. J Natl Cancer Inst 1996;88(8):530–5.
[34] Schantz SP, Spitz MR, Hsu TC. Mutagen sensitivity in patients with head and neck
cancers: a biologic marker for risk of multiple primary malignancies. J Natl Cancer Inst
1990;21;82(22):1773–5.
[35] Wang LE, Sturgis EM, Eicher SA, Spitz MR, Hong WK, Wei Q. Mutagen sensitivity to
benzo(a)pyrene diol epoxide and the risk of squamous cell carcinoma of the head and neck.
Clin Cancer Res 1998;4(7):1773–8.
[36] Sturgis EM, Castillo EJ, Li L, et al. Polymorphisms of DNA repair gene XRCC1 in
squamous cell carcinoma of the head and neck. Carcinogenesis 1999;20(11):2125–9.
[37] Gollin SM. Chromosomal alterations in squamous cell carcinomas of the head and neck:
window to the biology of disease. Head Neck 2001;23(3):238–53.
[38] Steinhart H, Bohlender J, Iro H, et al. DNA amplification on chromosome 7q in squamous
cell carcinoma of the tongue. Int J Oncol 2001;19(4):851–5.
[39] Singh B, Gogineni SK, Sacks PG, et al. Molecular cytogenetic characterization of head
and neck squamous cell carcinoma and refinement of 3q amplification. Cancer Res 2001;
61(11):4506–13.
[40] Hashimoto Y, Oga A, Kawauchi S, et al. Amplification of 3q26 approximately qter
correlates with tumor progression in head and neck squamous cell carcinomas. Cancer
Genet Cytogenet 2001;129(1):52–6.
[41] Piccinin S, Gasparotto D, Vukosavljevic T, et al. Microsatellite instability in squamous cell
carcinomas of the head and neck related to field cancerization phenomena. Br J Cancer
1998;78(9):1147–51.
[42] El-Naggar AK, Hurr K, Huff V, Clayman GL, Luna MA, Batsakis JG. Microsatellite
instability in preinvasive and invasive head and neck squamous carcinoma. Am J Pathol
1996;148(6):2067–72.
[43] Field JK, Kiaris H, Howard P, Vaughan ED, Spandidos DA, Jones AS. Microsatellite
instability in squamous cell carcinoma of the head and neck. Br J Cancer 1995;71(5):
1065–9.
[44] Field JK, Kiaris H, Risk JM, et al. Allelotype of squamous cell carcinoma of the head and
neck: fractional allele loss correlates with survival. Br J Cancer 1995;72(5):1180–8.
[45] Danahey DG, Tobin EJ, Schuller DE, Bier-Laning CM, Weghorst CM, Lang JC. p16
mutation frequency and clinical correlation in head and neck cancer. Acta Otolaryngol
1999;119(2):285–8.
[46] Yu KK, Zanation AM, Moss JR, Yarbrough WG. Familial head and neck cancer:
molecular analysis of a new clinical entity. Laryngoscope 2002;112(9):1587–93.
[47] Varley JM, Evans DG, Birch JM. Li-Fraumeni syndrome—a molecular and clinical review.
Br J Cancer 1997;76(1):1–14.
[48] Bargonetti J, Manfredi JJ. Multiple roles of the tumor suppressor p53. Curr Opin Oncol
2002;14(1):86–91.
[49] Steels E, Paesmans M, Berghmans T, et al. Role of p53 as a prognostic factor for survival in
lung cancer: a systematic review of the literature with a meta-analysis. Eur Respir J 2001;
18(4):705–19.

[50] Campo E, de la Calle-Martin O, Miquel R, et al. Loss of heterozygosity of p53 gene and
p53 protein expression in human colorectal carcinomas. Cancer Res 1991;51(16):4436–42.
[51] Marks JR, Humphrey PA, Wu K, Berry D, et al. Overexpression of p53 and HER-2/neu
proteins as prognostic markers in early stage breast cancer. Ann Surg 1994;219(4):332–41.
[52] Brennan JA, Boyle JO, Koch WM, et al. Association between cigarette smoking and
mutation of the p53 gene in squamous-cell carcinoma of the head and neck. N Engl J Med
1995;332(11):712–7.
[53] Somers KD, Merrick MA, Lopez ME, Incognito LS, Schechter GL, Casey G. Frequent
p53 mutations in head and neck cancer. Cancer Res 1992;52(21):5997–6000.
[54] Boyle JO, Hakim J, Koch W, et al. The incidence of p53 mutations increases with
progression of head and neck cancer. Cancer Res 1993;53(19):4477–80.
[55] Shin DM, Charuruks N, Lippman SM, et al. p53 protein accumulation and genomic
instability in head and neck multistep tumorigenesis. Cancer Epidemiol Biomarkers Prev
2001;10(6):603–9.
[56] Koch WM, Lango M, Sewell D, Zahurak M, Sidransky D. Head and neck cancer in
nonsmokers: a distinct clinical and molecular entity. Laryngoscope 1999;109(10):1544–51.
[57] Field JK, Zoumpourlis V, Spandidos DA, Jones AS. p53 expression and mutations in
squamous cell carcinoma of the head and neck: expression correlates with the patients’
use of tobacco and alcohol. Cancer Detect Prev 1994;18(3):197–208.
[58] Takes RP, Baatenburg de Jong RJ, Schuuring E, et al. Differences in expression of
oncogenes and tumor suppressor genes in different sites of head and neck squamous cell.
Anticancer Res 1998;18(6B):4793–800.
[59] Rodrigo JP, Suarez C, Gonzalez MV, et al. Variability of genetic alterations in different
sites of head and neck cancer. Laryngoscope 2001;111(7):1297–301.
[60] Jacob R, Welkoborsky HJ, Hans J, et al. Heterogeneity of squamous cell carcinomas of the
head and neck—analysis of tumor biologic factors and proliferation rates. Laryngoscope
1996;106(9):1170–5.
[61] Mao L, Hruban RH, Boyle JO, Tockman M, Sidransky D. Detection of oncogene
mutations in sputum precedes diagnosis of lung cancer. Cancer Res 1994;54(7):1634–7.
[62] Boyle JO, Mao L, Brennan JA, et al. Gene mutations in saliva as molecular markers for
head and neck squamous cell carcinomas. Am J Surg 1994;168(5):429–32.
[63] Brennan JA, Mao L, Hruban RH, et al. Molecular assessment of histopathological staging
in squamous-cell carcinoma of the head and neck. N Engl J Med 1995;332(7):429–35.
[64] Patel V, Leethanakul C, Gutkind JS. New approaches to the understanding of the molec-
ular basis of oral cancer. Crit Rev Oral Biol Med 2001;12(1):55–63.
[65] Leethanakul C, Patel V, Gillespie J, et al. Gene expression profiles in squamous cell
carcinomas of the oral cavity: use of laser capture microdissection for the construction
and analysis of stage-specific cDNA libraries. Oral Oncol 2000;36(5):474–83.

