Artificial Intelligence Meets Medicines Regulation

Artificial Intelligence Meets Medicines Regulation

As a medicines authority, the Paul-Ehrlich-Institut performs two important functions in the field of digital transformation of public health:

• Evaluating the use of the latest data-driven therapy methods and co-developing guidelines that contribute significantly to the safety and effectiveness of these methods. These functions are fulfilled in close cooperation with international partners, including the HMA/EMA Big Data Task Force, of which the Paul-Ehrlich-Institut was a member, and participation in its successor, the HMA/EMA Big Data Steering Group.
• Using artificial intelligence (AI) in the production of pharmaceuticals accelerates manufacturing processes and creates new therapeutic approaches, such as personalised vaccination approaches for cancer treatment. This has led to a continuous increase in ongoing AI-related procedures at the Paul-Ehrlich-Institut (see figure below). In order to meet the increased regulatory demand, the Paul-Ehrlich-Institut also relies on digital and AI-based tools. Automation, digitisation through data-driven analyses and evaluations, and using AI as a tool are not only unavoidable, but also indispensable to fulfil the Institute's tasks.

Development of Big Data and AI Methods

The research group "Cellular Aspects of Host-Pathogen Interactions" at the Paul-Ehrlich-Institut has already built up extensive expertise in the development of methods for analysing big data, bioinformatics, and AI applications for the targeted analysis of large data sets. Their expertise has been shared with EU-wide bodies and the Paul-Ehrlich-Institut is building upon the research group's work as the foundation for the advancement of its AI strategy.

Three research projects funded by the Federal Ministry of Health (Bundesministerium für Gesundheit, BMG) show how expertise in big data and AI methods can be expanded at the Paul-Ehrlich-Institut. The BLOODVIR project aims to develop new AI methods to improve the safety of blood products. The RENUBIA project investigates how to evaluate AI methods used in pharmaceuticals. The KIMERBA project explores how AI can support the Paul-Ehrlich-Institut's regulatory processes.

Use of AI at the Paul-Ehrlich-Institut

• Medicinal products that were developed with the help of AI applications and evaluated by the Institute are in clinical trials.
• AI-based data analysis and visualisation can be used in regulatory processes for multiple purposes, including identifying patterns and relationships in large datasets, such as those found in pharmacovigilance reports of suspected adverse events.
• There are intelligent automation solutions available within the Paul-Ehrlich-Institut that allow assessors to focus more on the analysis of the submitted data with regards to the quality, robustness and consistency of their tasks. These solutions increase the speed at which procedures are finalised.
• The Paul-Ehrlich-Institut is expanding its expertise in regulatory research in the field of AI in biomedicines.