PET/CT in Oesophageal, Gastric and GI Stromal Tumours

1. The clinical problem

Oesophageal squamous cell carcinoma (SCC) and adenocarcinoma behave as two diseases that share an organ. SCC arises mainly in the upper and middle oesophagus and responds well to chemoradiotherapy; adenocarcinoma arises in the distal oesophagus and at the oesophagogastric junction (OGJ, also written GOJ), usually on a background of Barrett's oesophagus. Gastric adenocarcinoma is divided by the Lauren classification into intestinal and diffuse types, and the diffuse, poorly cohesive (signet-ring) tumours spread along the wall and across the peritoneum. Gastrointestinal stromal tumours (GISTs) are sarcomas driven, in most cases, by activating mutations in KIT or PDGFRA, and are treated with tyrosine kinase inhibitors such as imatinib.

At the tumour board the questions are practical. Is there distant disease that makes surgery or radical chemoradiotherapy futile? Which nodes are involved, and are they regional? Is the tumour responding to neoadjuvant treatment, and has anything new appeared before the operation? After chemoradiotherapy, is there any residual tumour, or can the patient avoid oesophagectomy? In GIST, is imatinib working, and how quickly can we know?

Staging system in force: AJCC/UICC eighth edition

The eighth edition of the American Joint Committee on Cancer (AJCC) and Union for International Cancer Control (UICC) classification, still in force for the oesophagus and stomach, introduced separate clinical (cTNM), pathological (pTNM) and post-neoadjuvant pathological (ypTNM) stage groups. The cTNM groups are what a PET/CT report feeds. They differ by histology, whereas the ypTNM groups are identical for SCC and adenocarcinoma.

Junctional adenocarcinomas are also described by the Siewert classification: type I has its centre 1–5 cm above the anatomical cardia (distal oesophageal adenocarcinoma), type II from 1 cm above to 2 cm below it (true cardia carcinoma), and type III from 2 to 5 cm below it (subcardial gastric carcinoma). The type guides the surgical approach and, for type III tumours, the need for staging laparoscopy. GISTs are not staged with TNM; the British Sarcoma Group (BSG) guideline recommends risk assessment by mitotic rate, tumour size, site and rupture instead.

Where nuclear medicine changes management

• Selection for radical treatment: detecting distant metastases that CT has missed spares patients futile oesophagectomy.
• Restaging after neoadjuvant therapy: finding interval metastases, and contributing to the clinical response evaluation that allows active surveillance.
• Radiotherapy planning: defining the length of the tumour and involved nodes.
• Response-adapted treatment: an early metabolic response scan in adenocarcinoma, currently a trial tool.
• GIST: early proof that imatinib is working, especially before surgery, and problem-solving when CT response is ambiguous.

2. Tracers and why they work

¹⁸F-fluorodeoxyglucose (FDG) is taken up by glucose transporters, phosphorylated by hexokinase and trapped. Most oesophageal SCCs and adenocarcinomas beyond the earliest stages are avid. Uptake reflects cell density and glycolysis, so tumours with abundant mucin, diffuse single-cell infiltration or small volume are harder to see, and inflammation in the oesophagus, junction and stomach produces the recurring false positives.

The Lauren effect is well documented. In 40 locally advanced gastric carcinomas, FDG PET detected 60% overall: 83% of intestinal-type tumours against 41% of non-intestinal tumours. Mean SUV was 6.7 against 4.8, and mucus-containing tumours averaged 3.9 against 7.2 for those without mucus. FDG uptake did not predict survival in that series. Newer scanners detect more primaries, but the direction of the difference holds, and the practical rule is unchanged: a faint or absent primary on PET does not reassure.

When FDG fails and what replaces it

• T stage: EUS, which images the wall layers; PET cannot measure depth of invasion.
• Regional nodes, especially those next to the primary: EUS with fine-needle aspiration (FNA); avid nodes blend into an avid primary.
• Peritoneum: staging laparoscopy with peritoneal washings; PET is insensitive to small peritoneal deposits.
• Locoregional residual disease after chemoradiotherapy: endoscopy with bite-on-bite biopsies and EUS with FNA of suspicious nodes.
• Fibroblast activation protein inhibitor (FAPI) PET: an investigational stromal tracer with promising results in the peritoneum (Section 8).

