Speaker
Description
Quantitative Systems Pharmacology (QSP) models are increasingly employed to generate virtual populations (VPop), particularly in the field of immuno-oncology (IO). These populations can be integrated into existing drug development frameworks to reduce the number of clinical trials, support the selection and validation of novel therapeutic targets, and optimize combination therapies.
Leveraging dynamical systems analysis, we have developed an original methodology that characterizes therapeutic efficacy through phase trajectory analysis, grounded in the concept of attraction basins. This approach enables the identification of false positives and false negatives.
We also address the challenge of non-identifiability by introducing a framework that uniquely associates virtual patient responses to therapeutic protocols with their underlying parametric signatures.
Finally, we demonstrate the value of employing advanced sampling techniques to enhance the quality and representativeness of virtual populations.
Bibliography
1) Schirru, M; Brier, T; Petit, M; Zugaj, D; Tremblay, PO; Nekka, F. (2025). Generation of Virtual Populations for Quantitative Systems Pharmacology Through Advanced Sampling Methods. Bulletin of Mathematical Biology. 87(11): 165-187.
2) Zugaj, D; Nekka, F. (2025). Identification and characterization of virtual sub-populations through phenotype-guided filtering. The challenging case of nonidentifiable models in the context of therapeutic evaluation. Journal of Pharmacokinetics and Pharmacodynamics. 53(1): 1-18.
3) Schirru, M, Charef, H, Ismaili, KE et al. Predicting efficacy assessment of combined treatment of radiotherapy and nivolumab for NSCLC patients through virtual clinical trials using QSP modeling. J Pharmacokinet Pharmacodyn 51, 319–333 (2024). https://doi.org/10.1007/s10928-024-09903-0.
4) Zugaj D, Fenneteau F, Tremblay PO, Nekka F. Dynamical behavior-based approach for the evaluation of treatment efficacy. The case of immuno-oncology. Chaos. 2024 Jan 1;34(1):013142. doi: 10.1063/5.0170329. PMID: 38277131.
