Biotech Square: Transforming Clinical Development Failures into Lessons
by Max Yaghchi
Medical innovations, whether pharmaceuticals, biologics or medical devices, have the potential to advance healthcare and improve patient outcomes. However, the path to regulatory approval is rife with failure. For example, in drug development, only 1 in 10 candidates survive clinical research to meet the health authorities’ approval and reach patients. This unfortunate reality leaves many without treatment options, and has disproportionately affected those with rare conditions, high comorbidity, and members of marginalized patient populations. These failures also negatively impact sponsors, which can face inflated development costs and missed investment opportunities valued between $800M and $1.4B per new molecule [1].
Factors that contribute to clinical research failures are diverse and include poor patient recruitment, inappropriate cohort composition, and lack of regulatory guidance during study execution. Biotech Square applies natural language processing technologies on historical data hidden in regulatory documents, clinical trial registries, scientific literature, and over 25 other databases to uncover lessons that can benefit clinical development teams.
Natural Language Processing: A subfield of artificial intelligence (AI) that is concerned with processing and analyzing large amounts of information written in human language [1].
In order to reduce clinical development failures, Biotech Square is deploying an AI software that guides medical innovators through regulatory approval processing, study design planning, and execution team selection for a de-risked clinical development path.
The potential for AI in guiding clinical trials for improved success rates has been explored in scientific literature. Examples of clinical development challenges along with how AI can help address them are listed below [1].
The Patient Recruitment Challenge: Automatic assessment of patient demographics and medical history to determine recruitment eligibility for clinical trials
The Cohort Composition Challenge: Intelligent matching of patient cohorts for reduced population heterogeneity and better observed treatment effect
The Trial Execution Challenge: Investigator assistance with reminders, warnings about protocol deviations, and recommendations throughout the trial
With an initial focus on operational risks, Biotech Square is demonstrating a novel application of AI in the prevention of clinical development failures. In 2019 alone, 14 high profile pharmaceutical failures were reported, including treatments for aggressive brain tumors, heart failure, and major depression [2]. While these failures were setbacks for both patients and the pharmaceutical industry, information reported on them is actively analyzed by Biotech Square’s AI technologies to improve the chances of success of future innovations.
Learn more about Biotech Square’s technology and schedule a demo at www.BiotechSquare.com
References
[1] Harrer, S., Shah, P., Antony, B., & Hu, J. (2019). Artificial Intelligence for Clinical Trial Design. Trends in Pharmacological Sciences, 40(8), 577–591. doi: 10.1016/j.tips.2019.05.005
[2] Taylor, P. (2019). 2019’s top clinical trial flops (and a flip-flop). FierceBiotech. Source: https://www.fiercebiotech.com/special-report/2019-s-top-15-clinical-trial-flops-and-a-dishonorable-mention
