Translational PBPK-QSP modeling platform for antibody-drug conjugates (ADC): within-target and cross-pathway validation to bridge preclinical and clinical results

We designed a platform model that integrates physiologically-based pharmacokinetic (PBPK) modeling with quantitative systems pharmacology (QSP) to bridge translational challenges in antibody-drug conjugate (ADC) development. The PBPK-QSP platform model was developed for the ADC trastuzumab emtansine (T-DM1) in breast cancer patients. This mechanistic framework facilitates translation across preclinical in vitro experiments, in vivo studies, and clinical trials, supporting decision-making for novel ADCs. The PBPK-QSP model adequately predicts preclinical and clinical PK and PD data from two additional ADCs: trastuzumab deruxtecan (T-Dxd) and tusamitamab ravtansine. For within-target validation with T-Dxd in breast cancer, despite extensive preclinical calibration, efficacy predictions were initially overly optimistic compared to T-DM1 validation experience with the model and aggregated phase II trial data. Individual patient data from a phase II T-Dxd trial allowed evaluation of model performance and quantification of translational uncertainty in predicting clinical outcomes using preclinical experiments. Cross-pathway validation with tusamitamab ravtansine in non-small cell lung cancer has revealed the importance of incorporating a resistance module to describe clinical efficacy adequately. Clinical trial simulations for tusamitamab ravtansine subsequently inform that alternative fractional dosing could offer a potential efficacy advantage compared to existing clinical dosing. We integrated these insights into a practical recommended workflow for translational development programs, which addresses the key challenges in parameter estimation, data requirements, and uncertainty quantification in the key system parameters for each indication and cancer type. Ultimately, integrating an interactive modeling platform with a structured workflow to mitigate the risks of human translation and to potentially improve the clinical benefits of novel ADCs in oncology drug development.