TNM Staging Cancer

2005;55;242-258 CA Cancer J ClinSnehal G. Patel and Jatin P. Shah

DiversityTNM Staging of Cancers of the Head and Neck: Striving for Uniformity Among

This information is current as of July 14, 2005

http://caonline.amcancersoc.org/cgi/content/full/55/4/242located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://caonline.amcancersoc.org/subscriptions/individuals only): , go to (USCA: A Cancer Journal for CliniciansTo subscribe to the print issue of

1542-4863. American Cancer Society, Inc.) All rights reserved. Print ISSN: 0007-9235. Online ISSN:Cancer Society, 1599 Clifton road, NE, Atlanta, Georgia 30329.(Copyright © 2005 by the continuously since November 1950. CA is owned, published, and trademarked by the AmericanSociety by Lippincott Williams & Wilkins. A bimonthly publication, it has been published

is published six times per year for the American CancerCA: A Cancer Journal for Clinicians

by guest on July 14, 2005 (© by the A

merican C

ancer Society, Inc.)

caonline.amcancersoc.org

Dow

nloaded from

http://caonline.amcancersoc.org/cgi/content/full/55/4/242

http://caonline.amcancersoc.org/subscriptions/

http://caonline.amcancersoc.org

TNM Staging of Cancers of the Headand Neck: Striving for UniformityAmong DiversitySnehal G. Patel, MD, FRCS (Glasg); Jatin P. Shah, MD, FACS

ABSTRACT The sixth edition of the tumor-node-metastasis staging system for head and neck

cancers incorporates some significant shifts in philosophy. As treatment paradigms shift and

data from ongoing clinical and basic research become available, further revisions may be

expected in the future. The purpose of this review is to highlight the complexities involved in

developing a user-friendly staging system and to report the major changes in the new version.

The authors also discuss some areas of current interest that may have the potential to lead to

future modifications. (CA Cancer J Clin 2005;55:242–258.) © American Cancer Society, Inc.,

2005.

INTRODUCTION

The tumor-node-metastasis (TNM) staging system was first reported by Pierre Denoix in the 1940s.1 TheInternational Union Against Cancer (UICC) eventually adapted the system and compiled the first edition of theTNM staging system in 1968 for 23 body sites. It is important to realize that the TNM staging system is simply ananatomic staging system that describes the anatomic extent of the primary tumor as well as the involvement ofregional lymph nodes and distant metastasis.2 Over the years, revisions to the TNM staging system have been basedon an improved understanding of the natural history of tumors at various sites aided by advances in technology thathave allowed clinicians to better assess the extent of tumors. Changes or revisions of any staging system should ideallybe supported by evidence-based information generated from large cohorts of patients, preferably in the setting ofprospective randomized clinical trials. Therefore, Level I evidence is available to support changes in the staging systemfor certain cancers that are common enough to provide large patient cohorts for meaningful data analysis (eg,melanoma, cancers of the colon, breast, and lung).

On the other hand, the head and neck region comprises a variety of anatomic sites. For example, tumors arisingfrom the skin, nasal cavity and paranasal sinuses, oral cavity, nasopharynx, oropharynx, hypopharynx, larynx,esophagus, thyroid gland, salivary glands, soft-tissue tumors, bone sarcomas, and miscellaneous tumors such asneurogenic tumors and paragangliomas are all generically included in the head and neck region. More importantly,these tumors have diverse clinical behavior and outcomes. It is therefore literally impossible to generate a uniformstaging system that would be relevant for all tumors arising in the head and neck region. Another effect of thisdiversity among head and neck tumors is that patient cohorts available for outcomes analysis for most sites are limitedin numbers, and on the whole, no Level I evidence is available to support important clinical questions. Thus, changesin the TNM staging system have to be developed based on expert opinions and published reports in the literaturewhile keeping in sight advances in technology for improved assessment of tumor extent and shifting paradigms intherapeutic strategies. To this end, the UICC TNM Committee has established a structured process for continuousimprovement of the TNM classification.3 A panel of experts continuously reviews manuscripts published in theliterature that may affect future changes in the TNM staging system.

Like most other cancers, prognosis depends largely on the stage of the tumor, but other factors related to lifestylesuch as smoking and alcohol consumption and medical comorbidity also affect outcome. Accurate and reliablestratification of head and neck cancers (HNC) for prediction of outcomes has been challenging, mainly because of

Dr. Patel is Attending Surgeon, Headand Neck Service, Department ofSurgery, Memorial Sloan-KetteringCancer Center, New York, NY.

Dr. Shah is Chief, Head and NeckService, Department of Surgery, Me-morial Sloan-Kettering Cancer Cen-ter, New York, NY.

This article is available online athttp://CAonline.AmCancerSoc.org

TNM Staging of Cancers of the Head and Neck

242 CA A Cancer Journal for Clinicians


merican C



Dow

nloaded from


the numerous anatomic sites and subsites fromwhich tumors can arise and the diversity ofhistologic types of tumors in these locations.The overwhelming majority of mucosal HNCsare squamous cell carcinomas. Therefore, theTNM classification of the UICC and theAmerican Joint Committee on Cancer (AJCC)for most mucosal anatomic sites is designed forsquamous cell carcinoma and minor salivarygland cancers. This variety and heterogeneityplace a considerable onus on the clinician orcancer registrar involved in staging these tu-mors to be aware of the principles involved andthe basis for modification of accepted stagingcategories.

Treatment paradigms for head and neck car-cinoma, especially for tumors involving certainsites such as the larynx and pharynx, have under-gone radical change over the past decade. Pres-ervation of function is a major endpoint ofinterest, and nonsurgical treatment options usingchemotherapy and radiation therapy are nowconsidered as standards of care. Not all patients,however, are suited to organ preservation ther-apy, and a percentage of them that do not re-spond ultimately end up having surgical resectionof the affected organ. Therapeutic decisionmaking and selection of patients for appropriatetherapy need considerably more research. Con-ventional TNM information has so far provedinadequate in predicting response to nonsurgicaltherapy. In current practice, information obtainedfrom clinical examination and radiologic imagingis used to assign a clinical stage (cTNM), which isthen used to stratify patients for selection of ther-apy and to report outcomes of treatment. If thepatient undergoes surgical resection, the patho-logic stage (pTNM) derived from histopathologicexamination of the tumor and/or regional lymphnodes is useful in selecting postoperative adjuvanttherapy and for estimating prognosis. As nonsur-gical therapy gains wider acceptance as the initialdefinitive treatment of certain head and neckcarcinomas, the importance of pathologic infor-mation is likely to diminish because the tumorand regional nodes in these patients will not beavailable for comprehensive pathologic analysis.Better and more reliable methods of pretreatmenttumor assessment are therefore crucial to ensurethat the clinical assessment of tumor approximates

its actual pathologic extent. This is especially rel-evant for tumors of certain anatomic sites such asthe larynx, where considerable heterogeneity intumor characteristics and response to therapyhave been documented within the same T stage.4

Additionally, future comparisons of clinical out-comes may be expected to experience the effectsof this inconsistency between cTNM and pTNMstaging.

