Abstract
Prostate cancer (PC) remains a worldwide challenge, as does the question of how to distinguish its indolent from its aggressive form to reconcile proper management of the disease with age-related life expectations. This study aimed to differentiate the Notch-driven course of PC regarding patients’ ages and stage of their disease. We analyzed 397 PC samples split into age subgroups of ≦55, 60−70, and >70 years old, as well as early vs. late stage. The clinical association of Notch signaling was evaluated by DFS and UpSet analyses. The clustering of downstream effectors was performed with ExpressCluster. Finally, for the most relevant findings, functional networks were constructed with MCODE and stringApp. The results have been validated with an independent cohort. We identified specific patterns of Notch expression associated with unfavorable outcomes, which were reflected by entering into a hybrid epithelial/mesenchymal state and thus reaching tumor plasticity with its all consequences. We characterized the molecular determinants of the age-related clinical behavior of prostate tumors that stem from different invasive properties depending on the route of the EMT program. Of the utmost relevance is the discovery of age- and stage-specific combinations of the Notch molecules predicting unfavorable outcomes and constituting a new prognostic and therapeutic approach for PCs.