In this insightful interview, I spoke with Gen Li, CEO of Phesi, a pioneering company focusing on AI in trials. We discussed the innovative use of AI-driven digital twins, data privacy measures, efforts to ensure diverse patient populations, and how their technology streamlines clinical trials and reduces costs.
Moe Alsumidaie: How does using AI-driven digital twins address ethical concerns compared to traditional clinical trial methods?
Gen Li: Our approach to using AI-driven digital twins is unique because we don’t directly handle patient records ourselves, which helps us avoid many ethical issues. We leverage AI to gather data through legally approved means. Starting with a digital patient profile, we apply inclusion and exclusion criteria and purify the data to define baseline characteristics, safety, and outcomes. This method maintains transparency and ensures validation compared to traditional methods. A collaboration with Harvard Medical School recently showcased this process effectively.
Additionally, by avoiding direct handling of patient records and focusing on AI-driven data collection and processing, we address concerns related to the transparency and validation of these digital twins. Our method allows us to maintain ethical standards while providing reliable and accurate data for clinical trials. This approach not only streamlines the process but also enhances the overall integrity of the trial data.
Moe Alsumidaie: How do sponsors ensure the privacy and security of sensitive data in using technology platforms?
Gen Li: Sponsors should check to ensure that technology systems comply with GDPR and are certified. Patient data should be fully contextualized, meaning we know its source, collection methods, and design. This thorough documentation ensures data integrity and regulatory compliance. Technology-detailed contextualization processes should allow sponsors to maintain high privacy and security standards while handling data from over 120 million patients in our case.
Technology service providers’ commitment to privacy and security is fundamental. It is critical to ensure that sensitive patient data remains secure by adhering to stringent regulatory standards and implementing robust data contextualization practices. This approach meets regulatory requirements and builds trust with our clients and patients, ensuring that their data is handled with the utmost care and respect.
Moe Alsumidaie: How do data-driven approaches help select diverse patient groups for clinical trials?
Gen Li: Contextualized data addresses diversity issues by analyzing ethnic compositions in data cohorts. For example, we can recommend sites and investigators with better diversity representation in non-small cell lung cancer. This approach improves trial results and helps understand the geographical prevalence of diseases, aiding community development programs. One client aimed to include 13% African American participants in a cardiovascular trial but found only 3% representation. Using our data-driven approach, we identified and corrected such disparities.
By leveraging detailed data analysis, we ensure our trials include diverse patient populations, crucial for obtaining accurate and representative results. This commitment to diversity improves the quality of our trials and aligns with regulatory priorities, ensuring our methods meet the highest standards of inclusivity and representation in clinical research.
Moe Alsumidaie: What are the next steps for expanding Phesi’s digital patient profile catalog and its role in trial design?
Gen Li: The expansion of our catalog is client-driven rather than proactive. It serves as a communication tool to showcase our capabilities, providing top-line summaries of the data we handle. However, its real value comes from applying this data in specific development contexts. For example, the catalog helps potential clients understand the depth and breadth of our data, allowing them to make informed decisions about their clinical trial designs. This client-centric approach ensures that our expansions are relevant and directly meet market needs.
Our catalog’s role in transforming clinical trial designs is significant. We enable more precise and effective trial designs by offering detailed and contextualized patient data. This improves trial outcomes and reduces the need for amendments and regulatory issues, streamlining the entire clinical trial process. Our focus on client needs ensures that our catalog remains valuable for developing innovative and effective clinical trial strategies.
Moe Alsumidaie: Can you provide examples of how technology has streamlined clinical trials and reduced costs?
Gen Li: Technology integrates multiple datasets to improve site selection, trial design, and patient alignment, reducing amendments and improving regulatory discussions. For instance, we might recommend reducing the number of trial sites in rescue trials to avoid crowding and enhance enrollment. Digital twins can replace or partially replace control arms, saving costs and simplifying trial designs. We have supported many successful drug launches by providing actionable insights and improving trial efficiency. A notable example is our involvement in the KEYTRUDA and Dinutuximab trials, where we provided critical data that led to significant cost savings and more effective trial designs.
By streamlining these processes, technology use ensures faster and more cost-effective clinical trials. Our technology reduces the overall time and cost and improves the quality and reliability of the trial outcomes. This efficiency is critical in bringing new treatments to market faster and more efficiently, benefiting patients and the healthcare industry.
Moe Alsumidaie: How can sponsors measure success and the impact of its technologies on clinical trial outcomes?
Gen Li: Sponsors can measure impact in three main areas: site and country selection driven by digital patient profiles, trial design alignment to avoid amendments, and using digital twins to replace control arms or provide historical controls. These measures improve enrollment rates, shorten cycle times, and enhance regulatory compliance. Our successful track record in supporting drug launches and achieving cost savings highlights the effectiveness of our solutions. For example, our digital patient profiles help optimize site selection and patient enrollment, while our balanced scorecard approach ensures all datasets are meaningful and correlated, driving better trial outcomes.
Success is also measured by service providers’ ability to improve regulatory discussions and align trial designs with patient populations. By providing quantifiable evidence and objective insights, we facilitate more effective communication with regulatory bodies, ensuring smoother approval processes. Our impact is evident in the successful drug launches we have supported, showcasing the tangible benefits of our innovative approaches.
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.