Staging information must be recorded in detailat the source by the clinician. The absence ofdetailed and accurate staging information makes itdifficult for cancer registrars to interpret clinicalrecords and may increase the chance for error inTNM staging. By the same measure, it is insuf-ficient for the clinician to assign a cTNM stagewithout documenting the exact anatomic extentof the tumor. Failure to record the characteristicsof the tumor from which the TNM stage isderived not only precludes comparisons with pre-vious versions of the staging system and analysis ofstatistical systematic variations such as stage mi-gration but also prevents use of the data set togenerate evidence for modification of the existingsystem. “Stage migration” occurs when patientsare assigned to different clinical stages because ofdifferences in the accuracy of the staging methodand the staging method used rather than actualdifferences in the extent of disease. For example,introduction of newer diagnostics tests such ascomputed tomography (CT) and magnetic reso-nance imaging (MRI) or positron emission to-mography scans can result in assignment of ahigher clinical stage and “improved” outcome forpatients in a particular stage group purely becauseof the increased accuracy of the diagnostic testcompared with conventional methods of tumorassessment such as clinical examination. This es-sentially results in a redistribution of patients toeach clinical stage without any change in theoverall results of treatment. To this end, the de-velopment of “collaborative staging” by theAJCC is an important step forward. Collaborativestaging incorporates all features of the tumor thatmay or may not be currently used in the TNMstaging system such as clinical, radiologic, patho-logic, biologic, and molecular features and co-morbidity information. A more comprehensiveand accurate staging/prognostic system that isable to incorporate all this information may be-

CA Cancer J Clin 2005;55:242–258

Volume 55 Y Number 4 Y July/August 2005 243


merican C



Dow

nloaded from


come a reality in the future with advances inbiocomputational techniques such as artificialneural networks. Until then, anatomic stagingwill remain the only practical and widely availablemethod for clinicians throughout the world. It isimportant to realize that an overly detailed andcomplex system that is very accurate in determin-ing prognosis will by definition be limited in itsutility and user-friendliness. Conversely, a verysimple staging system would have a high compli-ance rate but would experience poor prognosticability. Therefore, any revisions to existing stag-ing systems should be considered a compromisebetween the ideal and the practical.

For many decades, the AJCC-UICC TNMstaging system has been used for staging HNCworldwide. The system has been periodicallyrevised not only to incorporate informationavailable from advances in diagnosis (eg, endos-copy and radiologic imaging) but also fromimproved understanding of the biologic behav-ior of the numerous tumors that occur in thisanatomic area. The latest version of the TNMsystem5 became effective in January 2003, andunlike many other cancers, the staging systemfor head and neck carcinoma has undergonesignificant modification. This report, which isthe fourth in the series,6 is aimed at summariz-ing the salient modifications in staging strategyand the reasons behind these changes.

DIFFERENCES BETWEEN THE FIFTH (1997) ANDSIXTH (2002) EDITIONS OF THE AJCCSTAGING SYSTEM

Global Changes

The fifth edition of the AJCC staging sys-tem7 for HNC was felt to be deficient incertain areas listed in Table 1. One of themajor concerns was that the staging systemlacked sensitivity to stratify patients whowere placed in Stage IV. For example, thefifth edition placed patients with an advancedprimary tumor with no evidence of regionalor distant metastasis (T4N0M0) in Stage IValong with those who had clearly incurabledisease by virtue of extensive nodal metasta-

ses (N2b or N3) or widespread diffuse meta-static disease (M1).

It is well known that a majority of patientswith cancer of the head and neck have recur-rent disease at the primary site or in the neck atthe time of death. In many patients, local/regional recurrence is indeed the cause ofdeath. Thus, control of tumor above the clav-icles in the head and neck region is of para-mount importance in the overall managementof patients with HNC. Within Stage IV, there-fore, there are patients who are salvageable byaggressive surgical treatment for an extensiveprimary tumor that is felt to be surgically re-sectable. These patients clearly have a betterprognosis compared with those with wide-spread metastatic disease at distant sites. Simi-larly, patients with advanced disease at theprimary site and/or regional lymph nodes,which is felt to be technically not resectable,are currently managed with a treatment pro-gram of radiation therapy and chemotherapywith a reasonable expectation of tumor controland longevity. This is quite in contrast to pal-liative therapy offered to patients with distantmetastatic disease.

Thus, Stage IV has been revised to reflect thesetherapeutic strategies by defining subcategories:Stage IVA includes tumors that are locally ad-vanced but surgically resectable and therefore sal-vageable; Stage IVB includes tumors that arelocally advanced and surgically unresectable butpotentially treatable with a reasonable expectationof locoregional control with chemoradiotherapy;and Stage IVC is assigned to patients with distantmetastatic disease that is incurable and thereforeonly suitable for palliative treatment. While thesechanges are not supported by evidence-based in-formation, they are based on the opinion of ex-perts involved in the treatment of HNC andclearly reflect the shifting paradigms in manage-ment of advanced cancers of the head and neck aspracticed today in the United States and world-wide.

Discrepancies in the size descriptor for Tstaging among various sites in the fifth editionhave been eliminated, and uniform parameters(�2 cm, 2 to 4 cm, and �4 cm) are now usedfor several anatomic sites of the primary tumor(oral cavity, thyroid, and salivary glands).




merican C



Dow

nloaded from


The fifth edition did not provide any stagingsystem for tumors of the nasal cavity and ethmoidsinuses, and the only staging system available wasfor tumors of the maxillary sinus. In recognitionof the fact that tumors of the nasal cavity andethmoid sinuses have a natural history, routes ofprogression, and therapeutic strategies that aredistinct from maxillary sinus tumors, it was essen-tial that a separate staging system be introducedfor these two sites. The introduction of craniofa-cial surgery in the management of tumors thatapproach the skull base has provided prognosticinformation to support development of the Tstaging system of the ethmoid sinuses.8

The neck node staging system takes into ac-count the size, multiplicity, and laterality of nodalmetastasis. The latest edition of the staging systemmakes no changes in the N staging for any sitesexcept the thyroid, and a descriptor has beenadded for nodal metastasis in the upper neck or inthe lower neck, designated by (U) and (L). Therisk of distant metastasis, and thus outcome, isthought to be related to the location of the nodalmetastases in the neck. Although the locationdescriptor is not mandatory for nodal staging inthe current edition, the hope is that it will allowcollection of information to test this clinical ob-servation in large cohorts of patients for consid-eration in future revisions.

T Category Issues

Classification of tumors into T stages basedon the size of the primary tumor has beenlargely arbitrary. The size criteria for the T

category are based on the inverse relationshipof tumor volume to outcome for most HNCs.In the previous editions, the cutoff limits forsize criteria used for T staging were differentfor oral cavity and thyroid and salivary glandtumors. In the sixth edition, a uniform descrip-tor for size for the above head and neck siteswas introduced. In addition, for T4 tumors(larger than 4 cm), subcategories a and b wereintroduced based on involvement of vital struc-tures and thus their suitability for surgical re-section. T4a implies locally advanced butresectable tumor, while T4b implies tumor thatis not technically resectable but is suitable fornonsurgical options such as chemoradiother-apy. The obvious problem in using feasibilityof surgical resection as a staging parameter isthat it is subject to the variability of the level ofavailable expertise. However, the criteria usedto define “unresectability” are generally ac-knowledged as indicative of extension of tumorto vital anatomic structures, where surgical re-section is either technically not feasible with acurative intent or not recommended.

Site-specific T Category Changes

Table 2 is a compilation of the modifica-tions that appear in the T category. Figures 1through 6 illustrate the anatomic details ofthe head and neck sites, and Tables 3 through15 list the parameters for T, N, and M stagingas defined in the sixth edition of the stagingsystem.5

TABLE 1 The Rationale for Major Modifications in the Sixth Edition of the AJCC Staging System

Deficiencies Observed in the Fifth Edition Action Taken for the Sixth Edition

Patients with Stage IV disease represent a group with a wide spectrum ofbiological and clinical behavior and are suited for different treatment modalitieswith considerably different outcome expectations.

Patients with locoregionally advanced stage disease are therefore stratified intosurgically resectable (T4a) (Stage IVA) versus unresectable (T4b) (Stage IVB)disease. Patients with distant metastasis are classified as Stage group IVC.

