Staging of oesophageal and gastric cancer

Acknowledgement

This chapter in the sixth edition was written by Graeme Couper and we are grateful to him for those parts of the chapter which we have kept in this edition.

Introduction

Survival after a diagnosis of gastric and oesophageal cancer is highly variable, even with curative treatment. The need for accurate staging is no longer limited to identifying patients who may be cured and those for whom cure is not possible, but provides material for a more informed discussion about treatment options through stratification of predicted outcomes and measurement of factors that influence treatment decisions. Taken together, this information can therefore help inform patients’ decisions regarding the treatment of their disease and manage the expectations of treatment outcomes. Increasing staging accuracy has resulted in a more complex staging process, with multiple possible investigations and results requiring a range of expertise to perform and interpret. In the context of a cancer diagnosis, these investigations should happen expediently, which has in part driven regional specialisation to higher volume centres and standardisation of reporting so that outcomes can be compared on a like-for-like basis nationally and internationally.

Staging classifications

Patients should be staged according to the TNM (tumour–node–metastasis) staging system proposed by Pierre Denoix in 1946. T represents the extent of the primary tumour, N the extent of regional lymph node involvement, and M the presence or absence of distant metastases. A single TNM staging classification was agreed between the American Joint Committee on Cancer (AJCC), the Japanese Joint Committee (JJC), and the International Union Against Cancer (UICC) in 1987, and multiple subsequent editions have been agreed. Currently, the Royal College of Pathologists use the American Joint Committee on Cancer TNM eighth edition (TNM8) classification as the reporting standard for gastric and oesophageal carcinoma, supported by the Association of Upper Gastrointestinal Surgeons. Accordingly, TNM8 is used throughout this chapter.

The eighth and most recent edition of the classification of gastric and oesophageal cancer was published in 2016 and takes into account advances in staging, diagnosis, and treatment that occurred since the last edition in 2009. The eighth edition of TNM is used in the UK as standard practice.

The prefixes of the TNM system are important to understand, especially for oesophagogastric cancer in which patients have frequently undergone neoadjuvant treatment.

c: clinical
p: pathological
yc: post neoadjuvant (radiation or systemic) therapy – clinical
yp: post neoadjuvant (radiation or systemic) therapy – pathological
a: autopsy

All oesophagogastric cancer should be confirmed histologically. Then, findings on clinical examination, imaging, direct luminal inspection, or laparoscopic inspection can be assessed and a clinical TNM staging assigned, which is designated with a ‘c’ prefix and is used to determine suitability for particular treatment options. The ‘p’ prefix refers to all the information from the clinical classification combined with the information gained from histopathological analysis. Pathological grading is more accurate in predicting prognosis and is therefore considered when planning adjuvant treatment. Discrepancy between cTNM and pTNM is common, ^, particularly in gastric cancer staging, and can therefore help evaluate the accuracy of staging methodologies. The ‘y’ prefix relates to patients who have undergone neoadjuvant treatment, and therefore acknowledges that the previous staging classification may have changed. TNM classifications are grouped into disease stages, and once the stage is established it should remain consistent in the medical records to allow for the selection and assessment of therapeutic interventions, and later audit. Final pathological stage grouping also provides the most accurate assessment of prognosis. In cases of doubt, where two classifications are given for T or N, then the lowest category is chosen. In situations of single-organ synchronous tumours, the TNM classification is based on the most advanced tumour, with either the number of tumours or ‘m’ recorded in parenthesis after the T stage. The prefix ‘r’ can be added in cases of recurrent cancer. The recent HERO consensus recommendations support the use of the Royal College of Pathologists Guidelines as a minimum dataset and provide explicit guidance on sample preparation and assessment, which will help benchmark pathological staging after oesophagogastric resection.

Oesophageal cancer staging

As described earlier, adequate and accurate staging of oesophageal cancer is crucial in planning treatment and to avoid unnecessary surgery, considering the negative impact resectional surgery has on an individual’s subsequent quality of life. Clinical assessment is an important early step in assessment, since some patients may require nutritional support, clearly lack the physiological reserve for major surgery, or do not want radical treatment, and recognising this from the outset is kinder and allows a more efficient delivery of healthcare.

