Artificial Intelligence Cracks the Code on Treatment-Resistant Bowel Cancer Using Repurposed Melanoma Drug
New AI mapping technique identifies drug vulnerabilities in the 85% of colorectal cancers that resist standard therapies, potentially fast-tracking clinical trials through drug repurposing.
Approximately 80-85% of bowel cancer patients face a particularly stubborn adversary: microsatellite stable (MSS) colorectal cancer, a subtype notorious for its resistance to conventional treatments. Now, an international research collaboration has deployed artificial intelligence to map these tumours at unprecedented resolution, uncovering unexpected drug vulnerabilities that could transform treatment options for thousands of patients annually.
The study, published in Cancer Cell, analyzed 2,000 patient tumour samples using a novel AI methodology called iPhenMap. Led by researchers from RCSI University of Medicine and Health Sciences and The Institute of Cancer Research, London, the work integrates genetic, metabolic, protein, and clinical data to reveal therapeutic targets that traditional analysis methods miss entirely.
AI Integration Reveals Hidden Treatment Pathways
The iPhenMap platform represents a significant advance in multiomic data integration, combining DNA, RNA, epigenetic markers, protein expression, metabolic profiles, single-cell analysis, and clinical outcomes into a unified biological map. According to Professor Annette Byrne, Head of RCSI's Precision Cancer Medicine Group, this comprehensive approach identifies connections between tumour features that isolated data streams cannot detect.
The research specifically targeted MSS RAS-mutant metastatic colorectal cancer, a patient population with particularly limited treatment options. While clinical trials combining chemotherapy with targeted drugs or immunotherapies have shown only modest improvements over standard treatments, the AI analysis identified specific drug combinations with substantially greater promise in preclinical models.
Melanoma Drug Shows Promise in Bowel Cancer Models
Among the most significant findings: trametinib, currently approved for treating melanoma, lung cancer, and thyroid cancers with specific mutations, demonstrated significant tumour growth reduction and survival extension in preclinical models when used in combination therapy for MSS colorectal cancer. This drug repurposing discovery could substantially accelerate clinical testing timelines, as trametinib's safety profile is already established in other cancer contexts.
The research team included Professor Anguraj Sadanandam from The Institute of Cancer Research, London, and Dr. Rodrigo Dienstmann from Vall d'Hebron Institute of Oncology in Barcelona and the University of Vic–Central University of Catalonia. The collaborative effort spanned institutions across Europe, including VIB KU Leuven in Belgium, Candiolo Cancer Institute in Italy, and multiple Irish clinical sites including Beaumont Hospital, Tallaght University Hospital, St. Vincent's University Hospital, and Bon Secours Cork Cancer Centre.
Implications for Precision Oncology
For Ireland alone, where approximately 2,500 bowel cancer cases are diagnosed annually, this research could impact 2,000-2,125 patients each year who develop the MSS subtype. Globally, the patient population is proportionally larger, representing a substantial unmet medical need.
The study was funded through the European Union Horizon 2020 Research and Innovation Programme's COLOSSUS project and a Research Ireland Project Award to Professor Byrne. The consortium included private sector partners Veracyte SAS, Epigenomics AG, and geneXplain GmbH, signaling commercial interest in translating these AI methodologies into clinical applications.
The iPhenMap approach could extend beyond colorectal cancer, offering a scalable framework for identifying drug repurposing opportunities across multiple treatment-resistant tumour types. For health technology companies developing AI-driven precision oncology platforms or companion diagnostics, this research demonstrates the commercial viability of multiomic integration for drug discovery and patient stratification.
Clinical validation remains necessary before these findings translate into patient access, but the identification of already-approved drugs as potential combination therapy components could significantly compress development timelines compared to novel drug discovery pathways.
Sources
Companies in this article
Google Search
Follow Health AI Central on Google
Choose Health AI Central as a preferred source, and Google is more likely to show our reporting in Top Stories.
Reporting & accountability
How we maintain accuracy
Health AI Central separates reported facts, analysis, and public-source data. Readers can review our standards, request a correction, or consult the relevant research methodology before relying on a claim.