MHRA Launches £2M AI Initiative to Predict Drug Interactions and Accelerate Medicine Approvals
Key Insights
The UK's MHRA has launched three AI-powered projects totaling over £2 million to predict harmful drug interactions (search) before they reach patients and streamline regulatory processes.
The flagship study will use AI to analyze anonymized NHS data to identify dangerous drug combinations, focusing initially on cardiovascular medicines (search), with lab validation using human-based models.
Drug interaction side effects currently cause approximately one in six hospital admissions in England and cost the NHS more than £2 billion annually.
The UK's Medicines and Healthcare products Regulatory Agency (MHRA) has announced three government-backed artificial intelligence projects worth over £2 million aimed at predicting harmful drug interactions (search) before treatments reach patients and modernizing medicine approval processes. The initiative, announced on October 22, 2025, represents a significant step toward embedding AI into pharmaceutical regulation while maintaining safety standards.
Addressing the Polypharmacy Challenge
The projects address a growing healthcare challenge: millions of people in the UK take multiple medicines daily, with approximately 1 in 7 people in England (8.4 million) regularly prescribed five or more medicines. While most combinations are safe, some interact in ways that cause harmful side effects, leading to repeated GP visits, prescription changes, or hospital admissions that strain patients, carers, and the NHS.
Drug interaction side effects are estimated to cause around one in six hospital admissions in England and cost the NHS more than £2 billion every year, highlighting the urgent need for better predictive tools.
AI-Powered Drug Interaction Detection
The flagship project, funded with £859,650 from the UK Government's Regulatory Innovation Office's AI Capability Fund, will use artificial intelligence to identify early signals of adverse interactions between medicines by analyzing anonymized NHS data. Scientists from the MHRA, working with PhaSER Biomedical (search) and the University of Dundee, will initially focus on cardiovascular medicines (search).
The system will look for patterns in NHS data showing how different medicines behave when used together. Crucially, any AI-flagged risks won't stop at pattern detection—the team plans to test these signals in laboratory systems using human-based models that mimic how drugs are processed in the body, providing biological confirmation that predicted interactions are clinically meaningful.
"The launch of this project will demonstrate how AI and advanced modelling can be built into drug development to design smarter, more efficient clinical trials," said Julian Beach, Interim Executive Director Healthcare Quality and Access at the MHRA, who is supervising the study. "By understanding how medicines work together, we can generate stronger, more realistic evidence to support new treatments and ultimately reduce avoidable harm."
Transforming Drug Development
The initiative could address a persistent industry challenge: around nine in ten promising drugs fail late in development because early trials can't always predict how they'll work in real patients. By using AI and real-world health data that reflect patient diversity and actual medicine-taking patterns, scientists can spot risks and successes earlier, providing regulators with stronger evidence for faster, well-informed decisions.
Chris Wardhaugh, Chief Executive of PhaSER Biomedical (search), emphasized the collaboration's potential: "This collaboration demonstrates a shared commitment between regulators and innovators to redefine how medicines are developed. Our model provides a uniquely human-relevant lens on drug metabolism (search) and safety, while our work with the MHRA ensures that these insights can translate directly into regulatory practice."
AI-Assisted Regulatory Processes
The second project, the AI for Regulatory Insight, Safety, and Efficiency (ARISE) programme, receives £1,000,000 via the Regulators' Pioneer Fund. This initiative will pilot AI-assisted tools to support experts in scientific advice, clinical trial assessments, and licensing decisions, aiming to improve efficiency and consistency while keeping all final decisions in human hands.
The programme aligns with the Life Sciences Sector Plan's goals to create faster, risk-proportionate, and predictable routes to regulatory approval.
Synthetic Data for Rare Conditions
The third project, funded with £259,250 from the Regulators' Pioneer Fund, will pilot the use of synthetic patient data to support clinical trials in cancer (search), inflammatory bowel disease (search), and rare pediatric seizure conditions (search). This approach could strengthen evidence from clinical trials in rare and underrepresented populations where recruiting large patient cohorts is challenging.
The project will operate within a regulatory sandbox to test how synthetic data can safely and responsibly supplement evidence for regulatory decisions.
Leadership Perspective on Modern Regulation
MHRA Chief Executive Lawrence Tallon positioned the initiative as essential for modern healthcare: "People are living longer and managing more conditions, often with multiple medicines, so our safety systems must keep up. By using new tools and real-world health data, the MHRA is delivering practical solutions that protect patients and speed access to effective treatments, making regulation safer, smarter and more inclusive."
Broader Innovation Framework
The projects contribute to the government's ambition to make the UK the most AI-enabled health system in the world, as outlined in the 10 Year Health Plan for England, while reinforcing the UK's position as a global hub for life sciences innovation under the Life Sciences Sector Plan.
The work will also inform the MHRA's National Commission into the Regulation of AI in Healthcare, which brings together patient advocates, clinicians, regulators, and technology companies to advise on AI regulation in healthcare.
Maintaining Safety Standards
Officials emphasize that all final regulatory decisions will remain with human experts, who will use AI as decision support rather than replacement. The use of real-world NHS data is designed to reflect patient diversity more accurately than some clinical trials, potentially identifying safety issues that might otherwise be missed.
The findings from these projects will help shape the next generation of clinical trials and approvals, producing practical guidance for developers on using AI and real-world data alongside traditional trial evidence. This comprehensive approach aims to balance innovation with caution, ensuring that new treatments reach patients faster while maintaining the highest safety standards.