Imaging in head and neck oncology
Ronald A. Alberico, MDa,b,*
Syed Hamed S. Husain, DOb
, Igor Sirotkin, MDb
a
Department of Radiology, Roswell Park Cancer Institute, Elm and Carlton Streets,
Buffalo, NY 14263, USA
b
State University of Buffalo School of Medicine and Biomedical Sciences, Buffalo VA
Medical Center 3495 Bailey Avenue, Buffalo, NY 14215, USA
Since the initial description of the pathologic distribution and patterns of
spread of tumors in the head and neck, the pretreatment assessment of the
size, extent, and pattern of spread has been necessary for optimal treatment
planning. It has subsequently become apparent that decisions, including the
operative approach, possibility of organ preservation or functional
preservation of tissue, and the appropriateness of an operative role in
patient care, hinge on these important pieces of information. Clinical
examination alone, however, is limited in its ability to properly assess the
extent and size of head and neck tumors, especially for submucosal
extension of disease and extent of nodal metastasis.
Modern radiologic imaging has provided the means to maximize
information available to clinicians during the treatment-planning pro-
cess. The combination of CT, MRI, ultrasound, and positron emission
tomography (PET) has enabled clinicians to obtain a great deal of
information about the patient before planning the surgical approach. Both
CT and MRI have been shown to be superior to clinical examination in
evaluating the size and extent of head and neck tumors and in detailing the
extent of nodal metastases [1–4]. Imaging also has added to the cost and, in
some cases, the controversy of the preoperative assessment and post-
treatment follow-up period. The choice of which imaging modality is
preferred remains controversial and remains part of an ongoing discussion
[5]. This article describes a strategy for imaging head and neck neoplasia in
an effort to simplify the process and emphasizes the strengths and
weaknesses of the available imaging modalities. In addition, the article
emphasizes techniques for imaging and reporting on patients who have head
* Corresponding author.
E-mail address: ronald.alberico@roswellpark.org (R.A. Alberico).
doi:10.1016/S1055-3207(03)00124-8

and neck cancer in a manner that maximizes the clinician’s ability to make
appropriate treatment plans and avoid unnecessary complications.
The primary goal in imaging of head and neck oncology is to answer the
pertinent clinical questions. Too often, the radiologist can get caught up in
the collateral findings and provide information that is confusing or
superfluous while omitting key points needed for the treatment plan.
Although the clinician may feel comfortable filling in the blanks, the scan
may not be optimally designed to answer the clinical question, particularly
in postoperative patients. Frequently, the radiology requests provide
insufficient clinical information to adequately plan the scan, possibly
resulting in exclusion of anatomy crucial to the diagnosis. The solution to
these problems is knowledge and communication. The radiologist must be
familiar with the surgical procedures available and the anatomic criteria that
exclude various procedures from consideration. In addition, the radiologist
must be made aware of the clinical findings and concerns to select the
appropriate imaging modality and optimize the imaging technique. The
radiologist’s goal should be not only to answer the questions of size and
extent of tumor but to point out potential surgical complications resulting
from vascular relationships to the tumor, and individual anatomic variants
that may complicate the procedure. The formation of a differential diagnosis
based on lesion location and imaging characteristics plays an important, but
secondary, role in this process. Even with the best modern imaging
available, the radiologist is still relegated to the role of gross pathologist,
with some limited physiologic data, and, as always, the final answer is in the
histology.
Imaging techniques
The radiology and head/neck surgery literature over the last decade has
supported either CT or MRI as the primary technique for evaluating
patients who have head and neck cancer. This situation has divided the
radiology community into two groups, each of which feels passionately
about their respective choices. CT has been shown to be superior to MRI in
evaluating necrosis in nodal metastases [6], whereas MRI is better for
detecting perineural extent of disease and disease at the skull base [5,7,8].
Other authors have shown improved lymph node detection with MRI [3].
Both modalities have advantages and disadvantages in the evaluation of
head and neck cancer. CT has the advantage of increased speed and
availability and better patient tolerance. The bony framework is better
evaluated with CT and small calcifications are more apparent. CT has the
disadvantage of requiring ionizing radiation and iodinated contrast agents.
MRI is more sensitive for subtle spread of disease along nerves and into the
skull base. In addition, MRI has higher soft tissue contrast resolution and
direct multiplanar imaging capability. Disadvantages of MRI include lower
patient tolerance and dangers associated with metallic implants, pacemakers
14 R.A. Alberico et al / Surg Oncol Clin N Am 13 (2004) 13–35

and other hardware, and increased expense. MRI is also subject to many
artifacts that can make interpretation more difficult. Patient motion is
always a concern in MRI, particularly in patients who have difficulty
suspending swallowing and lying flat.
PET scanning and ultrasound take a definite back seat to CT and MRI in
evaluating the head and neck. Ultrasound is useful for image-guided biopsy
and can provide the fastest, easiest means to guide the needle to the
appropriate target. Doppler sonography has shown some ability to improve
the specificity and sensitivity of nodal staging in clinically N0 neck disease,
as has PET imaging, but the clinical criteria for exploring N0 neck disease
frequently obviates the need to use PET or ultrasound for this purpose [9].
The current authors have found that, for most patients, CT, when
properly performed, provides a readily available and easily tolerated
assessment of head and neck neoplasia. It is easier to interpret for nodal
staging and successfully completed more often than MRI. Multidetector CT
obtained with thin images (2.5 mm) and contrast is able to detect
perineural disease and is readily reformatted into multiple imaging planes.
The current authors typically assess patients initially with CT and will
obtain MRI only if perineural spread of disease is suspected or ambiguous
on CT, or to better evaluate cartilage or marrow invasion. MRI is also
useful in patients who have tumors that are typically lower in attenuation on
CT, such as liposarcomas, and may provide additional information in
patients with this type of tumor. The current authors also use MRI for
thyroid tumors that may potentially be treated with radio-iodine therapy to
avoid the iodine load inherent in CT contrast media. Gadolinium contrast
agents, which are usually used in MRI, can be used as an alternative for
specific patients in CT who are allergic to iodinated contrast and who have
contraindications to MRI evaluation. When necessary, the current authors
use gadolinium as an alternative contrast agent in CT.
All scans are not equal, and to answer the pertinent clinical questions,
properly performed scans are needed. The current authors begin all CT
imaging for head and neck cancer above the orbit to include the skull base
foramina and pterygopalatine foramen. The authors previously used single
detector helical scanners with 5-mm thick sections at 5 mm intervals with 3
mm sections through the larynx. Currently, with multidetector scanners, it is
possible to scan with 2.5 mm section thickness and 2.5 mm section interval
through the entire neck without significant time constraints. Multiplanar
and three-dimensional models can be readily obtained from these data,
including CT angiography as needed to assess vessel–tumor relationships.
Artifacts on CT at the oral cavity can limit the evaluation of the intrinsic
tongue and hard palate; sections angled through the oral pharynx using
a coronal oblique orientation can result in improved visualization of these
areas with a minimum of effort (Fig. 1). It is important to find an imaging
center that uses techniques such as these, with an effort to guide the
treatment plan of the individual patient in the proper direction.
15R.A. Alberico et al / Surg Oncol Clin N Am 13 (2004) 13–35