Tumors arising from the nasal cavity or ethmoid sinuses have a very differentclinical course from those arising in the maxillary sinuses.

The nasoethmoid complex is now recognized as a distinct anatomic subsite withtwo regions: nasal cavity and ethmoid sinuses.

Categorization of T stage by size criteria is largely arbitrary and is based on theinverse relationship of tumor volume to prognosis. The lack of uniform sizecriteria for different anatomic sites resulted in confusion in their use.

Size descriptor for T categories are now uniform across all sites (oral cavity,oropharynx, salivary glands, and thyroid glands).

Anaplastic thyroid carcinomas behave very distinctly and had been grouped withother differentiated carcinomas of the thyroid.

Anaplastic carcinomas are considered T4a or T4b irrespective of other featuressuch as size.

The implication of involvement of central compartment nodes compared withlateral cervical and superior mediastinal nodes was not considered in Nstaging of differentiated thyroid cancer.

Metastasis to central compartment nodes is classified as N1a, and involvementof lateral cervical or superior mediastinal nodes is classified N1b.




merican C



Dow

nloaded from


Pharynx

The T classification for nasopharynx tumorsremains unchanged after the major revisions un-dertaken in the fifth edition. The description ofT2 has, however, been clarified to include exten-sion of tumor into soft tissue outside the naso-pharynx other than the parapharyngeal space.

The posterior pharyngeal wall is anatomi-cally continuous across the landmarks (plane ofthe superior surface of the soft palate and thesuperior surface of the hyoid bone) that areused to separate the nasopharynx, oropharynx,and the hypopharynx. Reliable identificationof the origin of a tumor that straddles thesethree regions of the pharynx may often beimpossible and lends some degree of unreliabil-ity to outcomes reporting for this subsite.

Tumors that invade the prevertebral fasciaare categorized T4b because the likelihood of

curative surgical resection is minimal, if any.However, delivery of a meaningful dose ofradiation to this area may be feasible usingintensity-modulated radiation therapy in spiteof its proximity to the spinal cord. Nonetheless,it is important to point out that the inherentdifficulty in detecting early invasion of the pre-vertebral fascia on radiologic imaging is likelyto cause some staging inhomogeneity withinthis category because most patients will un-dergo nonsurgical therapy.

Larynx

The prognostic significance of involvementof the paraglottic space by glottic as well assupraglottic tumors is acknowledged in thesixth edition, and these tumors are now stagedT3. There was considerable ambiguity in theprevious edition regarding the extent of carti-

TABLE 2 Site-specific Changes in Description of T4a and T4b Categories

Site Description

Lip and Oral CavityLip T4a: Tumor invades through cortical bone, inferior alveolar nerve, floor of mouth, or skin of face (ie, chin or nose)Oral cavity T4a: Tumor invades adjacent structures (eg, through cortical bone, into deep �extrinsic� muscle of the tongue �genioglossus,

hyoglossus, palatoglossus, and styloglossus�, maxillary sinus, skin of face) Superficial erosion alone of bone/tooth socketby gingival primary is not sufficient to classify as T4

Oral cavity and lip T4b: Tumor invades masticator space, pterygoid plates, or skull base and/or encases internal carotid arteryOropharynx T4a: Tumor invades the larynx, deep/extrinsic muscle of tongue, medial pterygoid, hard palate, or mandible

T4b: Tumor invades lateral pterygoid muscle, pterygoid plates, lateral nasopharynx, or skull base or encases carotid arteryNasopharynx No subdivision of T4

Stage groupings same as in fifth editionHypopharynx T4a: Tumor invades thyroid/cricoid cartilage, hyoid bone, thyroid gland, esophagus, or central compartment soft tissue

(including prelaryngeal strap muscles and subcutaneous fat)T4b: Tumor invades prevertebral fascia, encases carotid artery, or involves mediastinal structures

Larynx T4a: Tumor invades through the thyroid cartilage (glottic and supraglottic tumors) or cricoid cartilage (subglottic tumors) and/orinvades tissues beyond the larynx (eg, trachea, soft tissues of neck including deep extrinsic muscle of the tongue, strapmuscles, thyroid, or esophagus)

T4b: Tumor invades prevertebral space, encases carotid artery, or invades mediastinal structuresNasal cavity and ethmoid sinus T4a: Tumor invades any of the following: anterior orbital contents, skin of nose or cheek, minimal extension to anterior fossa,

pterygoid plates, sphenoid or frontal sinusesT4b: Tumor invades any of the following: orbital apex, dura, brain, middle cranial fossa, cranial nerves other than V2,

nasopharynx, or clivusMaxillary sinus T4a: Tumor invades anterior orbital contents, skin of cheek, pterygoid plates, infratemporal fossa, cribriform plate, sphenoid or

frontal sinusesT4b: Tumor invades any of the following: orbital apex, dura, brain, middle cranial fossa, cranial nerves other than maxillary

division of the trigeminal nerve V2, nasopharynx, or clivusMajor salivary glands T4a: Tumor invades skin, mandible, ear canal, and/or facial nerve

T4b: Tumor invades skull base and/or pterygoid plates and/or encases carotid arteryThyroid gland

Follicular, papillary, and medullary T4a: Tumor of any size extending beyond thyroid capsule to invade subcutaneous tissue, larynx, trachea, esophagus, orrecurrent laryngeal nerve

T4b: Tumor invades prevertebral fascia or encases carotid artery or mediastinal vesselsAnaplastic All anaplastic carcinomas are considered T4 tumors

T4a: Intrathyroidal anaplastic carcinoma—surgically resectableT4b: Extrathyroidal anaplastic carcinoma—surgically unresectable




merican C



Dow

nloaded from


lage invasion required for assigning Stage T4.Tumors that have caused minor cartilage ero-sion (eg, inner cortex of the thyroid lamina) arenow staged T3, while the T4a category is re-served for tumors that actually penetratethrough the cartilage.

Nasal Cavity and Paranasal Sinuses

In recognition that tumors of the nasal cavityand ethmoid sinuses are distinct from those of themaxillary sinus, two new anatomic subsites havebeen added under the category of paranasal si-nuses. This revision includes modifications thatallow appropriate classification of both nasal cav-

ity and ethmoid sinus tumors. Regarding tumorsof the maxillary sinuses, the description of T2 hasbeen reworded to clarify that extension of thetumor to the posterior wall of the maxillary sinus,and involvement of the pterygoid plates qualifiesas T3, a distinction that was not explicit andresulted in some ambiguity between the T2 andT3 categories in the previous edition.

Major Salivary Glands

The significant change in staging for majorsalivary gland tumors is the revision of thesize criteria for T3 tumors to eliminate theupper limit of 6 cm. The rationale for thisdecision was that because of the anatomicdimensions of the major salivary glands, tu-

FIGURE 1 Anatomic Subsites of the Lip and Oral Cav-ity. Reproduced with the permission of MedImmune On-cology, Inc., Gaithersburg, MD.

FIGURE 2 Anatomic Subsites of the Pharynx. Repro-duced with the permission of MedImmune Oncology,Inc., Gaithersburg, MD.




merican C



Dow

nloaded from


mors that are larger than 4 cm in dimensionnearly always have extension of the tumorbeyond the capsule of the gland. Extension ofthe tumor beyond the confines of the salivarygland clearly has adverse prognostic signifi-cance. With this change, the size criteria forsalivary gland tumors are now the same as forother anatomic sites (�2 cm, 2 to 4 cm, and�4 cm). The other major modification is inthe reorganization of the T4 category. Tu-mors that involve the skin of the preauricularregion, ear canal, mandible, and/or the facialnerve are amenable to potentially curativesurgical resection and are thus classified asT4a. More extensive tumors that involve theskull base and/or the pterygoid plates orthose that encase the carotid artery aredeemed surgically unresectable for cure andidentified as T4b.

