At the ASCO 2026 Annual Meeting, Agenus Inc. (Nasdaq: AGEN) and Noetik showcased groundbreaking data linking artificial intelligence (AI) analysis of routine pretreatment tumor pathology images to treatment responses and survival outcomes in patients with microsatellite stable (MSS) metastatic colorectal cancer (mCRC). Utilizing Noetik's TARIO-2 platform, researchers analyzed hematoxylin and eosin (H&E) images from 113 patients treated with the investigational immunotherapy combination of botensilimab (BOT) and balstilimab (BAL). The AI model identified specific spatial tumor microenvironment patterns that correlated with clinical outcomes, revealing a 64% response rate in the AI-identified subgroup compared to only 9% in the remaining cohort. Furthermore, this subgroup exhibited significantly improved overall survival, with a hazard ratio of 0.18. These findings underscore the potential of AI to refine patient selection for immunotherapy, particularly in tumor types resistant to conventional treatments. The data, presented by Ryan Dalton, Ph.D., during a poster session, paves the way for prospective validation of TARIO-2 as a practical biomarker strategy. As Agenus continues to develop BOT+BAL, this innovative approach may enhance treatment efficacy for patients with limited options.