Discordance between mismatch repair immunohistochemistry and molecular microsatellite instability testing in endometrial carcinoma: a systematic review and meta-analysis
In brief
Mismatch repair staining and microsatellite tests disagree in about 5% of endometrial cancers
Across 22 studies and 5,870 tumours, mismatch repair staining and molecular microsatellite testing disagreed in about 5% of cases. Most discordant tumours showed loss of mismatch repair proteins without high microsatellite instability; after review or repeat testing, only 18 of 109 became concordant. Results varied widely across studies, and further investigation did not resolve every mismatch.
- Journal
- Virchows Archiv : an international journal of pathology (Q1)
- Published
- 21 November 2026
- Study design
- Systematic review of cohort studies
- Evidence level
- Level 2, Moderate (CEBM 2a)
- Authors
- Manuel Pérez-Pérez, Carmen García de Sola-Llamas, Laura Macías-García
- PMID
- 43064825
- DOI
- 10.1007/s00428-026-04792-9
Why clinicians should know about it
- Picked for Pathology and Forensic Medicine (paper of the day, 22 November 2026): Direct comparison of MMR-IHC and molecular MSI classification
Abstract
Discordance between mismatch repair immunohistochemistry (MMR-IHC) and molecular microsatellite instability (MSI) testing may affect biomarker interpretation in endometrial carcinoma. We estimated its frequency, predominant pattern and outcomes of additional investigation. PubMed/MEDLINE, Embase and Scopus were searched through 30 August 2026. Study-specific discordance proportions were pooled using random-effects models, retaining initial valid paired results before discordance-triggered review. Sensitivity and specificity of MMR-IHC relative to molecular MSI classification were estimated jointly. Additional investigations and germline findings were described by study. Twenty-two studies included 5,870 tumours and 331 discordances. Pooled discordance was 5.49% (95% confidence interval [CI], 4.09-7.33%; 95% prediction interval [PI], 1.46-18.54%; I²=84.3%). Among discordant tumours, the pooled proportion with MMR protein loss without high-level MSI (MSI-H) was 80.63% (95% CI, 66.05-89.90%; 20 studies). Among all tumours with MMR protein loss, the pooled proportion that was molecularly microsatellite-stable or MSI-low was 15.47% (95% CI, 10.54-22.14%; 95% PI, 2.81-53.71%; 19 studies). Sensitivity of MMR-IHC for MSI-H was 97.88% and specificity was 95.49%. Following review or repeat testing of 109 discordant tumours in seven cohorts, 18 became concordant and 91 remained discordant. MMR-IHC and molecular MSI testing are not completely interchangeable. Additional investigation may change classification or identify relevant alterations without eliminating discordance. The molecular classification was the reference for accuracy estimates, not an independent measure of functional repair deficiency.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.