AI-derived tumor microenvironment features and recurrence risk in microsatellite-stable colon cancer after adjuvant chemotherapy.

Key Points

• This research aims to enhance risk prediction for recurrence in microsatellite-stable colon cancer post-chemotherapy using AI-derived tumor microenvironment features.
• Retrospective analysis of tumor microenvironment in stage II and III colon cancer patients after adjuvant chemotherapy.
• Quantification of tumor and stromal areas, and TME cell types using AI-based Lunit SCOPE IO.
• Development of a generalized linear model integrating AI-derived features with conventional clinicopathological variables.
• High stromal lymphocyte density associated with favorable disease-free survival (DFS) (HR 0.42, 95% CI 0.26 – 0.68).
• The TME-integrated model stratified patients into high- and low-risk groups, showing 3-year DFS rates of 72.3% and 93.8%, respectively (adjusted HR 3.29, 95% CI 1.98 – 5.46).
• Validation in an independent cohort showed the model's robustness with adjusted HR for high-risk at 3.68 (95% CI 1.60 – 8.48).