Imaging the primary tumor
The most important role of imaging in head and neck cancer is to
evaluate the primary tumor and its extent. Although T1 lesions are much
more conspicuous on clinical examination than in images (Fig. 2),
submucosal disease and the extent of tumor across tissue planes and along
Fig. 1. (A) Scout image from a CT scan of the neck and skull base for SCC shows dental
hardware and the usual scan section orientation. (B) The axial section from the angle scanned in
A at the level of the oral pharynx has extensive artifact from dental hardware, which obscures
the pharyngeal and parapharyngeal structures (arrow). (C) Scout image with coronal oblique
sections planned to avoid dental artifact through the oral pharynx. (D) The oral pharyngeal
walls and tonsils are now visible with associated left-sided mass (arrow).

nerves is best seen with imaging [10,11]. The findings may affect the choice of
radiation field needed to cover a lesion and can affect the surgical options
offered to the patient. The extent of tumors can frequently be observed
through the submucosal spaces on the images, resulting in higher tumor
staging than is suspected on clinical grounds, whereas in other cases the
clinical staging is confirmed. The images are key in defining the final extent
of the tumor (Figs. 3 and 4). Invasion of tumor into adjacent structures,
such as the mandible, or along perineural pathways may be clinically
inconspicuous. The sensitivity and specificity of imaging in detecting these
patterns of disease is well described in the literature; MRI is the preferred
method for detecting perineural disease and mandibular invasion [10–19].
Most reports to date have not accounted for recent advances in CT
technology, including multidetector scanning. The current authors have
found that perineural spread, although more obvious on a high-quality,
motion-free MRI, is detectable on CT, in most cases, by loss of the normal
fat signal at the foramen [10,18–19]. Perineural spread of tumor is usually
the result of squamous cell carcinoma (SCC), although this finding is likely
caused by the prevalence of this tumor in the population. Perineural spread
is also commonly seen in adenoid cystic carcinoma, followed by
mucoepidermoid carcinoma [10,18]. Because perineural spread is present
in a higher percentage of cases in these relatively rare tumors, MRI may
provide a more sensitive assessment of the extent of disease for salivary
Fig. 2. Axial section from a contrast-enhanced CT scan with a subtle high-attenuation lesion
(arrow) that represents a T1 SCC. This finding was much more apparent on clinical
examination.

malignancy. MRI offers advantages in detecting marrow invasion in the
mandible and the cartilage of the larynx (Figs. 5–9).
SCC, which originates in the mucosa, comprises most head and neck
cancer. Imaging can play a roll in the preoperative diagnosis of different
histologic subtypes by placing tumors in different spaces in the neck. The
suprahyoid neck is typically divided into muscosal, parapharyngeal, parotid,
and masticator spaces, with the parapharyngeal space further divided into
pre- and poststyloid components. The mucosal space is composed of the
mucosal surfaces of the nasal and oral pharynx. Lesions in this space are
most likely SCC with minor salivary tumors, including benign, mixed
tumors; mucoepidermoid carcinoma; and possibly adenoid cystic carci-
noma. The mucosal spaces of the oral pharynx, specifically the soft palate,
may provide perineural pathways of tumor spread along the greater palatine
or lesser palatine nerves (Fig. 10). The parotid space includes superficial and
deep lobes of the parotid and involves the space between the styloid process
and posterior mandibular ramus (stylomandibular tunnel). Tumors of the
parotid include the primary salivary tumors listed previously and metastatic
disease and lymphoma involving intraparotid nodes. The parotid space
provides a pathway of perineural spread along cranial nerve VII to the
stylomastoid foramen (see Fig. 9). The poststyloid parapharyngeal space
(carotid space) is defined by the styloid process and fascia anteriorly,
paraspinal musculature posteriorly and medially, and the sternocleidomas-
toid (SCM) muscle laterally. Tumors of this space include schwannomas,
glomus tumors, metastatic adenopathy or lymphoma, and lipomas or
liposarcomas. Perineural spread along the vagus nerve or direct spread
along the carotid artery or jugular vein can lead into the skull base. A mass
within this space can also result in vocal cord paralysis by means of its effect
Fig. 3. (A) This axial contrast-enhanced CT section reveals a typical-appearing high-
attenuation mass of the right floor of mouth and tongue (arrows). (B) The coronal reformatted
image from the scan in A demonstrates the superior inferior extent of the mass (arrows) and
confirms the lack of extension across the midline.

on the vagus nerve. The prestyloid parapharyngeal space borders the
masticator space anteriorly, the mucosal space medially, and the styloid
process posteriorly. It contains fat and lymphatics and is rarely directly
involved as a primary tumor site. Displacement of this fat by large masses
can give insight as to which space a large mass is originating from, thus
affecting the differential diagnosis. This space communicates with the
pterygopalatine fossa (see Fig. 8) and has access to all perineural routes
associated with the fossa, including spread along the vidian and rotundum
canals and into the inferior orbital fissure. The masticator space is defined
by the muscles of mastication and is affected primarily by sarcomas and
nerve sheath tumors, including rhabdomyosarcoma, liposarcoma, and
schwannomas. Metastatic disease and lymphoma can affect this space as
Fig. 4. (A) Axial contrast-enhanced CT of the neck reveals a high-attenuation mass in the right
pharyngeal tonsil (arrow). (B) A section lower in patient shown in A demonstrates involvement
of the tongue as the tumor spreads anteriorly along the palatoglossus muscle (arrows). (C) A
section higher than that shown in A reveals some early spread to the soft palate as well (between
arrows).