3. Indications by clinical scenario

The two UK and European documents agree for oesophageal and junctional cancer and differ slightly in tone for gastric cancer. NICE restricts PET in gastric cancer to suspected metastatic disease; ESMO 2022 does not recommend it routinely and puts its weight on laparoscopy. Single-centre series argue for broader use: in a UK cohort of 105 patients who had PET/CT, 86% of gastric primaries were avid (93% of intestinal against 78% of non-intestinal), and PET/CT upstaged 19%, mostly by finding distant metastases. Local practice varies, and the report should say plainly what a negative scan can and cannot exclude.

4. Evidence at a glance

5. Patient preparation and acquisition

Standard oncological FDG preparation applies: fasting, glucose check, a quiet uptake period of about 60 minutes, and the same scanner, protocol and reconstruction whenever scans will be compared quantitatively. Points specific to these tumours:

• Order of tests: for oesophageal and junctional cancer, PET/CT follows endoscopy and contrast CT and is reserved for patients still suitable for radical treatment (NICE NG83). Record the date of any endoscopy, EUS, biopsy, dilatation or stent insertion, because instrumentation adds inflammatory uptake.
• Coverage: skull base to mid-thigh, which includes the neck; for cervical and upper-thoracic SCC make sure the whole neck is in the field of view.
• Early response studies: the trial scans were at baseline and 14 days after starting chemotherapy (Weber, MUNICON, CALGB 80803), with the percentage change in SUV as the measure. Use the same scanner and uptake time, or the percentage change is meaningless.
• After chemoradiotherapy: in the SANO protocol the first response evaluation (endoscopy with bite-on-bite biopsies) is 4–6 weeks after treatment and the second, including PET/CT, at 10–14 weeks. Earlier scans are dominated by radiation oesophagitis.
• Radiotherapy planning: a planning scan in the treatment position lets PET and planning CT register without deformation; say so in the request.
• GIST: the baseline scan must precede the first imatinib dose, and a baseline without uptake makes PET useless for response. A response scan can be informative within the first week (ACRIN 6665) and within a few weeks in practice (BSG). Record the mutation if known.

6. Interpretation

Oesophagus and junction

Describe the primary by level, length of the avid segment and relation to the junction, and give SUVmax as a baseline for any later response assessment. Do not assign a T category from PET. For nodes, count the avid regional nodes and name their stations, because the N category rests on the number (N1 1–2, N2 3–6, N3 7 or more). Periesophageal nodes beside the primary often merge with it on PET, one reason nodal sensitivity is low; say when a node cannot be separated. Remember that coeliac nodes are regional. Distant disease is usually in the liver, lungs, bone, adrenals or non-regional nodes; a single lesion that would remove curative options needs tissue or targeted imaging before treatment is changed.

Stomach

The normal stomach takes up FDG to a variable and sometimes marked degree, so a primary is called when uptake is focal, clearly above adjacent gastric wall and matched by wall thickening on CT. A poorly avid primary predicts poor sensitivity for its nodes and metastases. State explicitly that PET does not exclude peritoneal disease: in PLASTIC, PET/CT had a sensitivity of 33% for distant metastases, and laparoscopy found peritoneal or locally unresectable disease in almost one patient in five.

GIST

Most untreated GISTs are avid, often heterogeneous with a photopenic necrotic or cystic centre. Report SUVmax of the primary and representative metastases at baseline. On treatment, compare the same lesions on the same scanner. Because response on CT is recognised by falling density as well as size, read the PET together with the contrast CT, and look for the progression pattern of a new enhancing, avid nodule inside a previously responding mass.

Criteria in use

Pitfalls and mimics

Wording the conclusion

• Give the clinical TNM elements PET can support and say what it cannot, for example: Avid distal oesophageal primary; one avid coeliac node (regional, cN1 by count); two liver lesions consistent with metastases (cM1, cStage IVB). T category requires EUS.
• Name the lesion that would change intent and how to confirm it.
• For gastric cancer, state the avidity of the primary and that peritoneal disease is not excluded.
• For response scans, give the percentage change in SUV, the measure used (SUVmax or SUVpeak) and whether acquisition matched the baseline.

