AI & Clinical Trials: Progress and Path of Innovation

Introduction

Similar to broad AI trends, clinical trial AI landscape is active, progressive, and everchanging.  While there is a lot of unknown implications of AI implementation within clinical trial design, technologies, real-world data analytics, there are many potential advancements and paths for permanent changes presently. However, there are challenges in basic AI integration related to timing, engagement, and overall AI adaptive strategy which are being considered globally and monitored with trial integrity and patient safety top of mind.

Active Landscape

In many ways, AI tools have been actively involved from design to real-world insights. There are algorithms used to help HCPs with trial placement and assistance with analyzing large amounts of data to aid HCP education and medicine advancements. The organization of data from health records and medical claims allows burdens to be lifted from staff and available for use in strategic and progressive ways. The intricate and delicate implementation of AI across all clinical trial landscape will be an ongoing endeavor with a very unknown path.

The help from AI to organize data and help identify ways to better deliver the information to patients will also lift burden to patient to clinical trial participation and ongoing involvements. The dual effect allows AI influence to benefit both HCPs and patient positively benefit from quicker and more efficient ways to collect data. An example of this is seen in trials that are using wearable devices and now able to share information quicker and more clearly to trial site staff. The introduction of these innovations will continue to enhance recruitment and patient education.

There are also many indications that AI tools, algorithms and insights could help improve clinical trial diversity. In MEDiSTRAVA Clinical Trial Diversity, Equity, and Inclusion 2024 overview, [linked here], the overall importance of diversity in clinical trial is outlined in detail and reiterates the crucial demand for diversity engagement and outcomes. Appropriately diverse trials outcomes are essential for global medicine advancement and human progress. In terms of organic community tracking, AI helps us understand and synthesize social media influence data and capture meaningful insights for use throughout study activity. There are current products being used such as chat bots to help connect patients to available resources, information, and other details that would aid participation and enrollment.

Challenges & The Unknown

AI’s path to implementation across all clinical trials will require human acceptance and engagement throughout all stages, from piloting to rollout and execution. With AI integrated mock trials, HCPs are able to extract pattern analysis for assistance in clinical development plan and protocol design. In utilizing AI collaborated data, there are important questions and details to consider related to information bias, reliability, and fit for use. It will be imperative for AI data to be quality checked for accuracy and completeness and will require consistent HCP influence.

There are other serious factors to consider when analyzing AI’s impact in clinical trial participation and performance. While AI tools and algorithms can begin to lift burdens on HCPs and site staff in ways such as synthesizing large amounts of data related to recruitment, enrollment, or retention, it can lead to inability to point to how the outcome was derived. The information received from AI tools and products can also lead to inability to directly quantify data received, leaving it useless for human engagement. For example, there could be changes in environment or population which may cause data drifts, less accurate information and inaccurate predictions.

Conclusion

As AI integration activity and impact to clinical trials is monitored by global health organizations, patient safety and the integrity of data and information will be most important to protect. While AI is dynamically imbedded from end-to-end trial design and actively impacting drug development and patient-centric visions, there is a need to consistently track over time to gauge deviations, patterns, and insights in close collaboration with HCPs, patients and study site staff to identify key gaps and guidance.

Citations

1. World Health Organization. WHO outlines considerations for regulation of artificial intelligence for health. World Health Organization (WHO). https://www.who.int/news/item/19-10-2023-who-outlines-considerations-for-regulation-of-artificial-intelligence-for-health. Published October 19, 2023. Accessed January 28, 2025.
2. Application of AI to Adaptive Enrichment Clinical Trials. U.S. Food and Drug Administration. Published 2023. Accessed January 14, 2025. https://www.fda.gov/science-research/advancing-regulatory-science/evaluate-application-artificial-intelligence-adaptive-enrichment-clinical-trials
3. Food & Drug Administration (FDA). FDA Announces Additional Steps to Modernize Clinical Trials. FDA. Published June 6, 2023. Accessed June 15, 2023. https://www.fda.gov/news-events/press-announcements/fda-announces-additional-steps-modernize-clinical-trials
4. ElZarrad K. The Role of Artificial Intelligence in Clinical Trial Design and Research with Dr. ElZarrad. Podcast. Published online May 24, 2024. Accessed February 12, 2025. https://www.fda.gov/media/178657/download