Prevalence and clinicopathological correlates of MSI/dMMR in colorectal cancer: a prospective multicenter registry study from Russia

Résumé fourni par la source

Background: Microsatellite instability/mismatch repair deficiency are well-established predictive biomarkers for immune checkpoint inhibitor therapy and are used to identify candidates for Lynch syndrome screening in colorectal cancer. In Russia, population-level data on their prevalence are limited - most published series are small, retrospective, or enriched for metastatic disease. Methods: We performed a pre-specified cross-sectional analysis within a prospective non-interventional registry that enrolled consecutive, unselected patients with primary colorectal adenocarcinoma at six specialised centres in Moscow between August 2022 and December 2025. Of 2,140 patients registered, 2,101 were included after exclusion of synchronous multiple primary tumours. Mismatch repair status was determined by immunohistochemistry, polymerase chain reaction-based microsatellite analysis, or both, following each centre's standard diagnostic protocol. Logistic regression was used to identify independent clinicopathological predictors of mismatch repair deficiency. Results: Mismatch repair deficiency was detected in 220 patients, giving an overall prevalence of 10.5% (95% confidence interval 9.2-11.8%). Prevalence was 23.2% in right-sided tumours, 6.2% in left-sided tumours, and 2.8% in rectal tumours (p<0.001). Among graded tumours, poorly differentiated cases had a rate of 32.4% versus 8.4% in low-grade tumours (p<0.001). Female patients accounted for 70.0% of mismatch repair-deficient cases versus 50.1% of proficient cases (p<0.001). On multivariate analysis, right-sided location (odds ratio 3.67; 95% confidence interval 2.57-5.31), high-grade histology (odds ratio 4.15; 95% confidence interval 2.79-6.13), and female sex (odds ratio for male sex 0.49; 95% confidence interval 0.34-0.69) were independently associated with mismatch repair deficiency, while advanced stage (odds ratio 0.64; 95% confidence interval 0.46-0.89) and older age (odds ratio 0.98 per year; 95% confidence interval 0.97-1.00) showed inverse associations. Despite these associations, a meaningful proportion of mismatch repair-deficient cases occurred outside conventional high-risk clinicopathological groups. Conclusion: Mismatch repair deficiency prevalence in unselected Russian colorectal cancer patients matches international estimates. Clinicopathological features alone are not sufficient to identify all affected patients. To our knowledge, this is the largest prospective multicentre series from Russia to date. Our data support universal mismatch repair testing rather than selective testing based on clinical or pathological criteria (ClinicalTrials.gov, NCT05495776).

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.