7. Response assessment and follow-up

Early metabolic response in adenocarcinoma

The Munich group established the principle. In 40 junctional adenocarcinomas, FDG uptake fell by a mean of 54% at day 14 in tumours that went on to respond and by 15% in those that did not; a 35% threshold predicted clinical response with sensitivity 93% and specificity 95%. MUNICON then used the threshold to act: 54 of 110 evaluable patients (49%) were metabolic responders and completed 12 weeks of chemotherapy before surgery, while non-responders stopped after 2 weeks and went straight to surgery. Major histological remission (less than 10% residual tumour) occurred in 58% of responders and in no non-responder, and median event-free survival was 29.7 against 14.1 months. MUNICON II gave non-responders salvage chemoradiotherapy: histological response improved (26%), but the R0 rate and survival did not, suggesting that metabolic non-response marks aggressive biology as well as drug resistance.

CALGB 80803 tested switching rather than stopping. After induction with FOLFOX (oxaliplatin, leucovorin and fluorouracil) or carboplatin and paclitaxel, non-responders (SUVmax fall under 35%) crossed to the other regimen during chemoradiotherapy. Their pathological complete response (pCR) rates were 18% and 20%, higher than historically expected for non-responders; responders to FOLFOX who continued it reached 40% pCR and a 5-year survival of 53%. Median survival was 48.8 months for responders and 27.4 months for non-responders.

The picture is different in SCC. In the SCOPE2 PET substudy of definitive chemoradiotherapy, 61% of patients were non-responders at day 14 of cisplatin and capecitabine. Switching SCC non-responders to carboplatin and paclitaxel gave worse treatment-failure-free survival and worse overall survival (20.4 against 42.5 months), and early metabolic response was not prognostic. The substudy was closed for futility and possible harm. The lesson for the report: give the percentage change, but do not imply that it should change treatment outside a protocol.

Restaging and organ preservation after chemoradiotherapy

After the CROSS regimen (carboplatin, paclitaxel and 41.4 Gy), roughly half of SCCs and a quarter of adenocarcinomas have no residual tumour in the primary. preSANO asked how to find the rest. Endoscopy with regular biopsies and FNA missed 31% of tumours with more than 10% residual carcinoma; bite-on-bite biopsies with FNA missed 10%; EUS measurement of maximum tumour thickness missed 28%; PET/CT missed 15%. PET/CT's distinctive contribution was distant: it found histologically proven interval metastases in 18 of 190 patients (9%), 17 of them with adenocarcinoma.

SANO then compared active surveillance with standard oesophagectomy in 309 patients with a clinical complete response, defined by biopsies, EUS and PET/CT. Two-year overall survival was 74% with surveillance and 71% with surgery, within the non-inferiority margin. During surveillance, response evaluations are repeated every 3 months in the first year, every 4 months in the second, every 6 months in the third and yearly in the fourth and fifth; PET/CT was built into the sixth and ninth evaluations (about 16 and 30 months). The nuclear medicine role in this pathway is specific: detect distant disease and flag locoregional regrowth, while tissue decides locoregional status.

GIST

Imatinib switches off glycolysis in sensitive GIST within days. In ACRIN 6665, 36 of 44 patients had a complete or partial metabolic response at week 1, and 23 of 27 with KIT exon 11 mutations had responded metabolically by then; by contrast RECIST showed a partial response in only 2 of 39. On CT, a falling density often precedes shrinkage, and lesions may even grow transiently. The Choi criteria were built against PET: in 40 patients a size fall over 10% or a density fall over 15% at 2 months identified PET responders with sensitivity 97% and specificity 100%. BSG adds that absence of progression at 6 months is equivalent to response, and that the 'nodule within the mass' is the typical pattern of secondary progression.

