Concordance Analysis of Microsatellite Instability via NGS and Mismatch Repair Deficiency via IHC in Endometrial and Colorectal Cancer
Assessment of mismatch repair (MMR) function provides critical guidance for diagnosis, prognosis, and therapeutic decision making in colorectal and endometrial cancers. Mismatch repair immunohistochemistry (IHC) is the routine clinical test for identifying MMR deficiency, while microsatellite instability (MSI) serves as its surrogate, detected by polymerase chain reaction or next-generation sequencing (NGS). Available data indicate a high concordance rate between these approaches in colon cancer, whereas a lower concordance has been reported in endometrial cancer. We aimed to assess the concordance rate between MMR-IHC and MSI-NGS from patients with colorectal or endometrial cancer, using IHC as the gold standard. A cohort of 520 patients (352 with endometrial cancer and 168 with colorectal cancer) were included. MMR‑IHC assessed MLH1, MSH2, MSH6, and PMS2 expression, while MSI‑NGS was determined by profiling 130 homopolymer repeat loci using the TruSight Oncology 500 panel from Illumina. While concordance was high in the colorectal cancer cohort (99%, 95% confidence interval 96-100), a lower level of agreement was observed in endometrial cancer cases (85%, 95% confidence interval 81-89). Fifty-two of 53 discordant cases exhibited MMR deficiency by IHC in the absence of detectable MSI. Forty percent of discordant cases could be explained by factors previously associated with reduced MSI levels, including mutations in DNA polymerase genes (n = 5), isolated MSH6 loss (n = 10), atypical IHC staining patterns (n = 8), and germline variants (n = 6). Additionally, the presence of genetic and epigenetic alterations (specifically, 19 cases with MLH1 promoter hypermethylation and ten with somatic or germline MMR variants) supports the interpretation that MSI calls were missed in a subset of cases. Finally, optimizing the MSI threshold enhanced detection accuracy in endometrial tumors. These findings confirm the lower concordance between MMR-IHC and MSI-NGS in endometrial cancer compared with colorectal cancer when broad panels are used, underscoring the importance of tumor-specific interpretation even within tumor-agnostic assays. Although cut-off optimization improved agreement, the evidence remains insufficient for clinical implementation, and further validation studies are needed.
