Drug discovery and development most frequently fail during clinical trials as in vitro and pre-clinical studies do not effectively translate into success in humans, and attaining regulatory approval, many fail their commercialization goals. Medicinal chemists and scientists in drug discovery can benefit from understanding the cause of these failures. This project will create a database of such failures in a specific disease area and evaluate a computational framework to examine the sources of these failures in an objective and comprehensive manner. This is intended to show the value of such a generalized approach for drug discovery and development. Multiple Sclerosis will be the initial target.
This project establishes a process that compares real-world evidence, e.g. protocols and inclusion/exclusion criteria, from clinical trials and commercial failures and successes to objectively analyze the basis for these results. The goal of this is to improve the clinical development and commercialization potential but ultimately optimize early drug discovery efforts, i.e. medicinal chemistry. A common data model/database that enables all clinical trial parameters, and potentially data, to be developed will enable both comparison and evaluation across multiple trials.
The initial focus involve inclusion/exclusion criteria and comparison with real-world patient populations but be generalized to all parameters. Of primary concern will be the quality of diagnostic criteria and the potential need for disease stratification based on clinical presentation. This data will serve in an “observational” manner to identify potential weak aspects of trial design to aid in improving trial performance and clinical success. This differs from the current use of real-world data to support regulatory submissions. Additional issues will be included: Is off-target selectivity given? Are PK properties supporting target engagement? Is posology in line with compliance standards? Are tolerability data in line with application and in line with target patient population requirements?
The resultant database will enable improving trial design, redefinition of important parameters for high-quality drug candidates and re-purposing of existing drugs. In addition, identification of these factors can be used in earlier discovery to better align medicinal chemistry with stratified diseases that can improve future trial success and commercialization, while providing better guidance for physicians in treatment decisions. A secondary impact will be the potential long-term impact on reducing drug development costs. The initial application to multiple sclerosis will address the many challenges to drug development in this critical area focusing on small molecules, e.g. Tecfidera, Aubagio, Gilenya, Copaxane, Novantrone, Ampyra, Decadron
Collaboration with SC on Toxicology:
The project will include evaluation of potential non-intended activities of these drugs, e.g. side-effects and drug-drug interactions, that could have been identified and evaluated early in both clinical trials and post-approval patient use.
Collaboration with SC on Nomenclature and Properties:
The project will closely monitor the use, misuse and/or inadequate use of standards used in clinical trial studies and/or post-market surveillance that could impact evaluation in meta-analyses.
Project announcement published in Chem Int Jan 2020, p. 29
Page last updated 12 Feb 2020