Abstract 2499: Precision oncology empowered: A web platform for population pharmacokinetic modeling

Abstract Background: More than 85% of oncology clinical trial failures are due to problems in pharmacokinetics, pharmacodynamics, and inter-patient variability. In cancer research, therapeutic windows are narrow, and inability to predict drug responses across genetically and clinically diverse populations further complicates the problem. Population pharmacokinetics thus provides a solution by characterizing variability and optimizing dosing; however, it is not yet widely adopted in translational oncology because of technical complexity. Herein, we describe an advanced web-based platform that streamlines PPK model creation, validation, and simulation to optimize trial design and therapy selection in oncology. Methods: Advanced computational approaches tailored for cancer drug development are integrated into the platform. Hierarchical parameterizations for fixed effects, random effects, and residual error structures enable precise modeling of drug behavior in oncology settings. Hybrid estimation techniques combine stochastic approximation and nonlinear mixed-effects optimization for robust parameter convergence in high-dimensional cancer datasets. Real-time diagnostic plots and multi-dimensional parameter exploration refine models iteratively. Virtual trial simulations incorporate demographic and genetic heterogeneity to predict drug responses and optimize dosing strategies. Results: The platform validated, with high accuracy, a one-compartment PK model for an anti-cancer agent. Virtual trial simulations demonstrated variability in dose-response across subgroups of cancer patients and identified strategies to optimize therapeutic windows. The platform reduced computational complexity, improved precision, and enabled rapid iteration of complex models. Simulations of alternative trial scenarios identified opportunities to refine inclusion criteria and dosing strategies, offering potential to reduce oncology trial failure rates. Conclusion: This platform now offers sophisticated tools to tackle the most relevant challenges in drug development against cancer, thus allowing more successful clinical trials and personalized treatments. It is a quantum leap in precision oncology through the in-silico simulation of drug responses and optimization of trial parameters. Currently available internally, the tool will be broadly distributed following further validation. Citation Format: Ahmed Fadiel. Precision oncology empowered: A web platform for population pharmacokinetic modeling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2499.