well. Perineural spread from the masticator space usually involves cranial
nerve V and specifically its third division (see Figs. 6 and 7).
Imaging of the infrahyoid neck is less complex overall but requires
knowledge of laryngeal anatomy and operative approaches. Most infra-
hyoid head and neck tumors are SCC or metastatic disease to the lymph
nodes. Three-dimensional and multiplanar modeling of the CT data can
provide the surgeon with a better appreciation of the anatomy pre-
operatively, providing a more surgically oriented perspective of the
pathology and, in some cases, allowing for production of synthetic
prostheses to be prepared preoperatively to fit the patient’s anticipated
surgical defect [20–22]. The normal distribution of adipose tissue in the
larynx allows clinicians to differentiate the false from the true vocal cords on
CT and to see the paraglottic space (see Fig. 10; Fig. 11). The various
surgical approaches to laryngeal cancer include supraglottic and supra-
cricoid laryngectomy and vertical hemilaryngectomy and total laryngecto-
my. Diagrams of these procedures can be modeled from modern CT images
(see Fig. 11; Figs. 12 and 13). For patients with laryngeal tumors, the images
can define extension of a primary neoplasm in the paraglottic space across
the laryngeal ventricle or across the midline that would render supraglottic
or vertical hemilaryngectomy unlikely to provide tumor-free margins. This
finding would affect the potential for operative cure in these patients. With
this information, the head and neck surgeon can have a more informed
discussion with the patient regarding potential operative options and
prognosis. CT provides the best, most rapid, and consistently motion-free
images in this population. MRI is more sensitive for invasion of the thyroid
Fig. 5. (A) Sagittal T2-weighted MRI image reveals a large intracranial component to this
esthesioneuroblastoma (arrows). Note the cystic and solid components of the mass, which is
a characteristic of these tumors. (B) The coronal T1 fat-saturated gadolinium-enhanced image
reveals the heterogeneous enhancement of this lesion and its sharp demarcation from the brain,
which is not yet invaded (arrows).

cartilage and may be useful in certain patients for evaluation of direct
cartilage invasion.
The hypopharynx is readily seen on MRI and CT, and patterns of tumor
spread to the pyriform sinus and retropharyngeal tissues typically can be
deﬁned by high attenuation on contrast-enhanced CT and enhancement on
MRI. Direct invasion of the cervical spine or perivertebral space is best
evaluated with MRI because it is more sensitive for bony invasion than CT.
There is currently no well-deﬁned role for PET or ultrasound in evaluating
the primary tumor site in the infrahyoid neck.
Less common tumors in the infrahyoid neck also include tumors of the
perivertebral space, which includes the cervical spine and cord, and the
perivertebral muscles. These must include all primary and metastatic bone
Fig. 6. (A) Axial contrast-enhanced fat-saturated T1 image of the suprahyoid neck reveals
a nodule of enhancing tissue (arrow) in the mandibular foramen of this patient who has
a retromolar trigone SCC. (B) Contrast-enhanced axial CT section in the same patient shows
the lack of fat signal typical of perineural spread in the same mandibular foramen imaged 3
weeks earlier (curved arrow). Note the normal fat signal in the foramen of the contralateral side
(straight arrow). (C, D) The invasion of the mandibular marrow space is clearly seen in these
coronal and axial contrast-enhanced T1-weighted MR images from the same patient (arrows).

tumors, and neurofibromas, schwannomas, and other central nervous
system tumors. Tumors of the thyroid gland include adenomas and thyroid
malignancies of all subtypes. A detailed discussion of thyroid neoplasm is
beyond the scope of this article; however, imaging of the thyroid is best
obtained with a combination of nuclear medicine thyroid scanning,
ultrasound, and MRI. CT is excellent as well; however, the iodine load
from contrast material can decrease uptake for potential nuclear medicine
scanning and result in delayed therapy, so care must be taken to avoid
unnecessary iodine loads before diagnosis. The poststyloid parapharyngeal
space continues into the infrahyoid neck as the carotid space and is a site for
metastatic disease and lymphoma. Glomus tumors and schwannomas also
can be found in this space.
Nodal staging
Staging of nodal disease in the neck traditionally has been based on
clinical examination; however, limitations in the clinical examination result
in relatively low sensitivity and specificity (60%–70%), leading to an
unacceptably low negative predictive value [3,4,23]. Improved negative
predictive value is important in defining a population that would benefit
from surgery without the need for neck dissection and radiation. Imaging,
including CT and MRI, uses a threshold size to determine if a node is
abnormal. Depending on the reference, this size varies between 1 and 1.5
cm. Morphology of the node is also considered in determining the likelihood
of metastasis, including the transverse-to-longitudinal ratio and the
Fig. 7. (A) An axial contrast-enhanced T1-weighted image from an MRI of the neck reveals an
enhancing mass at the top of the right masticator space just below foramen ovale (between
arrows). (B) A coronal contrast-enhanced T1-weighted image reveals the perineural spread of
the tumor into foramen ovale along cranial nerve V3 to involve Meckel’s cave (long arrow).
Compare this to the normal contralateral side (short arrow).

Fig. 8. An axial section from a contrast-enhanced CT of the neck reveals loss of the normal fat
attenuation within the left pterygopalatine fossa (arrow). This ﬁnding was conﬁrmed to be
perineural spread from the patient’s left tonsillar fossa SCC.
Fig. 9. An axial section from a contrast-enhanced CT of the neck in a patient with
mucoepidermoid carcinoma of the left parotid gland reveals loss of the normal fat attenuation
in the left stylomastoid foramen (thick arrow). Note the normal low attenuation of the
contralateral side (thin arrow).

attenuation of the node [24,25]. Even with a combined approach, the
literature varies widely on the specificity and sensitivity of nodal staging
with MRI and CT, with sensitivity varying from 40% to 80% and specificity
from 90% to 95%. In a large-scale study by Curtin et al [3], attempts were
made to obtain a negative predictive value of 90% with CT and MRI using
size criteria alone or size criteria in combination with internal morphology.
Although CT could achieve this 90% negative predictive value, it required
a size threshold of 5 mm, which decreased positive predictive value to 44%.
MRI did not achieve a 90% negative predictive criterion in that study,
regardless of size threshold used. Attempts at increasing sensitivity with
PET scanning or Doppler sonography to detect malignancy in normal-size
Fig. 10. (A) Axial CT section of the larynx a t the level of the false cords (arrows). Note the low
attenuation of the paraglottic fat. (B) Axial CT section at the level of the paraglottic space
shows the fat within the space to better advantage (arrows). (C) Coronal reformation from the
same scan shows the fatty attenuation in the paraglottic space (long arrow) and false cords (short
arrow) compared with the muscle attenuation of the true vocal cords (curved arrow). (D) Off-
midline sagittal reformat from the same patient clearly shows the air within the laryngeal
ventricle (long arrow). The false cords are above the ventricle with the muscular true cords
below.

