Gap-App: A sex-distinct AI-based predictor for pancreatic ductal adenocarcinoma survival as a web application open to patients and physicians.

In this study, using RNA-Seq gene expression data and advanced machine learning techniques, we identified distinct gene expression profiles between male and female pancreatic ductal adenocarcinoma (PDAC) patients. Building on this insight, we developed sex-specific 3-year survival predictive models alongside a single comprehensive model. Despite smaller sample sizes, the sex-specific models outperformed the general model. We further refined our models by selecting the most important features from the initial models. The refined sex-specific predictive models achieved higher accuracy and consistently outperformed the refined comprehensive model, highlighting the value of sex-specific analysis. To ensure robustness, all refined sex-specific models were calibrated and then evaluated using an independent dataset. Random Forest models emerged as the most effective predictors, achieving accuracies of 90.33 % for males and 90.40 % for females on the training dataset, and 81.25 % for males and 89.47 % for females on the independent test dataset. These top-performing models were integrated into Gap-App, a web application that leverages individual gene expression profiles and sex information for personalized survival predictions. As the first online tool bridging complex genomic data with clinical application, Gap-App facilitates more precise, individualized cancer care, marking a significant step in personalized prognosis prediction. This study underscores the importance of incorporating sex differences in predictive modeling and sets the stage for the shift from traditional one-size-fits-all to more personalized and targeted medicine. The Gap-App service is freely available for patients and clinicians at www.gap-app.org.