Discovery of a functional biomarker for sensitivity to poly ADP ribose polymerase inhibitors (PARPi) in colorectal cancer (CRC).

150 Background: Precision oncology is commonly defined as using information encoded by the tumor genome for the prediction of therapy response. PARPi are standard therapy for multiple tumor types with homologous recombination repair (HRR) defects, but these genomic alterations poorly predict PARPi response in CRC. TP53 status may influence PARPi sensitivity. Three-dimensional patient-derived tumoroids (PDTs) could identify CRC patients likely to benefit from PARPi beyond traditional genetic biomarkers. Methods: The TargetCRC study (NCT05401318) is a prospective biomarker investigation study involving CRC patients eligible for curative surgery. In collaboration with the biotechnology company Oncosyne, we employed their iCAN drug testing platform on a cohort of PDTs for pharmacoproteomic and pharmacogenetic profiling. This included targeted gene sequencing with the TruSight Oncology 500 gene panel (TSO500), microscopy, and multiplex immunohistochemistry. Drug sensitivity was measured by half maximal inhibitory concentration (IC 50 ). Results: Between March 2022 and December 2024, we tested 56 PDT models from individual patients for sensitivity to the PARPi olaparib and talazoparib. The median age of the 56 patients was 68 years (range, 38-86) with 19 (34%) females. Standard diagnostic evaluations revealed that 31 tumors (55%) had KRAS/NRAS mutations, while one tumor harbored BRAF V600E and three samples (5%) exhibited microsatellite instability. 5/56 (8.9%) PDT models showed sensitivity to olaparib higher than that of a positive control cell line (HGC-27, BRCA2 mut , TP53 mut ; IC 50 1.3 µM). Overall, median IC 50 was 6.0 µM (P25-P75, 2.9-NA). For the three most sensitive PDTs, olaparib IC 50 was 0.29-0.69 µM. Talazoparib sensitivity was concordant. All three PDTs carried alterations in HRR genes and 2/3 were TP53 mut . Specifically, we identified a likely germline BRCA1 E1060X variant in one patient while the two remaining harbored somatic BRIP1 D791V and BCRA1 L147F , respectively. One PDT with high olaparib sensitivity was also sensitive towards multiple TOP-inhibitors, including SN38 (irinotecan). There was a negative, but non-significant association between TP53 protein expression and PDT response to olaparib. Conclusions: We have developed a functional precision medicine platform for identifying tumors sensitive to PARPi. Our findings show that HRR deficiencies create therapeutic vulnerabilities to PARPi in CRC and suggest that the PDT findings might serve as predictive biomarkers with significant clinical potential. Given that single-agent PARPi show minimal activity in CRC, combination approaches may be necessary for clinical efficacy. The utility of this platform will be validated in a randomized clinical trial launched in September 2025.