Abstract C052: Integrated pharmacological analysis of patient-derived colorectal cancer tumoroids reveals strong predictive associations to approved therapies and novel combination therapies

Abstract Background. Colorectal cancer (CRC) ranks as the third most diagnosed cancer and it is the second most common cause of cancer death wordwide. The incidence is at 1.9 million cases with 900 000 deaths annually, and is increasing dramatically due to aging and growing populations, and changing lifestyle patterns. Despite recent advances, over- and undertreatment are major obstacles to significantly improve the outcome of these patients. Hence, there is an urgent need for more effective and personalized treatments. Patient-derived tumoroid (PDT) models are representative of the clinical landscape of CRC and can be used to individualize therapy and discover and develop new therapies. Since a significant number of patients do not respond to standard-of-care drugs we hypothesised that a functional precision oncology approach using clinical-grade drug screening could be used to identify and quantify the percentage of patients who are likely to benefit fom available therapies and to design effective combination therapies for patients with multiresistant tumors. Materials and Methods. We created a biobank of PDTs from primary CRC (N=56) to evaluate the pharmacological landscape of standard-of-care, and clinically relevant off-label and investigational drugs (N=35). Quantitative pharmacoproteomic and pharmacogenetic profiling was performed using the iCAN platform to identify predictive relationships using digital image analysis of fluorescence-based high-content imaging coupled with spatial profiling of functional biomarkers stained with multiplex immunocytochemistry. A set of positive control cell lines was used to document the clinical validity the functional platform. Results. The drug screening quality across the dataset of unselected PDT models was excellent, with a median Z-factor at 0.70 (range 0.27-0.85) and a mean RMSE at 0.076. Known pharmacogenetic associations to microsatellite instability, various KRAS mutations, BRAF, TP53, FGFR, HER2, ALK and NTRK mutation status were validated. Exeptional responders to the proteasome inhibitor carfilzomib and the CDK2 inhibitor fadraciblib were identified. A significant number of models were only sensitive to off-label or investigational drugs and about 25% of models were multi-resistant. Functional testing of rational combinational therapies identified a strong synergistic relationship between the chemotherapy methotrexate and the HMG-CoA reductase inhibitor atorvastatin. Pharmacoproteomic analyses of a set of potential predictive protein biomarkers (N∼30) showed a strong association between expression of integrin beta 4 (ITGB4) and sensitivity to the vinca alkaloid vinorelbine. Conclusions. Clinical-grade drug screening of PDTs confirmed pharmacogenetic relationships for a range of targeted drugs and identified atorvastatin and methotrexate as a promising combination therapy for CRC. ITGB4 protein expression was identified as a predictive biomarker for response to vinorelbine in CRC. Citation Format: Jarle Bruun, Peter W. Eide, Christer A. Andreassen, Katja Valimäki, Jose LS. Camacho, Silje S. Landmark, Øyvind Bruland, Anne H. Ree, Knut Magne Augestad, Teijo Pellinen, Sebastian Meltzer. Integrated pharmacological analysis of patient-derived colorectal cancer tumoroids reveals strong predictive associations to approved therapies and novel combination therapies [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2025 Oct 22-26; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2025;24(10 Suppl):Abstract nr C052.