stormTB: SimulaTOR of a murine Minimal-pbpk model for anti-TB drugs
Roberto Visintainer1, Anna Fochesato1,+, Daniele Boaretti1, Shayne Watson2, Micha Levi2, Federico Reali1, and Luca Marchetti1,3
1Fondazione The Microsoft Research – University of Trento Centre for Computational and Systems Biology (COSBI), Italy. 2Gates Medical Research Institute, USA. 3University of Trento, Department of Cellular, Computational and Integrative Biology (CIBIO), Italy. +current affiliation: Roche Pharma Research and Early Development, Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd., Basel, Switzerland
Introduction/Objectives:
In 2022 Tuberculosis (TB) was the second cause of death due to a single infectious agent after COVID-19 and caused almost twice as many deaths as HIV/AIDS. With more than 10 million people falling ill every year, TB is a global health threat [1]. Therefore, creating new treatment regimens that are both shorter and safer represents a current and urgent challenge. To help understand the drug exposure at the TB site of action, we implemented a simulation tool that provides qualitative and quantitative comparisons of pharmacokinetic profiles, aiding in assessing drug-dose combinations. stormTB: SimulaTOR of a murine Minimal-pbpk model for anti-TB drugs is a web-based user-friendly interface that allows the simulation of a murine minimal physiologically based pharmacokinetics (mPBPK) model to analyze the characteristics of the chosen anti-tuberculosis (TB) compounds [2]. A data-driven physiologically based unified platform for the analysis of possible treatment scenarios involving 11 novel and historical compounds was developed. The user does not need to have programming expertise to gain insights into exposure and efficacy for a selected drug combination during the early preclinical development stage.
Methods:
stormTB leverages an mPBPK model consisting of nine ordinary differential equations obtained by streamlining a whole-body mPBPK model via the identification of the tissues least involved in the TB site of action and the combination of relative compartments to obtain a smaller set of equations. Our user interface is implemented with the R libraries Shiny and shinyjs, plots are generated and rendered with ggplot2 and ggiraph. To generate the dynamics of virtual populations, stormTB particularly benefits from the C translation of the model, which is then solved via deSolve [3]. This setup allows us to simulate a virtual population of 100 mice in under 0.8 seconds, considering a simulated treatment with rifampicin, four days of simulation, and default model parameters. This contrasts to over 5 minutes required using a pure R implementation, resulting in a 350-fold decrease in average computational time.
Results:
The typical analysis carried out with stormTB allows the user to define a treatment scenario made by drug, dose, and duration, obtaining the abundance of the drug in 9 compartments. Descriptive statistics regarding Tmax, Cmax, AUC, and time of the treatment above defined potency thresholds, namely, MIC, MBC, MacIC, and WCC [4], are reported in tables. A virtual population generator can be activated to compute confidence intervals for each value reported. Each simulated scenario can be saved and recalled, being later visualized singularly or in combination to compare the drug PK profiles and the different performances in terms of achieved efficacy.
Conclusions:
stormTB is a web-based interface for a novel minimal physiologically based pharmacokinetic (mPBPK) platform designed to simulate custom treatment scenarios for tuberculosis in murine models. The app facilitates visual comparisons of pharmacokinetic (PK) profiles, aiding in assessing drug-dose combinations. It offers a unified perspective, overcoming the potential inconsistencies arising from diverse modeling efforts. The app provides a user-friendly environment for researchers to conduct what-if analyses and contribute to collective TB eradication efforts. By generating comprehensive visualizations of drug concentration profiles and pharmacokinetic/pharmacodynamic (PK/PD) indices for TB-relevant tissues, this tool empowers researchers to pursue more efficacious tuberculosis treatments.
References:
[1] W. H. Organization, Global Tuberculosis Report 2023. 2023.
[2] F. Reali et al., “A minimal PBPK model to accelerate preclinical development of drugs against tuberculosis,” Frontiers in Pharmacology, vol. 14, p. 1272091, 2024.
[3] K. Soetaert et al., “Solving Differential Equations in R: Package deSolve”, Journal of Statistical Software, 33(9), 2010.
[4] L.G. Wayne et al., “An in vitro model for sequential study of shiftdown of Mycobacterium tuberculosis through two stages of nonreplicating persistence”, Infection and Immunity, 64(6), pp. 2062– 2069, 1996.
