MTAP does not stand in for CDKN2A in the peritoneum

Design

retrospective comparison of immunohistochemistry against fluorescence in situ hybridisation, cases selected by known marker status

Population

24 peritoneal mesotheliomas, with 25 pleural mesotheliomas and 10 reactive mesothelial proliferations as controls

Primary outcome

concordance of MTAP immunohistochemistry and FISH with CDKN2A FISH (Cohen's kappa)

Effect

peritoneal: MTAP IHC vs CDKN2A FISH kappa 0.21 (P = 0.36); pleural: kappa 0.613 (P = 0.002)

Loss of MTAP by immunohistochemistry is used as a surrogate for CDKN2A homozygous deletion, and in pleural mesothelioma it works. This study asked whether the weaker performance observed in peritoneal mesothelioma reflects a real difference in how often the two genes are deleted together. Twenty-four peritoneal mesotheliomas underwent MTAP fluorescence in situ hybridisation with a custom probe, with 25 pleural mesotheliomas and 10 reactive mesothelial proliferations as external controls.

In the pleural cases the surrogate held. MTAP immunohistochemistry agreed perfectly with MTAP FISH (kappa 1.00) and acceptably with CDKN2A FISH (kappa 0.613), giving 100% specificity and 68.8% sensitivity for CDKN2A homozygous deletion. In the peritoneal cases the immunohistochemistry still tracked MTAP FISH closely (kappa 0.83) — the stain is reporting the gene correctly — but agreement with CDKN2A FISH collapsed to kappa 0.21 and was not statistically significant. Specificity for CDKN2A deletion fell to 75% and sensitivity to 50%. All reactive proliferations were correctly classified by every assay.

The mechanism this implies is site-specific: MTAP and CDKN2A are adjacent on chromosome 9p and are usually lost together, and the finding is that in the peritoneum they more often are not. The stain is not failing; the assumption behind it is.

So in a peritoneal specimen MTAP loss should be reported as MTAP loss, and used as an adjunct marker of malignancy in its own right rather than as evidence of CDKN2A deletion. Where CDKN2A status matters — for diagnosis in a difficult case, or because a protocol calls for it — it needs testing directly. That is a meaningful cost difference in a laboratory where FISH is not routine, and it is worth knowing before a report is written rather than after.

• In peritoneal specimens, report MTAP loss as MTAP loss and do not infer CDKN2A homozygous deletion from it
• Use MTAP loss in the peritoneum as an adjunct marker supporting malignancy in its own right
• Test CDKN2A directly by FISH where its status is needed for diagnosis in a peritoneal case
• Continue to use MTAP immunohistochemistry as a CDKN2A surrogate in pleural mesothelioma, where concordance was good
• Remember that retained MTAP does not exclude malignancy — sensitivity for CDKN2A deletion was 68.8% even in the pleura

Why it matters

A surrogate marker that holds in one site and not another is worse than no surrogate, because the report reads identically either way.

Don't overread it

Twenty-four deliberately selected cases cannot give the real-world accuracy of MTAP immunohistochemistry on consecutive peritoneal specimens.

The statistics, in plain English

Cohen's kappa measures agreement beyond chance: 1.00 is perfect, 0.613 acceptable, and 0.21 close to none. The peritoneal kappa of 0.21 came with a P value of 0.36, so agreement was not distinguishable from chance — but with only 24 cases this is also a small sample, and the estimate is imprecise. The cases were deliberately selected by known CDKN2A and MTAP status rather than consecutively, so the sensitivity and specificity figures describe this designed comparison and not the performance you would see on an unselected run of peritoneal specimens.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.