Thyroid Gland

Differentiated and Medullary Thyroid Cancer

For differentiated (papillary and follicular)and medullary tumors confined to the paren-chyma of the thyroid gland without extrathy-roidal extension, there is no evidence to suggestthat using a size cutoff of 1 cm provides better

FIGURE 3 Anatomic Subsites of the Larynx. Repro-duced with the permission of MedImmune Oncology,Inc., Gaithersburg, MD.

FIGURE 4 Saggital Section Showing the Nasal Cavityand Paranasal Sinuses. Reproduced with the permissionof MedImmune Oncology, Inc., Gaithersburg, MD.

FIGURE 5 Coronal Section Showing the Relationshipof the Paranasal Sinuses to the Nasal Cavity and SkullBase. Reproduced with the permission of MedImmuneOncology, Inc., Gaithersburg, MD.




merican C



Dow

nloaded from


prognostic stratification compared with the2-cm cutoff used for other head and neck sites.Therefore, the T1 through T3 categories wererevised to maintain consistency with other

sites. Similarly, the fifth edition did not dis-criminate between minor extrathyroidal exten-sion of tumor (into the sternothyroid muscle(s)that can be easily encompassed in a routinethyroidectomy without any adverse prognosticinfluence) and more extensive extrathyroidalinvolvement of structures such as the sternohy-oid muscle, larynx, trachea, recurrent laryngealnerve, or esophagus. The fact that diverse out-comes may be expected in these two groups ofpatients is now recognized in the sixth edition:tumors that involve the sternothyroid muscleare classified as T3, while extension to larynx,trachea, esophagus, recurrent laryngeal nerve,or subcutaneous soft tissue, all of which aresurgically resectable, is classified as T4a. Tu-mors that invade the prevertebral fascia or en-case the carotid artery or mediastinal greatvessels are not resectable for cure, and thesepatients are staged T4b.

Anaplastic Thyroid Carcinoma

Patients with anaplastic thyroid carcinomahave an almost uniformly dismal prognosis irre-spective of local features of the primary tumor.Therefore, the only T category available for stag-ing these tumors has been T4. In keeping withthe theme of classifying T4 tumors by surgicalresectability, the sixth edition divides anaplastictumors into T4a (a small tumor confined withinthe thyroid parenchyma) and T4b (extrathyroidaltumor that is surgically unresectable). These twogroups of patients are, however, biologically dis-tinct, and the clinician should be aware that ananaplastic tumor confined to the thyroid gland istruly rare and is most often discovered as anincidental histologic finding in the surgical spec-imen after thyroidectomy for an otherwise unre-markable tumor. The diagnosis of anaplasticcarcinoma under these circumstances is by nomeans comparable to that of the more typicalpatient who presents with a rapidly growingsymptomatic thyroid mass. The T staging systemfor anaplastic thyroid carcinoma should thereforebe tempered with an appreciation of the biologyof the tumor and the background in which it hasarisen.

FIGURE 6 The Thyroid Gland and Its Lymphatic Drain-age. Reproduced with the permission of MedImmuneOncology, Inc., Gaithersburg, MD.

TABLE 3 T Staging for Tumors of the Lip andOral Cavity

TX Primary tumor cannot be assessedT0 No evidence of primary tumorTis Carcinoma in situT1 Tumor 2 cm or less in greatest dimensionT2 Tumor more than 2 cm but not more than 4 cm in greatest

dimensionT3 Tumor more than 4 cm in greatest dimensionT4a

Lip Tumor invades through cortical bone, inferior alveolarnerve, floor of mouth, or skin of face (ie, chin or nose)*

OralCavity

Tumor invades through cortical bone, into deep �extrinsic�muscle of tongue (genioglossus, hyoglossus,palatoglossus, and styloglossus), maxillary sinus, or skinof face

T4b Tumor involves masticator space, pterygoid plates, or skullbase and/or encases internal carotid artery

*Superficial erosion alone of bone/tooth socket by gin-gival primary is not sufficient to classify as T4. Used withthe permission of the American Joint Committee onCancer (AJCC), Chicago, Illinois. The original source forthis material is the AJCC Cancer Staging Manual, SixthEdition (2002) published by Springer-Verlag New York,www.springeronline.com.




merican C



Dow

nloaded from


TABLE 4 T Staging for Tumors of the Pharynx

TX Primary tumor cannot be assessedT0 No evidence of primary tumorTis Carcinoma in situNasopharynxT1 Tumor confined to the nasopharynxT2 Tumor extends to soft tissues

T2a Tumor extends to the oropharynx and/or nasal cavity without parapharyngeal extension*T2b Any tumor with parapharyngeal extension*

T3 Tumor involves bony structures and/or paranasal sinusesT4 Tumor with intracranial extension and/or involvement of cranial nerves, infratemporal fossa, hypopharynx, orbit, or masticator

spaceOropharynxT1 Tumor 2 cm or less in greatest dimensionT2 Tumor more than 2 cm but not more than 4 cm in greatest dimensionT3 Tumor more than 4 cm in greatest dimensionT4a Tumor invades the larynx, deep/extrinsic muscle of tongue, medial pterygoid, hard palate, or mandibleT4b Tumor invades lateral pterygoid muscle, pterygoid plates, lateral nasopharynx, or skull base or encases carotid arteryHypopharynxT1 Tumor limited to 1 subsite of hypopharynx and 2 cm or less in greatest dimensionT2 Tumor invades more than 1 subsite of hypopharynx or an adjacent site, or measures more than 2 cm but not more than

4 cm in greatest diameter without fixation of hemilarynxT3 Tumor measures more than 4 cm in greatest dimension or with fixation of hemilarynxT4a Tumor invades thyroid/cricoid cartilage, hyoid bone, thyroid gland, esophagus, or central compartment soft tissue†T4b Tumor invades prevertebral fascia, encases carotid artery, or involves mediastinal structures

*Parapharyngeal extension denotes posterolateral infiltration beyond the pharyngobasilar fascia.†Central compartment soft tissue includes prelaryngeal strap muscles and subcutaneous fat. Used with the permission ofthe American Joint Committee on Cancer (AJCC), Chicago, Illinois. The original source for this material is the AJCC CancerStaging Manual, Sixth Edition (2002) published by Springer-Verlag New York, www.springeronline.com.

TABLE 5 T Staging for Tumors of the Larynx

TX Primary tumor cannot be assessedT0 No evidence of primary tumorTis Carcinoma in situSupraglottisT1 Tumor limited to one subsite of supraglottis with normal vocal cord mobilityT2 Tumor invades mucosa of more than one adjacent subsite of supraglottis or glottis or region outside the supraglottis (eg, mucosa of

base of tongue, vallecula, medial wall of pyriform sinus) without fixation of the larynxT3 Tumor limited to larynx with vocal cord fixation and/or invades any of the following: postcricoid area, preepiglottic tissues,

paraglottic space, and/or minor thyroid cartilage erosion (eg, inner cortex)T4a Tumor invades through the thyroid cartilage and/or invades tissues beyond the larynx (eg, trachea, soft tissues of neck including

deep extrinsic muscle of the tongue, strap muscles, thyroid, or esophagus)T4b Tumor invades prevertebral space, encases carotid artery, or invades mediastinal structuresGlottisT1 Tumor limited to the vocal cord(s) (may involve anterior or posterior commissure) with normal mobility

