Cancer DiscovJClub
A Foundation Model of Cancer Genotype Enables Precise Predictions of Therapeutic Response.
Cancer Discovery · · Journal Article · Open access
Kong, Lee + more
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
MutationProjector, a pre-trained genomic model, predicts therapeutic resistance and sensitivity across multiple cancer types with performance matching or exceeding state-of-the-art methods.
- Why it matters: Accurately translating complex tumor mutation profiles into treatment decisions remains a major challenge in cancer care, limiting personalized therapy options.
- What they did: The model was trained on over 30,000 tumors, integrating extensive molecular knowledge to reveal altered pathways, reconstruct mutations, and predict resistance to immunotherapy and chemotherapy.
- The result: It successfully identifies novel biomarkers like KMT2D mutation for immunotherapy sensitivity and combined SMARCA4 and STK11 alterations for resistance, establishing a framework linking genotypes to outcomes.
The findingWhy it mattersWhat they didThe result
- Open access