AI Integration Opportunities for Sponsors
The integration of artificial intelligence (AI) is revolutionizing clinical research, offering unique opportunities to enhance efficiency and effectiveness, as well as to improve patient outcomes. Sponsors are increasingly turning to the capacity of AI to address a range of challenges and achieve greater success in their pursuit to bring novel and life-saving therapies to market.
In this new era of AI, it can be difficult to navigate beyond the hype to pragmatically apply this technology. Generative or extractive? ChatGPT, Llama, Palm, Gemini, or BERT? Pre-trained or fine-tuned? The sheer volume of options and applications can be daunting.
As a starting point, it is often pragmatic to identify the clinical trial use cases most likely to benefit from AI. The following describes some of the top areas we believe have the greatest potential today.
Revolutionizing Patient Recruitment and Engagement
AI is redefining patient recruitment by enabling targeted identification and engagement of individuals meeting specific study criteria. This approach can utilize large amounts of information, including electronic health records (EHR), real-world data (RWD), and social media to streamline the recruitment process and to foster enrollment of the most suitable participants.
Having the ability to analyze and evaluate large volumes of data quickly, AI models have the potential to predict inclusion/exclusion attributes and to assist in the identification of subjects who may benefit from enrollment in a specific trial. These same models can also be utilized to reach more diverse and underserved populations.
In a recent example, Memorial Sloan Kettering Cancer Center utilized AI/ML to identify subject profiles as candidates for studies by classifying patient records for the prevalence of a relevant protein with a reported accuracy of 93%1.
Optimizing Trial Design and Execution
Using AI to assist with trial design and execution can have an enormous positive impact on your study. Analyses of historical data, real-world evidence, and genomic information provide insights into patterns and associations, informing the design of more effective and efficient trials with optimized patient selection, testing regimens, and endpoints.
Specially trained models can be utilized to assist in drafting protocols and identifying opportunities for efficiencies without sacrificing the effectiveness of the trial2. AI can be employed to find ways to best structure subject visits to maximize time spent with the patient. Utilizing this technology can also assist in more rapidly identifying potential confounders to ensure you are collecting all of the data pertinent to your trial.
Accelerated IVD and Device Development
AI has the potential to streamline the development and path-to-market for IVDs and devices, from early-stage discovery to regulatory clearance. It facilitates the identification of promising IVD and device candidates, can help to predict performance and safety profiles, and optimizes clinical trial design. This predictive capability expedites the development process, reducing time and resources required, and paving the way for faster access to innovative diagnostic tools and therapeutic devices.
AI is a leading driver for novel diagnostic devices, allowing researchers the ability to comb through large stores of data to identify meaningful biomarkers and therapeutic targets.
AI is a leading driver for novel diagnostic devices, allowing researchers the ability to comb through large stores of data to identify meaningful biomarkers and therapeutic targets. It is driving a revolution in early cancer detection with new opportunities emerging on an almost-daily basis.
According to an article published by the American Hospital Association, AI has led to “nearly 400 Food and Drug Administration approvals of AI algorithms for the radiology field.”3
Enhancing the Clinical Trial Experience
AI fosters a patient-centric approach through personalized support, timely response to inquiries, and close monitoring of adherence and potential adverse events. This empowers patients to actively participate in trials and is another tool to promote their safety and well-being, ultimately leading to a more positive clinical trial experience.
AI can be utilized to develop meaningful patient applications to answer questions and can assist investigators in drafting responses in a way that is more easily accessible and understood by study participants.
Navigating Ethical and Responsible AI Implementation
As AI integration progresses, concerns regarding data privacy, security, and transparency remain paramount. Ethical considerations abound in this new paradigm. Robust safeguards and transparent, interpretable, and validated AI algorithms are essential to ensure patient confidentiality and reliable, unbiased results.
AI does not replace the need for expert clinical study staff, nor does it reduce the requirements for diligence and ensuring subject safety. While regulatory bodies are working to address the impact and use of AI within clinical trials, there are still many open questions. Early and frequent interactions with ethics and review boards are of the utmost importance.
It remains critical to partner with organizations that are current with the latest regulatory guidance available and that have experience transiting these complex areas.
Beaufort CRO: Connecting Sponsors to AI Expertise
Navigating the intricate world of AI-powered clinical trials can add a layer of complexity that can be too burdensome for some sponsors. Beaufort’s team works with select AI partners to bring together the right expertise and solutions to power clinical research programs for sponsors.
Through a collaborative process between all stakeholders, we help ensure compatibility, validate accuracy, and manage data security, privacy, and compliance. Beaufort can also provide regulatory guidance, ensuring adherence to the latest governing body guidelines.
If you have questions about how we can help to integrate AI into your trial, we’d welcome the opportunity to connect.
1Snorkel AI. (n.d.). Memorial Sloan Kettering Cancer Center customer story. Snorkel AI. https://snorkel.ai/memorial-sloan-kettering-cancer-center-customer-story/
2 Greg Licholai, M. (2023, October 5). AI in clinical research: Now and beyond. Forbes. https://www.forbes.com/sites/greglicholai/2023/09/18/ai-in-clinical-research-now-and-beyond
3 How AI is improving diagnostics, decision-making and care: AHA. American Hospital Association. (n.d.). https://www.aha.org/aha-center-health-innovation-market-scan/2023-05-09-how-ai-improving-diagnostics-decision-making-and-care