Predicting Drug Response and Synergy Using a Deep Learning Model of Human Cancer Cells

• Published in: Cancer cell 2020-11, Vol.38 (5), p.672-684.e6
• Main Authors: Kuenzi, Brent M., Park, Jisoo, Fong, Samson H., Sanchez, Kyle S., Lee, John, Kreisberg, Jason F., Ma, Jianzhu, Ideker, Trey
• Format: Article
• Language: English
• Subjects: Antineoplastic Agents - pharmacology; Antineoplastic Agents - therapeutic use; cancer; Cell Line, Tumor; Computational Biology - methods; Databases, Factual; Deep Learning; Drug Screening Assays, Antitumor; Drug Synergism; drug synergy; Genotype; Humans; interpretable deep learning; machine learning; Neoplasms - drug therapy; Neoplasms - genetics; network modeling; Patient-Specific Modeling; precision medicine
• ISSN: 1535-6108; 1878-3686
• DOI: 10.1016/j.ccell.2020.09.014

Most drugs entering clinical trials fail, often related to an incomplete understanding of the mechanisms governing drug response. Machine learning techniques hold immense promise for better drug response predictions, but most have not reached clinical practice due to their lack of interpretability and their focus on monotherapies. We address these challenges by developing DrugCell, an interpretable deep learning model of human cancer cells trained on the responses of 1,235 tumor cell lines to 684 drugs. Tumor genotypes induce states in cellular subsystems that are integrated with drug structure to predict response to therapy and, simultaneously, learn biological mechanisms underlying the drug response. DrugCell predictions are accurate in cell lines and also stratify clinical outcomes. Analysis of DrugCell mechanisms leads directly to the design of synergistic drug combinations, which we validate systematically by combinatorial CRISPR, drug-drug screening in vitro, and patient-derived xenografts. DrugCell provides a blueprint for constructing interpretable models for predictive medicine. [Display omitted] •Development of an interpretable deep learning model of human cancer cells•Model interpretations represent synergistic drug combination opportunities•Predicted combinations improve progression-free survival in PDX models•Response predictions stratify ER-positive breast cancer patient clinical outcomes Kuenzi et al. develop DrugCell, an interpretable deep learning model that simulates the response of human cancer cells to therapy. DrugCell predictions might generalize to patient tumors and can be used to design synergistic drug combinations that significantly improve treatment outcomes.