Currently in the UK, core staging investigations include endoscopy with biopsy, multidetector contrast-enhanced computed tomography (CT), endoscopic ultrasound (EUS) with or without fine-needle aspiration (FNA) of suspicious lymph nodes, and positron emission tomography combined with CT (PET/CT). Laparoscopy is also commonly used in patients with infradiaphragmatic oesophagogastric tumours. Other staging investigations that may be used in selected patients, usually to investigate further findings identified in the core staging investigations, are magnetic resonance imaging (MRI), ultrasound (US) imaging with or without FNA, endoscopic mucosal resection (EMR) for T1 lesions, endobronchial US (EBUS), and laparoscopic US.

The T, N, and M criteria for oesophageal cancer are listed in Table 3.1 . In cases where a category cannot be established this is denoted with an ‘x’, with the exception of the M stage, since this is a binary output. The categories are the same for adenocarcinoma and squamous cell carcinomas. The T, N, and M criteria are required to establish the clinical stage grouping, as detailed in Table 3.2 . In addition to the T, N, and M stages, TNM8 pathological stage groupings also include assessment of oesophageal cancer grade and location, with the latter relevant only to squamous cell carcinoma.

Table 3.1

TNM8 cancer staging categories for oesophageal cancer

Category	Criteria
T category
TX	Tumour cannot be assessed
T0	No evidence of primary tumour
Tis	High-grade dysplasia, defined as malignant cells confined by the basement membrane
T1	Tumour invades the lamina propria, muscularis mucosae, or submucosa
T1a	Tumour invades the lamina propria or muscularis mucosae
T1b	Tumour invades the submucosa
T2	Tumour invades the muscularis propria
T3	Tumour invades the adventitia
T4	Tumour invades the adjacent structures
T4a	Tumour invades the pleura, pericardium, azygos vein, diaphragm, or peritoneum
T4b	Tumour invades other adjacent structures, such as the aorta, vertebral body, or trachea
N category
NX	Regional lymph nodes cannot be assessed
N0	No regional lymph node metastasis
N1	Metastasis in 1–2 regional lymph nodes
N2	Metastasis in 3–6 regional lymph nodes
N3	Metastasis in 7 or more regional lymph nodes
M category
M0	No distant metastasis
M1	Distant metastasis

Table 3.2

TNM8 clinical stage groupings for oesophageal cancer

Clinical stage group	cT	cN	cM
Squamous cell carcinoma
0	Tis	N0	M0
I	T1	N0–1	M0
II	T2	N0–1	M0
	T3	N0	M0
III	T3	N1	M0
	T1–3	N2	M0
IVA	T4	N0–2	M0
	T1–4	N3	M0
IVB	T1–4	N0–3	M1
Adenocarcinoma
0	Tis	N0	M0
I	T1	N0	M0
IIA	T1	N1	M0
IIB	T2	N0	M0
III	T2	N1	M0
	T3–4a	N0–1	M0
IVA	T1–4a	N2	M0
	T4b	N0–2	M0
	T1–4	N3	M0
IVB	T1–4	N0–3	M1

T stage

TNM8 includes T1 subcategories T1a and T1b to reflect the increasing role of endoscopic resection in early cancer, and the prognostic importance of submucosal invasion. T2 is invasion into the muscularis propria, and T3 is invasion into the oesophageal adventitia. In T4 oesophageal cancers, TNM8 considers the azygos vein, pleura, diaphragm, pericardium, and peritoneum within the T4a subcategory. T4b structures are the major blood vessels, vertebra, adjacent organs, and airways.

N stage

Confident N staging requires examination of at least seven lymph nodes, but in recognition that increasing lymph node yields are associated with improved survival, the National Oesophago-Gastric Cancer Audit (NOGCA) recommends a yield of at least 15 nodes, which is helpfully consistent with the standards of lymphadenectomy in gastric cancer. In cases where fewer than seven lymph nodes are examined and found to be clear of cancer the staging would still be reported as pN0, but the possibility of understaging should be considered in treatment planning. Regional lymph nodes are confined to those within the oesophageal drainage zones – paratracheal, subcarinal, pulmonary ligament, paraoesophageal, diaphragmatic, paracardial, left gastric, common hepatic, splenic, celiac, and cervical paraesophageal level VI/VII nodes. Lymph nodes in other distant anatomical locations, such as supraclavicular nodes, are considered as metastatic disease.

