Weishen Pan

3 weeks ago | nature.com | Weishen Pan |Zhenxing Xu |Qiannan Zhang |Ying Li |Fei Wang |Deep Hathi

Predicting treatment response is an important problem in real-world applications, where the heterogeneity of the treatment response remains a significant challenge in practice. Unsupervised machine learning methods have been proposed to address this challenge by clustering patients with similar electronic health record (EHR) data. However, they cannot guarantee coherent outcomes within the groups. Here, we propose Graph-Encoded Mixture Survival (GEMS) as a general machine learning framework to identify distinct predictive subphenotypes that guarantee coherent survival and baseline characteristics within each subphenotype. We apply our method to a real-world dataset of advanced non-small cell lung cancer (aNSCLC) patients receiving first-line immune checkpoint inhibitor (ICI) therapy to predict overall survival (OS). Our method outperforms baseline methods for predicting OS and identifies three reproducible subphenotypes associated with distinct baseline clinical characteristics and OS. Our results demonstrate that our method can provide insights in the heterogeneity of treatment response and potentially influence treatment selection. Response to cancer therapeutics is heterogenous making it hard predict despite advances in machine learning approaches. Here, the authors develop a graph neural network-based approach, Graph-Encoded Mixture Survival (GEMS), to identify ‘predictive subphenotypes’ of patients with similar baseline characteristics and survival outcomes of cancer patients using electronic health records to predict patient response to therapy.