Interactive applications using Shiny in R transform pharmacometric model simulations by enabling dynamic user interfaces that update simulations in real-time. Despite advancements in R packages like mrgsolve and rxode2, challenges persist, particularly for mechanistic and AI/ML models. The InSilicoTrials (IST) platform bridges these gaps by integrating diverse programming languages (NONMEM, R, Python, Matlab) into a seamless drag-and-drop workflow. It eliminates coding barriers while supporting methodologies like PKPD, QSP, and ML models. We demonstrate the platform’s impact through a case study in obesity.
Methods: Clinical trial simulations were performed on the IST platform to validate the adequacy of the model and to calculate the probability of success for various study designs and titration schemes. The use case involves Tirzepatide, leveraging published PK-PD and PK-Safety model1,2. In order to perform additional validation, we used the published data of the Phase 3 study, that the model was based on3. To ensure that the model was adequately to study alternative titration schemes than employed in the Phase 3 study, we additionally performed external validations based on data not used for model building4. Simulations were implemented using R 4.4.2 and the mrgsolve package.
Results: The model prediction adequately captured the clinical study data used for model validation. The interactive Input User Interface (UI) empowers project teams to explore alternative titration schemes and clinical study designs dynamically. For our case study, we optimized the titrant scheme for Phase 3 by setting realistic threshold – 20% body weight reduction and 20% incidence of nausea – and performed 1000 simulations with 100 subjects per arm. The joint probability of success based on titration scheme and maintenance doses was interactively visualized using the R Markdown-based Output UI, which requires no additional coding knowledge.
Conclusion: By streamlining workflows and integrating diverse methodologies, the InSilicoTrials platform empowers data-driven decisions in drug development, optimizing outcomes while reducing risks and investments, it represents a transformative tool in pharmacometric modeling.
Citations: [1] FDA. CENTER FOR DRUG EVALUATION AND RESEARCH. Application number 217806Orig1s000, https://www.accessdata.fda.gov/drugsatfda_docs/nda/2024/217806Orig1s000ClinPharmR.pdf [2] Schneck, K. & Urva, S. Population pharmacokinetics of the GIP/GLP receptor agonist tirzepatide. CPT Pharmacometrics Syst Pharmacol 13, 494–503 (2024). [3] Jastreboff, A. M. et al. Tirzepatide Once Weekly for the Treatment of Obesity. New England Journal of Medicine 387, 205–216 (2022). [4] Frias, J. P. et al. Efficacy and tolerability of tirzepatide, a dual glucose-dependent insulinotropic peptide and glucagon-like peptide-1 receptor agonist in patients with type 2 diabetes: A 12-week, randomized, double-blind, placebo-controlled study to evaluate different dose-escalation regimens. Diabetes Obes Metab 22, 938–946 (2020).
