Mismatch-repair IHC alone is not enough in GI and pancreaticobiliary cancers

Design

Single-centre real-world case series with multimodal molecular review

Population

91 mismatch-repair-deficient upper gastrointestinal and pancreaticobiliary cancers

Primary outcome

Mismatch-repair mechanisms and concordance of IHC with ancillary tests

Effect

IHC-MSI discordant in 16%; MLH1 methylation in 80% of MLH1/PMS2-lost tumours; Lynch in 17% tested

Mismatch-repair status is both a diagnostic and a predictive biomarker, increasingly tied to immunotherapy decisions. This real-world series of 91 mismatch-repair-deficient upper gastrointestinal and pancreaticobiliary cancers examined how a single-modality workup performs.

Immunohistochemistry alone proved insufficient. Among cases with both tests, mismatch-repair immunohistochemistry and microsatellite-instability PCR were discordant in 16% (3 of 19). The deficiency arose through several mechanisms: MLH1 promoter methylation in 80% of tested MLH1/PMS2-lost tumours, Lynch syndrome in 17% of those with germline testing, and somatic mismatch-repair gene variants in 43% of tumours sequenced.

The practice point is to treat mismatch-repair testing as multimodal. Pair immunohistochemistry with microsatellite-instability testing where the result is equivocal or the clinical stakes are high, use MLH1 methylation to triage sporadic from heritable MLH1 loss, and route appropriate cases to germline testing — so neither a false biomarker call nor a missed Lynch diagnosis slips through.

• Series of 91 mismatch-repair-deficient upper gastrointestinal and pancreaticobiliary cancers.
• Immunohistochemistry and microsatellite-instability PCR were discordant in 16% of tested cases.
• MLH1 promoter methylation explained 80% of MLH1/PMS2-lost tumours; Lynch syndrome in 17% of those tested.
• Somatic mismatch-repair variants were found in 43% of sequenced tumours.
• Pair immunohistochemistry with MSI, MLH1 methylation and, where indicated, germline testing.

Why it matters

A mismatch-repair call now drives immunotherapy eligibility and Lynch screening, so a single-test error has direct consequences for treatment and family risk.

Don't overread it

This is a single-centre descriptive series; it shows single-modality testing is unreliable here, not a precise discordance rate for every setting.

The statistics, in plain English

A 16% discordance between immunohistochemistry and MSI means roughly one in six single-test calls could be wrong; the varied mechanisms explain why no single assay captures every case.

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.