WALTHAM, Mass.– BostonGene, a leader in AI-powered solutions for drug discovery and development, today announced the publication of a collaborative study, “AI-driven multimodal algorithm predicts immunotherapy and targeted therapy outcomes in clear cell renal cell carcinoma,” in Cell Reports Medicine. The research introduced the largest harmonized transcriptomic and clinical dataset in kidney cancer and showcased how multimodal AI can identify new determinants of response to immunotherapy and VEGF inhibitor combinations in solid tumors.
While immune checkpoint inhibitors (ICIs) and VEGF inhibitor tyrosine kinase inhibitors (TKIs) have reshaped treatment for metastatic clear cell renal cell carcinoma (ccRCC), patient outcomes remain variable, with frequent disease progression, treatment-related toxicities, and little guidance on when to use combination versus single-agent therapy.
Leveraging more than 3,600 patient samples and harmonized clinical data, researchers developed a multimodal foundation model integrating genomics, transcriptomics, and tumor microenvironment profiling. The model uncovered five Harmonized Immune Tumor Microenvironment (HiTME) subtypes, defined by immune infiltration patterns, genomic alterations, and prognostic outcomes. Validation with spatial proteomics confirmed that predictions were grounded in tumor biology rather than functioning as a “black box.”
By applying these HiTME subtypes, the study generated clinically interpretable responder scores that correlated with survival outcomes for ICIs and TKIs across independent cohorts. This approach produced a decision-tree tool that stratifies patients into ICI-preferred, TKI-preferred, or non-responder categories, offering a new framework for therapy selection, efficient trial design, and drug development. The model also identified a resistant subgroup with immune-desert phenotypes and angiogenic signaling, highlighting unmet needs and opportunities for novel therapies.
“This study demonstrates how multimodal foundation models can reshape oncology,” said Nathan Fowler, M.D., Chief Medical Officer at BostonGene. “By grounding predictions in tumor biology rather than opaque algorithms, we enable clinicians to understand why patients may or may not respond to treatment — making AI both clinically actionable and scientifically trustworthy.”