nodes have had some success, but the sensitivity is only marginally improved
over that of MRI or CT [25,26]. Combined MRI and CT have approached
90% sensitivity for metastatic node detection in one study [1]. It would seem
that, regardless of imaging technique, negative predictive value has not
achieved a level that would be clinically useful in excluding clinically N0
neck disease without unacceptably low positive predictive values.
Until negative and positive predictive values of nodal disease are
improved, exclusion of patients from treatment of the neck with radiation
or neck dissection based on imaging is not appropriate; however, there are
other uses for nodal assessment with imaging that can affect patient care and
prognosis. The location of nonpalpable adenopathy in the neck in patients
who have disease of any nodal stage can affect the size and extent of
radiation fields and the side and extent of neck dissection. Given the
increased sensitivity of image-based nodal staging compared with clinical
staging, this is sufficient to warrant nodal evaluation with imaging.
The description of nodal locations in the neck requires precise language
to facilitate communication between the head and neck radiologist, the
surgeon, and the pathologist. Without such a system, patterns of nodal
disease and their relationships to tumor prognosis and location would lack
precision and result in inaccuracies in clinicians’ knowledge of disease
prognosis and patterns of spread, limiting their ability to treat patients.
Various classification systems for nodal disease in the neck therefore have
been used in the past, including those of the American Academy of
Fig. 11. (A) An axial contrast-enhanced CT scan at the level of the true vocal cords (arrow).
Note the high attenuation (muscle) of the true vocal cords compared with the false cords seen in
Fig. 10. The line delineates the surgical resection for a vertical hemilaryngectomy. Because the
cricoid cartilage is preserved by the surgery, extension of tumor into the cricoid or arytenoid
would contraindicate this type of voice-sparing procedure. (B) A color three-dimensional
diagram of the larynx, again showing the surgical plan for a supracricoid laryngectomy. The
thyroid cartilage is blue, the cricoid cartilage is light blue, the epiglottis is red, and the hyoid
bone is white.

Otolaryngology–Head and Neck Surgery and the American Joint Commit-
tee on Cancer. These classification systems did not always precisely define
nodal locations, however, and did not account for retropharyngeal nodes
described in the original anatomic system proposed by Rouviere [27]. The
previous nodal classification systems also have been based on anatomic
landmarks that are not necessarily conspicuous in the axial plane. Because
modern nodal assessment almost always includes imaging, a modern
classification system should refer to anatomic landmarks that are reliably
identified in the axial plane and at the time of surgery.
In 1999, Som et al [27] undertook this considerable task. They defined
level I as submental (IA) and submandibular (IB), with both levels anterior
to the posterior margin of the submandibular gland, above the hyoid bone,
Fig. 12. (A) Sagittal CT reformatted image of the neck reveals the resection plan for
a supraglottic laryngectomy. Note the line goes through the laryngeal ventricle and spares the
vocal cords. (B) Coronal section reformatted from the same scan shows the surgical plan
through the laryngeal ventricle between the true and false vocal cords. (C) Color volume–
rendered model of the larynx again reveals the surgical margin. The thyroid cartilage is blue, the
cricoid cartilage light blue, the hyoid bone is white, and the epiglottis is red.

and below the mylohyoid muscle. Level IA is between the anterior bellies of
the digastric muscles, with level IB lateral to the digastric muscle anterior
belly. Level II extends from the skull base to the bottom of the hyoid bone,
posterior to the back of the submandibular gland, and anterior to the back
of the SCM muscle (Fig. 14). Level IIA consists of nodes in the level II
region that are inseparable from the jugular vein by a fat plane, with level
IIB nodes posterior to the vein and separable from it by a fat plane. Level III
nodes are anterior to the back of the SCM muscle and between the bottom
of the hyoid bone and the bottom of the cricoid arch (Fig. 15). Level IV
nodes are located below the bottom of the cricoid arch but above the
clavicles. They are anterior to the line joining the back of the SCM muscle
with the posterolateral margin of the anterior scalene muscle, and lateral to
the common carotid arteries. Level V nodes are posterior to the back of the
SCM muscle from the skull base to the bottom of the cricoid arch (level VA)
and continue posteriorly to the line connecting the back of the SCM muscle
and the posterolateral margin of the anterior scalene muscle to the level of
the clavicle (level VB). Level VI nodes are between the common carotid
arteries from the bottom of the hyoid bone to the top edge of the
manubrium, with level VII nodes located between the carotid arteries below
the top edge of the manubrium to the level of the brachiocephalic vein. The
level of the clavicles is defined as the first axial section in which the clavicles
are visible. The supraclavicular nodes are at the level of the clavicles lateral
to the common carotid arteries. Retropharyngeal nodes are defined as
medial to the internal carotid arteries, within 2 cm of the skull base.
This system provides the precise framework needed to facilitate
communication among and between surgeons, pathologists, and radiolog-
ists. Although all nodal levels may not be commonly used by surgeons in all
Fig. 13. (A) An axial CT reformatted image at the level of the true vocal cords demonstrates the
surgical plan for a supracricoid laryngectomy. Note the cricoid cartilage is spared and only one
arytenoid cartilage is resected. (B) A sagittal section of the same scan shows the two possible
plans for the supracricoid laryngectomy, with and without preservation of the epiglottis.

locations, an effort to adhere to this classification system should improve the
quality of surgical–pathologic correlation and result in research in head and
neck cancer, which has historically been difficult to study because of low
numbers of patients and inconsistencies in language used in the literature
and pathology and radiology reports.
Other applications of head and neck imaging in malignant disease
Other applications of head and neck imaging include CT or ultrasound-
guided biopsy of suspected recurrent or primary disease, evaluation of the
neck post treatment, and assessment of anatomic variants that may impact
the surgical approach. CT-guided percutaneous biopsy has been widely
studied in the literature for virtually all potential targets, including the
brain. The application of this technology to head and neck cancer can result
in safe and efficacious tissue sampling of retropharyngeal, parapharyngeal,
and other deep or difficult-to-palpate regions of the neck [28,29]. With
proper techniques, nodal biopsy with ultrasound or CT guidance can be
performed with minimal risk to the patient [23]. The current authors’
preferred CT technique is to access the face through the buccal space using
a short guide needle to the posterior edge of the pterygoid muscle, allowing
Fig. 14. An axial contrast-enhanced CT section reveals metastatic adenopathy at level IIA on
the right and the left (arrows). Nodes are above the bottom of the hyoid bone, posterior to the
back of the submandibular gland, and anterior to the back of the SCM muscle, inseparable
from the jugular vein. Note that the nodes are enlarged, abnormal in shape (rounded), and
abnormally low in attenuation.

