Author: Heather Nonnemacher, Associate Director, Services Solutions
AI's transformative potential for clinical trials
In my previous blog post, I discussed the potential of Artificial Intelligence (AI) in redefining clinical trials. Its capability to augment processes, from improving patient comprehension in informed consent documents to analyzing unstructured data in different trial phases, signals a powerful shift towards increased efficiency in clinical research.
We have plenty of reasons to be optimistic about the future of AI in clinical trials. At the same time, embarking on this pathway requires a cautious approach to safely and ethically harness AI’s capabilities.
Challenges in clinical trials during critical moments
The path of clinical trials is filled with 'urgent moments' – scenarios demanding swift, effective responses. In recent years, AI has been an important assistive tool to respond to these urgencies.
For example, AI accelerated COVID-19 vaccine development to just 12-16 months by enabling rapid protein structure prediction, vaccine candidate design, and clinical trial optimization.1 AI is also playing a role in addressing standard of care and comparator drug shortages in clinical trials; up and coming technology-companies like iethico are working to resolve drug shortages through AI prediction, pre-empting crises before they affect patient care.2
Future applications of AI in clinical trials
AI is already supporting trial sponsors and sites to manage urgent clinical trial challenges, and its potential extends beyond crisis management.
Promising areas where AI could transform the future of clinical trials include:
Research and development
- Compound development and strategy: Traditional clinical trials often include a broad range of patients, including patients unlikely to respond to treatment. AI, when mining rich patient data, can pinpoint subgroups most likely to benefit from a specific treatment, leading to more efficient trials and faster development of effective therapies.3
Patient recruitment, retention and participation
- Enhancing patient recruitment: AI algorithms can identify eligible trial participants quickly and efficiently by analyzing raw data, reducing workloads, and streamlining the recruitment process. For instance, a study of pediatric oncology trial recruitment found AI systems could reduce physicians’ trial recruitment workload between 85-90%.4
- Medication management: Almost 40% of patients become nonadherent to investigational medical products (IMP) in clinical trials lasting longer than 150 days.5 Smartphones, together with AI, could help patients adapt to their medication schedules as needed. AI can create personalized suggestions based on data from patient smart devices, including reminders to speak to a physician if concerning trends are found.6
- Facilitating care as part of the clinical team: AI-powered solutions can guide patients through eligibility assessments and consent processes remotely, enhancing patient comprehension and access to trials, as well as reducing administrative burdens on sites and sponsors. And, people like AI: in an evaluation of AI and physician responses to patient questions, published in JAMA Internal Medicine, licensed health care professionals were three times more likely to rate AI responses as “good” or “very good” compared to physician responses.7
Study design optimization
- Adaptive trial designs: AI can facilitate adaptive trial designs, analyzing interim results data in real-time to modify study parameters, improving resource utilization and patient outcomes.
Outcomes analysis
- Data optimization and analysis: AI can optimize trial endpoints by identifying patient characteristics that are good proxies for treatment effectiveness.
The journey forward with AI
Artificial intelligence is poised to create substantial impact in clinical trials. From enhancing patient comprehension across diversified and broad-scale trials to refining recruitment and compliance strategies, AI has potential to fundamentally shift clinical trial processes and outcomes.
However, it's crucial to approach this promising horizon pragmatically by taking measured steps to integrate AI technologies responsibly. When applied methodically, rooting each step forward in practicality and patient-need, AI offers a compelling vision for how our industry can bring medicines to patients as safely and quickly as possible.
References
- Ashwani Sharma et al., “Artificial Intelligence-Based Data-Driven Strategy to Accelerate Research, Development, and Clinical Trials of COVID Vaccine,” BioMed Research International 2022 (July 6, 2022): 1–16, https://doi.org/10.1155/2022/7205241.
- iethico, accessed April 12, 2024, https://iethico.com/.
- “How Artificial Intelligence Can Power Clinical Development,” McKinsey & Company, November 22, 2023, https://www.mckinsey.com/industries/life-sciences/our-insights/how-artificial-intelligence-can-power-clinical-development.
- 1. Yizhao Ni et al., “Increasing the Efficiency of Trial-Patient Matching: Automated Clinical Trial Eligibility Pre-Screening for Pediatric Oncology Patients,” BMC Medical Informatics and Decision Making 15, no. 1 (April 14, 2015), https://doi.org/10.1186/s12911-015-0149-3.
- “Non-Adherence: A Direct Influence on Clinical Trial Duration and Cost,” Applied Clinical Trials, November 12, 2020, https://www.appliedclinicaltrialsonline.com/view/non-adherence-direct-influence-clinical-trial-duration-and-cost.
- “Catalyzing Commitment: The Symphony of AI-Powered Medication Adherence,” Pharmacy Times, May 15, 2024, https://www.pharmacytimes.com/view/catalyzing-commitment-the-symphony-of-ai-powered-medication-adherence.
- John W. Ayers et al., “Comparing Physician and Artificial Intelligence Chatbot Responses to Patient Questions Posted to a Public Social Media Forum,” JAMA Internal Medicine 183, no. 6 (June 1, 2023): 589, https://doi.org/10.1001/jamainternmed.2023.1838.
Heather Nonnemacher
Associate Director, Services Solutions, Suvoda