Oesophageal & Gastric Cancer

Meet the patient

A 68-year-old man arrives in clinic with three months of progressive solid-food dysphagia — meat first, then rice, now bread — and eight kilograms of unplanned weight loss. He has smoked for forty years, drinks spirits daily, and was told years ago he had Barrett's oesophagus but never kept surveillance. He looks thinner than his photograph.[1]

The two questions that decide the rest of his care are the two that decide every upper-GI cancer: is this resectable? (CT, EUS, PET and staging laparoscopy answer that) and what is the histology? (biopsy answers that, and it sets the chemotherapy). Hold those two questions and the whole page slots into place.[1]

One tube, two cancers — split oesophageal disease by histology first

Oesophageal cancer is not one disease but two that happen to share a tube, and the histology decides everything downstream — site, geography, and which chemotherapy fits. Adenocarcinoma rises in the distal oesophagus and GOJ from Barrett's intestinal metaplasia, tracks the obesity and reflux epidemics, and predominates in overweight white men in their sixties. Squamous cell carcinoma rises in the upper and middle thirds, driven by alcohol, tobacco, very hot drinks, achalasia and nitrosamines, and dominates globally along the oesophageal cancer belt of Central and East Asia, East Africa and the Caspian.[1]

[1] [3]

The discriminator to carry into the viva: Barrett's plus obesity points to adenocarcinoma of the lower oesophagus; the smoking-and-alcohol belt points to squamous carcinoma of the upper and middle. Get the histology before you choose the chemotherapy.[1]

Gastric cancer is overwhelmingly adenocarcinoma (about 95 percent), the remainder being MALT and other lymphomas, GIST and neuroendocrine tumours. The Lauren classification splits it into an intestinal type (gland-forming, environment-driven, Correa cascade, older males, better prognosis) and a diffuse type (poorly cohesive signet-ring cells infiltrating the wall, CDH1 or E-cadherin loss, the rigid leather-bottle stomach of linitis plastica, worse prognosis).[2]

How common, who, and the levers that prevent it

Oesophageal cancer is the eighth most commonly diagnosed cancer and the sixth leading cause of cancer death worldwide; gastric cancer is the fifth diagnosed and the fourth deadliest. Together they cause more than 1.6 million new cases and 1.2 million deaths each year, and most of that burden is preventable.[3]

[3]

The risk-factor clusters sort cleanly by histology, and naming them earns the marks:[1]

• Oesophageal adenocarcinoma — Barrett's oesophagus (a 30- to 60-fold risk), chronic GORD (5- to 7-fold), central obesity, male sex, white ethnicity, age over 50, smoking.
• Oesophageal SCC — alcohol and tobacco together (5- to 10-fold), very hot beverages above about 65 deg C (IARC Group 2A), achalasia (up to 30-fold), caustic or lye injury, Plummer-Vinson syndrome, tylosis (Howel-Evans, near-certain SCC), nitrosamine-rich diet, HPV, low socioeconomic status, prior radiation, coeliac disease, head-and-neck field change.
• Gastric adenocarcinoma — H. pylori (IARC Group 1, Class I carcinogen; CagA-positive most virulent), atrophic gastritis and intestinal metaplasia, blood group A, salted and smoked and pickled diet low in fresh fruit, smoking (1.5- to 2-fold), family history and hereditary syndromes (CDH1 hereditary diffuse gastric cancer, Lynch, FAP, Peutz-Jeghers, BRCA2, Li-Fraumeni), previous partial gastrectomy 15 to 20 years on, pernicious anaemia, Menetrier's disease.[1][2]

The CDH1 family is the one to refer. A germline CDH1 mutation confers a 70 to 80 percent lifetime risk of diffuse gastric cancer and a high risk of lobular breast cancer, so asymptomatic adult carriers are offered prophylactic total gastrectomy — the paradigm of precision cancer prevention, because diffuse cancer is submucosal and patchy and endoscopic surveillance cannot be relied upon.[2]

The preventable-harm list — the public-health angle every final-prof candidate should carry:[2]