T1a Tumor limited to one vocal cordT1b Tumor involves both vocal cords

T2 Tumor extends to supraglottis and/or subglottis, or with impaired vocal cord mobilityT3 Tumor limited to larynx with vocal cord fixationT4a Tumor invades cricoid or thyroid cartilage and/or invades tissues beyond the larynx (eg, trachea, soft tissues of neck including

deep extrinsic muscles of the tongue, strap muscles, thyroid, or esophagus)T4b Tumor invades prevertebral space, encases carotid artery or invades mediastinal structuresSubglottisT1 Tumor limited to the subglottisT2 Tumor extends to vocal cord(s) with normal or impaired mobilityT3 Tumor limited to larynx with vocal cord fixationT4a Tumor invades cricoid or thyroid cartilage and/or invades tissues beyond the larynx (eg, trachea, soft tissues of neck including

deep extrinsic muscles of the tongue, strap muscles, thyroid, or esophagus)T4b Tumor invades prevertebral space, encases carotid artery, or involves mediastinal structures

Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois. The original sourcefor this material is the AJCC Cancer Staging Manual, Sixth Edition (2002) published by Springer-Verlag New York,www.springeronline.com.




merican C



Dow

nloaded from


N Category Issues

There have been no modifications to the Nstaging system except for those on thyroid car-cinoma. Figure 7 illustrates the nodal levels thatare used to describe regional lymphatic meta-static spread to the neck.9 Although there was

considerable discussion by the AJCC TaskForce for Head and Neck Sites regarding theprognostic implications of certain features suchas the location of the involved nodes (upper

TABLE 6 T Staging for Tumors of the Nasal Cavity and Paranasal Sinuses

TX Primary tumor cannot be assessedT0 No evidence of primary tumorTis Carcinoma in situMaxillary sinusT1 Tumor limited to the maxillary sinus mucosa with no erosion or destruction of boneT2 Tumor causing bone erosion or destruction including extension into the hard palate and/or middle nasal meatus, except

extension to posterior wall of maxillary sinus, subcutaneous tissues, floor or medial wall of orbit, pterygoid fossa,ethmoid sinuses

T3 Tumor invades any of the following: bone of the posterior wall of maxillary sinus, subcutaneous tissues, floor or medial wallof orbit, pterygoid fossa, ethmoid sinuses

T4a Tumor invades anterior orbital contents, skin of cheek, pterygoid plates, infratemporal fossa, cribriform plate, sphenoid orfrontal sinuses

T4b Tumor invades any of the following: orbital apex, dura, brain, middle cranial fossa, cranial nerves other than maxillarydivision of trigeminal nerve V2, nasopharynx, or clivus

Nasal cavity and ethmoid sinusT1 Tumor restricted to any one subsite, with or without bony invasionT2 Tumor invading two subsites in a single region or extending to involve an adjacent region within the nasoethmoidal

complex, with or without bony invasionT3 Tumor extends to invade the medial wall or floor of the orbit, maxillary sinus, palate, or cribriform plateT4a Tumor invades any of the following: anterior orbital contents, skin of nose or cheek, minimal extension to anterior cranial

fossa, pterygoid plates, sphenoid or frontal sinusesT4b Tumor invades any of the following: orbital apex, brain, middle cranial fossa, cranial nerves other than V2, nasopharynx, or

clivus

Used with the permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois. The original sourcefor this material is the AJCC Cancer Staging Manual, Sixth Edition (2002) published by Springer-Verlag New York,www.springeronline.com.

TABLE 7 T Staging for Tumors of the MajorSalivary Glands

TX Primary tumor cannot be assessedT0 No evidence of primary tumorT1 Tumor 2 cm or less in greatest dimension without

extraparenchymal extension*T2 Tumor more than 2 cm but not more than 4 cm in

greatest dimension without extraparenchymalextension*

T3 Tumor more than 4 cm and/or tumor havingextraparenchymal extension*

T4a Tumor invades skin, mandible, ear canal, and/or facialnerve

T4b Tumor invades skull base and/or pterygoid plates and/orencases carotid artery

*Extraparenchymal extension is clinical or macroscopicevidence of invasion of soft tissues. Microscopic evidencealone does not constitute extraparenchymal extension forclassification purposes. Used with the permission of theAmerican Joint Committee on Cancer (AJCC), Chicago,Illinois. The original source for this material is the AJCCCancer Staginger Manual, Sixth Edition (2002) publishedby Springer-Verlag New York, www.springeronline.com.

TABLE 8 T Staging for Tumors of the Thyroid

TX Primary tumor cannot be assessedT0 No evidence of primary tumorT1 Tumor 2 cm or less in greatest dimension limited to the

thyroidT2 Tumor more than 2 cm but not more than 4 cm in

greatest dimension limited to the thyroidT3 Tumor more than 4 cm limited to the thyroid or any tumor

with minimal extrathyroid extension (eg, extension tosternothyroid muscle or perithyroid soft tissues)

T4a Tumor of any size extending beyond the thyroid capsuleto invade subcutaneous soft tissues, larynx, trachea,esophagus, or recurrent laryngeal nerve

T4b Tumor invades prevertebral fascia or encases carotidartery or mediastinal vessels

Anaplastic carcinomas†T4a Intrathyroidal anaplastic carcinoma—surgically resectableT4b Extrathyroidal anaplastic carcinoma—surgically

unresectable

All categories may be subdivided: (a) solitary tumor, (b)multifocal tumor (the largest determines the classification).†All anaplastic carcinomas are considered T4 tumors.Used with the permission of the American Joint Committeeon Cancer (AJCC), Chicago, Illinois. The original sourcefor this material is the AJCC Cancer Staging Manual, SixthEdition (2002) published by Springer-Verlag New York,www.springeronline.com.




merican C



Dow

nloaded from


neck versus lower neck) and extracapsular ex-tension, the Committee agreed that insufficientevidence exists at the current time to meritinclusion of these criteria in this staging system.It was proposed that the location of nodal me-tastasis be recorded separately (“U” for nodeslocated in the neck superior to the plane of the

lower border of the cricoid cartilage, and “L”for those located inferior to it) from the man-datory N category information in order thatthis information be available for future analysis.

TABLE 9 N Staging for All Head and Neck SitesExcept the Nasopharynx and Thyroid

Nx Regional lymph nodes cannot be assessedN0 No regional lymph node metastasisN1 Metastasis in a single ipsilateral lymph node, 3 cm

or less in greatest dimensionN2 Metastasis in a single ipsilateral lymph node, more

than 3 cm but not more than 6 cm in greatestdimension; or in multiple ipsilateral lymph nodes,none more than 6 cm in greatest dimension; or inbilateral or contralateral lymph nodes, none morethan 6 cm in greatest dimension

N2a Metastasis in a single ipsilateral lymph node morethan 3 cm but not more than 6 cm in greatestdimension

N2b Metastasis in multiple ipsilateral lymph nodes, nonemore than 6 cm in greatest dimension

N2c Metastasis in bilateral or contralateral lymph nodes,none more than 6 cm in greatest dimension

N3 Metastasis in a lymph more than 6 cm in greatestdimension

Used with the permission of the American Joint Committeeon Cancer (AJCC), Chicago, Illinois. The original sourcefor this material is the AJCC Cancer Staging Manual, SixthEdition (2002) published by Springer-Verlag New York,www.springeronline.com.