for multiple passes with a 22-guage needle into the substance of the mass,
without additional percutaneous passes and with good maintenance of an
entry point close to the tumor margin (Fig. 16). From this approach, the
guide needle can be angled slightly at the skin surface to obtain samples at
different locations within the tumor mass, with minimal additional risk. The
current authors have performed biopsies in 38 patients using this technique
over the last 4 years without complication. Sufficient tissue for diagnosis was
obtained in 98% of patients who underwent biopsy.
The use of imaging in postoperative patients is perhaps the most difficult
part of head and neck imaging interpretation for the radiologist. MRI in the
postoperative setting frequently proves difficult for patients, because motion
and suspension of swallowing can be difficult to control. MRI has been
shown to have a high false-positive rate after radiation, which increases over
time to as high as 58%; CT has a specificity and sensitivity of 80% to 90%
[30,31]. Knowledge of the surgical procedure performed and the type or
location of any operative flap reconstruction, and history of radiation
treatment, will decrease the false-positive rate in MRI. This again
emphasizes the need for communication between surgeons and radiologists
to obtain accurate evaluation of the patient and the optimal imaging
technique [32]. Recognition of the post-treatment appearance of head and
neck cancer on CT is an acquired skill that requires practice and readily
available follow-up information for the radiologist to become proficient.
Fig. 15. An axial contrast-enhanced CT section reveals an enlarged, rounded hypoattenuating
node at level III on the right (arrow). The node is between the bottom of the hyoid bone and the
bottom of the cricoid cartilage, lateral to the carotid artery, and anterior to the back of the
SCM muscle.

The operative flaps typically contain fat and muscle, with identifiable
vascular pedicles. The postradiation density of tumor is intermediate
between muscle and fat but may maintain the shape of the original neoplasm
(Fig. 17). The initial evaluation of PET scanning in recurrent disease and for
tumor response to chemotherapy has been promising [33–35], with some
articles stating improved sensitivity of PET over MRI and CT for
evaluation of recurrent disease. Overall patient numbers have been low in
these studies, however. Other studies have implied PET has a role during
initiation of chemotherapy to evaluate initial tumor response using glucose
metabolism as an indicator of tumor response. A lack of metabolic change
with initiation of therapy implies therapy may not be effective [31]. This
finding could potentially provide an early indicator that a new therapeutic
regimen should be considered.
The role of any radiologist is to provide the clinician with important
anatomic details about the patient that may affect the difficulty or feasibility
of the planned therapeutic approach. This communication is particularly
important with head and neck malignancies. The use of three-dimensional
and multiplanar reformatted images in CT or MRI to help define tumor
relationships to vessels and the likelihood of vascular, spinal, perineural, or
tracheal invasion is crucial. Controversy still exists as to whether CT or
MRI performs these perspective tasks with higher sensitivity, but the well-
trained eye, and the clinician who is informed of the planned clinical
procedure and is aware of the surgical approach and risks, is the best tool
for alerting the surgeon to potential pitfalls related to anatomic variants in
a specific patient (Figs. 18 and 19).
Fig. 16. (A) An axial CT section of the neck without contrast reveals a lateral pharyngeal node
on the left (arrow). The section includes part of the biopsy guide needle in the buccal space.
(B) This section shows the biopsy needle piercing the lateral pharyngeal node. Note the initial
scan was obtained with contrast to locate the carotid artery, which was clearly lateral to the
node before biopsy.

Summary
Evaluation of head and neck cancer with imaging is a topic that is far
more extensive than can be covered in this article. The main reason for head
and neck imaging is to evaluate the true extent of disease to best determine
surgical and therapeutic options. This process includes evaluation of the
size, location, and extent of tumor infiltration into surrounding vascular and
visceral structures. Important anatomic variants must be pointed out so the
surgeon can avoid potential intraoperative complications. These variants
can be evaluated with the appropriate multiplanar and three-dimensional
images to provide as much information as possible to the surgeon
Fig. 17. (A) An axial contrast-enhanced CT section through the larynx at the level of the
paraglottic space reveals a high-attenuation mass crossing the laryngeal ventricle through that
space (arrow). (B) The coronal reformatted image from A confirms the paraglottic spread of the
tumor (arrow). This tumor also extended across the midline anteriorly, excluding the patient
from voice-preservation surgery. (C) A follow-up CT scan after radiation therapy in the same
patient reveals the typical low attenuation of treated tumor, which is between muscle and fat
density (arrow). (D) A coronal reformation of C confirms that the paraglottic space also shows
evidence of radiation effect (arrow). Note that the mass effect from the tumor has not yet
subsided and continues to indent the supraglottic airway.

preoperatively. Second, nodal staging should be assessed in an effort to
increase the number of abnormal nodes detected by physical examination
and, more important, to precisely define their location by a standard
classification system that can be understood and consistently applied by the
radiologist, surgeon, radiation oncologist, and pathologist. Although
secondary to the previously described tasks, imaging frequently enables
a limitation of the diagnostic and histologic possibilities based on lesion
location and signal-attenuation characteristics, which may lead the clinical
Fig. 18. (A) An axial contrast-enhanced CT section at the level of the hard palate in a patient
with a superficial palatal carcinoma on clinical examination reveals an asymmetry in
attenuation and size of the greater palatine foramen (solid arrow). Compare this to the normal
side (open arrow) in which the attenuation is normal and the foramen is comparatively small.
(B) A bone window of the same section demonstrates the size asymmetry to better advantage
(arrows). (C) Sagittal reformatted image of the mass demonstrates the lack of fatty attenuation
at the opening of the palatine foramen on the left (arrow). (D) Sagittal reformatted image of the
normal side reveals the expected normal fatty attenuation. This finding is consistent with
perineural spread on the CT and changed the surgical plan from intraoral resection to a split
mandible procedure with partial resection of the maxilla. Perineural tumor in the greater
palatine foramen was found pathologically.

investigation along a diﬀerent path, saving the patient unnecessary risk and
shortening the time to diagnosis and ultimate treatment.
This article has attempted to detail the current state of the controversy
between CT, MRI, and other modalities, and has emphasized the constant
evolution of this controversy because of the evolving imaging technology.
Although CT and MRI are both well suited to evaluation of the deep spaces
and submucosal spaces of the head and neck, each has some limitations.
MRI has the advantages of higher soft tissue contrast resolution, the lack of
iodine-based contrast agents, and high sensitivity for perineural and
intracranial disease. The disadvantages of MRI include lower patient
tolerance, contraindications in pacemakers and certain other implanted
metallic devices, and artifacts related to multiple causes, not the least of
which is motion. CT is fast, well tolerated, and readily available but has
Fig. 19. (A) A volume-rendered CT laryngoscopy view reveals a posterior bulge in the wall of
the upper hypopharynx (arrows). (B) The cutaway view of the same image reveals a densely
enhancing structure in the submucosa of the retropharyngeal space (arrow). (C) An axial
contrast-enhanced CT section reveals the mass to be secondary to a tortuous carotid artery,
which resulted in a retropharyngeal position of the carotid bifurcation on the right (arrows).
This ﬁnding was brought to the attention of the head and neck surgeon.