Genotype shapes the question. Mutational analysis is mandatory before neoadjuvant imatinib: PDGFRA exon 18 D842V GISTs are insensitive to imatinib (avapritinib is the most active drug), KIT exon 9 tumours need 800 mg daily, and NF1-related and many SDH-deficient 'wild-type' GISTs respond poorly. BSG notes that in the absence of molecular analysis, FDG PET/CT can show response very rapidly, within a few weeks, so that surgery is not delayed in non-responding disease.

8. Theranostics and emerging tracers

FAPI is attractive precisely where FDG is weakest: diffuse and signet-ring gastric cancer and small peritoneal deposits, because its signal comes from the tumour stroma rather than tumour glycolysis. The studies so far are small, mostly single-centre and at high pre-test probability, and fibrosis and inflammation also take up FAPI. It should be described to referrers as a research test until prospective data show that it changes outcome.

9. Structured report example

10. Take-home points

1. FDG PET/CT belongs in the staging of every oesophageal or junctional cancer being considered for radical treatment (except T1a): its job is distant disease.
2. EUS stages T and is more sensitive for regional nodes; coeliac nodes are regional, and the N category depends on the number of involved nodes.
3. In gastric cancer, use PET selectively. Diffuse, signet-ring and mucinous tumours are often poorly avid, and staging laparoscopy, not PET, finds peritoneal disease.
4. A 35% fall in SUV at day 14 predicts response in adenocarcinoma, and switching chemotherapy helped non-responders in CALGB 80803; the same strategy harmed SCC patients in SCOPE2. Response-adapted treatment is still a trial tool.
5. Restage with PET/CT before surgery: about 1 in 11 patients has interval metastases after chemoradiotherapy. Confirm before cancelling surgery, remembering radiation-induced liver injury.
6. After chemoradiotherapy, clinical complete response needs biopsies, EUS and PET/CT together; active surveillance was non-inferior at 2 years in SANO.
7. In GIST, check the genotype first, record baseline avidity, and expect metabolic response within days; on CT use Choi, not RECIST alone.

Test yourself

5 quick questions. Pick an answer to see the explanation.