TABLE 10 N Staging for Tumors of theNasopharynx

Nx Regional lymph nodes cannot be assessedN0 No regional lymph node metastasisN1 Unilateral metastasis in lymph node(s), 6 cm

or less in greatest dimension, above thesupraclavicular fossa*

N2 Bilateral metastasis in lymph node(s), 6 cm orless in greatest dimension, above thesupraclavicular fossa*

N3 Metastasis in a lymph node(s) �6 cm and/orto supraclavicular fossa

N3a Greater than 6 cm in dimensionN3b Extension to the supraclavicular fossa*

*Midline nodes are considered ipsilateral nodes. Usedwith the permission of the American Joint Committee onCancer (AJCC), Chicago, Illinois. The original source forthis material is the AJCC Cancer Staging Manual, SixthEdition (2002) published by Springer-Verlag New York,www.springeronline.com.

TABLE 11 N Staging for Tumors of the Thyroid

Nx Regional lymph nodes* cannot be assessedN0 No regional lymph node metastasisN1 Regional lymph node metastasis

N1a Metastasis to Level VI (pretracheal, paratracheal, andprelaryngeal/Delphian lymph nodes)

N1b Metastasis to unilateral, bilateral, or contralateral cervicalor superior mediastinal lymph nodes

*Regional nodes are the central compartment, lateral cer-vical, and upper mediastinal lymph nodes. Used with thepermission of the American Joint Committee on Cancer(AJCC), Chicago, Illinois. The original source for this ma-terial is the AJCC Cancer Staging Manual, Sixth Edi-tion (2002) published by Springer-Verlag New York,www.springeronline.com.

TABLE 12 M Staging for Head and NeckTumors

Mx Distant metastasis cannot be assessedM0 No distant metastasisM1 Distant metastasis


TABLE 13 Stage Grouping for All Head andNeck Sites Except the Nasopharynx and Thyroid

StageGroup

TStage

NStage

MStage

0 Tis N0 M0I T1 N0 M0II T2 N0 M0III T3 N0 M0

T1 N1 M0T2 N1 M0T3 N1 M0

IVA T4a N0 M0T4a N1 M0T1 N2 M0T2 N2 M0T3 N2 M0T4a N2 M0

IVB T4b Any N M0Any T N3 M0

IVC Any T Any N M1





merican C



Dow

nloaded from


Extracapsular extension of nodal metastasis, onthe other hand, can be most reliably detectedonly on pathologic examination of the node inquestion because current radiologic imagingtechniques are not adequately accurate. Addi-tionally, the prognostic impact of extracapsularnodal extension independent of metastatic tu-mor volume has not been examined satisfacto-rily. More data will therefore be necessarybefore this feature can be considered for inclu-sion in the staging criteria.

Lymph node metastases in general have alimited prognostic impact in differentiat-ed thyroid carcinoma, especially in patientsyounger than 45 years. The worst possible Ncategory that can be assigned in these patientsis therefore only N1, which is now subclas-sified into N1a (metastasis to first echelonlymph nodes at Level VI) and N1b (metas-tasis to lateral cervical or superior mediastinalnodes). In recognition of the adverse prog-nostic impact of nodal metastasis in patientsolder than 45 years, N1b disease is includedunder Stage Group IVA for all tumors thatare smaller than Stage T4b.

M Category Issues

There have been not modifications in thiscategory except the clarification that all patientswith distant metastases are now classified asStage IVC.

Stage Grouping Issues

The stage grouping strategy has undergone amajor change in the sixth edition. As described

TABLE 14 Stage Grouping for Tumors of theNasopharynx

StageGroup

TStage

NStage

MStage

0 Tis N0 M0I T1 N0 M0IIA T2a N0 M0IIB T1 N1 M0

T2a N1 M0T2b N0 M0T2b N1 M0

III T1 N2 M0T2a N2 M0T2b N2 M0T3 N0 M0T3 N1 M0T3 N2 M0

IVA T4 N0 M0T4 N1 M0T4 N2 M0

IVB Any T N3 M0IVC Any T Any N M1


TABLE 15 Stage Grouping for Tumors of theThyroid*

Papillary or Follicular CarcinomaUnder 45 years

Stage Group T stage N stage M stageI Any T Any N M0II Any T Any N M1

45 years and olderStage Group T stage N stage M stageI T1 N0 M0II T2 N0 M0III T3 N0 M0

T1 N1a M0T2 N1a M0T3 N1a M0

IVA T4a N0 M0T4a N1a M0T1 N1b M0T2 N1b M0T3 N1b M0T4a N1b M0

IVB T4b Any N M0IVC Any T Any N M1

Medullary CarcinomaStage Group T stage N stage M stageI T1 N0 M0II T2 N0 M0III T3 N0 M0

T1 N1a M0T2 N1a M0T3 N1a M0

IVA T4a N0 M0T4a N1a M0T1 N1b M0T2 N1b M0T3 N1b M0T4a N1b M0

IVB T4b Any N M0IVC Any T Any N M1

Anaplastic CarcinomaStage Group T stage N stage M stageIVA T4a Any N M0IVB T4b Any N M0IVC Any T Any N M1

*Separate stage groupings are recommended for papillaryor follicular, medullary, and anaplastic (undifferentiated)carcinoma. Used with the permission of the AmericanJoint Committee on Cancer (AJCC), Chicago, Illinois. Theoriginal source for this material is the AJCC Cancer Stag-ing Manual, Sixth Edition (2002) published by Springer-Verlag New York, www.springeronline.com.




merican C



Dow

nloaded from


above, Stage IV now includes three subcatego-ries based largely on reorganization of the Tcategory. Therefore, unlike the fifth edition,tumors staged T3 or lower do not qualify forStage IV in the absence of other adverse fea-tures such as N2 or higher neck stage. Tomaintain consistency with this rule, T3N1 tu-mors of the major salivary glands are nowgrouped under Stage III instead of Stage IV asin the previous edition.

Like its predecessor, the sixth edition alsorecognizes the fact that the age of the host playsa major prognostic role in differentiated thy-roid cancer. Papillary and follicular carcinomasof the thyroid in individuals younger than 45years in age have an excellent outcome on thewhole, and therefore Stage Groups III and IV

are not applicable to this cohort irrespective ofT, N, or M stage of the tumor.

Pretreatment Staging Investigations

The importance of accurate mapping of thetumor before commencing therapy has alwaysbeen recognized by the AJCC-UICC. Thoroughphysical examination, including endoscopy if ap-propriate, should be combined with radiologicimaging to record the precise local (T), regionalnodal (N), and distant (M) extent of the tumor. Inmost instances, the information derived is thenassembled by the clinician to assign a pretreat-ment TNM stage. While recording this informa-tion, it is vital that a detailed description of thetumor, including tumor diagrams or photo-graphs, be archived to allow future analysis ofmodifications to the staging system or compari-sons with other systems based on the raw data.While abstraction of stage from clinical records bycancer registrars is an accepted practice, this in-volves a considerable amount of interpretation.Accurate data collection and pretreatment stagingof the tumor should be the responsibility of thephysician providing or coordinating the patient’scare. The accuracy of these data depends on theexpertise and experience of the physician.

