261 Background: Clinical trials are one of the major barriers in contemporary drug development. Functional assays based on patient-derived organoids (PDO) are a promising new tool for derisking clinical development of new therapies, but their use has been limited due to small cohort sizes and the absence of systematic validation studies. Using the largest cohort of matched PDOs, longitudinal clinical data, transcriptomic and WES data in CRC, we explore the use of large PDOs collections to characterize ADC efficacy and affinity through systematic mapping of the relationship between target antigen affinity and payload efficacy. Methods: We have assembled a collection of 85 PDOs from colorectal cancer (CRC), generated from primary tumour samples and metastatic biopsies. Patients ranged from 32 to 89 years old and had experienced an average of 1.54 lines of treatment prior to the PDO-generating tissue sampling. Using gene expression levels of well known target antigens such as CEACAM5, ERBB2, MSLN and TROP2 as a proxy for protein concentration, we explore the relationship between the target antigen affinity and the efficacy of a subset of common payloads including topoisomerase and microtubule inhibitors. This allows us to identify the relationship between two of the three main vectors of ADC efficacy in CRC. Results: We show that topoismerase inhibitor and microtubule inhibitor efficacy relates to the RNA expression level of multiple target antigens commonly used in active clinical trials in CRC. These data match the target antigen and payload couples currently under development or approved in the clinic. We are developing a tool to optimize clinical trial design. This approach will allow us to match patients more accurately with the appropriate payload, ensuring a higher likelihood of response. Conclusions: We report that large-scale PDO-based assays can be a useful tool to anticipate clinical readouts. This work suggests that well-characterized PDO collections could also help redefine patient populations, increasing the likelihood of identifying those more likely to respond to a given ADC, and thus improving the success rates of clinical studies for new treatments in CRC.
Read full abstract