• Eradicate H. pylori — the only bacterium classified a Class I human carcinogen; screen-and-treat in high-incidence countries (Japan, Korea, China) cuts incident gastric cancer.
• Stop tobacco and alcohol — the two synergistic drivers of oesophageal SCC.
• Lower dietary salt and nitrites; raise fresh fruit and vegetables.
• Tackle obesity and reflux — the engine of the adenocarcinoma epidemic.
• Surveil Barrett's oesophagus and ablate dysplasia before it becomes cancer.[1]

Three carcinogenic cascades — the biology that decides the treatment

Each upper-GI cancer has a named multistep pathway, and each pathway has a point where prevention or interception works. Know the three cascades and the molecular driver of each.[1]

The Barrett's cascade (oesophageal adenocarcinoma). Chronic acid and bile reflux reprogrammes the distal squamous epithelium into specialised intestinal metaplasia — columnar cells with goblet cells, the Barrett's mucosa — through CDX1 and CDX2. That metaplastic mucosa then accumulates TP53 mutations, chromosomal instability and methylation, advancing through low-grade to high-grade dysplasia to invasive adenocarcinoma. Non-dysplastic Barrett's progresses at roughly 0.3 to 0.5 percent per year; high-grade dysplasia carries a much higher yearly risk and is the trigger for endoscopic resection and radiofrequency ablation.[1]

The injury cascade (oesophageal SCC). Ethanol (metabolised by CYP2E1 and alcohol dehydrogenase to the DNA-crosslinking acetaldehyde), tobacco nitrosamines and very hot drinks drive basal-cell hyperplasia through dysplasia to carcinoma. East-Asian carriers of the ALDH2 star-2 variant accumulate acetaldehyde and carry a markedly raised SCC risk with alcohol — a classic gene-environment interaction. In achalasia, stasis and fermentation chronically irritate the dilated segment.[1]

The Correa cascade (intestinal-type gastric adenocarcinoma). Over decades: normal mucosa to chronic superficial gastritis to atrophic gastritis to intestinal metaplasia to dysplasia to invasive adenocarcinoma. H. pylori (especially CagA-positive) injects CagA to drive ERK and MAPK and NF-kB signalling; its urease raises local pH; reactive oxygen species damage DNA. Loss of parietal cells (atrophy) permits bacterial overgrowth that converts dietary nitrate to N-nitroso compounds. The cascade is largely preventable — H. pylori eradication interrupts it.[2]

The CDH1 pathway (diffuse gastric adenocarcinoma). Loss of E-cadherin (CDH1 on 16q22), germline or somatic, dissolves the epithelial adherens junction and frees discohesive, mucin-laden signet-ring cells to infiltrate singly through the wall. Submucosal infiltration with desmoplastic fibrosis produces the rigid, non-distensible leather-bottle stomach of linitis plastica, which presents as early satiety despite a small tumour burden.[2]

The Cancer Genome Atlas sorts gastric adenocarcinoma into four molecular subtypes that increasingly guide therapy:[2]

• EBV-positive (about 9 percent) — PIK3CA mutations, PD-L1 and PD-L2 overexpression; best prognosis, highly immunotherapy-responsive.
• MSI-high (about 22 percent) — defective mismatch repair, hypermutation; good prognosis, responsive to pembrolizumab.
• Genomically stable (about 20 percent) — enriched for diffuse histology and CDH1 or RHOA mutations; poor prognosis.
• Chromosomal instability (about 50 percent) — HER2 amplification and RTK amplifications; HER2-targeted therapy relevant.[1]

HER2 and PD-L1 are the two markers that change therapy, so test them in every advanced gastric and GOJ adenocarcinoma. HER2 (ERBB2) amplification occurs in roughly 15 to 20 percent (more in intestinal and GOJ tumours) and is the target of trastuzumab. PD-L1, measured as the Combined Positive Score (CPS), predicts checkpoint-inhibitor response, with the greatest benefit at CPS at least 5.[8][9]

Dysphagia is a late alarm — recognise the red flags

The dominant presentation is late. Symptoms arise only when the tumour obstructs, bleeds or metastasises, which is the central reason the overall prognosis is poor and the rationale for urgent endoscopy on any red flag.[1][2]

The two highest-yield alarm features are progressive dysphagia — solids, then soft foods, then liquids, a mechanical pattern that means the lumen has lost roughly two-thirds of its diameter — and new or worsening dyspepsia at age 55 or over with any of weight loss, early satiety, vomiting, dysphagia, anaemia or a palpable mass. Either is a 2-week-wait OGD.[1]