M stage

M staging refers to the presence of distant disease, including disease in non-regional lymph nodes. In TNM8, inability to establish M stage is regarded as M0, rather than Mx as in earlier editions of TNM.

Grade

Tumour grade is reported using the standardised WHO histological grading system and is subdivided into well differentiated (G1), moderately differentiated (G2), and poorly differentiated (G3) ( Table 3.3 ). Grade reporting is of particular importance in both early cancers, when planning perioperative treatment or anticipating the risk of recurrence after endoscopic treatment, and in advanced cancers, since G3 lesions or those with signet ring morphology have a particularly poor prognosis. Consequently, if the lesion is deemed G3 or signet ring type from the original diagnostic biopsy this should be recorded, since it may be difficult to assess later after neoadjuvant treatment. Poor inter-observer reliability in the pathological reporting of tumour grade and dysplasia is well known, ^, with one study showing that TNM7 staging was altered in 21% of patients with T1 or T2 node-negative oesophageal cancer. Therefore, TNM8 only includes grade in the pTNM rather than cTNM stage groupings, thus avoiding reliance on endoscopic sampling to establish grade.

Table 3.3

TNM8 pathological grade of oesophageal cancer

Adenocarcinoma
GX	Differentiation cannot be assessed
G1	Well differentiated. > 95% of tumour is composed of well-formed glands
G2	Moderately differentiated. 50–95% of tumour shows gland formation
G3 ^†	Poorly differentiated. Tumours composed of nests and sheets of cells, with <50% of tumours demonstrating glandular formation
Squamous cell carcinoma
GX	Differentiation cannot be assessed
G1	Well differentiated. Prominent keratinization with pearl formation and a minor component of nonkeratinizing basal-like cells. Tumour cells are arranged in sheets, and mitotic counts are low
G2	Moderately differentiated. Variable histologic features, ranging from parakeratotic to poorly keratinizing lesions. Generally, pearl formation is absent
G3 ^‡	Poorly differentiated. Consists predominantly of basal-like cells forming large and small nests with frequent central necrosis. The nests consist of sheets or pavement-like arrangements of tumour cells, and are occasionally punctuated by small numbers of parakeratotic or keratinizing cells

† If further testing of “undifferentiated” cancers reveals a glandular component, categorise as adenocarcinoma G3.

‡ If further testing of “undifferentiated” cancers reveals a squamous cell component, or if after further testing they remain undifferentiated, categorise as squamous cell carcinoma G3.

Location

According to TNM8 definitions, oesophageal cancers occur throughout the length of the oesophagus and the proximal 2 cm of the stomach. Proximal overlap of cancer into the pharynx is rarely of surgical importance since the primary treatment is usually chemoradiotherapy, but distal overlap of oesophageal cancer into the stomach can alter the surgical approach. In the case of oesophageal squamous cell carcinoma (OSCC) tumour location influences the stage, and TNM8 specifies the cancer epicentre as the reference point and uses common radiological landmarks to define anatomical subsites ( Box 3.1 ), whereas in adenocarcinoma most tumours are distal.

Box 3.1

TNM8 anatomical sites for location staging of oesophageal squamous cell carcinoma

Upper: Cervical oesophagus to lower border of azygos vein
Middle: Lower border of azygos vein to lower border of inferior pulmonary vein
Lower: Lower border of inferior pulmonary vein to stomach, including the oesophagogastric junction

For surgical planning, further anatomical descriptions of tumour location, assessed by imaging and endoscopically, are often used in multidisciplinary team (MDT) discussions since these help to plan the surgical approach and suitability for radiotherapy. It is useful to include these anatomical definitions here, to recognise that they differ from the TNM8 anatomical subsite descriptions and avoid confusion when terms may be used interchangeably in MDT:

Cervical oesophagus: lower border of the cricoid to thoracic inlet, marked at the suprasternal notch and approximately 18 cm from incisors.
Upper thoracic oesophagus: thoracic inlet to tracheal bifurcation (24 cm from the incisors).
Middle thoracic oesophagus: tracheal bifurcation to the halfway point between tracheal bifurcation and the oesophagogastric junction (32 cm from the incisors).
Lower thoracic oesophagus: the halfway point between tracheal bifurcation and the oesophagogastric junction to the oesophagogastric junction (40 cm from the incisors).
Oesophagogastric junction.

