Predicting Survival in Prostate Cancer Patients with Interpretable Artificial Intelligence

Background: The management of prostate cancer (PCa) that has been diagnosed after PSA testing relies on risk stratification. Treatments have significant urinary, digestive and sexual toxicities that can impact the patients’ quality of life. Many patients do not benefit from treatment because their cancer is indolent or because they die from other competitive causes. New models are needed to stratify patients at diagnosis and decide whether treatment could have an interest or not. Artificial Intelligence (AI) techniques can provide solutions in that setting. Methods: We used data from the prospective clinical trial PLCO and selected patients who were diagnosed with PCa during follow-up. We trained two gradient-boosting models to predict 10-year cancer-specific (CSS) and overall survival (OS) with features that described the general health status of the patients and PCa. Using shapley values, we provided a graphical and interpretable prediction, at the individual level. Findings: During follow-up, 8,776 patients were diagnosed with PCa. The dataset was split into a training (n=7,021) and a testing (n=1,755) dataset. Accuracy was 0.87 (± 0.02) and 0.98 (± 0.01) for OS and CSS respectively. The area under the receiver operating characteristic was 0.84 (± 0.02) and 0.81 (± 0.04) and the area under precision-recall was 0.6 (± 0.03) and 0.55 (± 0.07) for OS and CSS respectively. The models are available online. Interpretation: The clinical question of who to treat and who to observe is a major unmet need in prostate cancer. Using prospective data, we trained two models to predict 10-year CSS and OS with high accuracy. These models can be used online to provide predictions and support informed decision-making in PCa treatment. AI interpretability provides a novel understanding of the predictions to the clinicians and the patients using the models. Funding Statement: None. Declaration of Interests: The authors declare no conflict of interest. Ethics Approval Statement: The authors stated that since this was a secondary analysis of a previously approved clinical trial, no ethics committee approval was required.