Early lesions of the oral cavity and oropharynxare easily amenable to clinical examination in-cluding palpation. Clinical assessment of the su-perficial mucosal extent of these lesions isgenerally easy, and the depth of invasion intounderlying tissue can be approximated by palpa-tion. Radiologic imaging can be helpful in assess-ing tumors of the tongue for deep extension andfor evaluating the status of the mandible in tumorsthat are in close proximity to the bone. Imagingis also helpful in assessing tumors of the posteriorpharyngeal wall for their relationship to the pre-vertebral fascia. A particular advantage of cross-sectional imaging is its ability to detect nodalmetastases that cannot be appreciated on clinicalexamination, especially in patients who are notamenable to satisfactory examination (eg, thick,muscular neck) or in those who have at-risknodal basins that cannot be assessed clinically (eg,retropharyngeal nodes in patients with tonsil car-cinoma). Most experts consider CT superior toMRI at detecting extracapsular nodal extension,

FIGURE 7 Location of Lymphatic Nodal Levels in theNeck. Reproduced with the permission of MedImmuneOncology, Inc., Gaithersburg, MD.




merican C



Dow

nloaded from


and there are reports that imaging criteria such asinvolvement of internodal fat or spiculated mar-gins of metastatic disease are reliable indicators.10

Accurate staging of early larynx cancer requiresdirect visualization of the vocal cords during pho-nation to assess mobility. The distinction betweenimpaired mobility and vocal cord fixation deter-mines whether a patient with glottic cancer isstaged T2 or T3. While this distinction may ap-pear to be relatively straightforward, it does in-volve a degree of subjectivity that is impossible tostandardize. Consequently, T staging in thisgroup of patients involves an inherent heteroge-neity that is difficult to remedy. Thus, for glottictumors that are staged T2 from impaired mobilityof the vocal cord, a certain amount of observer-related subjectivity must be anticipated. Theprognostic implication of impaired mobility froma bulky tumor that is largely exophytic is appre-ciably different from a tumor that restricts cordmovement secondary to infiltration of the vo-calis muscle or arytenoid cartilage. Advancesin technology in the future may allow betterobjective quantification of some of these is-sues and help standardize staging data. Underthe current circumstances, however, it isdoubtful that any staging system can be com-prehensive enough to account for such vari-ations and still maintain broad usefulness inclinical practice. Radiologic imaging pro-vides little additional staging information forassessment of the mucosal extent of laryngealtumors. It can, however, provide valuableinformation to define T staging descriptorsrelated to the depth of the tumor such asinvasion of the base of the tongue, preepi-glottic and paraglottic spaces, or cartilage in-vasion.

Contrary to evaluation of tumors of themucosal surfaces of the remainder of the upperaerodigestive tract, cross-sectional radiologicimaging is mandatory for accurate assessment oftumors of the nasopharynx and paranasal si-nuses. Staging information related to therapeu-tic decision making such as involvement of thedura in an ethmoid tumor or parapharyngealextension of nasopharynx carcinoma cannot beobtained by clinical or endoscopic examina-tion. CT is excellent in assessing bone, whileMRI provides excellent delineation of soft-

tissue descriptors such as involvement of thedura or perineural extension of tumor. Theseimaging modalities should therefore be used tocomplement each other to answer specificquestions related to staging and therapy.

Assessment of tumors of the major salivaryglands and the thyroid gland benefits from radio-logic imaging if the extent of the tumor in itsthird dimension cannot be determined clinically.While cross-sectional imaging of a small, well-localized tumor of the parotid gland can provideaccurate dimensions for tumor staging, it has littlepotential to change the treatment algorithm.Conversely, radiologic imaging obtained duringthe course of investigation of unrelated illness hasresulted in an increasing incidence of incidentallydiscovered tumors (“incidentalomas”) of the thy-roid and parotid glands. Indeed, only a smallpercentage of these tumors are malignant, butwith the widespread use of imaging for evaluationof common ailments, early detection of some ofthese malignant tumors will likely result in im-proved outcomes and the inevitable debate of theeffect of “lead-time” bias.11

FUTURE MODIFICATIONS

T-category staging for tumors of certain ana-tomic sites will certainly undergo modification asinformation from well-conducted clinical studiesbecomes available. The increasing use of imagingalong with improved technology will hopefullybe able to provide more accurate data on the thirddimension or depth of the tumor. The depth ofinvasion of the primary tumor is already recog-nized as an important predictor for risk of nodalmetastases in some tumors, such as those of theoral cavity and especially the tongue.12,13 Othertumor attributes of prognostic relevance includethe morphology (exophytic versus endophytic)and the nature of the host-tumor interface (push-ing versus infiltrating).14 Histologic characteris-tics, such as perineural involvement, vascular andlymphatic invasion, the nature of cellular infiltratewithin the tumor, and numerous molecularmarkers,15 have also been reported to be indica-tive of outcome. If one adds to this mix thecountless reports supporting the “significance” ofa particular biologic or molecular characteristic of




merican C



Dow

nloaded from


a tumor to predict outcome, it is easy to see theneed for discretion in using this information forany future modifications to the staging system.Nonetheless, it is also important to recognize thatthe anatomic/morphologic features of the tumorthat we currently use for staging are only a re-flection of underlying genetic and molecularevents. While clinical and imaging characteristicsof tumors are undoubtedly more vulnerable tosubjective interpretation, the incorporation of ge-netic and molecular characteristics into the stag-ing system will have to await much betterunderstanding of basic tumor and host biology.

As nonsurgical organ-preserving therapeuticapproaches gain general acceptance, the relativeimportance of noninvasive radiologic assessmentof the extent of the tumor has increased. Unfor-tunately, correlative data comparing the accuracyof these approaches to the definitive standard ofpathologic examination do not exist for mosthead and neck sites. This situation is unlikely tochange in view of the trend toward nonsurgicaltherapy for most advanced head and neck tumors.Considerable research is underway to developchemosensitivity assays16 and biological and/ormolecular predictors of response to nonsurgicaltherapy.17 However, a reliable method of iden-tifying patients who are likely to respond to che-moradiation therapy remains elusive. Advancedlarynx cancer provides a classic example of thisdilemma. It is well recognized that not all patientswith advanced laryngeal tumors benefit fromorgan-preserving chemoradiation therapy. Al-though there have been attempts to identify clin-ical (patient- and host-related)18 and radiologicpredictors19 of response, there is considerable am-biguity regarding their accuracy. Specifically, pa-tients who have involvement of the laryngealcartilage framework are thought to be at risk forpoor response to chemoradiation therapy. A ma-jor drawback of all clinical and radiologic studiesthat evaluate this question in the contemporaryera is the lack of pathologic confirmation of theiraccuracy because index total laryngectomy is nolonger accepted as standard practice except inpatients with advanced laryngopharyngeal tumorsand gross thyroid cartilage destruction. It remainsto be seen how clinical staging will have an im-pact on future analysis of outcomes, and at leastfor some anatomic sites, the staging criteria may

need modification to account for the effects ofthis conundrum.

Certain tumors, such as malignant tumors,that involve the skull base or temporal bone arerelatively uncommon, and because individualinstitutions are unable to muster enough datafor outcomes analysis, staging systems are eithernonexistent or relatively basic.20 The onlypractical solution to this problem is multiinsti-tutional, preferably international, collaborativeefforts at data collection and reporting.8 Col-lection, audit, analysis, and reporting of datafrom around the world are certainly facilitatedby the use of the Internet, and this approachmay also be useful to consolidate the stagingsystem for the more common tumors of thehead and neck.

Nodal metastasis is the most important pre-dictor of outcome for squamous cell cancer ofthe head and neck. As discussed above, therelative impact of certain characteristics ofnodal metastatic disease such as extracapsularspread and their location in the neck needmore investigation. While the prognostic im-plication of clinically recognized metastases isbeyond doubt, the controversy regarding theadverse effect of subclinical nodal disease hasnot been settled. Sentinel node biopsy is widelyaccepted as the standard of care in the manage-ment of cutaneous malignant melanoma. Theprocedure is currently under investigation formucosal HNC, and immunohistochemistryand assessment of molecular markers of meta-static disease in sentinel nodes will identify“submicroscopic” or “molecular” metastases in“sentinel” nodes. The clinical implication ofthis entity is currently under investigation formalignant melanoma and a similar trend can beexpected for HNC.