Oesophageal cancer also presents with odynophagia, retrosternal chest pain, regurgitation of undigested food (with aspiration risk), long-standing reflux that may paradoxically improve as the tumour scars the GOJ, hoarseness from recurrent laryngeal nerve involvement, and cough on swallowing from a tracheo-oesophageal fistula.[1]

Gastric cancer presents with dyspepsia unresponsive to a PPI or to H. pylori eradication, early satiety (the hallmark of linitis plastica), nausea and vomiting from antral or outlet obstruction, iron-deficiency anaemia from occult bleeding, haematemesis or melaena, and an epigastric mass in advanced disease.[2]

The named metastatic and paraneoplastic signs every candidate must recognise — they signal incurable disease at the bedside:[2]

[1]

Paraneoplastic phenomena in gastric cancer include acanthosis nigricans (with tripe palms and the sign of Leser-Trelat — a sudden eruption of seborrhoeic keratoses), Trousseau's migratory thrombophlebitis, membranous nephropathy and rarely polyarteritis nodosa. Acanthosis nigricans or Leser-Trelat in an older patient is gastric cancer until excluded.[2]

The classic trap: progressive dyspepsia chalked up to GORD or a peptic ulcer in a patient over 55 with weight loss. The history alone — new, progressive, with alarm features — earns the urgent OGD before any empirical trial.[1]

The differential fork — solids-then-liquids versus liquids-and-solids

One question in the dysphagia history separates mechanical obstruction from neuromuscular failure, and the answer points to cancer or to achalasia. Solids-then-liquids, progressing over weeks to months, is mechanical — tumour or stricture. Liquids-and-solids together from the onset is neuromuscular — achalasia or a motility disorder.[1]

[1]

The dyspepsia differential runs gastric cancer against peptic ulcer disease (food-relieved duodenal or food-worsened gastric pain, H. pylori or NSAID, no weight loss), GORD (postprandial heartburn, no weight loss), functional dyspepsia (chronic, no alarm features, normal OGD), pancreatic cancer (epigastric pain to the back, painless jaundice, new diabetes), gallstone disease (right-upper-quadrant, fatty-food trigger, Murphy's sign), and angina (exertional, no relationship to food).[1]

Every gastric ulcer is malignant until proven otherwise. Biopsy a minimum of six sites (four from the rim, two from the base), and repeat the OGD at 8 to 12 weeks to document complete healing — failure to heal is malignancy until disproven. Endoscopic features favouring malignancy are heaped-up irregular edges, a friable necrotic base, folded surrounding mucosa, and a greater-curve location.[2]

OGD plus biopsy is the gateway — then stage to find the resectable minority

Diagnosis is OGD with targeted biopsy; everything else is staging. For an oesophageal tumour take six or more biopsies from the tumour and the mucosal margin. For a gastric ulcer take at least six (four rim, two base) and repeat at 8 to 12 weeks. Narrow-band imaging and chromoendoscopy (acetic acid; Lugol's iodine for SCC) sharpen detection, and endoscopic ultrasound at the same setting reads depth and regional nodes.[1]

Each staging test answers one question, and the team must ask each explicitly:[1][2]

[1]

Staging laparoscopy is non-negotiable before gastrectomy. CT misses peritoneal disease in 10 to 30 percent of patients who look resectable; laparoscopy with peritoneal washings detects it, upstages to M1 on positive cytology, spares a futile laparotomy, and re-routes the patient to systemic therapy. Omitting it is one of the classic, examinable errors.[2]

TNM staging, framed as what decides resectability. The T and N detail the examiner expects, verbatim:[1][2]

Oesophageal cancer — T1 lamina propria, muscularis mucosae or submucosa; T2 muscularis propria; T3 adventitia; T4 adjacent structures, split as T4a resectable (pleura, pericardium, azygos vein, diaphragm) and T4b unresectable (aorta, vertebral body, trachea). N0 to N3 by number of involved regional nodes; M1 is distant metastasis (non-regional nodes or organ).[1]

Gastric cancer — T1 lamina propria or submucosa; T2 muscularis propria; T3 subserosa; T4a serosa (visceral peritoneum); T4b adjacent organs. N0 to N3b by number of nodes; M1 is distant metastasis, including positive peritoneal cytology.[1]

