In the evolving world of clinical trials, AI and decentralization are key drivers of change. Iddo Peleg, CEO of Yonalink, shares insights on how these innovations enhance trial efficiency, reduce costs, and improve patient recruitment. Peleg discusses the challenges and successes of implementing AI-driven solutions and the future of decentralized trials, offering a glimpse into the transformative potential of these advancements.
Moe: You explored AI in trials. Can you share data or cases showing AI’s impact on efficiency, cost, and recruitment?
Indeed, in my experience, the data streaming from Electronic Health Records (EHR) to Electronic Data Capture (EDC) systems exemplifies AI’s transformative power. Traditionally, this data transfer was manual, consuming for example, around 200 hours per patient in oncology trials. With AI, this process takes less than an hour and has significantly higher accuracy. For instance, in pilot studies conducted globally, including sites in Israel and the U.S., we reduced data transfer time from 6-8 weeks to less than a day. This cuts costs and accelerates the entire trial process, allowing for quicker decision-making. The AI-driven automation ensures that data is harmonized and filtered efficiently, eliminating human error and enhancing the overall quality of the trial data. This transformation is crucial for sponsors who need reliable and timely data to make informed decisions, ultimately leading to faster and more effective clinical trials.
Moe: Decentralized trials are popular, but adoption varies. What evidence shows their impact on success, adherence, and compliance?
While complete decentralization remains rare, hybrid models are gaining traction. These models integrate smaller medical centers into the trial process, especially in rural areas. This approach significantly increases access to clinical trials, available to less than 7% of Americans. By bringing trials closer to patients, we enhance participation and adherence. For example, in the U.S., we are working to include smaller clinics outside major urban centers, effectively turning them into satellite sites that contribute to the main trial, thus improving protocol adherence and compliance. This model broadens the reach of clinical trials and ensures that diverse patient populations are represented, leading to more comprehensive and inclusive research outcomes. The hybrid approach balances the benefits of decentralization with the need for centralized oversight, ensuring that trials remain compliant with regulatory standards while maximizing patient engagement.
Moe: Many claim seamless EHR to EDC integration. Can you share a real-world example of resolving data transfer inefficiencies?
In my experience, successful EHR to EDC integration requires addressing four key challenges: data extraction, harmonization, filtering, and transfer. Our approach is flexible, allowing us to work independently or with partners. For instance, we compared manual data capture with automated transfer in a pediatric oncology trial at Stanford. The automated system improved accuracy and significantly reduced time and effort. This trial demonstrated that we could achieve seamless data integration with minimal interaction from medical centers, setting a new standard for efficiency and reliability in clinical trials. Automating these processes eliminates the bottlenecks associated with manual data entry, ensuring that data is transferred quickly and accurately. This enhances the data’s quality and frees up valuable resources that can be redirected towards more critical aspects of the trial, such as patient care and protocol optimization.
Moe: With AI in trial decision-making, what safeguards ensure unbiased patient selection and protocol optimization?
From my perspective, accuracy is paramount in AI-driven systems. We encourage sponsors to verify our system’s accuracy through initial 100% source data verification (SDV) for the first patient, gradually reducing as confidence builds. This ensures transparency and trust. Additionally, site feasibility now includes criteria for data streaming capabilities, ensuring sites can handle automated data transfers effectively. This approach mitigates bias and ensures that the data collected is reliable and actionable, paving the way for more informed decision-making in trial design and execution. By implementing these safeguards, we ensure that AI is used responsibly and ethically, maintaining the integrity of the trial process. Our commitment to accuracy and transparency builds confidence among stakeholders, fostering a collaborative environment where innovation can thrive without compromising patient safety or data quality.
Moe: How should the industry balance data interoperability, compliance, and patient autonomy in clinical research?
In my view, this is a complex issue, as regulations vary globally. In the U.S., we leverage the 21st Century Cures Act, where patients consent and establish data connections. Internationally, we work with medical centers to ensure compliance. For example, in Israel, we have successfully implemented a system where the study coordinator establishes the connection, providing compliance and scalability. Our system’s flexibility allows us to adapt to different regulatory environments, ensuring patient autonomy is respected while maintaining data integrity and interoperability. By prioritizing patient consent and involvement, we empower individuals to take control of their health data, fostering trust and engagement in the research process. This approach aligns with regulatory requirements and enhances the overall quality and reliability of the data collected, ultimately leading to more effective and patient-centered clinical trials.
Moe: Are traditional KPIs like enrollment speed and data quality the best measures of trial success, or should we consider alternatives?
In my opinion, while traditional KPIs like enrollment speed and data quality are essential, the focus should shift to the percentage of data points captured electronically, as this ensures accuracy. In oncology trials, we achieve 85-88% electronic data streaming, aiming for over 90%. This shift in metrics better reflects the quality and efficiency of modern trials. By prioritizing electronic data streaming, we can ensure that the data is accurate and timely, providing a more comprehensive view of trial success. This approach allows us to identify and address potential issues early in the trial process, improving overall outcomes and reducing the risk of costly delays. By embracing new metrics that reflect the realities of modern clinical research, we can drive innovation and improve the efficiency and effectiveness of clinical trials, ultimately benefiting patients and sponsors alike.
Moe: Is there anything else you’d like to add?
I believe it’s crucial to adopt a new mindset in clinical trials. We have a responsibility to patients to bring therapies to market faster. Embracing innovation and data streaming is essential. This change is happening in other industries and must extend to clinical trials to improve outcomes and efficiency. By fostering a culture of innovation and collaboration, we can overcome the traditional barriers that have hindered progress and ensure that clinical trials are conducted in a way that is both efficient and patient-centric. It’s time for the industry to move beyond outdated practices and embrace the potential of new technologies to transform how we conduct research. By doing so, we can accelerate the development of new therapies and improve patients’ lives.
Moe Alsumidaie is Chief Editor of The Clinical Trial Vanguard. Moe holds decades of experience in the clinical trials industry. Moe also serves as Head of Research at CliniBiz and Chief Data Scientist at Annex Clinical Corporation.
