- Design
- Retrospective single-centre cohort
- Population
- 66 patients with oesophagogastric adenocarcinoma restaged after neoadjuvant chemotherapy
- Primary outcome
- Accuracy for ypT, ypN, pathological response and metastases
- Effect
- Response AUC 0.72 (PET-CT) vs 0.51 (CT); metastases 8/8 vs 4/8
This single-centre cohort compared contrast-enhanced CT and FDG PET-CT, performed together before and after neoadjuvant chemotherapy, in 66 patients with oesophagogastric adenocarcinoma who went on to resection. Response was measured as change in oesophageal thickness on CT and change in SUVmax on PET-CT.
The two were similar, and poor, at predicting pathological T stage (AUC 0.65 vs 0.62) and N stage (0.52 vs 0.53). PET-CT predicted pathological tumour response better (AUC 0.72 vs 0.51, P = .041). Of 8 patients with metastases found at restaging, PET-CT detected all 8 and CT 4 (P = .076).
Neither test should be used to call nodal status after chemotherapy. PET-CT adds most where response assessment would change management or occult metastasis is likely. The sample is small and the metastasis comparison is not statistically significant, so cost and access will decide much of the choice.
- Do not rely on CT or PET-CT to predict nodal stage after neoadjuvant chemotherapy; both performed at chance.
- Prefer PET-CT restaging when response would change the surgical or oncological plan.
- Consider PET-CT for patients at high risk of occult metastases before resection.
- Report SUVmax change from baseline, not only the post-treatment value.
Why it matters
Contrast CT gave no useful information on pathological response in this cohort.
Don't overread it
Single-centre, 66 patients; the metastasis difference (8 vs 4) was not statistically significant.
The statistics, in plain English
An AUC of 0.5 is a coin toss; 0.72 is moderate. CT's 0.51 for response and both tests' 0.52 to 0.53 for nodes mean they did not discriminate. The metastasis result rests on eight patients, so it is suggestive only.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for radiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free