Stage grouping, conceptually: Stage I is localised (T1 to T2, N0, M0) and surgically curable; Stage II to III is locally advanced (deeper wall and/or nodes, M0) and needs multimodality therapy; Stage IV is metastatic (M1) and is palliative. That three-way split is the fork the whole management section hangs on.[1]

Test HER2 and PD-L1 in every advanced or recurrent gastric and GOJ adenocarcinoma. HER2 by immunohistochemistry and in-situ hybridisation (IHC 3+ or ISH-positive defines HER2-positivity for trastuzumab); PD-L1 as the Combined Positive Score (benefit greatest at CPS at least 5); plus MSI and MMR testing (MSI-H tumours respond to pembrolizumab and carry a better prognosis). CA 19-9 and CEA are not diagnostic but track response and recurrence.[2][8][9]

Curative intent — match the chemo to the histology

Surgery is the only cure, and only the resectable minority (Stage I to III, M0) benefit. Multidisciplinary staging decides who is resectable; perioperative systemic therapy improves survival in locally advanced disease; M1 disease is palliative. Within curative intent, the regimen is chosen by site and histology.[1][2]

Early mucosal disease — endoscopic resection preserves the organ. Endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) are curative for T1a disease with low nodal risk: Barrett's high-grade dysplasia and T1a adenocarcinoma (EMR of visible lesions plus radiofrequency ablation of the remaining Barrett's, over 95 percent 5-year survival), and early intestinal-type gastric cancer (well-differentiated, under 2 cm, no ulceration, T1a, no lymphovascular invasion — ESD, the Japanese standard). Deep submucosal invasion, lymphovascular invasion or poor differentiation upgrades to surgical resection with lymphadenectomy.[1]

Resectable oesophageal cancer (Stage II to III, M0) — neoadjuvant therapy then oesophagectomy. Two evidence-based options:[1]

• Neoadjuvant chemoradiotherapy, the CROSS regimen — carboplatin (AUC 2) plus paclitaxel (50 mg/m²) weekly for 5 weeks with concurrent radiotherapy 41.4 Gy in 23 fractions, then oesophagectomy at 6 to 8 weeks. CROSS improved R0 resection (92 versus 69 percent) and median survival (49 versus 24 months), with benefit in both SCC and adenocarcinoma.[6]
• Neoadjuvant chemotherapy, the JCOG9907 approach — cisplatin plus 5-fluorouracil (CF) for two cycles then surgery; JCOG9907 established neoadjuvant CF as the Japanese standard for thoracic SCC, with better survival than adjuvant CF.[7]

CROSS is widely preferred in Europe and for Siewert I and adenocarcinoma; perioperative FLOT increasingly covers Siewert II and III and GOJ adenocarcinoma. Definitive chemoradiotherapy (without surgery) is preferred for cervical oesophageal SCC, for organ preservation, and for patients unfit for surgery.[1]

The oesophagectomy. Ivor Lewis (two-phase, right thoracotomy plus laparotomy) is the standard for the mid and lower oesophagus and GOJ — a gastric conduit is pulled up and anastomosed in the right chest. McKeown (three-stage, cervical anastomosis) suits upper thoracic tumours and allows cervical lymphadenectomy. Transhiatal avoids thoracotomy (lower respiratory morbidity, more anastomotic leaks). Minimally invasive thoracoscopic, laparoscopic and robotic approaches are now standard in many centres.[1]

Resectable gastric and GOJ adenocarcinoma (Stage II to III, M0) — gastrectomy with D2 lymphadenectomy plus perioperative FLOT.[2]

Perioperative FLOT, the current global standard: 5-Fluorouracil plus Leucovorin (folinic acid) plus Oxaliplatin plus Docetaxel, four cycles pre-operatively and four post-operatively (each cycle every 2 weeks). FLOT4 showed improved median overall survival (50 versus 35 months) and 5-year survival (45 versus 36 percent) over the anthracycline-based ECF and ECX regimen (epirubicin, cisplatin, capecitabine or 5-FU), and replaced it.[5] The principle of perioperative chemotherapy itself was established by MAGIC, in which perioperative ECF lifted 5-year survival from 23 to 36 percent.[4]

