- Design
- Retrospective cohort with 1:3 propensity matching, plus molecular analyses
- Population
- 666 resected gastric cancers: 27 with a neuroendocrine carcinoma component, 639 adenocarcinomas
- Primary outcome
- Overall and recurrence-free survival
- Effect
- Worse OS and RFS with an NEC component after matching; nodal metastases mostly NEC
This retrospective single-centre cohort compared 27 resected gastric adeno-neuroendocrine carcinomas with 639 conventional adenocarcinomas (2011–2022), using 1:3 propensity matching, component-level immunohistochemistry and public transcriptomic data.
After matching, cancers with a neuroendocrine carcinoma component had worse overall and recurrence-free survival and more recurrence and liver metastasis. In the mixed tumours, lymph node metastases were mostly made of the neuroendocrine component. That component also showed more ASCL1 and phosphorylated STAT3 and fewer CD8-positive T cells than the paired adenocarcinoma.
The molecular findings are exploratory, but the morphological message is direct: a neuroendocrine carcinoma component in a gastric adenocarcinoma carries prognostic weight, and its share of the tumour, and of the nodal deposits, is worth reporting.
- Look for a neuroendocrine carcinoma component in poorly differentiated or solid gastric carcinomas.
- Confirm with synaptophysin, chromogranin and INSM1, and assess Ki-67 in that component.
- Report the estimated percentage of the neuroendocrine component.
- When lymph nodes are positive, state which component the metastases are made of.
Why it matters
The nodal disease may reflect the neuroendocrine component, which could matter for treatment choice.
Don't overread it
A small single-centre series; the ASCL1 and STAT3 findings are exploratory and do not yet guide treatment.
The statistics, in plain English
Only 27 patients had the mixed tumour, so the survival estimates are imprecise even after matching. The molecular associations come from one centre plus a public dataset and a cell-line experiment — they suggest mechanisms rather than prove them.
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 pathology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free