{"version":1,"type":"story","url":"https://digestai.news/story/ai-virtual-cell-model-predicts-effective-drugs-for-triplenegative-breast-cancer","json":"https://digestai.news/story/ai-virtual-cell-model-predicts-effective-drugs-for-triplenegative-breast-cancer.json","markdown":"https://digestai.news/story/ai-virtual-cell-model-predicts-effective-drugs-for-triplenegative-breast-cancer.md","slug":"ai-virtual-cell-model-predicts-effective-drugs-for-triplenegative-breast-cancer","headline":"AI virtual cell model predicts effective drugs for triple‑negative breast cancer","summary":"Triple‑negative breast cancer (TNBC) lacks hormone receptors, making treatment difficult. Researchers at Westlake University created a virtual cell model that uses proteomics data to predict how individual tumour cells respond to drugs. The model was trained on 38 million protein measurements from 18 cell lines, including 16 TNBC lines, and tested on 63 FDA‑approved drugs and 59 combinations. It achieved 88 % accuracy on 81 unseen drugs and correctly forecasted clinical outcomes for 501 patient biopsies, matching the drugs that ultimately proved effective.\n\nThis is the first time a virtual cell model has been validated in a real‑world clinical setting, offering a path toward personalized therapy for the 15–20 % of breast‑cancer patients with TNBC. By pinpointing effective drugs early, the approach could reduce trial‑and‑error treatment cycles and improve survival rates.\n\nThe study demonstrates how large‑scale proteomics and time‑course data can power AI to guide precision oncology, potentially reshaping how aggressive cancers are managed.","keyPoints":["Model trained on 38M protein measurements from 18 cell lines, 16 TNBC, predicting drug response with 88% accuracy","Validated on 501 patient biopsies, matching real‑world treatment outcomes","First virtual cell model tested clinically, enabling personalized TNBC therapy"],"whyItMatters":"By accurately forecasting drug efficacy, the AI model could streamline treatment selection for triple‑negative breast cancer, reducing ineffective therapies and improving patient outcomes in a disease that accounts for 15–20% of cases.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["Westlake University","Arc Institute"],"models":[],"people":["Tiannan Guo","Hani Goodarzi"]},"firstPublishedAt":"2026-09-09T00:00:00Z","updatedAt":"2026-09-09T00:00:00Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"Nature Machine Learning","title":"AI model predicts which breast-cancer drugs work best","url":"https://nature.com/articles/d41586-026-02845-2","publishedAt":"2026-09-09T00:00:00Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"AI virtual cell model predicts effective drugs for triple‑negative breast cancer\", 9 September 2026, https://digestai.news/story/ai-virtual-cell-model-predicts-effective-drugs-for-triplenegative-breast-cancer","publisher":"Digest AI","title":"AI virtual cell model predicts effective drugs for triple‑negative breast cancer","datePublished":"2026-09-09T00:00:00Z","url":"https://digestai.news/story/ai-virtual-cell-model-predicts-effective-drugs-for-triplenegative-breast-cancer"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}