Adjuvant chemoradiotherapy (the Macdonald, Intergroup 0116 approach) is used mainly in North America where D2 uptake was historically low: 5-FU plus leucovorin with 45 Gy of radiotherapy after a D0 or D1 gastrectomy. Intergroup 0116 showed improved survival with adjuvant chemoradiotherapy after gastrectomy.[10] In D2-operated patients (Asian and European practice), adjuvant chemotherapy — S-1 or XELOX in the CLASSIC and ARTIST trials — is preferred over chemoradiotherapy.[2]

The gastrectomy and D2 lymphadenectomy. Total gastrectomy (for proximal, diffuse or linitis plastica tumours; Roux-en-Y oesophagojejunostomy) or subtotal distal gastrectomy (for distal tumours with a 5 cm proximal margin; Billroth II or Roux-en-Y). D2 adds to D1 (perigastric nodes) the nodes along the left gastric, common hepatic, splenic, coeliac and (for some) hepatoduodenal ligament arteries, with at least 15 nodes examined. The Dutch D1D2 trial at 15-year follow-up established D2 as standard — lower gastric-cancer death and recurrence — with a spleen-preserving technique that avoids the morbidity of historical distal pancreateatosplenectomy. Multivisceral resection (spleen, distal pancreas, transverse colon) is only for direct invasion aiming at R0.[1]

[1]

Advanced disease — test HER2 and PD-L1 before you choose

Metastatic (Stage IV) disease is palliative. The goals are symptom control, quality of life and survival extension, and the regimen is guided by performance status, HER2 status and PD-L1 CPS.[2][8][9]

• First-line chemotherapy — a fluoropyrimidine (5-FU or capecitabine) plus a platinum (cisplatin or oxaliplatin), plus or minus a taxane (FLOT, FOLFOX, CAPOX, ECF), typically to progression or toxicity.
• HER2-positive (IHC 3+ or ISH-positive) — add trastuzumab (8 mg/kg loading then 6 mg/kg IV every 3 weeks) to first-line chemotherapy. ToGA improved median overall survival from 11.1 to 13.8 months — the first targeted-therapy success in gastric cancer.[8]
• PD-L1 CPS at least 5 (or MSI-H) — add nivolumab (360 mg/m² every 3 weeks) or pembrolizumab to first-line chemotherapy. CheckMate 649 showed improved survival with nivolumab plus chemotherapy, benefit greatest at CPS at least 5.[9]
• Second line — ramucirumab (anti-VEGFR2) plus or minus paclitaxel, irinotecan, or pembrolizumab for MSI-H.
• Third line and beyond — trifluridine and tipiracil (TAS-102) and best supportive care.[1]

Advanced oesophageal SCC is treated with cisplatin plus 5-FU (CF) or paclitaxel, immune checkpoint inhibitors for CPS at least 10 or after chemotherapy, palliative radiotherapy for dysphagia, bleeding or bone pain, and an oesophageal stent for obstruction.[1]

Supportive and palliative care is active treatment, not an afterthought. A self-expanding metal stent rapidly relieves malignant dysphagia (watch for perforation, bleeding, migration, reflux, food impaction); brachytherapy or single-fraction external beam gives slower but more durable relief; laser recanalisation or argon plasma coagulation manages exophytic bleeding or obstruction. Layer on the WHO analgesic ladder (a coeliac-plexus block for gastric pain), nutritional support, paracentesis for ascites, and early palliative-care involvement with advance care planning.[2]

The four classic traps

These are the recurring errors that fail patients and fail exams — name them by heart.[1]

• Trap 1 — missing the alarm feature and presenting late. Dysphagia is usually late; the lumen is already two-thirds gone. Any progressive dysphagia, or new dyspepsia at 55 plus with weight loss, is a 2-week-wait OGD.[1]
• Trap 2 — not testing HER2 and PD-L1 in advanced disease. The result changes therapy (trastuzumab, nivolumab). Test every metastatic or recurrent gastric and GOJ adenocarcinoma.[8][9]
• Trap 3 — conflating oesophageal SCC and adenocarcinoma. CROSS serves both, but cervical SCC gets definitive chemoradiotherapy, GOJ adenocarcinoma gets perioperative FLOT, and the histology sets the geography and the regimen.[1]
• Trap 4 — operating without staging and laparoscopy. In gastric cancer, CT misses peritoneal disease in 10 to 30 percent; a positive washout is M1 and re-routes to systemic therapy. Stage before you cut.[2]

