Principles of Cancer Classification

Why should we classify cancer tumours ?

The classification of the tumours has several goals :

• · Predicting prognosis,
  
• · Adapting therapy to the clinical situation,
  
• · Comparing therapeutic results between relatively homogeneous groups of patients,
  
• · Enabling therapeutic studies which are essential in proving any therapeutic progress.

Classification enables the definition of therapeutic groups, for which therapeutic protocols can be elaborated, taking into account all treatment possibilities.
It is essential for physicians to establish the classification of a tumour before any treatment can be administered to the patient in order to:

• · avoid proposing unnecessary treatment (for instance mutilating surgery when the patient unfortunately has metastases),
  
• · propose the most appropriate treatment (for instance: general treatment when localised treatment might be more appropriate).

Criteria used for tumour classifications

Very early in the twentieth century, physicians who exchanged data about their therapeutic results felt the necessity to base their common classification on objective, easy to understand and easy to implement factors.

Most classifications are based on clinical data. However, other criteria are sometimes considered.
The most determining factors are :

• the degree of local invasion,
  
• the degree of remote invasion,
  
• histological types of cancer with specific grading for each type of cancer,
  
• possibly various tumour markers,
  
• (in the near future, gene markers and other proteomic abnormalities may become determining factors),
  
• general status of the patient.

General methodology

A great number of consensus meetings were held among experts in order to set up classification standards.

The need for a common nomenclature led to the clinical classification of cancer by the League of Nations Health Organisation in 1929 and then by the UICC.

Just after the Second World War, Pr Pierre DENOIX (surgeon and Director of the Institut Gustave ROUSSY, near Paris), proposed to the UICC (International Union Against Cancer) Board, the TNM classification, which collects data from local tumour invasion (T), node invasion (N) and remote metastases invasion (M).

Other bodies (for example: AJCC: American Joint Committee on Cancer in 1959, FIGO: Fédération Internationale de Gynécologie et Obstétrique since 1937) proposed slightly different classifications and with time all classifications progressively converged towards an "improved TNM classification".

As the results of therapeutic studies progress according to these classifications, the classifications are improved, taking into account demonstrated prognostic factors. At regular intervals, in view of clinical and biological results, new modifications are discussed and accepted by UICC bodies.

These considerations explain the utmost importance of long-term clinical follow-up, which is the only way to verify classification value, the various proposed prognostic factors, the real impact of any new therapeutic measures on patient survival and the onset of late side-effects which could strongly modify the final result.

Result data collection enables a clear view of overall results and trends. In France, the Federation of Comprehensive Cancer Centres has collected therapeutic results for every patient treated in these hospitals for over 50 years. At an international level, the Fédération Internationale de Gynécologie et d'Obstétrique (FIGO) collects, every 2 to 3 years, the results of hundreds of clinical departments around the world.

The therapeutic protocol

The retrospective analysis of results enables the regular adjustment of cancer therapy and the definition of decision trees according to classification. At every stage, each therapy is motivated by observed results and complications. This clear protocol definition requires:

• · daily updating of literature knowledge,
  
• · great precision in the methodology of protocol drafting,
  
• · constant and rigorous application of the pre-defined protocol,
  
• · wide participation in clinical trials,
  
• · well organised follow-up of each patient,
  
• · voluntary, permanent and exhaustive collection, up-dating and presentation of therapeutic results.

In a subsequent chapter on multidisciplinarity, the importance in confronting the results and the experiences of various physicians and scientists from different specialities in order to elaborate efficient therapeutic protocols will be explained.

Staging assessment procedures
Histo-pathological classifications,
The main classification systems,
Several classifications and their prognostic value
General performance status classification

Local invasion Assessment :


Local invasion assessment begins with the initial complete clinical examination during which the para-clinical examinations necessary to correctly assess local cancer extension and propose a correctly applicable therapy are determined.

If, in theory, all cancers can give birth to any kind of metastases, in reality the elective occurrence of metastases according to the primitive tumour allows the definition of a standard chart for the majority of cancer localisations.

Local extension

The goal of the study of local extension is to identify tumour localisation and dimension and its relation with neighbouring structures. Its evaluation is based on tumour size and using various methods depending on tumour localisation.

Evaluating a superficial tumour

In most cases, the clinical examination can offer substantial data.
For instance, considering breast carcinoma:

• measurement of the diameters of the mammary nodule,
  
• identification of its localisation in relation to the nipple,
  
• verification of its relationship with skin: adhesion, skin invasion ('peau d'orange' phenomenon or orange peel phenomenon),
  
• verification of its mobility with regards to the pectoral muscle and the chest wall,
  
• mammography (and potentially RMI) allows a more objective, accurate evaluation and other specific radiological criteria typical of a cancer: structure irregularity, stellar aspect, micro-calcifications.

The same clinical evaluation can be carried out for other tumours such as thyroid, testis, soft tissue tumours: with a well practised clinical examination, a good classification can be established.

Local evaluation by clinical examination

Combining a quality clinical examination with a few paraclinical tests allows an accurate evaluation of local invasion.

For instance, for cancer of the cervix uteri,

• speculum examination allows a good measurement of the lesion and a definition of vaginal invasion,
  
• pelvic examination (vaginal but especially rectal examination) permits an appreciation of the infiltration of the vaginal wall and of parameters,
  
• the intravenous pyelography detects any ureteral compression (or a non functioning kidney) by a parametrial invasion,
  
• cystoscopy is mandatory to exclude a vesical invasion (and in the case of doubt, a rectoscopy for studying a rectal invasion).

