Abstract
Abstract Colorectal cancer (CRC) exhibits high genetic heterogeneity and frequent resistance to single-agent therapies, highlighting the need for rationally designed drug combinations. We apply a mechanistic modelling framework to prioritise synergistic combinations in 16 CRC cell lines. Simulations of 210 double perturbations among 21 targeted inhibitors guided the selection of 17 combinations for experimental validation. Of 704 tested conditions, 68% showed synergy, most involving PI3K/AKT/MTOR-related inhibitors. We report six novel compound combinations that inhibit CRC cell-line growth by targeting MTOR and ERK, or MTOR and BCL2. In silico analysis indicates that feedback suppression via S6K–FOXO3 mediates the MTORi–ERKi synergy, whereas apoptotic reinforcement explains the MTORi–BCL2i effect. These results demonstrate that model-guided prioritisation enables the efficient discovery of synergistic drug pairs, mechanistic insight into their action, and a scalable route to rational combination-therapy design in CRC.