And the smaller pitfalls that still cost marks: failing to biopsy a gastric ulcer and confirm healing at 8 to 12 weeks; forgetting lifelong B12 after total gastrectomy; mislabelling achalasia as cancer (or vice versa) on a careless dysphagia history; and underestimating the young patient, in whom CDH1 hereditary diffuse gastric cancer or Lynch must be considered.[1]

Complications after the surgeon's knife — by the clock

Major upper-GI surgery is high-morbidity, and the complications an examiner asks about are the ones a registrar must catch early.[1]

After oesophagectomy: anastomotic leak (5 to 15 percent — drain, antibiotics, stent; cervical leaks are commoner but better tolerated), anastomotic stricture (endoscopic dilatation), gastric-conduit necrosis (catastrophic, re-operation), chylothorax (medium-chain triglyceride diet, octreotide, duct ligation if persistent), recurrent laryngeal nerve palsy, atrial fibrillation, and respiratory complications — atelectasis and pneumonia, the commonest cause of early mortality. Reflux, dumping and early satiety follow loss of the lower oesophageal sphincter and conduit denervation.[1]

After gastrectomy — the dumping syndrome is the one to explain. Early dumping at 15 to 30 minutes is vasomotor — palpitations, sweating, dizziness, diarrhoea — from rapid osmotic fluid shift; late dumping at 1 to 3 hours is hypoglycaemia from an exaggerated insulin response to the glucose load. Manage with small, frequent, low-carbohydrate, high-protein meals, separating fluids and solids, and octreotide for refractory cases. Add afferent and efferent loop syndrome (Billroth II), alkaline reflux gastritis, anaemia (B12 from loss of intrinsic factor, iron from loss of acid), calcium and vitamin D malabsorption causing osteoporosis, internal hernia after Roux-en-Y, marginal ulcer, gastric-remnant cancer 15 to 20 years on, and postgastrectomy malnutrition.[2]

Chemotherapy and chemoradiotherapy toxicities sort by regimen: FLOT brings neutropenia and febrile neutropenia (G-CSF prophylaxis often used), diarrhoea, mucositis, oxaliplatin and docetaxel neuropathy, alopecia and hypersensitivity; CROSS brings oesophagitis, radiation pneumonitis and cardiac toxicity; trastuzumab brings LVEF decline; checkpoint inhibitors bring immune-related adverse events — colitis, pneumonitis, hepatitis, and the endocrinopathies (thyroiditis, hypophysitis, type 1 diabetes, adrenal insufficiency). Gastric and pancreatic cancer carry a high thromboembolism risk; consider prophylactic LMWH.[4][5][6]

Prognosis, and the numbers that frame the conversation

Overall survival is poor because presentation is late; survival by stage is the frame for the honest conversation with the family.[1][2]

• Oesophageal cancer — overall 5-year survival roughly 15 to 20 percent; localised (Stage I) about 45 to 50 percent; regional (Stage II to III with nodes) about 25 percent; distant (Stage IV) under 5 percent.
• Gastric cancer — overall 5-year survival roughly 25 to 30 percent; localised about 65 to 70 percent; regional about 30 percent; distant under 6 percent. In Japan and Korea, where endoscopic screening finds early disease, overall 5-year survival exceeds 60 to 70 percent.[1]

Predictors of poor outcome: advanced stage, signet-ring or diffuse histology, linitis plastica, peritoneal disease, R1 or R2 resection, low albumin or cachexia, and poor performance status (ECOG 2 or more). A high PD-L1 CPS paradoxically predicts a better immunotherapy response.[1]

The survival gains from modern perioperative therapy, in one place: CROSS median overall survival 49 versus 24 months;[6] FLOT median 50 versus 35 months with 5-year survival 45 versus 36 percent;[5] trastuzumab in HER2-positive metastatic disease median 13.8 versus 11.1 months;[8] nivolumab plus chemotherapy at CPS at least 5 median roughly 14 versus 11 months.[9]

