Multiple sclerosis (MS) is a critical area for clinical trials. It has seen both a boom in research focus over the past 30 years and a recent string of disappointments. A surprising number of phase 2 and phase 3 trials have failed,[1] causing major investments to collapse and sending a shockwave through the field.
The risk of failure, or more precisely, the failure to end informatively, is a key concern. When a trial fails to provide valuable insights to guide clinical, policy, or research decisions, it deprives researchers, patients, and investors of the information needed to advance the field.
In this context, an “informative” clinical trial is that proposed by Zarin et al in Harms From Uninformative Clinical Trials, JAMA 2019. You can read more about how the Clinical Trial Risk Tool quantifies the risk of a trial ending without delivering informative results in our article in Clinical Leader.
Fast Data Science’s Clinical Trial Risk Tool quantifies the risk of trial ending uninformatively and outputs design recommendations.
The Clinical Trial Risk Tool uses a machine learning algorithm fed with data from past trials, both successful and not. It analyses a multitude of factors, such as the number of trial participants, and the length of the trial. If it identifies a problem in the trial design which could lead to an uninformative outcome, this is given a score and a recommendation is displayed to the user.
By providing a risk estimate directly from the protocol text, Fast Data Science’s Clinical Trial Risk Tool empowers researchers, policymakers, and investors to consider the pitfalls and potential areas of improvement before the trial is underway. It holds significant potential for improving the efficiency and efficacy of future MS clinical trials, and ultimately, boosting the informativeness of trial outcomes.
The end goal is to assist the industry in learning from earlier, unsuccessful trials and adopt more effective strategies for innovative therapy development. With enhanced preparation and insight, the risk of failure can be decreased, leading to more informative and successful clinical trials.