The Siewert-Stein classification is used for oesophagogastric junctional tumours, with the tumour epicentre defining the type, separating them into type I (1–5 cm above the oesophagogastric junction), type II (from 1 cm above the oesophagogastric junction to 2 cm below it), and type III (more than 2 cm below the oesophagogastric junction). Since TNM8, type III oesophagogastric junction cancers are classified and staged as gastric cancers.

The pathological stage groupings for oesophageal adenocarcinoma (OAC) and OSCC are different to reflect that the prognosis of early OSCC is worse than in early OAC. ^, While data supporting survival differences for OSCC in different locations is mixed, ^, TNM7 has recognised tumour location as a prognostic factor and it remains in pathological stage groupings for TNM8 ( Table 3.4 ). Arguably, the pure pTNM stage groups are losing relevance for oesophageal cancer due to the widespread use of neoadjuvant treatment with alternative prognostic factors such as tumour regression and complete pathological response becoming increasingly important. ^, This has led to a new stage grouping specific to post neoadjuvant patients in TNM8, with the same groupings for oesophageal squamous cell and adenocarcinoma ( Table 3.5 ).

In 2 045 and 5 686 patients with OSCC and adenocarcinoma, respectively, there was no residual cancer (ypT0N0M0) in 25% of squamous cell carcinomas and 13% of adenocarcinomas treated with neoadjuvant therapy, prompting the need for specific post neoadjuvant stage groupings.

Table 3.4

TNM8 pathological stage groupings for oesophageal cancer

Pathological stage group	pT	pN	pM	pGrade	pLocation
Squamous cell carcinoma
0	Tis	N0	M0	N/A	Any
IA	T1a	N0	M0	G1, X	Any
IB	T1b	N0	M0	G1, X	Any
	T1	N0	M0	G2–3	Any
	T2	N0	M0	G1	Any
IIA	T2	N0	M0	G2–3, X	Any
	T3	N0	M0	Any	Lower
	T3	N0	M0	G1	Upper/middle
IIB	T3	N0	M0	G2–3	Upper/middle
	T3	N0	M0	X	Any
	T3	N0	M0	Any	X
	T1	N1	M0	Any	Any
IIIA	T1	N2	M0	Any	Any
	T2	N1	M0	Any	Any
IIIB	T4a	N0–1	M0	Any	Any
	T3	N1	M0	Any	Any
	T2–3	N2	M0	Any	Any
IVA	T4a	N2	M0	Any	Any
	T4b	N0–2	M0	Any	Any
	T1–4	N3	M0	Any	Any
IVB	T1–4	N0–3	M1	Any	Any
Adenocarcinoma
0	Tis	N0	M0	N/A
IA	T1a	N0	M0	G1, X
IB	T1a	N0	M0	G2
	T1b	N0	M0	G1–2, X
IC	T1	N0	M0	G3
	T2	N0	M0	G1–2
IIA	T2	N0	M0	G3, X
IIB	T1	N1	M0	Any
	T3	N0	M0	Any
IIIA	T1	N2	M0	Any
	T2	N1	M0	Any
IIIB	T4a	N0–1	M0	Any
	T3	N1	M0	Any
	T2–3	N2	M0	Any
IVA	T4a	N2	M0	Any
	T4b	N0–2	M0	Any
	T1–4	N3	M0	Any
	T1–4	N0–3	M1	Any

N/A , Not applicable; X, not defined.