However, to obtain an accurate classification of a cervix uteri tumour, according to FIGO criteria, a gynaecological examination under general anaesthesia should be performed in order to avoid any painful patient reaction and to offer the physician sufficient time to confirm the invasion diagnosis.

Similar situations may be observed for tongue cancer, head and neck cancers, prostate carcinoma and rectal carcinoma. Generally, the clinical examination enables a quality evaluation and classification.

However, very often, a calmer and uninterrupted examination can be performed under general anaesthesia (notably to obtain further biopsies).

Profound cancer evaluation

For those cancers, many other complex examinations should be performed:

For colon carcinoma, coloscopy is the most important examination for diagnosis and evaluation of superficial invasion. However, most often, the depth invasion will be evaluated after colectomy on the surgical specimen.

For lung cancer, bronchoscopy allows a good definition of the superficial tumour extension, while a thoracic scanner is mandatory for exploring infiltration and tumour size; a mediastinoscopy may be mandatory. In spite of all these examinations, surprising surgical discoveries are quite frequent.
For ovarian carcinoma, only a well conducted exploratory laparotomy with precise specific surgical exereses will allow an accurate classification.

Node extension assessment:

Clinical evaluation

According to tumour lymphatic drainage, node extension assessment may be done (at least partly) by clinical evaluation.

An invaded node has a very typical appearance : its size is increased, its consistency is indurate without any pain, its mobility is more or less reduced relating to the adjacent tissues; usually there is no inflammatory reaction (except in inflammatory forms of breast cancer or when the node or the tumour is super infected.

With careful palpation of node areas, one can find:

• the axillary nodes of a breast carcinoma, (but of course the intern mammary nodes need a thoracic exploratory examination such as a scan),
• the cervical nodes of head and neck cancers,
• the inguinal nodes of vulvae or penis cancer, or of melanoma on the leg,
• the supraclavicular nodes from distant metastases (we are no longer speaking of regional involvement).

Paraclinical evaluation

Most nodes need paraclinical evaluation in order to be assessed:

• pelvic lymphography, usually done for lymph node diseases but still, according to certain medical teams, for cervix uteri cancers, showing node hypertrophy, the presence of lacunae and a jamming or derivation of the contrasting agent in the case of massive involvement,
• thoraco-abdomino-pelvic scan, shows an increased node size (any volume above 1 cm begins to be significant) which is not always in relation to a cancerous nodal involvement,
• abdominal ultrasonography,
• RMI,
• positron scintigraphy seems most interesting for many tumours (PET scan).

Surgical evaluation

Most clinical and paraclinical methods do not offer a definitive conclusion.

For this reason, in most cancer surgical procedures, exploratory satellite lymph node surgery is mandatory with systematic histological study of the specimens.

Metastasis assessment :

 The risk of distant metastases exists for any type of cancer, however it is almost inexistent for skin basal cell carcinoma and is rare for brain tumours.

The clinical check-up will try to find the most frequent and preferential metastases as defined in the 'General History' chapter.

Four organs are the most frequently studied:

• the lungs: with a single thoracic radiography but nowadays, with a one centimentre sliced pulmonary scanner,

• the liver: studied with sonography, sometimes scan, hepatic biology in particular enzymes showing biliary retention (γ GT),

• the bones: bone scintigraphy, one the most sensitive ways to detect either dense or lytic lesions, with bone radiographies if necessary. RMI might be another way to largely explore the skeleton and also for studying vertebrae and the spinal chord.

• the brain: by scann or RMI, with lumbar puncture for certain tumours (breast cancer meningitis).

Some tumours will necessitate specific examination : bone marrow count or bone marrow biopsy for lymphoma, for instance.

Tumour markers :


The search for distant metastases should be more careful when tumour markers are elevated (especially after initial surgery):

• Ca 19-9 or CEA for colo-rectal tumours,

• Ca 15-3 for mammary carcinoma,

• PSA for prostate cancer,

• βHCG et αFP for testis tumours.

Some classifications use tumour marker levels thus modifying the therapeutic indications (noticeably for testicular, ovarian and extra-gonadal germinal tumours).

For testis cancer, the serum value of tumour markers (the S criteria) will classify the tumour from SX to S3.

In prostate cancer, although not yet integrated in the TNM classification system, a high PSA value(above 20 ng/ml) suggests involvement beyond the prostate and should modify the therapeutic decision (for instance, no radical prostatectomy).


Histo-pathology study:

Histopathology classification is one of the most essential criteria for tumour classification. Here are a few examples:

Lungs

Small cell lung carcinoma has a very distinct prognosis as compared to other histopathologic forms, like for instance epidermoid tumours. In the past, its prognosis was catastrophic (survival of only a few months). Its sensitivity to chemotherapy is very strong, thus offering a few cases of long remission. On the contrary, pulmonary adenocarcinomas are generally peripheral tumours with a slower development than epidermoid cancer, and that can often be treated by surgery.
Thus, lung cancer per se does not exist and studies should only compare similar histologies (with a simple differentiation between 'small cell lung carcinoma' and 'non-small cell lung carcinoma').

