Development of a computational histology artificial intelligence (CHAI)–powered predictive biomarker for first-line chemotherapy intensification in metastatic colorectal cancer (mCRC) and validation in prospective randomized phase III trials (RCT).

Key Points

• To develop and validate a histology-based biomarker using AI for predicting outcomes in metastatic colorectal cancer treated with different chemotherapy regimens.
• Used H&E-stained whole slide images from participants in the MRC FOCUS randomized clinical trial.
• Developed a predictive biomarker by optimizing histologic feature associations with progression-free survival and overall survival.
• Evaluated the biomarker on independent validation cohorts from TRIBE and TRIBE2 trials using multivariable Cox models.
• Biomarker (+) cases had superior 3-year progression-free survival (17% vs 7%, HR = 0.51 [0.37, 0.70], p < 0.001) and overall survival (43% vs 22%, HR = 0.51 [0.35, 0.73], p < 0.001) with triplet therapy.
• Biomarker (-) cases showed no significant differences in progression-free survival (5% vs 15%, HR = 1.30 [0.96, 1.74], p = 0.09) or overall survival (31% vs 40%, HR = 1.35 [0.94, 1.92], p = 0.1).
• The biomarker-treatment interaction was significant (p < 0.001), indicating a predictive association between the biomarker and treatment regimen.