Abstract 891: Turning data into information: Using PD3D® models to guide colorectal cancer therapy by Optim.AITM

Abstract

Abstract Colorectal cancer (CRC) is one of the most prevalent and lethal malignancies globally with up to 50 % of patients eventually progressing to metastatic disease. The mitogen-activated-protein kinase (MAPK) pathway emerges as a key player, being one of the most frequently mutated signaling pathways in the oncogenesis of CRC. However, given the numerous genomic aberrations and tumor heterogeneity in CRC, patients may benefit from combinatorial therapies, particularly those who have inoperable or metastatic tumors. In the present study, we investigated the feasibility of combining two platforms, patient-derived 3D (PD3D®) models with Optim.AI™, to identify more effective cancer therapies. PD3D® models can robustly retain the genotypic and histopathologic features of the primary patient tumor, model tumor heterogeneity and were shown to predict a patient’s drug response. Optim.AI™ is a hybrid computational-experimental platform that uses small data sets to rationally converge to optimal drug combinations within a defined drug search space. By mapping experimental data points to a second-order quadratic function, Optim.AI™ can predict every possible 531k data points and thus the cell-killing efficacy for all other possible combinations without testing each individual drug-dose combination. Two CRC PD3D® models with different mutation profiles were tested with 155 different combinations of 12 drugs at variant concentrations. The post-treatment cell viability was measured and used for Optim.AI™ analysis to evaluate and compare the best therapies. Optim.AI™ analysis revealed differential drug sensitivity between tested CRC PD3Ds®. The top-ranked drug combinations included SN-38, active compound of commonly used chemotherapeutic irinotecan, paired with MEK-inhibitors trametinib or cobimetinib which we could confirm with both platforms. With this study, we successfully demonstrated the feasibility, the robustness, and the efficiency of combining PD3Ds® and Optim.AI™ in identifying effective drug combination therapies, here for CRC, within one month. This combined technology provides precise insights into tumor treatability and its functional causes of treatment outcomes, leading to new treatment combinations and accelerating the development of new cancer drugs in a time- and cost-effective manner. Citation Format: Ulrike Pfohl, Masturah Mohd Abdul Rashid, Jhin Jieh Lim, Juergen Loskutov, Lena Wedeken, Edward Kai-Hua Chow, Hugo Saavedra, Christoph Reinhard, Christian R. Regenbrecht. Turning data into information: Using PD3D® models to guide colorectal cancer therapy by Optim.AITM [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 891.