Thyroid

There are two very distinct histological forms of thyroid carcinoma: papillary carcinomas (more or less differentiated) are generally slow developing tumours which are, for a prolonged period of time, even when they give birth to metastases, sensitive to metabolic irradiation through radioactive Iodine doses, whereas medullary thyroid carcinoma is a rarer disease with a very characteristic elevation of calcitonin secretion, a relatively rapid evolution, and that is not particularly reactive to any kind of therapy.

Testis tumours

Testis tumours have a similar spontaneous evolution whatever their histology. However seminoma are very sensitive to low doses of radiotherapy (and in fact to chemotherapy) whereas 'non-seminomatous' testis tumours (a very heterogeneous group) are not very sensitive to radiotherapy but generally very sensitive to chemotherapy. Some histological forms (in particular chorio-carinomatous forms) have an even quicker disease progression but seem to have differing sensitivity to chemotherapy. Tumour markers are very important for the classification of testis tumours.

Lymphoma

Lymphomas need a very precise pathological analysis in order to distinguish Hodgkin’s lymphoma from other types (non-Hodgkin’s lymphoma) with various classifications attempting to improve differentiation among a variety of prognoses. Molecular biology and immunochemistry are now very important tools for their classification.

Breast

Within the same general histological classification (adenocarcinoma), various degrees of differentiation exist. For breast cancer, the Scarff, Bloom and Richardson classification enables us to distinguish poorly differentiated tumours which are prone to quickly metastasise, even in apparently localised tumours, and which should lead to the systematic prescription of adjuvant chemotherapy.

Prostate

The same reasoning can be made for prostate cancer which has a great histological heterogeneity among specimens from the same patient: Gleason proposed a grading system which adds the most frequently observed differentiation of a patient's tumour to its less frequently observed differentiated aspect. For instance a Gleason grading (2 + 4) means that most of the tumour is well differentiated (grade II) but the least differentiated is grade IV.

Classification according to stages:


The various professional societies and institutions treating cancer have, for a long time, used the following classification according to cancer stage:

Stage	Description
Stage 0	In situ (non invasive) cancer
Stage 1	Very localised tumour with no remote metastases
Stage 2	Locally limited extension with/without minimal node satellite extension and with no remote metastases
Stage 3	Locally advanced extension with/without major node satellite extension and with no remote metastases
Stage 4	Locally advanced tumour and/or distant metastases

Many different classifications are proposed according to tumour type.

These classifications are generally very close to strategic therapies but they are not as clear as TNM classification (discussed later). With time, however, all of these classifications have converged thus unifying the daily practice of professional societies and the more universal systematic TNM classification.

UICC (International Union Against Cancer) which promotes TNM classification regularly asks expert clinicians to revise the classifications according to new prognostic factors.

TNM Classification:

This classification may be:

• either purely clinical with the prefix 'c' (cTNM) or simply TNM
• or after surgical treatment and pathological study of the surgical specimen with the prefix 'p' (pTNM)
• or after relapse with the prefix 'r' (rTNM).

T component

This criterion concerns the primitive tumour

T	Description
T x	The primitive tumour cannot be studied
T 0	There is no primitive tumour
T 1	Very limited primitive tumour
T 2	Larger tumour (generally more than 2 cm in diameter)
T 3	Tumour with extension to adjacent connective tissue (fixed tumour)
T 4	Extension to adjacent structures.

N component

This criterion concerns the regional nodes

N	Description
N x	It is not possible to describe the node status
N 0	The research for satellite nodes is negative
N 1	Minimal invasion of regional nodes
N 2	Major invasion of regional nodes
N 3	Node invasion beyond regional nodes

M component

This criterion describe the presence of metastases.

M	Description
M x	It is not possible to describe the metastasis status
M 0	There are no remote metastases
M 1	There are one or many remote metastases.

A prefix can be used to describe multiple metastases (mM).
The localisation of the metastases is described by suffix:

• pul (lung),
• oss (bone),
• hep (liver),
• bra (brain),
• lym (remote lymph node),
• pleu (pleura),
• per (peritoneum),
• ski (skin),
• oth (other site).

G component

This component describes the histological grading of epithelial tumours.

G	Description
G x	No precision about histological grading
G 1	Well differentiated tumour
G 2	Moderately differentiated tumour
G 3	Poorly differentiated or undifferentiated tumour

A few classification examples :


The following examples are illustrated by diagrams drawn by the author and based on UICC documents. Prognostic data are also given.

Cancer of the tongue:
        Tongue cancer T1 

The lesion is situated on the border of the tongue and measures less than 2 cm at its largest diameter.

Such lesions are often observed in front of dental stumps and are due to an unceasing local irritation.

Moreover, there is a heavy alcoholic and tobacco intoxication.

Tongue T2

  The lesion is situated on the border of the tongue. Its dimension is between 2 to 4 cm at its largest diameter.

Same aetiological conditions as for T1.

Such a lesion leads to haemorrhage and a very fetid breath.

Tongue T3

Lesions of a border of the tongue greater than 4 cm at its largest diameter.

 Cross sectional view

Tongue cancer T4

The lesion involves neighbouring structures.