After curative treatment, surveillance is history, examination, bloods (FBC, LFT, iron studies, B12, tumour markers) and CT at intervals (commonly 6-monthly for 2 years then annually to 5 years), with OGD to monitor the anastomosis and the gastric remnant. Every patient is discussed in a specialist upper-GI cancer MDT, operated on in a high-volume centre (the volume-outcome relationship is real), and palliative care is involved early in advanced disease.[1]

The trials that decide every management question

Seven landmark trials cover almost every curative and palliative decision in upper-GI cancer; match each to its management fork.[4][5][6][7][8][9][10]

[4] [5] [6] [7] [8] [9] [10]

Regional practice differs, but the framework is globally consistent.[1]

[1] [1] [1] [1]

Open controversies worth naming calmly: oesophagectomy approach (transthoracic Ivor Lewis versus transhiatal; open versus minimally invasive and robotic); neoadjuvant chemotherapy versus chemoradiotherapy for adenocarcinoma (ESOPEC and NEO-AEGIS comparing FLOT and CROSS — both acceptable); the role of radiation in gastric cancer after D2 (ARTIST and ARTIST-II, benefit confined to node-positive disease); immunotherapy moving earlier into perioperative disease (KEYNOTE-585, DANTE, MATTERHORN); and the timing and extent of prophylactic gastrectomy in CDH1 carriers.[1]

Special populations

Examiners reach for these scenarios because the presentation and the thresholds both shift.[1]

The young patient with gastric cancer (under 40). Suspect hereditary diffuse gastric cancer (CDH1) or Lynch syndrome; refer for genetic counselling and testing. Diffuse-type disease is aggressive and often advanced at presentation.[2]

The elderly or frail patient. Comorbidity and performance status, not age alone, drive treatment: prefer subtotal over total gastrectomy where possible, dose-reduced or single-agent chemotherapy, definitive chemoradiotherapy for oesophageal cancer in those unfit for surgery, and early palliative-care involvement.[1]

Barrett's surveillance. Document circumferential (C) and maximal (M) extent by the Prague criteria. Non-dysplastic Barrett's: surveillance endoscopy every 3 to 5 years; indefinite for dysplasia: repeat at 6 months on high-dose PPI; low-grade dysplasia: 6 to 12 months and consider RFA; high-grade dysplasia or T1a: EMR plus RFA. This is where adenocarcinoma is intercepted before it becomes incurable.[1]

Previous partial gastrectomy, pernicious anaemia and autoimmune gastritis. Cancer risk rises 15 to 20 years after partial gastrectomy (bile reflux, bacterial overgrowth, nitrosamines); pernicious anaemia and autoimmune gastritis warrant surveillance for gastric cancer and type I gastric neuroendocrine tumours.[2]

Pregnancy. Rare; stage with MRI and ultrasound where possible to avoid CT radiation, weigh maternal treatment timing against foetal safety, and involve the MDT with obstetric and neonatal teams.[1][2]

The memory devices, and the mantra

[1] [1] [2] [1] [2] [4] [5] [6] [7]

The Correa cascade, verbatim: normal mucosa, then chronic superficial gastritis, then atrophic gastritis, then intestinal metaplasia, then dysplasia, then invasive adenocarcinoma.[2]

The lines that decide a viva answer:[1]

• Dysphagia: solids then liquids is mechanical (tumour or stricture); liquids and solids together from onset is neuromuscular (achalasia).
• Every gastric ulcer is malignant until proven otherwise — six biopsies, repeat OGD at 8 to 12 weeks to confirm healing.
• Staging laparoscopy with peritoneal washout is non-negotiable before gastrectomy — it picks up 10 to 30 percent of CT-occult peritoneal disease.
• After total gastrectomy, lifelong IM hydroxocobalamin for B12 — the loss of intrinsic factor.
• H. pylori is the only bacterium classified a Class I human carcinogen — eradication is a population-level cancer-prevention intervention.[2]
• Acanthosis nigricans or the sign of Leser-Trelat in an older patient is gastric cancer until excluded.

