How is generative AI transforming clinical trials?

The pharmaceutical industry is on the cusp of a groundbreaking transformation with the integration of generative AI — especially in clinical trials. By addressing the complexities and challenges traditionally linked to developing trials, recruiting patients and supporting them throughout their care journey, AI-driven innovations are accelerating the process of bringing drugs to market.

We recently attended the excellent DPHARM conference (16–17 September 2024, Philadelphia, USA), where many of these developments were discussed as potential methods of improving clinical research — both for patients and for pharmaceutical companies.

And as you can see from a few of the interesting statistics that we collected across the 2-day event, this innovation couldn't come at a better time:

• 80% of trials face delays due to recruitment issues
• 11% of sites fail to enrol a single patient
• 55% of trials are terminated due to low accrual rates, with an average completion delay of 12 months
• 50% of patients live at least 125 miles from a trial site (this statistic is even more stark when we also know that they are typically willing to travel an average of 20.4 miles to attend appointments)

So how can generative AI help to solve these issues?

Site selection

Generative AI can significantly enhance site selection for clinical trials. By analysing extensive datasets such as demographic, historical trial performance and logistical data, AI is able to identify the most suitable trial locations.

Increasing the likelihood of choosing a site that matches the needs of the study ensures that the selected sites have the necessary resources and patient availability, which can streamline the trial process and reduce delays.

Finding diverse patient populations

Recruiting a diverse patient population for clinical trials is a major challenge. Generative AI tackles this by analysing large-scale health records and demographic databases to identify underserved communities.This approach not only enhances trial diversity, but also improves study generalisability (that is, the degree to which a study’s findings can be applied to other situations or groups of people).

Patient outreach

Effective communication with potential trial participants is essential. AI-powered platforms facilitate personalised and automated outreach, using predictive analytics to identify individuals who are likely to benefit from participating in particular studies and trials. This targeted approach boosts enrolment rates and maintains a steady flow of suitable participants.

Ethical and privacy concerns

So, those are the positives – but we also need to take a balanced view. While generative AI holds tremendous promise for transforming clinical trials, it also raises important ethical and privacy concerns. For example, the extensive use of health records and demographic data can potentially infringe on patient privacy if not handled with the utmost care. Which is why it's crucial for pharmaceutical companies and AI developers to implement stringent data protection and anonymisation protocols to safeguard patient information.

There is also a risk of bias in AI algorithms if they're not properly trained on diverse datasets, meaning that trial outcomes could be skewed, thereby limiting their applicability across different populations.Maintaining transparency in AI processes and ensuring equitable access to trials are imperative when addressing these concerns effectively.

But will the use of generative AI in clinical trials become widespread?

The FDA has recently issued guidance emphasising the importance of diversity in clinical trials. This guidance aims to ensure that clinical trials reflect the demographic makeup of the population that will eventually use the drug. By encouraging the inclusion of underrepresented groups, the FDA hopes to improve trial results and address healthcare disparities.

Due to its ability to efficiently identify and recruit diverse patient populations, AI could quickly become a valuable tool and play a pivotal role in ensuring the new FDA guidelines are upheld.  By integrating the FDA's diversity recommendations into AI-driven recruitment strategies, pharmaceutical companies will be able to foster more representative and equitable clinical research and align with regulatory expectations at the same time.

What does the future hold?

Isn’t that the question we all want answered?! One thing’s for sure though, it's certainly an exciting time for the clinical trial space. And with the advent of new technologies and approaches, we think it likely that change and progress will be rapid and exponential. We’ll certainly be keeping a close eye on things.

‍