Table 3.5

TNM8 post neoadjuvant therapy (ypTNM) stage groupings

ypStage group	ypT	ypN	ypM
I	T0–2	N0	M0
II	T3	N0	M0
IIIA	T0–2	N1	M0
IIIB	T4a	N0	M0
	T3	N1–2	M0
	T0–3	N2	M0
IVA	T4a	N1–2, X	M0
	T4b	N0–2	M0
	T1–4	N3	M0
IVB	T1–4	N0–3	M1

TNM8 also includes numerous other requirements for staging oesophageal cancer, which may influence clinical decisions in borderline cases and help guide therapeutic options. A complete list of these requirements is included in Box 3.2 .

Box 3.2

TNM8 registry data collection variables for oesophageal cancer

CT , Computed tomography; EUS , endoscopic ultrasound; FNA , fine-needle aspiration; HER2 , human epidermal growth factor receptor two; LVI , lymphovascular invasion; PET , positron emission tomography; ypTNM , post neoadjuvant (radiation or systemic) therapy – pathological tumour–node–metastasis staging.

Clinical staging modalities (endoscopy and biopsy, EUS, EUS/FNA, CT, PET/CT)
Tumour length
Depth of invasion
Number of nodes involved, clinical
Number of nodes involved, pathological
Location of nodal disease, clinical
Location of nodal disease, pathological
Sites of metastases, if applicable
Presence of skip lesions: T(m)
Perineural invasion
LVI (lymphatic, vascular, both)
Extranodal extension
Type of surgery
Chemotherapy
Chemoradiation therapy (for ypTNM)
Surgical margin (negative, microscopic, macroscopic)
HER2 status (positive or negative), adenocarcinoma only

Gastric cancer staging

As for oesophageal cancer, accurate gastric cancer staging is important to avoid unnecessary surgery, since outcomes from palliative gastrectomy cannot justify resection in most cases. International consensus on gastric cancer staging has been more challenging than in oesophageal cancer due to the different aetiology and anatomical distribution of gastric cancers in different parts of the world. ^, Unlike in oesophageal cancer, nearly all gastric cancers are adenocarcinomas and numerous histopathological classifications exist, ^, making globally unified subtyping difficult to apply. Consequently, there is no differentiation in the TNM staging groupings to recognise the clear variability in the biological behaviours of gastric cancers worldwide. Ultimately, anticipation of limited availability of the immunohistochemical and genetic assays which would be required to biologically subtype gastric cancers have restricted their inclusion in TNM8. In general, distal gastric cancers have a lower mortality and are more common in Asian countries, whereas proximal gastric cancers are more common in the West and have poorer survival. Whether very proximal lesions are staged as gastric or oesophageal cancers has evolved over the last iterations of TNM; with clinician’s judgement used in TNM6, all tumours with involvement of the oesophagogastric junction are staged as oesophageal in TNM7, to the current system in TNM8 where type III Siewert-Stein tumours (involving junction but tumour epicentre 2–5 cm below the oesophagogastric junction) are staged as gastric cancers. Note that gastric neuroendocrine tumours are also staged by TNM8. The TNM8 categories are shown in Table 3.6 .

Table 3.6

TNM8 cancer staging categories for gastric cancer

T category
TX	Primary tumour cannot be assessed
T0	No evidence of primary tumour
Tis	Carcinoma in situ: intraepithelial tumour without invasion of the lamina propria, high-grade dysplasia
T1	Tumour invades the lamina propria, muscularis mucosae, or submucosa
T1a	Tumour invades the lamina propria or muscularis mucosae
T1b	Tumour invades the submucosa
T2	Tumour invades the muscularis propria
T3	Tumour invades the subserosa
T4	Tumour perforates the serosa (visceral peritoneum) or invades the adjacent structures
T4a	Tumour perforates the serosa
T4b	Tumour invades the adjacent structures
N category
NX	Regional lymph nodes cannot be assessed
N0	No regional lymph node metastasis
N1	Metastasis in 1–2 regional lymph nodes
N2	Metastasis in 3–6 regional lymph nodes
N3a	Metastasis in 7–15 regional lymph nodes
N3b	Metastasis in 16 or more regional lymph nodes
M category
M0	No distant metastasis
M1	Distant metastasis

M1 includes positive peritoneal cytology, peritoneal seeding, and omental tumour not part of continuous extension.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here