Towards Realizing the Vision of Precision Medicine: AI Based Prediction of Clinical Drug Response

Accurate and individualized prediction of response to therapies is central to precision medicine. However, due to the generally complex and multi-faceted nature of clinical drug response, realizing this vision is highly challenging and requires integration of different types of clinical and genomic data from the same individual into one machine learning based prediction model. Outside the highly specific field of oncology, no such models currently exist. In our work, we used the anti-epileptic drug brivaracetam as a case study, and combined a hybrid data-/knowledge-driven feature extraction with advanced machine learning to systematically integrate clinical and genetic data from a clinical discovery dataset (n = 235 patients). As such, we constructed a model that successfully predicted clinical drug response (cross-validated AUC = 0.76; external validation dataset: AUC = 0.75), and thus showed that even with limited sample size, integrating high-dimensional genetics data with clinical data for informing drug response prediction is possible. We then explored the potential impact of our model on clinical trial design and demonstrated that, by enriching for probable responders, significant reductions in clinical study sizes may be achieved. To our knowledge, our model represents the first retrospectively validated machine learning model for clinical drug response prediction outside the field of oncology. Hence, it provides a blueprint for how machine learning-based multimodal data integration can act as a driver in achieving the goals of precision medicine. Funding Statement: This project was funded by UCB Pharma. UCB Pharma was responsible for the project design, and collection and analysis of the data. The authors, who are UCB Pharma employees, were responsible for data interpretation, revising the manuscript for intellectual content, and approving of the manuscript for submission. Declaration of Interests: All authors received salaries from UCB Pharma. UCB Pharma had no influence on the content of this work.