lower contrast resolution and requires iodinated contrast and ionizing
radiation. The current authors’ practice is heavily centered on CT for initial
evaluation, preoperative planning, biopsy targeting, and postoperative
follow-up. They reserve MRI for tumors that are suspicious for perineural,
cartilaginous, or bony invasion on CT, or for tumors such as adenoid cystic
carcinoma that are highly likely to spread by way of these routes. For
patients who have head and neck cancer, a radiologist who is educated in
the treatment options, patterns of tumor growth, and important surgical
landmarks, and who has a well-established pattern of communication with
the head and neck clinical services, including surgery, radiation oncology,
and pathology, is key in providing accurate and useful image interpretation.
References
[1] Hillsamer P, Schuller D, McGhee R, Chakeres D, Young D. Improving diagnostic
accuracy of cervical metastases with computed tomography and magnetic resonance
imaging. Arch Otolaryngol Head Neck Surg 1990;116:1297–301.
[2] Merritt R, Williams M, James T, Porubsky E. Detection of cervical metastasis, a meta-
analysis comparing computed tomography with physical examination. Arch Otolaryngol
Head Neck Surg 1997;123:149–52.
[3] Curtin H, Ishwaran H, Mancuso A, Dalley R, Caudry D, McNeil B. Comparison of CT
and MR imaging in staging of neck metastases. Radiology 1998;207:123–30.
[4] Hao S, Ng S. Magnetic resonance imaging versus clinical palpation in evaluating metastasis
from head and neck cancer. Otolaryngol Head Neck Surg 2000;123:324–7.
[5] Som P. The present controversy over the imaging method of choice for evaluating the soft
tissues of the neck. AJNR Am J Neuroradiol 1997;18:1869–72.
[6] Yousem D, Som P, Hackney D, Schwaibold F, Hendrix R. Central nodal necrosis and
extracapsular neoplastic spread in cervical lymph nodes: MR imaging versus CT.
Radiology 1992;182:753–9.
[7] Schmalfuss IM, Mancuso AA, Tart RP. Arytenoid cartilage sclerosis: normal variations
and clinical significance. AJNR Am J Neuroradiol 1998;19:719–22.
[8] NemzekW, Hecht S,Gandour-EdwardsR, DonaldP,McKennan K. Perineural spread of head
and neck tumors: how accurate is MR imaging? AJNR Am J Neuroradiol 1998;19:701–6.
[9] Lydiatt WM, Shah JP, Hoffman HT, Head and Neck Sites Task Force. American Joint
Committee on Cancer: AJCC stage groupings for head and neck cancer: should we look at
alternatives? A report of the Head and Neck Sites Task Force. Head Neck 2001;23(8):
607–12.
[10] Parker G, Harnsberger H. Clinical-radiologic issues in perineural tumor spread of
malignant diseases of the extracranial head and neck. Radiographics 1991;11:383–99.
[11] Woodruff W Jr, Yeates A, McLendon R. Perineural tumor extension to the cavernous sinus
from superficial facial carcinoma: CT manifestations. Radiology 1986;161:395–9.
[12] Tsue T, McCulloch T, Girod D, Couper D, Weymuller E Jr, Glenn M. Predictors of
carcinomatous invasion of the mandible. Head Neck 1994;16:116–26.
[13] Ator G, Abemayor E, Lufkin R, Hanafee W, Ward P. Evaluation of mandibular tumor inva-
sion with magnetic resonance imaging. Arch Otolaryngol Head Neck Surg 1990;116:454–9.
[14] Acton C, Layt C, Gwynne R, Cooke R, Seaton D. Investigative modalities of mandibular
invasion by squamous cell carcinoma. Laryngoscope 2000;110:2050–5.
[15] Kalavrezos ND, Gratz KW, Sailer HF, Stahel WA. Correlation of imaging and clinical
features in the assessment of mandibular invasion of oral carcinomas. Int J Oral Maxillofac
Surg 1996;25:439–45.

[16] Smyth DA, O’Dwyer TP, Keane CO, Stack J. Predicting mandibular invasion in mouth
cancer. Clin Otolaryngol 1996;21:265–8.
[17] Brown JS, Griffith JF, Phelps PD, Browne RM. A comparison of different imaging
modalities and direct inspection after periosteal stripping in predicting the invasion of the
mandible by oral squamous cell carcinoma. Br J Oral Maxillofac Surg 1994;32:347–59.
[18] Laine F, Braun I, Jensen M, Nadel L, Som P. Perineural tumor extention through the
foramen ovale: evaluation with MR imaging. Radiology 1990;174:65–71.
[19] Williams L. Advanced concepts in the imaging of perineural spread of tumor to the
trigeminal nerve. Top Magn Reson Imaging 1999;10(6):376–83.
[20] Cavalcanti M, Ruprecht A, Bonomie J, Vannier M. The validation of 3D spiral CT-based
measurements of simulated maxillofacial neoplasms. Oral Surg Oral Med Oral Pathol Oral
Radiol Endod 2000;89:753–8.
[21] Moharir V, Fried M, Vernick D, Janecka I, Zahajsky J, Hsu L, et al. Computer-assisted
three-dimensional reconstruction of head and neck tumors. Laryngoscope 1998;108:
1592–8.
[22] Greess H, Nomayr A, Tomandl B, Blank M, Lell M, Lenz M, et al. 2D and 3D
visualization of head and neck tumours from spiral-CT data. Eur J Radiol 2000;33:170–7.
[23] van den Brekel MWM. Lymph node metastases: CT and MRI. Eur J Radiol 2000;33:
230–8.
[24] Takashima S, Sone S, Takayama F, Wang Q, Kobayashi T, Horii A, et al. Papillary
thyroid carcinoma: MR diagnosis of lymph node metastasis. AJNR Am J Neuroradiol
1998;19:509–13.
[25] Jabour B, Choi Y, Hoh C, Rege S, Soong J, Lufkin R, et al. Extracranial head and neck:
PET imaging with 2-[F-18]fluoro-2-deoxy-D-glucose and MR imaging correlation.
Radiology 1993;186:27–35.
[26] Ariji Y, Kimura Y, Hayashi N, Onitsuka T, Yonetsu K, Hayashi K, et al. Power Doppler
sonography of cervical lymph nodes in patients with head and neck cancer. AJNR Am J
Neuroradiol 1998;19:303–7.
[27] Som P, Curtin H, Mancuso A. An imaging-based classification for the cervical nodes
designed as an adjunct to recent clinically based nodal classifications. Arch Otolaryngol
Head Neck Surg 1999;125:388–96.
[28] Hansen M, Anzai Y, Hamilton DR, Kangarloo F, Abemayor E, Castro D, et al.
Interventional computed tomography and MR imaging in the head and neck. Otolaryngol
Clin North Am 1995;28:651–65.
[29] Tu A, Geyer C, Mancall A, Baker R. The buccal space: a doorway for percutaneous CT-
guided biopsy of the parapharyngeal region. AJNR Am J Neuroradiol 1998;19:728.
[30] Lell M, Baum U, Greess H, Nomayr A, Nkenke E, Koester M, et al. Head and neck
tumors: imaging recurrent tumor and post-therapeutic changes with CT and MRI. Eur J
Radiol 2000;33:239–47.
[31] Nomayr A, Lell M, Sweeney R, Bautz W, Lukas P. MRI appearance of radiation-induced
changes of normal cervical tissues. Eur Radiol 2001;11:1807–17.
[32] Hudgins P, Burson J, Gussack G, Grist W. CT and MR appearance of recurrent malignant
head and neck neoplasms after resection and flap reconstruction. AJNR Am J Neuroradiol
1994;15:1689–94.
[33] Haberkorn U, Srauss L, Dimitrakopoulou A, Seiffert E, Oberdorfer F, Ziegler S, et al.
Fluorodeoxyglucose imaging of advanced head and neck cancer after chemotherapy. J Nucl
Med 1993;34:12–7.
[34] Anzai Y, Carroll W, Quint D, Bradford C, Minoshima S, Wolf G, et al. Recurrence of head
and neck cancer after surgery or irradiation: prospective comparison of 2-deoxy- 2-[F-18]
fluoro-D-glucose PET and MR imaging diagnoses. Radiology 1996;200:135–41.
[35] Greven K, Williams D, Keyes JW Jr, McGuirt W, Watson NE Jr, Randall M, et al.
Positron emission tomography of patients with head and neck carcinoma before and after
high dose irradiation. Cancer 1994;74(4):1355–9.