References

1. Rice TW, Ishwaran H, Ferguson MK, Blackstone EH, Goldstraw P. Cancer of the esophagus and esophagogastric junction: an eighth edition staging primer. J Thorac Oncol. 2017;12(1):36-42.
2. Siewert JR, Stein HJ. Classification of adenocarcinoma of the oesophagogastric junction. Br J Surg. 1998;85(11):1457-9.
3. Obermannová R, Alsina M, Cervantes A, et al. Oesophageal cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol. 2022;33(10):992-1004.
4. Obermannová RL, Leong T, Smyth EC, et al. ESMO Clinical Practice Guideline interim update on the treatment of locally advanced oesophageal and oesophagogastric junction adenocarcinoma and metastatic squamous-cell carcinoma. ESMO Open. 2025;10(2):104134.
5. Lordick F, Carneiro F, Cascinu S, et al. Gastric cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol. 2022;33(10):1005-20.
6. National Institute for Health and Care Excellence. Oesophago-gastric cancer: assessment and management in adults. NICE guideline NG83. London: NICE; 2018.
7. Casali PG, Blay JY, Abecassis N, et al. Gastrointestinal stromal tumours: ESMO–EURACAN–GENTURIS Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2022;33(1):20-33.
8. Judson I, Jones RL, Wong NACS, et al. Gastrointestinal stromal tumour (GIST): British Sarcoma Group clinical practice guidelines. Br J Cancer. 2025;132(1):1-10.
9. van Vliet EP, Heijenbrok-Kal MH, Hunink MG, Kuipers EJ, Siersema PD. Staging investigations for oesophageal cancer: a meta-analysis. Br J Cancer. 2008;98(3):547-57.
10. Gertsen EC, Brenkman HJF, van Hillegersberg R, et al. 18F-fludeoxyglucose-positron emission tomography/computed tomography and laparoscopy for staging of locally advanced gastric cancer: a multicenter prospective Dutch cohort study (PLASTIC). JAMA Surg. 2021;156(12):e215340.
11. Stahl A, Ott K, Weber WA, et al. FDG PET imaging of locally advanced gastric carcinomas: correlation with endoscopic and histopathological findings. Eur J Nucl Med Mol Imaging. 2003;30(2):288-95.
12. Bosch KD, Chicklore S, Cook GJ, et al. Staging FDG PET-CT changes management in patients with gastric adenocarcinoma who are eligible for radical treatment. Eur J Nucl Med Mol Imaging. 2020;47(4):759-67.
13. Weber WA, Ott K, Becker K, et al. Prediction of response to preoperative chemotherapy in adenocarcinomas of the esophagogastric junction by metabolic imaging. J Clin Oncol. 2001;19(12):3058-65.
14. Lordick F, Ott K, Krause BJ, et al. PET to assess early metabolic response and to guide treatment of adenocarcinoma of the oesophagogastric junction: the MUNICON phase II trial. Lancet Oncol. 2007;8(9):797-805.
15. zum Büschenfelde CM, Herrmann K, Schuster T, et al. 18F-FDG PET-guided salvage neoadjuvant radiochemotherapy of adenocarcinoma of the esophagogastric junction: the MUNICON II trial. J Nucl Med. 2011;52(8):1189-96.
16. Goodman KA, Ou FS, Hall NC, et al. Randomized phase II study of PET response-adapted combined modality therapy for esophageal cancer: mature results of the CALGB 80803 (Alliance) trial. J Clin Oncol. 2021;39(25):2803-15.
17. Mukherjee S, Hurt CN, Adams R, et al. Efficacy of early PET-CT directed switch to carboplatin and paclitaxel based definitive chemoradiotherapy in patients with oesophageal cancer who have a poor early response to induction cisplatin and capecitabine in the UK: a multi-centre randomised controlled phase II trial. EClinicalMedicine. 2023;61:102059.
18. Noordman BJ, Spaander MCW, Valkema R, et al. Detection of residual disease after neoadjuvant chemoradiotherapy for oesophageal cancer (preSANO): a prospective multicentre, diagnostic cohort study. Lancet Oncol. 2018;19(7):965-74.
19. Eyck BM, van der Wilk BJ, Noordman BJ, et al. Updated protocol of the SANO trial: a stepped-wedge cluster randomised trial comparing surgery with active surveillance after neoadjuvant chemoradiotherapy for oesophageal cancer. Trials. 2021;22(1):345.
20. van der Wilk BJ, Eyck BM, Wijnhoven BPL, et al. Neoadjuvant chemoradiotherapy followed by active surveillance versus standard surgery for oesophageal cancer (SANO trial): a multicentre, stepped-wedge, cluster-randomised, non-inferiority, phase 3 trial. Lancet Oncol. 2025;26(4):425-36.
21. Choi H, Charnsangavej C, Faria SC, et al. Correlation of computed tomography and positron emission tomography in patients with metastatic gastrointestinal stromal tumor treated at a single institution with imatinib mesylate: proposal of new computed tomography response criteria. J Clin Oncol. 2007;25(13):1753-9.
22. Van den Abbeele AD, Gatsonis C, de Vries DJ, et al. ACRIN 6665/RTOG 0132 phase II trial of neoadjuvant imatinib mesylate for operable malignant gastrointestinal stromal tumor: monitoring with 18F-FDG PET and correlation with genotype and GLUT4 expression. J Nucl Med. 2012;53(4):567-74.
23. Voncken FEM, Aleman BMP, van Dieren JM, et al. Radiation-induced liver injury mimicking liver metastases on FDG-PET-CT after chemoradiotherapy for esophageal cancer: a retrospective study and literature review. Strahlenther Onkol. 2018;194(2):156-63.
24. Muijs CT, Beukema JC, Pruim J, et al. A systematic review on the role of FDG-PET/CT in tumour delineation and radiotherapy planning in patients with esophageal cancer. Radiother Oncol. 2010;97(2):165-71.
25. Veldhuijzen van Zanten SEM, Pieterman KJ, Wijnhoven BPL, et al. FAPI PET versus FDG PET, CT or MRI for staging pancreatic-, gastric- and cholangiocarcinoma: systematic review and head-to-head comparisons of diagnostic performances. Diagnostics (Basel). 2022;12(8):1958.