- Design
- Retrospective clinicopathological and immunohistochemical study
- Population
- 42 biliary tract neuroendocrine carcinomas, 22 resected
- Primary outcome
- Subtype distribution and disease-specific survival
- Effect
- 40% NEUROD1, 26% POU2F3, 21% ASCL1, 12% null; null type HR 3.48 for shorter DSS
This retrospective study classified 42 biliary tract neuroendocrine carcinomas — gallbladder, intrahepatic and extrahepatic ducts and ampulla — by immunohistochemistry for the transcription factors ASCL1, NEUROD1 and POU2F3, as used in small-cell lung cancer.
The split was 21% ASCL1-type, 40% NEUROD1-type, 26% POU2F3-type and 12% null. POU2F3-type tumours clustered in the gallbladder (39% of gallbladder cases) and tended to be unresectable; NEUROD1-type tumours were more often resectable. The null type independently predicted shorter disease-specific survival (HR 3.48, P = .034). Most resected tumours were mixed neuroendocrine-non-neuroendocrine neoplasms, and metastases were mostly neuroendocrine carcinoma whatever the proportion in the primary.
The diagnostic point is the one to act on: several POU2F3-type tumours lacked chromogranin A, synaptophysin and INSM1. A poorly differentiated biliary carcinoma that is negative for the usual panel could still be a neuroendocrine carcinoma. With 42 cases, the prognostic findings are preliminary.
- Consider POU2F3 immunohistochemistry for a poorly differentiated biliary or gallbladder carcinoma with small-cell or solid morphology but a negative neuroendocrine panel.
- Report the proportion of neuroendocrine carcinoma in mixed tumours; metastases were mostly the neuroendocrine component.
- Look for an associated intraductal papillary neoplasm; four tumours arose from one.
- Treat the prognostic value of subtype as preliminary; the cohort was small.
Why it matters
Missing the neuroendocrine carcinoma changes the chemotherapy the patient receives.
Don't overread it
Only 42 cases; the survival differences by subtype need confirmation in larger cohorts.
The statistics, in plain English
A hazard ratio of 3.48 means the null-type tumours had about three and a half times the rate of disease-specific death. With so few patients in each subtype, the confidence interval is likely wide, and the P value of .034 is only just below the usual threshold.
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