Abstract A034: Multi-layer stratified oncology platform utilizing transcriptomics, prostate cancer organoids, and modeling of drug response

Key Points

• The aim is to develop a platform that integrates molecular profiling and functional drug screening to personalize prostate cancer treatment.
• Utilized a twin-biopsy strategy for molecular and functional analysis of lesions.
• Derived organoids from matched tumor and benign tissues with a success rate of 73%.
• Performed transcriptomic clustering to identify molecular subtypes.
• Conducted drug screening with 11 pharmacological compounds.
• Developed machine learning models to predict drug sensitivity based on transcriptomic data.
• Identified two stable molecular subtypes (C1 and C2) with distinct drug sensitivities.
• C2 lesions showed increased responsiveness to MET inhibitors, correlating with elevated MET phosphorylation.
• Machine learning models accurately predicted drug responses and stratified patients by subtype.
• Validation with TCGA data confirmed the clinical relevance of the stratification.