AI and machine learning: A new era for clinical trials

Fiona Maini, Senior Director Global Compliance & Strategy, at Medidata looks at the rise of artificial intelligence (AI) and machine learning (ML) within life sciences and the applications for the technology in clinical trials.

The benefits that AI and machine learning (ML) can bring to society and their potential to radically change the world we live in can sometimes feel a little intangible, futuristic or even far-fetched. In many respects, however, these technologies are already firmly embedded in our daily lives and, in the healthcare and life sciences sectors, they are making a meaningful and even life-saving difference.  

There has been an increasing focus over the past few years about the uses of AI and machine learning in drug discovery and development, particularly in their ability to identify molecules capable of treating disease and to simulate trial results. In the clinical trial management and planning systems space, we are already seeing how the practical implementation of these tools is accelerating drug development and bringing innovative medicines to patients faster. This new era of digitalisation is exciting and promises great opportunity, but it does introduce questions as to how regulators can keep up to combat concerns around how AI should be monitored, data ethics, data privacy, and safety with appropriate supporting legislation. 

Advanced technology

The wider use of advanced technology such as decentralised digital technologies has allowed for a more patient-centric approach, easing the burden on patients taking part in trials. Additionally, the use of wearable devices has increased the types of data collected, subsequently improving future recruitment levels.

These ‘smart’ devices can monitor vital signs and activity levels, as well as being able to detect unusual readings and adverse drug reactions that might otherwise have gone unnoticed. This real-time monitoring, coupled with other technology solutions, generates considerably more data than traditional clinical trial methods.

This technology poses an opportunity for researchers, but also a significant challenge for the teams managing trials: how do they process and manage this sensitive data? Here, AI and ML offer a unique solution: the ability to collect, manage and analyse the vast quantities of data that clinical trials now generate, which have incredible potential to accelerate the time it takes to produce results. As clinical trial sponsors continue to leverage these technologies, we are also beginning to see more advanced applications beyond simple data collection and towards meaningful analysis and actionable insights.

The rapid spread of these new technologies and the fast pace at which they develop and change has, however, left regulators playing catch up as they get to grips with the associated safety and ethical implications. Any tool that can successfully shave time off the lengthy drug development process and bring better-quality treatments to those who need them is good news for medical professionals, patients, and the general population, so it is vital that the industry and regulators continue to collaborate to encourage and enable progress in this area. 

The patient perspective 

One worry that has emerged with the advent of AI, ML and other digital advances is around data privacy and the appropriate use of data. Patient recruitment and retention can be particularly challenging for clinical trial organisers, so addressing patients’ concerns and keeping participants engaged and informed is crucial to a trial’s success. 

Sponsors, organisers, and regulators must work together to reassure patients by keeping them informed about how their data will be used and stored. Transparency here is paramount, particularly at a time when there is growing and widespread scrutiny about data privacy and security across multiple industries. For example, it is our collective responsibility to ensure that large language models (LLMs) designed for analysing large clinical datasets have been thoroughly evaluated to ensure they are unbiased, producing more reliable and accurate results.

The regulatory landscape today

Policymakers around the world have started to recognise that as technology goes either unregulated or inappropriately regulated, this can lead to distrust or misuse and can cause serious challenges. There is now growing pressure for regulators to take action. The result is a concerted effort to tackle these issues by debating them openly in an effort to create a fresh regulatory framework to police this new digital world.

For instance, the world’s first AI Summit in Bletchley Park last November brought leading players in the AI space together to discuss the challenges faced by the industry, while the EU AI Act, first proposed in 2021 and further updated last year, highlights encouraging regulatory progress. 

These are all positive developments, but it is worth highlighting that the EU’s AI Act is not industry-specific. The life sciences sector is complex and unique and requires specific regulatory oversight. The industry needs to understand how the AI Act will impact existing rules, such as the EU Medical Device Regulation (MDR) which currently regulates medical devices. 

Progress is ongoing on this front, with the European Medicines Agency (EMA) developing its own guidance. In 2023, the EMA published its “draft reflection paper” on the use of AI in the life cycle of medicines as part of this ongoing conversation, which concluded that a human-centric approach should guide all development and deployment of AI and ML, but that early regulatory support should be in place to support its implementation. In recent years, the Medicines and Healthcare products Regulatory Agency (MHRA) has developed a Software and AI as a Medical Device Change Programme, as well as the MHRA Innovation Office, which has been helping to support innovators by providing free and confidential expert regulatory information. 

Next steps

Regulators are embracing new innovation and collaborations but it can be a lengthy process as there are many complex cogs to consider. AI and machine learning have the potential to improve efficiency, productivity, access, diversity, and the design of clinical trials, resulting in a more dynamic and energised drug development cycle. For these tools to become more widely adopted and ‘legitimised’, patients and trial organisers alike need to be aware of the rules and confident that they are operating within a robust regulatory and legislative framework.

DDW Volume 25 – Issue 2, Spring 2024

About the author:

Fiona MainiFor the past 20 years, Fiona Maini has been providing advisory services consulting within the pharmaceutical arena and mobile health technologies for Clinical Trials. Currently Senior Director Global Compliance and Strategy at Medidata, she is active in the EU Artificial Intelligence Alliance and ACRO Working Party on Virtual Trials.


Related Articles

Join FREE today and become a member
of Drug Discovery World

Membership includes:

  • Full access to the website including free and gated premium content in news, articles, business, regulatory, cancer research, intelligence and more.
  • Unlimited App access: current and archived digital issues of DDW magazine with search functionality, special in App only content and links to the latest industry news and information.
  • Weekly e-newsletter, a round-up of the most interesting and pertinent industry news and developments.
  • Whitepapers, eBooks and information from trusted third parties.
Join For Free