Skip to main content

Real-world comparison of routine and companion diagnostic mismatch repair immunohistochemistry in a large Chinese pan-tumour cohort

In brief

A two-protein mismatch repair test could miss 33 deficient tumors

In 1,443 tumor specimens from five Chinese hospitals, routine mismatch repair results agreed less well with the approved four-protein test in some tumor types, and a two-protein approach was estimated to miss 33 deficient cases. A third test supported the approved classification in 31 of 34 disagreements, underscoring why full-panel testing and consistent interpretation matter for treatment selection.

Journal
Histopathology (Q1)
Published
19 November 2026
Study design
Retrospective cohort
Evidence level
Level 3, Low (CEBM 3b)
Authors
Shumei Yan, Dong Kuang, Yong Zhang, Qiang Liu, Hong Zou, Guoping Wang, et al.
PMID
43059199
DOI
10.1111/his.70315

Why clinicians should know about it

  • Picked for Pathology and Forensic Medicine (top studies of the week, 22 November 2026): Directly compares MMR immunohistochemistry classifications and interpretation

Abstract

AIMS: DNA mismatch repair (MMR) immunohistochemistry (IHC) is widely used for Lynch syndrome screening and selection of patients for immune checkpoint inhibitor (ICI) therapy. Standardized companion diagnostic (CDx) testing and interpretation frameworks support consistent patient classification. We evaluated routine clinical-record MMR IHC results against the Food and Drug Administration-approved CDx, VENTANA MMR RxDx Panel (Roche Diagnostics), and examined focal staining and reduced-protein testing strategies. METHODS AND RESULTS: This retrospective multicentre study included 1443 resection specimens from five hospitals in China: colorectal carcinoma (CRC; n = 402), endometrial carcinoma (EC; n = 413), gastric carcinoma (GC; n = 367) and other solid tumours (n = 261). Concordance between historical clinical-record MMR status and the CDx classification varied by tumour type. The highest agreement was observed in CRC and GC, whereas EC and the heterogeneous other-tumour subgroup showed the lowest positive per cent agreement (95.81%) and the lowest negative per cent agreement (92.98%), respectively. Third-panel assessment of 34 discordant cases supported the VENTANA classification in 31 cases, while the three remaining cases were unresolved without orthogonal molecular testing. Subclonal loss was identified in 74 of 525 dMMR cases (14.1%; most in MSH6 and MLH1), highlighting interpretative complexity beyond a binary framework. Reduced two-protein testing, although operationally attractive, may miss clinically meaningful dMMR cases: the best-performing PMS2/MSH6 combination was estimated to miss 33 dMMR cases when assessed against the VENTANA four-protein classification. CONCLUSIONS: These findings support full-panel MMR assessment and adherence to CDx-specific interpretation guidance to improve confidence in patient selection for ICI therapy.

Abstract as published, via PubMed.

View on PubMedFull text at the publisherOpen in the app

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.