Microsatellite Peak Shifts in Polymerase Chain Reaction-Based Fragment Length Data Correlate With Microsatellite Instability Degree and Vary With Mismatch Repair Gene Defects and Tumor Size.

Objective: Microsatellite instability (MSI) arises from mismatch repair-deficiency (MMR-d) and is a predictor of immune checkpoint blockade (ICB) therapy response. Although MSI diagnostics typically yield a binary classification (MSI or microsatellite stable), the molecular phenotype likely represents a continuum. We explored whether MSI follows a quantitative spectrum, reflecting the extent and severity of microsatellite alterations, and whether it holds clinical significance.

Methods: To quantitatively assess MSI, we evaluated the length of MSI peak shifts from polymerase chain reaction-based fragment length data in two cohorts (combined N = 237) of hereditary (Lynch syndrome) and sporadic MMR-d colorectal carcinomas (CRCs). We examined whether MSI peak shift lengths in diagnostic markers BAT25, CAT25, BAT26, and BAT40 correlate with specific MMR defects, histologic and clinical features, and coding microsatellite (cMS) mutations, potentially reflecting antigen load.

Results: MSI peak shift lengths varied among MMR-d CRCs and were influenced by the specific underlying MMR defect, such as germline mutations in MLH1, MSH2, MSH6, or PMS2, or somatic MLH1 promotor hypermethylation. Hereditary MSH6-deficient CRCs exhibited shorter peak shifts compared with hereditary MLH1-deficient and MSH2-deficient CRCs. Longer MSI peak shifts were associated with larger tumor sizes (odds ratio of 1.3 for every 1-cm increase in tumor diameter) and number of cMS mutations per tumor (P = 8.73e-09).

Conclusions: Our findings demonstrate that (1) MSI peak shift lengths differ among MMR-d CRCs, (2) this variation is influenced by the specific MMR defect, and (3) this variation correlates with tumor size and number of cMS mutations. These findings suggest that a quantitative MSI classification could enhance clinical utility, with potential, for instance, in predicting disease progression and ICB therapy response.