Predicting Drug Response and Synergy Using a Deep Learning Model of Human Cancer Cells
Brent M Kuenzi1, Jisoo Park1, Samson H Fong2
1Division of Genetics, Department of Medicine, University of California San Diego, La Jolla, CA 92093, USA.
Abstract:
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.
Insights
DrugCell, an interpretable deep learning model, accurately predicts cancer drug responses and identifies synergistic drug combinations by analyzing tumor genotypes and drug structures. This approach enhances predictive medicine and aids in clinical trial success.
Area of Science:
- Computational Biology
- Pharmacology
- Machine Learning
Background:
- High failure rates in drug clinical trials stem from poor understanding of drug response mechanisms.
- Current machine learning models lack interpretability and focus on monotherapies, limiting clinical application.
Purpose of the Study:
- To develop DrugCell, an interpretable deep learning model for predicting human cancer cell drug response.
- To elucidate the biological mechanisms underlying drug response and identify synergistic drug combinations.
Main Methods:
- Trained DrugCell on 1,235 tumor cell lines and 684 drugs, integrating tumor genotypes with drug structures.
- Utilized deep learning for predictive modeling and mechanism discovery.
- Validated synergistic drug combinations using combinatorial CRISPR, in vitro screening, and patient-derived xenografts.
Main Results:
- DrugCell demonstrated accurate drug response predictions in cell lines.
- Model predictions successfully stratified clinical outcomes.
- Identified and validated synergistic drug combinations with significant therapeutic potential.
Conclusions:
- DrugCell offers an interpretable framework for predictive medicine in oncology.
- The model facilitates the discovery of novel therapeutic strategies, including synergistic drug combinations.
- This approach provides a blueprint for developing interpretable AI models in clinical practice.
More Related Videos
07:46Microfluidic Co-Culture Models for Dissecting the Immune Response in in vitro Tumor Microenvironments
Published on: April 30, 2021
09:53Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography
Published on: August 16, 2020