 Vocal chord cancer:

The following table gives a general outline of vocal chord cancer classification.
Diagrams are available for the following:

T1 Vocal Chord Cancer T2 Vocal Chord Cancer T3 Vocal Chord Cancer T4 Vocal Chord Cancer

N1 Vocal Chord Cancer N2a Vocal Chord Cancer N2b Vocal Chord Cancer N2c Vocal Chord Cancer N3 Vocal Chord Cancer

T Classification

Classification	Description
T1 T1a T1b	Tumour limited to the vocal chord(s) (may involve anterior or posterior commissure) with normal mobility tumour limited to one vocal chord tumour involves both vocal chords
T2	Tumour extending to supraglottis and/or subglottis and/or with impaired vocal chord mobility
T3	Tumour limited to the larynx with vocal chord fixation and/or invading paraglottic space, and/or minor thyroid cartilage erosion (e.g., inner cortex)
T4 T4a T4b	T4 : extension beyond the larynx T4a: tumour invading through the thyroid cartilage and/or invading tissues beyond the larynx (e.g., trachea, soft tissues of neck, including deep extrinsic muscle of the tongue, strap muscles, thyroid, or oesophagus) T4b: tumour invading prevertebral space, encasing carotid artery, or invading mediastinal structures

N Classification

N.B. on the following link you will find the anatomy of cervical nodes.

Classification	Description
NX	Regional lymph nodes cannot be assessed
NO	No regional lymph node metastasis
N1	Metastasis in a single ipsilateral lymph node, 3 cm or less at its largest point
N2a	Metastasis in a single ipsilateral lymph node larger than 3 cm but no more than 6 cm at its largest point
N2b	Metastasis in multiple ipsilateral lymph nodes, no larger than 6 cm at its largest point
N2c	Metastasis in bilateral or contralateral lymph nodes, no larger than 6 cm at its largest point
N3	Metastasis in a lymph node more than 6 cm at its largest point

Usual therapeutic results

Supra glottic tumour
T	Local control	5 year survival
T1	80-90%	65-90%
T2	60-80%	50-65%
T3	35-70%	35-55%
T4	30-60%	15-40%

It is clear that despite more and more mutilating surgery, 5 year survival diminishes: this is due to the high frequency of remote metastases (and of monstrous nodes which cannot be cured by radiotherapy).

Vocal chord tumour
T	Local control	5 year survival
T1	85-95 %	80-95 %
T2	65-75 %	60-85 %
T3	20-35 %	35-60 %
T4	15-30 %	10-30 %

Compilation of results according to Laramore (Radiation therapy of Head and Neck Cancer. Berlin. Springer Verlag 1988).

Lung cancer:

The classification of lung (or bronchial) carcinoma is based on TNM or more surgical stages derived from TNM.
The following drawings illustrate these stages:

lung cancer stage I lung cancer stage II

lung cancer stage IIIa lung cancer stage IIIb

T criteria

T	Description
TX	Primary tumour cannot be assessed, or tumour is proven by the presence of malignant cells in sputum or bronchial washings but is not visualised by imaging or bronchoscopy.
T0	No evidence of primary tumour
Tis	Carcinoma in situ
T1	A tumour measuring less than 3 cm at its largest point, surrounded by lung or visceral pleura, and without bronchoscopic evidence of more proximal invasion than the lobar bronchus (i.e., not in the main bronchus).
T2	A tumour with any of the following features of size or extension: >3 cm at its largest point, involving the main bronchus and >=2 cm distal to the carina invading the visceral pleura, associated with atelectasis or obstructive pneumonitis extending to the hilar region but not involving the entire lung
T3	A tumour of any size directly invading any of the following: chest wall (including superior sulcus tumours), diaphragm, mediastinal pleura, parietal pericardium, or, tumour in the main bronchus <2 cm distal to the carina but without involvement of the carina, or, associated atelectasis or obstructive pneumonitis of the entire lung
T4	A tumour of any size invading any of the following: mediastinum, heart, great vessels, trachea, oesophagus, vertebral body, carina, or, separate tumour nodules in the same lobe, or, tumour with a malignant pleural effusion.

N criteria

N	Description
NX	Regional lymph nodes cannot be assessed
N0	No regional lymph node metastasis
N1	Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes, and intrapulmonary nodes including involvement by direct extension of the primary tumour
N2	Metastasis to ipsilateral mediastinal and/or subcarinal lymph node(s)
N3	Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s)

Staging grouping

Stages	Description
Stade 0	Tis, N0, M0
Stage 1	T1 N0 M0 or T2 N0 M0
Stage 2	T1 N1 M0 or T2 N1 M0 (no mediastinal abnormality)
Stage 3a	Extra-pulmonary invasion (T3 N0 M0 or T3 N1 M0, T3 N2 M0, T2 N2 M0)
Stage 3b	T1-4 N3 M0, T4 and N1-3 M0
Stade 4	Remote metastases

Surgical stage grouping (according to G.L. Walsh)

Usual therapeutic results for non-small cell lung cancers:

Stages	Description	5 year survival
Stage 1a	T1 N0 M0	> 70%
Stage 1b	T2 N0 M0	60%
Stage 2a	T1 N1 M0	50%
Stage 2b	T2 N1 M0	30-40%
Stage 2b	T3, N0-N1,M0	30-40%
Stage 3a	T3 N0 M0 or T3 N1 M0, T3 N2 M0, T2 N2 M0	10-30%
Stage 3b	T1-4 N3 M0, T4 and N1-3 M0	< 10%
Stage 4	Remote metastases	< 5%

Small cell lung cancer is a specific histological entity. The website of the Lorraine Cancer Network, Oncolor describes decision trees for this pathology.

