Deep survival learning for prognosis prediction in non-metastatic castration-resistant prostate cancer

BackgroundNon-metastatic, castration-resistant prostate cancer (nmCRPC) is an advanced state of prostate cancer with variable prognosis; early identification of patient risk is crucial, so that clinicians can recommend optimal treatment.ObjectiveCompare predictive models in identifying patient risk; evaluate the value of electronic healthcare record (EHR) time-series (TS) information in prediction.MethodsWe evaluated SurvTRACE, Weibull Time to Event Recurrent Neural Network (WTTE-RNN), and traditional Cox proportional hazards (CPH) models' performance on EHR data from 12,819 nmCRPC patients in the Veterans Health Administration, using area under the receiver operating characteristic curve and Brier score.ResultsWTTE-RNN, which intrinsically uses EHR TS information, outperformed the other models without TS information. Feature-engineered TS information improved performances of CPH and especially SurvTRACE; with TS information, SurvTRACE outperformed WTTE-RNN.ConclusionDeep learning methods, whether intrinsically able to handle TS data or enhanced with TS information, can outperform traditional survival analysis in predicting risk.