AI-discovered drugs in Phase I clinical trials have an 80-90 per cent success rate, far outpacing drugs discovered by humans, new research has found.
Human-discovered drugs have an average success rate of 40-65 per cent in Phase I.
The new BCG research is published in the journal Drug Discovery Today.
The study is the first analysis of the industry-wide pipeline of AI-discovered drugs and vaccines, focusing on clinical success rates.
The research also found that in Phase II, AI-discovered drugs achieved a success rate of ~40 per cent (albeit on a limited sample size), which is in line with industry averages.
Overall, the probability of a molecule succeeding across all clinical phases end-to-end would increase with AI from 5-10 per cent to 9-18 per cent, the researchers concluded.
They said: “Our analysis gives a first view of the potential of AI-discovered molecules in clinic trials, and perhaps provides a glimpse into the future of AI-powered R&D.
“This breakthrough represents almost a doubling of pharmaceutical R&D productivity overall, bringing innovative medicines to patients faster, better and cheaper.”
AI is transforming almost all areas of health tech, with regulation struggling to keep up.
Yesterday, the UK’s Medicines and Healthcare products Regulatory Agency (MHRA) launched AI Airlock – its new regulatory sandbox for AI as a Medical Device (AIaMD).
The Airlock aims to help the agency to identify and address the challenges for regulating standalone AIaMD, initially seeking out and supporting four to six virtual or real-world projects through simulation.
The new process will enable the health body to test a range of regulatory issues for these devices when they are used for direct clinical purposes within the NHS.