Whole body functional imaging using tech-niques such as 18-fluorodeoxyglucose positronemission tomography scanning is becoming in-creasingly popular in the metastatic workup ofpatients, especially those undergoing chemora-diation therapy for advanced HNC.21 One ofthe advantages of chemoradiation therapy re-ported in clinical trials completed over the pastdecade has been the reduction in deaths fromdistant metastases. It is, however, very likelythat some degree of discrepancy existed among




merican C



Dow

nloaded from


patient groups in these trials regarding theirpretreatment distant metastatic status in view ofthe fact that whole body functional imagingtechniques were not available for routine use.A proportion of these patients may have har-bored metastatic disease that was never re-corded, thus introducing bias in comparingtreatment groups. The availability of wholebody scanning and its use in patients with ad-vanced HNC may have the effect of homoge-nizing the “cM0” population that are subject toclinical trials.

Except for differentiated thyroid cancer, wherethe age of the patient is a critical descriptor, thecurrent TNM staging system for HNC does nottake into account any characteristic of the host. Itis well recognized that the typical squamous cellcarcinoma of the upper aerodigestive tract is eti-ologically related to the use of tobacco and/oralcohol. Consequently, these patients also havecomorbid conditions that although not a featureof the cancer itself, have the potential to impactthe type, efficacy, and outcome of therapy. Theprognostic influence of comorbidity on outcomesafter treatment of several HNCs has been re-ported in the recent literature.22 More researchand validation of these reports are desperatelyneeded so that the prognostic impact of this pa-rameter can be considered in future modificationsof the staging system. The status of the hostimmune system, on the other hand, is admittedlydifficult to quantify but does remain another rel-atively unexplored characteristic in terms of prog-nostic impact. The explosive clinical course of anearly stage cancer in a patient with human im-munodeficiency virus or drug suppression fol-lowing organ transplantation is the most extremeexample of the impact of a compromised im-mune system on outcome. Lesser degrees of im-mune dysfunction may be prevalent in somepatients with early tumors who have an atypicallyaggressive clinical course (eg, early stage tonguecancer in a young female who reports no use oftobacco and/or alcohol). As our understanding ofthe complex interactions that obviously occurbetween the host and the tumor improves, futurestaging systems will likely have the advantage ofcomputer technology such as artificial neural net-works and will benefit from inclusion of these and

other characteristics of the host. The develop-ment of collaborative staging by the AJCC is astep in this direction and will allow incorporationof relevant nonanatomic parameters in future re-visions of the TNM staging system.

CONCLUSION

The TNM staging system for HNCs is now inits sixth version and is an important tool not onlyfor reporting and comparing outcomes of ther-apy, but also for improving stratification of pa-tients for inclusion in clinical trials. As technologyadvances and data from ongoing clinical and basicresearch become available, in the coming yearsfurther revisions may be expected to incorporateas yet unexplored issues related to the host (eg,medical comorbidity and immune competence)and the tumor (eg, biological and molecularmarkers). Advances in computing technologysuch as artificial neural networks may facilitate thisprocess. The challenge then will be to maintainsight of the central theme of user-friendliness sothat the staging system maintains its role of pro-viding a common language for communicationbetween clinicians and scientists of all specialtiesaround the world.

It is important, however, to realize that nostaging system is perfect. The more comprehen-sive and detailed the staging system, the moreaccurate and more predictive of prognosis thesystem becomes. However, a very detailed andcomplex staging system becomes less user-friendly, and therefore its utility and compliancedrops. Thus, the AJCC/UICC staging system is acompromise between the ideal and the practical,and the current revision is no exception. Never-theless, the changes implemented in this revisionoffer an incremental gain toward a “perfect” stag-ing system.

ACKNOWLEDGMENTS

The authors would like to acknowledge Ms.Nancy Bennett for her editorial assistance.




merican C



Dow

nloaded from


REFERENCES

1. Denoix PF. Tumor, Node and Metastasis(TNM). Bull Inst Nat Hyg (Paris) 1944;1:1–69.

2. Sobin LH. TNM: evolution and relation toother prognostic factors. Semin Surg Oncol2003;21:3–7.

3. Gospodarowicz MK, Miller D, Groome PA,et al. The process for continuous improvement ofthe TNM classification. Cancer 2004;100:1–5.

4. Pameijer FA, Balm AJ, Hilgers FJ, Muller SH.Variability of tumor volumes in T3-staged headand neck tumors. Head Neck 1997;19:6–13.

5. Greene FL, Page DL, Fleming ID, et al., eds.AJCC Cancer Staging Manual. 6th ed. NewYork, NY: Springer; 2002.

6. Greene FL. TNM: our language of cancer.CA Cancer J Clin 2004;54:129–130.

7. Fleming ID, ed. AJCC Cancer StagingManual. Philadelphia, PA: Lippincott-Raven;1997.

8. Patel SG, Singh B, Polluri A, et al. Craniofa-cial surgery for malignant skull base tumors: re-port of an international collaborative study.Cancer 2003;98:1179–1187.

9. Robbins KT, Clayman G, Levine PA, et al.Neck dissection classification update: revisions pro-posed by the American Head and Neck Society and

the American Academy of Otolaryngology-Headand Neck Surgery. Arch Otolaryngol Head NeckSurg 2002;128:751–758.

10. Yousem DM, Som PM, Hackney DB, etal. Central nodal necrosis and extracapsularneoplastic spread in cervical lymph nodes: MRimaging versus CT. Radiology 1992;182:753–759.

11. Stockler MR, Boyd NF, Tannock IF.Guide to Studies of Diagnostic Tests, Prognos-tic Factors, and Treatments, in Tannock IF,Hill RF (eds). The Basic Science of Oncology.3rd ed. New York, NY: McGraw-Hill;1998:473.

12. Spiro RH, Huvos AG, Wong GY, et al.Predictive value of tumor thickness in squa-mous carcinoma confined to the tongue andfloor of the mouth. Am J Surg 1986;152:345–350.

13. O-charoenrat P, Pillai G, Patel S, et al. Tu-mour thickness predicts cervical nodal metastasesand survival in early oral tongue cancer. OralOncol 2003;39:386–390.

14. Spiro RH, Guillamondegui O Jr, PaulinoAF, Huvos AG. Pattern of invasion and marginassessment in patients with oral tongue cancer.Head Neck 1999;21:408–413.

15. Sotiriou C, Lothaire P, Dequanter D, et al.

Molecular profiling of head and neck tumors.Curr Opin Oncol 2004;16:211–214.

16. Singh B, Li R, Xu L, et al. Prediction ofsurvival in patients with head and neck cancerusing the histoculture drug response assay. HeadNeck 2002;24:437–442.

17. Koch W, Sidransky D. Molecular markersof radiation effectiveness in head and necksquamous cell carcinoma. Semin Radiat Oncol2004;14:130 –138.

18. Sherman EJ, Fisher SG, Aliff E, et al. TALKscore: validation of a tool for predicting larynxpreservation (LP) outcomes. Proc Am Soc ClinOncol 2003;22:497.

19. Zbaren P, Becker M, Lang H. Pretherapeuticstaging of laryngeal carcinoma. Clinical findings,computed tomography, and magnetic resonanceimaging compared with histopathology. Arch Oto-laryngol Head Neck Surg 1996;77:1263–1273.

20. Kadish S, Goodman M, Wang CC. Olfac-tory neuroblastoma. A clinical analysis of 17 cases.Cancer 1976;37:1571–1576.

21. Schoder H, Yeung HW. Positron emission im-aging of head and neck cancer, including thyroidcarcinoma. Semin Nucl Med 2004;34:180–197.

22. Piccirillo JF. Importance of comorbidity inhead and neck cancer. Laryngoscope 2000;110:593–602.




merican C



Dow

nloaded from


TNM Staging Cancer

Documents

Transcript of TNM Staging Cancer