[1]

The mantra: Scope the alarm, stage before you cut, match the chemo to the histology.[1]

[1] [2]

Ward-round test — three stems

Stem 1 — the Barrett's patient who now has dysphagia (answer)

A 68-year-old man with long-standing Barrett's oesophagus, lost to surveillance, presents with three months of progressive solid-food dysphagia and 8 kg weight loss. What is the diagnosis to exclude, and what is the first step? Model: This is oesophageal adenocarcinoma until proven otherwise — progressive mechanical dysphagia (solids then liquids) with weight loss in a man with the Barrett's risk factor is a 2-week-wait red flag. The first step is urgent OGD with targeted biopsy; the histology then sets the pathway. If adenocarcinoma is confirmed at the lower oesophagus or GOJ, stage with CT, EUS and PET-CT and, for GOJ disease, staging laparoscopy, then plan neoadjuvant CROSS chemoradiotherapy or perioperative FLOT ahead of an Ivor Lewis oesophagectomy. Do not settle for a PPI trial or attribute the dysphagia to GORD.[1]

Stem 2 — resectable GOJ adenocarcinoma, which perioperative regimen (answer)

A 60-year-old fit man has a Siewert II GOJ adenocarcinoma, staged T3 N1 M0 on CT, EUS and a clean staging laparoscopy. What perioperative regimen do you offer, and what surgery follows? Model: Perioperative FLOT — 5-Fluorouracil plus Leucovorin plus Oxaliplatin plus Docetaxel, four cycles before surgery and four after (each cycle every 2 weeks). FLOT4 showed median overall survival of 50 versus 35 months and 5-year survival of 45 versus 36 percent over the older ECF and ECX (MAGIC) regimen, and is the global standard. The principle of perioperative chemotherapy itself came from MAGIC. Surgery is a total or extended gastrectomy with D2 lymphadenectomy (at least 15 nodes), and after a total gastrectomy the patient needs lifelong intramuscular hydroxocobalamin 1 mg every 3 months.[5][4]

Stem 3 — advanced HER2-positive gastric cancer (answer)

A 58-year-old woman presents with epigastric pain, early satiety, iron-deficiency anaemia and a hard left supraclavicular node. OGD shows a gastric adenocarcinoma; staging confirms liver and peritoneal metastases. What two tests change her first-line therapy, and what do you add? Model: This is metastatic (Stage IV) gastric adenocarcinoma — the Virchow node (Troisier's sign) signals incurable disease.[2] Two tests change first-line therapy: HER2 by IHC and ISH and PD-L1 Combined Positive Score. If HER2-positive (IHC 3+ or ISH-positive), add trastuzumab (8 mg/kg loading then 6 mg/kg IV every 3 weeks) to first-line chemotherapy — ToGA improved median overall survival from 11.1 to 13.8 months.[8] If PD-L1 CPS is at least 5, add nivolumab (360 mg/m² every 3 weeks) to chemotherapy — CheckMate 649 showed the greatest survival benefit at CPS at least 5.[9] The goals are symptom control, quality of life and survival extension; bring in palliative care early.

References

1. [1]Smyth EC, Lagergren J, Fitzgerald RC, et al. Oesophageal cancer Nat Rev Dis Primers, 2017.PMID 28748917
2. [2]Ajani JA, Lee J, Sano T, et al. Gastric adenocarcinoma Nat Rev Dis Primers, 2017.PMID 28569272
3. [3]Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries CA Cancer J Clin, 2021.PMID 33538338
4. [4]Cunningham D, Allum WH, Stenning SP, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer N Engl J Med, 2006.PMID 16822992
5. [5]Al-Batran SE, Homann N, Pauligk C, et al. Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial Lancet, 2019.PMID 30982686
6. [6]van Hagen P, Hulshof MC, van Lanschot JJ, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer N Engl J Med, 2012.PMID 22646630
7. [7]Ando N, Kato H, Igaki H, et al. A randomized trial comparing postoperative adjuvant chemotherapy with cisplatin and 5-fluorouracil versus preoperative chemotherapy for localized advanced squamous cell carcinoma of the thoracic esophagus (JCOG9907) Ann Surg Oncol, 2012.PMID 21879261
8. [8]Bang YJ, Van Cutsem E, Feyereislova A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial Lancet, 2010.PMID 20728210
9. [9]Janjigian YY, Shitara K, Moehler M, et al. First-line nivolumab plus chemotherapy versus chemotherapy alone for advanced gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma (CheckMate 649): a randomised, open-label, phase 3 trial Lancet, 2021.PMID 34102137
10. [10]Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction N Engl J Med, 2001.PMID 11547741