Daniel Roeshammar1, Pauline Bambury1, Fianne Sips2, Niccolò Totis2, Jane Knöchel1, Roberta Coletti3, Lorena Leonardelli3, Luca Marchetti3,4
1InSilicoTrials Technologies S.p.A, 2InSilicoTrials Technologies B.V., 3Fondazione The Microsoft Research - University of Trento Centre for Computational and Systems Biology (COSBI), 4University of Trento, Department of Cellular, Computational and Integrative Biology (CIBIO)
Introduction: Despite advancements in novel immunotherapies in oncology improving survival, integrating these into specific treatment strategies remains challenging. In advanced castration-resistant prostate cancer (PCa), effective combination therapies are crucial, yet evaluating their efficacy and safety is complex. To address this, an in silico testing environment leveraging Quantitative Systems Pharmacology (QSP) models has been developed on the InSilicoTrials platform. Objectives: This work investigates implementation of mechanistic QSP models of advanced castration-resistant PCa drug therapy on the InSilicoTrials platform. This will allow for easy setup of clinical trial simulation scenarios for expert and non-expert users alike. Methods: Two established QSP models of PCa are being integrated on the platform for clinical trial simulation. A preclinical QSP model of PCa [1] combines a detailed description of the tumor microenvironment and the interaction between tumor cells and the local immune system, facilitating the integration of novel immunotherapy mechanisms of action. The model includes novel targets of immunotherapies such as myeloid-derived suppressor cells (MDSCs) and natural killer (NK) cells. The framework was calibrated on preclinical data from murine models. It allows the evaluation of the effect of combination therapy on tumor inhibition and the assessment of potential synergistic interactions. The model has previously been applied successfully to evaluate the efficacy of a wide range of therapeutic combinations. A second QSP model of PCa [2] was first established for anti-CTLA4 and sipuleucel-T immunotherapies. Based on a mechanistic description of key PCa immune interactions, this model was calibrated on clinical data and is applicable to predict not only efficacy but also adverse events. The model is suited to support the identification of optimized doses and schedules, accounting for both efficacy and safety of mono- and combination therapies. Moreover, it can be expanded to incorporate novel mechanisms of action and emerging clinical trial results. After integrating both models into the InSilicoTrials platform, user interfaces are built to facilitate clinical trial simulation also for the less technical user and for the wider project teams. The QSP models can be coupled as building blocks in workflows combining various pharmacometrics models and analytical modules. The platform makes it easy to connect the disease-specific QSP models to drug-specific pharmacokinetic models of new compounds. The analytical modules streamline the evaluation of the efficacy and safety of these novel therapies. The platform builds on Azure Cloud Services to facilitate secure, scalable computational workflows. It enables full compatibility between models developed in different programming languages (NONMEM, R, Python, MATLAB, etc.) and analytical modules for visualization, statistical analyses, and simulation outputs, yielding insights that can serve to guide decision-making. Results: The integration of the QSP models into the InSilicoTrials platform may offer scientists an intuitive tool for design of advanced castration-resistant PCa studies. Through customized simulation input interfaces, the user can specify alternative study designs, dosing strategies and explore the impact of varying model parameters. The platform enables the integration of data, coupling of drug-specific pharmacokinetic models with disease-specific QSP models directly from their original software and building of complex simulation workflows. The clinical trial simulator module facilitates detailed analyses of study designs, dosing strategies and the effect of combination treatment. The resulting computational environment provides a versatile set of tools to evaluate the potential of pharmacotherapies for advanced PCa. Finally, the platform allows the creation of interactive results dashboards, which can make the computational analysis and obtained insights accessible to a wide range of non-expert stakeholders directly, facilitating decision-making. Conclusion: Integration of the PCa QSP models on the platform may provide a powerful computational environment to assess the potential of novel mono- or combination therapies for advanced castration-resistant PCa. The simulation platform empowers both technical modelers and non-technical users from the wider clinical team to evaluate therapeutic potential and refine clinical trial designs for PCa therapies, assessing the impact of key model parameters and dosing strategies. References: [1] Coletti, R.; Leonardelli, L.; Parolo, S.; Marchetti, L. “A QSP model of prostate cancer immunotherapy to identify effective combination therapies.” Sci Rep 10, 9063 (2020). https://doi.org/10.1038/s41598-020-65590-0 [2] Coletti, R.; Pugliese, A.; Lunardi, A.; Caffo, O.; Marchetti, L. “A Model-Based Framework to Identify Optimal Administration Protocols for Immunotherapies in Castration-Resistance Prostate Cancer.” Cancers 2022, 14, 135. https://doi.org/10.3390/cancers14010135
Reference: PAGE 33 (2025) Abstr 11650 [www.page-meeting.org/?abstract=11650]
Poster: Methodology - Other topics