AI to help predict medicine side effects under MHRA projects

The Medicines and Healthcare products Regulatory Agency (MHRA) is leading three new government-funded projects that will use artificial intelligence (AI) and real-world NHS data to improve medicines safety and speed up access to new treatments.

One study, supported by £859,650 from the government’s AI Capability Fund, will use AI to predict adverse effects arising from drug combinations before they reach patients.

The project, run by the MHRA with PhaSER Biomedical and the University of St Andrews, will analyse anonymised NHS data to identify patterns in how medicines interact, initially focusing on cardiovascular drugs.

Related Article: New data reveals 'relentless' 27% hike in costs of prescription drugs

Findings will then be validated in the lab using human-based models that replicate how medicines are metabolised.

The aim of the project is to create a tool that enables clinicians to assess interactions and tailor prescriptions more safely and effectively.

According to the MHRA, the approach could help reduce medicine-related hospital admissions, which account for around one in six in England and cost the NHS more than £2bn annually.

Two further MHRA initiatives will also receive funding through the Regulators’ Pioneer Fund. The £1m AI for Regulatory Insight, Safety and Efficiency (ARISE) programme will test how AI can support experts in scientific advice, clinical trial assessment and licensing decisions.

A separate £259,250 project will explore the use of synthetic patient data to strengthen trial evidence, particularly in rare and under-represented conditions.

Related Article: Beyond the scales: reframing eating disorder recognition in primary care

Lawrence Tallon, MHRA chief executive, said the new programmes would make regulation ‘safer, smarter and more inclusive’, adding that the agency was committed to using real-world health data and advanced technologies to protect patients and support faster access to effective medicines.

Julian Beach, interim executive director for healthcare quality and access at the MHRA, said the work would show how AI and advanced modelling ‘can be built into drug development to design smarter, more efficient clinical trials’.

The announcement of the projects follows an NHS trial that found using AI to support administrative work could save an average of 43 minutes per day per staff member, which equates to five weeks annually.

Related Article: Short supply notice extended for blood pressure drug

Earlier this year, a major UK research funding programme was announced that will see new collaborations between academia and the pharmaceutical industry that will involve the use of AI to speed up drug manufacturing, develop next-generation therapies, and tackle environmental challenges in medicine production.