01Core principlesThe concepts and mechanisms needed to understand the subject.
Each modality has a distinct scale. CT surveys the body and identifies unresectable invasion or distant metastases. PET-CT adds metabolic detection but cannot reliably distinguish inflammation from tumour. EUS examines the immediate wall and regional nodes at high resolution. Laparoscopy directly inspects peritoneal surfaces and permits biopsy or cytology. No single result substitutes for the others in every patient.
Sequence follows the treatment question. After histological confirmation of invasive cancer, whole-body CT comes first. Add PET-CT before radical oesophageal or junctional treatment except T1a and use EUS selectively when it changes management, never solely to label T2 versus T3. Suspected T1N0 oesophageal adenocarcinoma is staged by endoscopic resection before routine CT or EUS is used merely to discriminate T1a from T1b. MAPS III similarly avoids routine cross-sectional, EUS or metabolic imaging before endoscopic resection of an early gastric lesion unless deep invasion or non-suitability for local treatment is suspected.
Staging is a decision process rather than a scan collection. A suspicious distant focus that would deny cure may need biopsy. Enlarged nodes can be reactive and normal-size nodes malignant. EUS T-stage after neoadjuvant therapy is distorted by inflammation and fibrosis. Document the clinical question, result, uncertainty and resulting MDT choice.
Key points
- Whole-body CT is the initial anatomical staging survey after histologically confirmed oesophageal, junctional or gastric cancer, but it is not routine before resection merely to distinguish T1a from T1b in suspected early oesophageal adenocarcinoma.
- Offer FDG PET-CT before radical treatment of oesophageal or junctional cancer except T1a; investigate a solitary unexpected focus when it would remove a curative option.
- Use EUS only when local depth or nodal information will guide management; NICE says not to offer it solely to distinguish T2 from T3, and a stenosis must not be forcibly traversed.
- Potentially curable gastric cancer in a radical-surgery pathway needs staging laparoscopy, whereas an early gastric lesion suitable for endoscopic resection does not routinely need CT, EUS, MRI, PET-CT or laparoscopy before local resection.
- Consider staging laparoscopy for potentially curable junctional tumours when it will help guide treatment.
- Order a test because its result can change management; completing every modality regardless of stage delays treatment and exposes patients to avoidable procedures.
02Mechanisms and patternsImportant relationships and how to distinguish them.
Unequivocal metastases often remove the need for invasive local staging unless tissue or symptom treatment remains necessary.
Gastric serosal disease, diffuse type and ascites increase concern for small peritoneal deposits.
Non-traversability limits EUS and raises perforation risk, requiring alternative staging.
A solitary avid lesion may be inflammatory and deserves confirmation if it changes curative intent.
Fibrosis and inflammation reduce the accuracy of local restaging and must be interpreted with baseline findings.
03Interpreting evidenceInformation, measurements and their limitations.
Consider the information, its meaning and its limitations before deciding what follows.
- 01
Whole-body contrast CT after cancer confirmation - Why
- Establish gross local, regional and distant anatomical stage for an invasive-cancer pathway.
- Interpretation and limitations
- Look for organ invasion, nodes and distant disease. Do not insert CT merely before endoscopic resection of suspected T1N0 oesophageal adenocarcinoma or a gastric lesion already suitable for local resection.
- 02
FDG PET-CT - Why
- Find occult spread before radical oesophageal or junctional treatment, except for T1a tumours.
- Interpretation and limitations
- Confirm unexpected solitary disease when feasible because inflammation and second primaries can mimic metastasis.
- 03
Selective endoscopic ultrasound - Why
- Assess wall or regional nodes when the result will change a curative strategy.
- Interpretation and limitations
- Do not use EUS solely to distinguish T2 from T3, do not force passage through a stenosis, and do not use it routinely before resection of an appropriate early gastric lesion.
- 04
Staging laparoscopy - Why
- Inspect and sample occult peritoneal or superficial liver disease.
- Interpretation and limitations
- A positive deposit or malignant cytology changes curative gastric surgery plans despite negative CT. This radical-surgery recommendation does not add routine laparoscopy before endoscopic resection of a suitable early gastric lesion.
- 05
Diagnostic biopsy of metastasis - Why
- Resolve a management-changing ambiguous distant lesion.
- Interpretation and limitations
- Select the safest accessible site and preserve enough tissue for histology and biomarkers.
04Applied reasoningWorked examples connecting principles to decisions.
01Worked case: gastric cancerFind spread below CT resolutionA fit 62-year-old has biopsy-proven antral adenocarcinoma; contrast CT suggests cT3N1 disease without liver, lung or visible peritoneal metastasis.+
- 1The specialist MDT confirms that curative treatment remains plausible on CT and offers staging laparoscopy before committing to the perioperative surgical pathway.
- 2Laparoscopy identifies a 4 mm nodule on the right hemidiaphragm that was not visible on CT; the surgeon biopsies it and obtains peritoneal washings.
- 3Histology confirms metastatic adenocarcinoma in the nodule, so the MDT cancels the proposed radical gastrectomy pathway and requests tumour biomarkers for systemic-treatment selection.
- 4Verify that the patient receives the revised stage and plan, that the avoided laparotomy is recorded, and that oncology and dietetic appointments are booked before discharge.
02Worked case: oesophageal cancerSearch for occult systemic spreadA distal oesophageal cancer appears localised on CT and radical treatment is contemplated.+
- 1Arrange PET-CT to identify occult metastatic disease and review any unexpected focus.
- 2Use EUS only if wall or nodal information will change the radical strategy and passage is safe.
- 3Obtain tissue from a management-changing distant lesion when feasible before abandoning curative intent.
- 4Confirm the final stage and treatment choice at the MDT.
03Worked case: obvious metastasesStop redundant local stagingCT shows multiple liver and lung metastases from biopsy-proven junctional cancer.+
- 1Recognise that detailed EUS wall staging will not restore resectability.
- 2Complete biomarker testing and assess dysphagia, nutrition and performance status.
- 3Plan systemic and local palliative treatment around goals and symptom urgency.
- 4Review response and emerging complications using investigations that can alter care.
05Checking understandingVerify the reasoning, revisit uncertainties and apply feedback.
- Record the explicit question for every staging test and the decision it changed.
- Track renal function and contrast risk around CT without withholding necessary imaging reflexively.
- Audit non-traversable EUS attempts and procedure-related perforation.
- Ensure laparoscopy pathology or cytology is incorporated before radical surgery.
- Revisit stage when new symptoms or interval progression appear during treatment.
06Special situationsVariants, exceptions and circumstances that change the usual approach.
Resolution has a location
EUS sees wall layers; CT sees the body; laparoscopy sees the peritoneal surface.
Negative is modality-specific
A negative CT means no CT-visible spread, not absence of microscopic peritoneal cancer.
A test can be correct and unhelpful
Precise local T stage adds little when distant metastases already determine palliation.
Unexpected findings need proportionate proof
One inflammatory PET focus should not automatically remove a realistic curative option.
07Common pitfallsFrequent interpretation and management errors.
- 01
Ordering EUS after metastatic disease is already established without a treatment question.
- 02
Assuming normal-size nodes are benign or enlarged nodes are malignant without context.
- 03
Omitting laparoscopy because potentially curable gastric cancer has a clear CT.
- 04
Forcing an echoendoscope across a near-occlusive lesion.