Another interesting page for lung pathology is the website Webpath from Florida University.
Of course, the NCI has published two interesting series of pages : for non-small cell lung cancer and for small-cell lung cancer.

Breast cancer:

The following links give access to diagrams of the various stages of breast cancer.

T1a  breastcancer T1b breast cancer T1c breast cancancer	N1 breast cancer N2a breast cancer N2b breast cancer
T2 breast cancer T3 breast cancer T4a breast cancer T4b breast cancer	N3a breast cancer N3b breast cancer N3c breast cancer

T Classification

Classification	Description
TX	Primary tumour cannot be assessed
T0	No evidence of primary tumour
Tis	In situ tumour
T1	Tumour <= 2cm
T1mic T1a T1b T1c	Microinvasion <=0.1 cm at its largest point Tumour >0.1 cm but <=0.5 cm at its largest point Tumour >0.5 cm but <=1.0 cm at its largest point Tumour >1.0 cm but <=2.0 cm at its largest point
T2	Tumour >2.0 cm but <=5.0 cm at its largest point
T3	Tumour >5.0 cm at its largest point
T4	Tumour of any size with direct extension to (a) chest wall or (b) skin
T4a T4b T4c T4d	Extension to chest wall, not including pectoralis muscle Oedema (including peau d’orange) or ulceration of the breast skin, or satellite skin nodules confined to the same breast Both T4a and T4b Inflammatory carcinoma

Classification N

This is a very complicated classification which has changed many times.

Classification	Description
NX	Regional lymph nodes cannot be assessed
N0	No regional lymph node metastasis
N1	Metastasis to movable ipsilateral axillary lymph node(s)
N2a	Metastasis in ipsilateral axillary lymph nodes fixed to one another (matted) or to other structures
N2b	Metastasis only in clinically apparent ipsilateral internal mammary nodes and in the absence of clinically evident axillary lymph node metastasis
N3a	Metastasis in ipsilateral infraclavicular lymph node(s)
N3b	Metastasis in ipsilateral internal mammary lymph node(s) and axillary lymph node(s)
N3c	Metastasis in ipsilateral supraclavicular lymph node(s)

Usual clinical results

In classical forms (except in inflammatory diseases), the T value is of relatively little consequence for 5 year survival (unless there is thoracic wall invasion) except that, generally speaking, the higher the T value, the greater the risk of having positive nodes.

For this indirect reason, here are the results of our Centre François Baclesse (from 1989 to 1999 - Work by Thierry Delozier - Hubert Crouet). The time scale is in months.

Data from Centre François Baclesse

However, the T value is very important for surgical treatment (since beyond T2 it becomes very difficult to save the breast due to the size of the tumour).

The N classification is very important for prognosis and justifies the numerous clinical trials (noticeably the intensification in chemotherapy) aimed at increasing survival. Here are the results of disease free survival observed by our team.

('1 to 3 positive nodes' describes stage pN1b; '4 to 9 positive nodes' describes stage pN2a; '10 and above positive nodes' describes stage pN3).

Data from Centre François Baclesse

Below is a classical representation of the importance of node status on 5 year survival (according to the number of positive nodes after surgery).
For French speaking readers, the website of Lorraine cancer network Oncolor describes breast carcinoma, its classification and decision trees according to the various and complex situations observed.
For English speaking readers, the NCI website also gives a lot of information.

 Colon cancer:

Many classification systems are in use, although they are in fact very similar and concern the importance of invasion of the colic wall and proximal nodes. Generally, classification is done after surgery and the study of the surgical specimen.

With the following links, you will find diagrams of the main stages:

T1 colon cancer T2 colon cancer T3 colon cancer T4 colon cancer

N1 colon cancer N2 colon cancer

The following TNM Classification is the basis for stages which seem closer to the clinical reality.

T Criterion

T	Description
TX	Primary tumour cannot be assessed
T0	No evidence of primary tumour
Tis	Carcinoma in situ: intraepithelial or invasion of the lamina propria
T1	Tumour invading submucosa
T2	Tumour invading muscularis propria
T3	Tumour invading through the muscularis propria into the subserosa, or into nonperitonealised pericolic or perirectal tissues
T4	Tumour directly invading other organs or structures, and/or perforating visceral peritoneum

N Criterion

N	Description
NX	Regional nodes cannot be assessed
N0	No regional lymph node metastasis
N1	Metastasis in 1 to 3 regional lymph nodes
N2	Metastasis in 4 or more regional lymph nodes

The American Joint Committee for Cancer Classification has proposed the following grouping of TNM stages:

Stages	Groups
Stage 0	T0, Tis, N0, M0
Stage 1A Stage 1B	T1, N0, M0 T2, N0, M0
Stage 2A Stage 2B	T3, N0, M0 T4, N0, M0
Stage 3A Stage 3B Stage 3 C	T1-T2, N1, M0 T3-T4, N1, M0 T1-T4, N2, M0
Stage 4	T1-T4, N1-N3, M1

In many studies (retrospective and long term studies), the Dukes classification is still in use (it was proposed before TNM stages by Dukes CE:Cancer of the rectum: ana analysis of 1,000 cases. J Pathol Bacteriol 50:527-539, 1940)

Dukes	Description
A	T1-T2, N0, M0
B	T3, N0, M0
C1	T1-T3,N1,M0
C2	T4,N0, M0
D	T1-T4, N1-N3, M1