The expanding role of dental oncology
in head and neck surgery
Maureen Sullivan, DDS
Department of Dentistry and Maxillofacial Prosthetics, Roswell Park Cancer Institute,
Dental oncology is the discipline within dentistry that combines general
dentistry, maxillofacial prosthetics, oral medicine, and oral pathology. To
provide state-of-the-art care for patients who have head and neck cancer,
the dental oncologist must play an integral role in all facets of treatment,
whether surgical or nonsurgical. Historically, the maxillofacial prostho-
dontist was the only dental specialist involved in the treatment and reha-
bilitation of patients with head and neck cancer; however, it became
apparent that early intervention by a dentist familiar with the complica-
tions associated with head and neck malignancies was required. The oral
complications secondary to the management of head and neck malignan-
cies have been well described [1–3]. To maximize the possibility of optimal
function and cosmesis, and limit the possibility of complications that can
lead to signiﬁcant morbidity, a comprehensive dental screening must be
performed during the pretreatment phase. Close communication between
the head and neck surgeon, radiation oncologist, dental oncologist, and
maxillofacial prosthodontist is paramount in achieving function and cure.
There are three treatment modalities involved in eradicating head and neck
cancer: (1) surgery; (2) radiation therapy, with or without chemotherapy;
and (3) combined treatment. Because all of these treatments dramati-
cally aﬀect the oral environment, close scrutiny of the dental condition is
crucial.
With a complete understanding of the role of the dental oncology team
involved in the pretreatment, treatment, and rehabilitation phases of treat-
ing patients with head and neck cancer, the goal of cure with an acceptable
quality of life can be obtained.
E-mail address: maureen.sullivan@roswellpark.org
doi:10.1016/S1055-3207(03)00121-2

Dental oncologic assessment
Because radiation therapy is used frequently in the treatment of head and
neck malignancies, prompt attention to the dental condition at the time of
diagnosis can allow time for adequate healing before the onset of radiation,
if oral surgery is required. Frequently, the patient will require procedures
under general anesthesia performed by the head and neck surgeon to
formulate an appropriate plan for tumor removal. These assessment
procedures also provide the dental oncologist an opportunity to perform
necessary dental procedures in the operating room setting, which is
generally preferred by the patient and, more important, allows for longer
healing time if radiation treatment is required.
The oral cavity undergoes monumental insult as a direct result of
radiation therapy to the head and neck. The immediate effects of radiation
include mucositis, pain, trismus, and hypoguesia. These effects are variable,
depending on the type of radiation used, the dose, and the field of involve-
ment. Furthermore, these effects can be minimized with close attention to
the dental condition. The long-term effects may include xerostomia,
rampant dental caries, trismus, soft tissue necrosis, and, potentially the
most devastating effect, osteoradionecrosis (ORN).
Comprehensive evaluation of the dental patient includes Panorex
(Orthopantomograph OP 100, Instrumentarium Corp, Tuusala, Finland)
radiograph, a full-mouth series of intraoral radiographs, and intraoral
examination, including periodontal probing. All teeth deemed unrestorable,
especially those in the field of radiation should be extracted. This procedure
has received much attention in the literature in an attempt to find a formula
for dental extraction requirements before radiation [4,5]. All teeth with
advanced dental decay, with or without pulpal involvement, and advanced
periodontal disease are generally extracted. Partially impacted third molars
with evidence of pericoronitis and any teeth with periapical pathology should
be extracted. The time required for adequate healing should be between 2 to 3
weeks [6]. If third molars are completely impacted without evidence of
pathology, they are left and simply monitored. Again, if oral surgery is ad-
dressed during the head and neck surgeons’ initial treatment-planning phase,
there should be adequate time for healing if radiation therapy is required.
After all obvious sources of infection have been eliminated, the need for
existing restorative dental work is evaluated and any necessary restorative
dentistry is completed. The specific type of restorative material used in
patients that will become xerostomic has been a consideration [7]. There
have been a few studies evaluating the efficacy of fluoride-releasing materials,
and therefore the type of restoration is not as critical as removing a potential
source of mechanical irritation during treatment.
If the patient has not had a thorough periodontal scaling and prophylaxis
within the 3 previous months to diagnosis, these procedures should be
completed. The dental hygienist must be familiar with the secondary
38 M. Sullivan / Surg Oncol Clin N Am 13 (2004) 37–46

Current diagnosis-and-therapy-for-head-and-neck-mal

Current diagnosis-and-therapy-for-head-and-neck-mal

Recomendados

Recomendados

Mais conteúdo relacionado

Mais procurados

Mais procurados (19)

Semelhante a Current diagnosis-and-therapy-for-head-and-neck-mal

Semelhante a Current diagnosis-and-therapy-for-head-and-neck-mal (20)

Mais de Burhan Khan

Mais de Burhan Khan (16)

Último

Último (20)

Current diagnosis-and-therapy-for-head-and-neck-mal