Bayesian Augmented Clinical Trials in TB Therapeutic Vaccination

Moving forward through the in silico modeling of multiple sclerosis: Treatment layer implementation and validation

Multiple sclerosis is an autoimmune inflammatory disease that affects the central nervous system through chronic demyelination and loss of oligodendrocytes. Since the relapsing-remitting form is the most prevalent, relapse-reducing therapies are a primary choice for specialists. Universal Immune System Simulator is an agent-based model that simulates the human immune system dynamics under physiological conditions and during several diseases, including multiple sclerosis. In this work, we extended the UISS-MS disease layer by adding two new treatments, i.e., cladribine and ocrelizumab, to show that UISS-MS can be potentially used to predict the effects of any existing or newly designed treatment against multiple sclerosis. To retrospectively validate UISS-MS with ocrelizumab and cladribine, we extracted the clinical and MRI data from patients included in two clinical trials, thus creating specific cohorts of digital patients for predicting and validating the effects of the considered drugs. The obtained results mirror those of the clinical trials, demonstrating that UISS-MS can correctly simulate the mechanisms of action and outcomes of the treatments. The successful retrospective validation concurred to confirm that UISS-MS can be considered a digital twin solution to be used as a support system to inform clinical decisions and predict disease course and therapeutic response at a single patient level.

A Credibility Assessment Plan for an In Silico Model that Predicts the Dose-Response Relationship of New Tuberculosis Treatments

Tuberculosis is one of the leading causes of death in several developing countries and a public health emergency of international concern. In Silico Trials can be used to support innovation in the context of drug development reducing the duration and the cost of the clinical experimentations, a particularly desirable goal for diseases such as tuberculosis. The agent-based Universal Immune System Simulator was used to develop an In Silico Trials environment that can predict the dose-response of new therapeutic vaccines against pulmonary tuberculosis, supporting the optimal design of clinical trials. But before such in silico methodology can be used in the evaluation of new treatments, it is mandatory to assess the credibility of this predictive model. This study presents a risk-informed credibility assessment plan inspired by the ASME V&V 40-2018 technical standard. Based on the selected context of use and regulatory impact of the technology, a detailed risk analysis is described together with the definition of all the verification and validation activities and related acceptability criteria. The work provides an example of the first steps required for the regulatory evaluation of an agent-based model used in the context of drug development.

Bayesian Inference-Based Gaussian Mixture Models With Optimal Components Estimation Towards Large-Scale Synthetic Data Generation for In Silico Clinical Trials

Goal: To develop a computationally efficient and unbiased synthetic data generator for large-scale in silico clinical trials (CTs). Methods: We propose the BGMM-OCE, an extension of the conventional BGMM (Bayesian Gaussian Mixture Models) algorithm to provide unbiased estimations regarding the optimal number of Gaussian components and yield high-quality, large-scale synthetic data at reduced computational complexity. Spectral clustering with efficient eigenvalue decomposition is applied to estimate the hyperparameters of the generator. A case study is conducted to compare the performance of BGMM-OCE against four straightforward synthetic data generators for in silico CTs in hypertrophic cardiomyopathy (HCM). Results: The BGMM-OCE generated 30000 virtual patient profiles having the lowest coefficient-of-variation (0.046), inter- and intra-correlation differences (0.017, and 0.016, respectively) with the real ones in reduced execution time. Conclusions: BGMM-OCE overcomes the lack of population size in HCM which obscures the development of targeted therapies and robust risk stratification models.