In the last year, drugmakers have talked up the prospect of medicines discovered or designed by artificial intelligence. AI could accelerate the pace of new drugs approved to treat everything from rare diseases to cancer to antibiotic-resistant infections, they tout.
It will take years in the research pipeline to see whether those claims bear out. Meanwhile, another crop of companies is deploying machine learning and artificial intelligence to the design and execution of clinical trials for existing drug candidates, hoping to see a much faster payoff.
The ultimate manifestation of artificial intelligence in clinical trials would be total simulation: No human participants, just predicted clinical and safety outcomes based on immense databases of previous clinical trials, real-world patient records, pharmacokinetic data, and more. Unsurprisingly, “a lot of the applications that we’re seeing are still around optimization, and not necessarily the wild and cool stuff that people are anticipating,” said Morgan Hanger, executive director of the Clinical Trials Transformation Initiative.
This article is exclusive to STAT+ subscribers
Unlock this article — plus in-depth analysis, newsletters, premium events, and news alerts.
Already have an account? Log in