Astler-Coller classification (or modified Dukes classification) 

 This is another post surgical classification (Astler VB, Coller FA: The prognostic significance of direct extension of carcinoma of the colon and rectum. Ann surg 139:846-852, 1954)

Astler-Coller	Description
A	T1, N0, M0
B	T2, N0, M0
B2	T3, N0,M0
C1	T1-T2, N1-N2, M0
C2	T3, N1-N2, M0
D	T1-T4, N1-N3, M1

Usual therapeutic results

These results are generally observed and are issued from compilations

Classification	Survival at 5 years
T1, N0, M0	> 95 %
T2,N0,M0	90 %
T3, N0, M0	> 75 %
T4, N0, M0	> 60 %
Duke’s A	82 %
Duke’s B	73 %

Relative risk of loco-regional relapse or metastases according to stage and localisation

Classification	Risk of relapse
Colon (n = 479)	28 %
Rectum (n = 430)	30 %
Dukes A (n = 81)	6 %
Dukes B (n = 558)	31 %
Dukes C (n = 270)	59 %

according the study by Adloff JP, Chirurgie, 1989, 115, 228-236 .

For Dukes C stages, it has been confirmed that adjuvant chemotherapy improves the prognosis. For stage A cancers(82% of cases), only surgery is needed. For Stage B, trial results are still uncertain.

Cervix Uteri Cancer:

One of the most frequently used classifications is the classification by the “Fédération Internationale de Gynécologie Obstétrique” (FIGO) which has allowed comparisons between hospital series for more than 40 years.

The following drawings are available:

Stage 1a Stage 1b Stages 2a and 2b

Stages 3a and 3b Stage 4 Node metastases

The following table compares the TNM and FIGO classifications.

In order to obtain an accurate clinical classification, the FIGO organisation recommends a gynaecological examination under general anaesthesia.

In the past, a pelvic lymphographia was recommended to determine node metastases; it has now been replaced by abdominopelvic CT scan. This examination reveals any ureteral dilatation in relation to a compression by parametrial tumour invasion, as well as kidney malfunction.

TNM	FIGO	Description
Tx		Primary tumour cannot be assessed
T0		No evidence of primary tumour
Tis	St 0	Carcinoma in situ
T1 T1a T1a1 T1a2 T1b T1b1 T1b2	St I Ia Ia1 Ia2 Ib 1b1 1b2	Cervical carcinoma confined to cervix uteri Invasive carcinoma diagnosed only by microscopy. Measured stromal invasion 3 mm or less in depth and 7 mm or less in horizontal spread Measured stromal invasion more than 3 mm and no more than 5 mm with a horizontal spread of 7 mm or less Clinically visible lesion confined to the cervix Clinically visible lesion 4 cm or less at largest point Clinically visible lesion more than 4 cm at largest point
T2 T2a T2b	St II IIa IIb	Cervical carcinoma invading beyond cervix uteri but not to pelvic wall or to the lower third of the vagina Tumour without parametrial involvement Tumour with parametrial involvement
T3 T3a T3b	St III IIIa IIIb	Tumour extending to the pelvic wall and/or involving the lower third of the vagina, and/or causing hydronephrosis or kidney malfunction. Tumour involving lower third of the vagina, no extension to pelvic wall Tumour extending to pelvic wall and/or causing hydronephrosis or kidney malfunction.
T4	St IVa	Tumour invading mucosa of the bladder or rectum, and/or extending beyond true pelvis
M1	St IVb	Distant metastasis

N Criteria

N	Description
NX	Regional nodes cannot be assessed
N0	No regional lymph node metastasis
N1	Metastasis to regional lymph nodes

Generally observed results

These results are issued from a recent publication by FIGO on 32,500 patients treated around the world. (Annual report on the results of treatment in gynecological cancer. Twenty-first volume. Statements of results obtained in patients treated from 1982 to 1986, inclusive 3 and 5-year survival up to 1990.

% of patients	Stage	5 year survival
38%	T1	82%
32%	T2	62%
26%	T3	37%
4%	T4	12%

As soon as T2, the prognosis of cervix uteri cancer is limited. Such forms would never be observed if efficient screening programs were set up.

The following table displays the extent to which node invasion is detrimental to prognosis.

Stage	5 year survival
T1a	99 %
T1b, N0	90 %
T1b, N1	60 %
T2b, N0	85 %
T2b, N1	49 %

For French speaking readers, the Lorraine oncology network website Oncolor is very interesting and describes classification and decision trees.

Corpus Uteri Cancer Classification:

One of the most frequently used classifications is the classification by the Fédération Internationale de Gynécologie Obstétrique (FIGO) which has allowed comparisons between hospital series for more than 40 years.

The following drawings are available:

stage 1a, stage1b , stage 1c, stage 2a, stage 2b

stage 3a, stage 3b, stage 3c stage 4

FIGO and TNM Classification

This classification requires surgery results and the study of surgical specimens to specify the invasion (when colpo-hysterectomy is feasible).

