The Role of AI in Solving Clinical Trial Diversity and Access Challenges

In a recent video interview with Applied Clinical Trials, Michel van Harten, MD, CEO, myTomorrows, discussed the integration of artificial intelligence (AI) in clinical trials, highlighting its potential to reduce costs, accelerate timelines, and improve inclusion. Looking forward, advancements with AI are expected to speed up drug discovery, predict drug efficacy, and develop digital twins for personalized treatment simulations, potentially reducing trial risks and costs.

ACT: How can AI be used to address underrepresentation in trials?

van Harten: I think in general, access to clinical trials is a big deal, especially for people from underrepresented communities, and we believe that artificial intelligence plays a big role in changing that. Right now, the way to clinical trials and the way clinical trials are set up, often creates barriers that are very tough to overcome. Many trial sites are concentrated in big cities or academic medical centers, so if you live in a rural or underserved area, just getting to a trial might be out of reach, and for some people, there are also a lot of financial challenges like treatment costs or logistical challenges like transportation, and these burdens hit underrepresented populations especially hard, as well as people with lower incomes. Also, these people in those areas are not always made aware of the opportunities, not educated on the concept of clinical trials, and physicians are also not equipped to look for trials for their patients, so this leads to a situation, for example, in the US, where I think racial and ethnic minorities represent only 2% of participants in clinical trials, and that's of course a problem, because when certain groups are left out, we also miss out on crucial data, including how different people from different backgrounds respond to medications, so how can AI help? At myTomorrows, we lower the barriers for physicians and patients to identify and access trials. At the moment, we have two examples built into our product, where AI is used for treatment matching, where we save about 90% of the time that a physician is spending on matching a patient's profile to all those eligibility criteria. Second is education. As you know, the healthcare space is full of terminology which is very difficult to understand for patients and when it comes to clinical trials, it's even more complex, so AI can really help to translate consent forms, educational materials, etc. into multiple languages and use simpler terms, make it more accessible to non-English speakers and those also with low health literacy.