Improving the generation and selection of virtual populations in quantitative systems pharmacology models

Abstract

Quantitative systems pharmacology (QSP) models aim to describe mechanistically the pathophysiology of disease and predict the effects of therapies on that disease. For most drug development applications, it is important to predict not only the mean response to an intervention but also the distribution of responses, due to inter-patient variability. Given the necessary complexity of QSP models, and the sparsity of relevant human data, the parameters of QSP models are often not well determined. One approach to overcome these limitations is to develop alternative virtual patients (VPs) and virtual populations (Vpops), which allow for the exploration of parametric uncertainty and reproduce inter-patient variability in response to perturbation. Here we evaluated approaches to improve the efficiency of generating Vpops. We aimed to generate Vpops without sacrificing diversity of the VPs' pathophysiologies and phenotypes. To do this, we built upon a previously published approach (Allen et al., 2016) by (a) incorporating alternative optimization algorithms (genetic algorithm and Metropolis-Hastings) or alternatively (b) augmenting the optimized objective function. Each method improved the baseline algorithm by requiring significantly fewer plausible patients (precursors to VPs) to create a reasonable Vpop.

Pharmacodynamic Drug–Drug Interactions

“…Employing this paradigm, only one treatment combination (or a small number of combinations) that was predicted to result in significantly altered clinical endpoints compared with single‐drug treatments would require clinical testing. Virtual patient populations, generated by simulation using identified covariates and proper parameter ranges, could be used to assess interpatient variability in such QSP models applied to clinical studies …”

Multiscale Agent-Based and Hybrid Modeling of the Tumor Immune Microenvironment

“…The challenge is that available data for individual patients is limited. To address this problem, machine learning approaches can be used to build statistical models based on available patient data, and these models can be employed to simulate virtual populations to predict the effects of therapies [145]. These approaches have already been expanded to identify biomarkers and find important mutations that affect response to treatment with drugs in cancer cell lines [146][147][148].…”

Conducting a Virtual Clinical Trial in HER2-Negative Breast Cancer Using a Quantitative Systems Pharmacology Model With an Epigenetic Modulator and Immune Checkpoint Inhibitors

“…It is built with a detailed MDSC module and pharmacokinetics and pharmacodynamics of entinostat, to investigate the effect of entinostat and its combination with nivolumab and ipilimumab by conducting an in silico virtual clinical trial. Virtual clinical trials aim to generate virtual patient cohorts with physiologically plausible parameters and predict efficacies of treatments of interest using in silico simulations with a QSP model (Allen et al, 2016;Cheng et al, 2017;Rieger et al, 2018). Due to the heterogeneity of patient cohorts enrolled in clinical trials and wide range of treatment strategies, in silico simulations using a virtual patient cohort that resembles the desired clinical population can provide insights into the potential therapeutic outcome even before the therapy begins.…”