TNM	FIGO	Description
Tx		Primary tumour cannot be assessed
T0		No evidence of primary tumour
Tis	St 0	Carcinoma in situ
T1 T1a T1b T1c	St I Ia Ib Ic	Tumour limited to corpus uteri tumour limited to endometrium invasion of less than one half of the myometrium. invasion greater than one half of the myometrium.
T2 T2a T2b	St II IIa IIb	Invasion of cervix uteri endocervical glandular involvement only. cervical stromal invasion.
T3 and/or N1 T3a T3b N1	St III IIIa IIIb IIIc	Tumour confined to the true pelvis tumour invades serosa and/or adnexa and/or positive peritoneal cytology vaginal metastases metastases to pelvic and/or para-aortic lymph nodes
T4	St IVa	Tumour invasion of bladder and/or bowel mucosa.
M1	St IV B	Distant metastases, including intra-abdominal and/or inguinal lymph nodes.

Non-operated patients

For non-operated patients, the classification is very similar. However for stage 2 cancers there are two classifications 2a for tumours situated in uteri within a small cavity (as observed on hysterography) and stage 2b for those situated within large uterine cavity.

Therapeutic results

According to FIGO publications: (Annual report on the results of treatment in gynaecological cancer. Twenty-first volume. Statements of results obtained in patients treated in 1982 to 1986, inclusive 3 and 5-year survival up to 1990.

Stage	5 year survival
St 1	90 - 95 %
St 2	78 - 85 %
St 3	50 - 60 %
St 4	0 - 25 %

For French speaking readers, the website of the Lorraine oncology network, Oncolor describes the classification of endometrial adenocarcinoma as well as decision trees.

Ovarian Carcinoma:

The TNM classification is not very well adapted to ovarian carcinoma due to the major role of surgery in obtaining an acurate classification. The FIGO (Fédération Internationale de Gynécologie et Obstétrique) classification is preferred.

The following drawings have been made to illustrate the various stages: 

stage 1a stage 1b stage 1c stage 2a stage 2b stage 2c

stage 3a stage 3b stage 3c - peritoneal invasion stage 3c - node invasion

The following table compares the two classification systems: a complete laparotomy and systematic histological samplings are mandatory for complete classification. Any incomplete surgery makes classification difficult, and for this reason (for instance when the lesion is apparently limited to one ovary but when no systematic samplings have been made), a new laparotomy is mandatory in order to confirm the localised characteristics of a tumour, thus avoiding unnecessary systematic chemotherapy.

TNM	FIGO	Description
Tx		Primary tumour cannot be assessed
T0		No evidence of primary ovarian tumour
T1 T1a T1b T1c	Stage I Ia Ib Ic	Tumour limited to ovary(ies) Tumour limited to 1 ovary; capsule intact, no tumour on ovarian surface. No malignant cells in ascites or peritoneal washings. Tumour limited to both ovaries; capsules intact, no tumour on ovarian surface. No malignant cells in ascites or peritoneal washings. Tumour limited to 1 or both ovaries with any of the following: capsule ruptured, tumour on ovarian surface, malignant cells in ascites or peritoneal washings.
T2 T2a T2b T2c	St II IIa IIb IIc	Tumour limited to pelvis Extension and/or implants on the uterus and/or fallopian tubes. No malignant cells in ascites or peritoneal washings. Extension to and/or implants on other pelvic tissues. No malignant cells in ascites or peritoneal washings. Pelvic extension and/or implants (stage IIA or stage IIB) with malignant cells in ascites or peritoneal washings.
T3 T3a T3b T3c or N1	St III IIIa IIIb IIIc-p et IIIc-g	Tumour limited to the abdominal cavity Microscopic peritoneal metastasis beyond pelvis (no macroscopic tumour). Macroscopic peritoneal metastasis beyond pelvis <=2 cm at largest diameter Peritoneal metastasis beyond pelvis >2 cm at largest diameter and/or Regional lymph node metastasis.
M1	St IV	Remote metastases Tumour involving 1 or both ovaries with distant metastasis. If pleural effusion is present, positive cytological test results must exist to designate a stage IV case. Parenchymal liver metastasis is equivalent to stage IV.

Therapeutic results

Here are the results of FIGO registries concerning 11,600 patients treated between 1982 and 1986. Therefore, we cannot evaluate possible recent progress (such as better use of platinum or paclitaxel). (Annual report on the results of treatment in gynecological cancer. Twenty-first volume. Statements of results obtained in patients treated in 1982 to 1986, inclusive 3 and 5-year survival up to 1990.

% of patients	FIGO stage	5 year Survival
27%	St I	79%
13%	St II	57%
20%	St III NP	22%
3%	St III a	49%
5%	St III b	33%
17%	St III c	19%
15%	St IV	8%

Please note the important figure of stage III NP which represents cases for which details on tumour mass size during laparotomy were not recorded. The following table shows the importance of complete and quality surgery since the size of residual disease is inversely correlated to survival.

Tumour residues	Stage 3a	Stage 3b	Stage 3c
Complete surgery	83%	60%	47%
Residues <= 2cm	50%	38%	30%
Residues > 2cm	20%	25%	15%

For French speaking readers, the website of Lorraine oncology network, Oncolor is a very interesting website which describes tumour classification as well decision trees. 

