Computational Modeling of Drug Response Identifies Mutant-Specific Constraints for Dosing panRAF and MEK Inhibitors in Melanoma

Purpose: This study investigates the potential of pre-clinical in vitro cell line response data and computational modeling to determine optimal dosage requirements for pan-RAF (Belvarafenib) and MEK (Cobimetinib) inhibitors in melanoma treatment. Our research is driven by the importance of drug combinations in enhancing anti-cancer responses and the need to address the knowledge gap surrounding effective dosing strategies to maximize their potential.

Results: In a drug combination screening of 43 melanoma cell lines, we identified specific dosage landscapes for pan-RAF and MEK inhibitors in NRAS versus BRAF mutant melanomas. Both types showed benefits, but NRAS mutant melanomas exhibited a notably more synergistic effect within a narrower dosage range (mean Bliss score of 0.27 for NRAS compared to 0.1 for BRAF mutants). Computational modeling and subsequent molecular experiments attributed this difference to a mechanism of adaptive resistance involving negative feedback. We validated the applicability of our in vitro dose-response maps in vivo by accurately predicting tumor growth in xenografts, capturing both cytostatic and cytotoxic responses. Additionally, we analyzed pharmacokinetic and tumor growth data from Phase 1 clinical trials of Belvarafenib combined with Cobimetinib, demonstrating that the synergy requirement imposes stricter dosing constraints for NRAS mutant melanoma patients.

Conclusion: By leveraging pre-clinical data and computational modeling, our approach proposes dosing strategies that optimize synergy in drug combinations while addressing the real-world challenges of maintaining precise dose ranges. Overall, this work provides a framework to assist in dose selection for drug combinations.