Multi-omics in urologic cancers

Urologic cancers are common globally with increasing incidence and mortality over the last three decades. Despite this, the current landscape in diagnosis, treatment, and prognosis of these cancers is growing exponentially. Single-omic approaches exploring the genomic, epigenomic, transcriptomic, proteomic, and metabolomic profiles of urologic cancers have identified numerous potential biomarkers but are limited in clinical use due to poor generalizability. The integration of multiomics with artificial intelligence (AI) and machine learning (ML) algorithms is a growing contemporary area in research that combines multiple single-omic datasets to construct more robust models that elucidate mechanisms of cancer development, predict survival outcomes, treatment response, and recurrence risk, and discover novel druggable targets to personalize care. This chapter summarizes the current state of multiomic approaches applied in urologic cancers, highlights the application of ML, a subfield of AI to multi-omics research, and provides insight into emerging omic technologies, including pathomics, spatial-omics, and radiomics in urologic cancers.