Prostate cancer:

The following drawings are available as diagrams illustrating the various stages:

Prostate Cancer T1a Prostate Cancer T1b Prostate Cancer T1c Prostate Cancer T2a Prostate Cancer T2b Prostate Cancer T2c

Prostate Cancer T3a Prostate Cancer T3b Prostate Cancer T4 Prostate Cancer N1

T Criterion

Classification	Description
TX	Primary tumour cannot be assessed
T0	No evidence of primary tumour
T1 T1a T1b T1c	Clinically imperceptible tumour neither palpable nor visible by imaging: Tumour incidental histological finding in <=5% of resected tissue Tumour incidental histological finding in >5% of resected tissue Tumour identified by needle biopsy (e.g., because of elevated PSA)
T2 T2a T2b T2c	Tumour confined to prostate Tumour involving 50% of 1 lobe or less Tumour involving >50% of only one lobe Tumour involving both lobes
T3 T3a T3b	Tumour extending through the prostate capsule Extracapsular extension (unilateral or bilateral) Tumour invading seminal vesicle(s)
T4 T4	Tumour is fixed or invades adjacent structures Tumour is fixed or invades adjacent structures other than seminal vesicles: bladder neck, external sphincter, rectum, levator muscles, and/or pelvic wall

N Criterion

Classification	Description
NX	Regional lymph nodes were not assessed
N0	No regional lymph node metastases
N1	Metastases in regional lymph node(s)

Histological classification

As well as this TNM classification, we also study the Gleason classification which describes histological grading.

Here are some examples of the Gleason classification:

• Grade 1
• Grade 2
• Grade 3
• Grade 4
• Grade 5

In fact, as many tumours, prostate cancer is very heterogeneous in its differentiation grade, from a very well differentiated grade to a totally undifferentiated grade. The latter grades are the most dangerous since they rapidly become hormone-independent and quickly give birth to remote metastases. Gleason adds two of the tumour's components (from 1 to 5):

• The most represented component of the tumour
• The most indefferentiated component of the tumour (which can be the same as the previous one).

So we speak of a Gleason tumour Grade (3 + 1) or (4 + 3), and so on..

Therapeutic results

The TNM classification has a major role in the survival of patients with prostate cancer:

	Survival
	at 5 years	at 10 years	at 15 years
T1	85%	65%	40%
T2	83%	55%	35%
T3	68%	38%	20%
T4	< 20%	< 5%	-

These figures are very interesting when correlating to the age of patients:

• at 65, half of the men will still live16 years,
• at 80, half of the men will still live 7 years.

Therefore treatment should be adapted to the stage and the age of the patient. Older patients with a limited T1 could benefit from watchful waiting and only be treated when major clinical symptoms appear, whereas young patients (65 and under) with more advanced disease should be actively treated since cancer will probably severely shorten their lifespan.

Among other important prognostic factors, are:

• the histological Gleason grading
• the initial PSA level.
  

Partin elaborated tables in order to determine if cancer goes beyond prostate limits according to T, PSA level and Gleason grading. The higher the PSA level, the more elevated the Gleason grade, and the greater the risk of the tumour going beyond prostate limits and quickly giving birth to node or remote metastases.
The graph below shows the 5 year survival according to Gleason Grade and T stage.

In this older graph, T2c stage is in reality the currently used T2b stage, the T2b is a bilateral T2a, and T2a is a unilateral T2a.

The knowledge of the risk of capsule invasion is very important for the indication of therapy. A prostate tumour extending beyond the capsule should not be operated (the limits of exeresis will never be free of disease) but can possibly be treated by radiotherapy. As from this stage (and higher), the risk of node and remote metastases is very high and trials have shown that the adjuvant use of hormone therapy (surgical or chemical castration, anti-androgens) increases patient survival.

Urinary bladder cancer:


The following diagrams can be consulted:

Urinary bladder cancerTis Urinary bladder cancer Ta Urinary bladder cancer T1 Urinary bladder cancer T2a Urinary bladder cancer T2b Urinary bladder cancer T3a Urinary bladder cancer T3b

Urinary bladder cancer T4a Urinary bladder cancer T4b Urinary bladder cancer N1 Urinary bladder cancer N2 Urinary bladder cancer N3

The clinical staging of bladder carcinoma is determined by the depth of invasion of the bladder wall by the tumour. This determination requires a cystoscopic examination that includes a biopsy, and examination under general anaesthesia to assess the size and mobility of palpable masses, the degree of induration of the bladder wall, and the presence of extravesical extension or invasion of adjacent organs

T Criterion

Classification	Description
TX	Primary tumour cannot be assessed
T0	No evidence of primary tumour
Tis	Carcinoma in situ (i.e., flat tumour)
Ta	Noninvasive papillary carcinoma
T1	Tumour invading subepithelial connective tissue
T2	Tumour invading bladder muscle
T2a	Tumour invading superficial muscle (inner half)
T2b	Tumour invading deep muscle (outer half)
T3	Tumour invading perivesical fat tissue
T3a	Microscopically
T3b	Macroscopically (extravesical mass)
T4	Tumour invading neighbouring organs
T4a	Tumour invading the prostate, uterus, vagina
T4b	Tumour invadin the pelvic wall, abdominal wall

N Criterion

Classification	Description
NX	Regional lymph nodes cannot be assessed
N0	No regional lymph node metastasis
N1	Metastasis in a single lymph node, <=2 cm at largest point
N2	Metastasis in a single lymph node, >2 cm but <=5 cm at largest point; or multiple lymph nodes, <=5 cm at largest point
N3	Metastasis in a lymph node, >5 cm at largest point

Therapeutic results

Stage	5 year survival
Tis	> 90%
T1	75%
T2	60%
T3a	50%
T3b	15%
N+	<5%
M1	< 5%