Pharmaceutical companies are increasingly turning to artificial intelligence and digital health technologies to transform oncology clinical trials, with experts emphasizing that AI's greatest impact lies in matching the right patient populations with appropriate therapies to improve trial success rates.
Speaking at the 15th Annual Clinical Trial in Oncology West Coast conference held May 14-15 in Burlingame, California, Steve Warner, Senior Vice President and Head of Translational Research & Medicine and Oncology at Sumitomo Pharma America, outlined how AI is enabling his company to address large and growing datasets to identify optimal patient populations and design trials that address their specific needs.
Warner revealed that Sumitomo has begun experimenting with digital twins, utilizing preclinical and clinical study data to simulate trials. He stressed the significance of AI use in early clinical development, highlighting the adaptability it brings to trial designs.
Telehealth Integration Reduces Patient Burden
Jennifer Willert, Global Clinical Development Medical Director at Novartis Cell and Gene/Hematology, shared her experience implementing telehealth solutions at institutions including University of California San Francisco and Stanford University. Even before the pandemic, due to California's large geographic spread of patients, Willert had incorporated telehealth into clinical trials through virtual consent conferences, making trial participation more accessible and reducing travel burden.
"There is often an unrealistic expectation that patients will regularly travel for tests and visits," Willert noted. "Clinical trials are also frequently designed without sufficient consideration for the patient's experience, leading to poor patient retention and incomplete data."
With telehealth integration, laboratory tests no longer need to be conducted exclusively at clinical study sites. Some institutions now allow participants to transition into a "survival follow-up" phase with fewer in-person requirements and virtual options provided.
Wearable Technology Shows Promise Despite Adoption Challenges
Warner's team has begun incorporating wearables into studies for indications such as myelofibrosis, where tracking symptoms like fatigue and night sweats is critical and traditionally relies on patient-reported outcomes. Wearables offer a more data-driven approach through step counts or heart rate variation, potentially improving both clinical insights and patient compliance.
However, implementation challenges persist. An audience member shared that in a breast cancer study from UCSF, only 16% of participants used wearables, emphasizing the need for better adherence strategies.
Warner acknowledged that while wearable data is not typically used as a primary endpoint, it is increasingly being considered valuable supplementary data. He noted that sponsors often do not need or want access to all collected data, particularly location and other personal information, addressing privacy concerns.
Strategic Drug Development and Site Engagement
Warner emphasized that one of the most effective ways to engage clinical sites and patients is to advance only therapeutic agents that demonstrate compelling and meaningful signals during development. Sites and patients want to participate in truly transformative trials, while engagement becomes problematic when potential benefits are ambiguous.
"Advancing agents into Phase II or III studies without compelling activity in early development can strain limited site and patient engagement capacity," Warner explained. "Resources could be more effectively allocated to more promising candidates."
Abhijit Ramachandran, Director of Clinical Development Operations of Oncology at Corcept Therapeutics, agreed that drug activity is crucial and emphasized "making life easy for the sites." He discussed the burden on sites from multiple systems and the need for single sign-on platforms and e-learning systems to streamline operations.
Technology Implementation Requires Site Input
Edward Owen, Therapeutic Area Director of Oncology and Early Clinical Development at Genentech, stressed the importance of site involvement in technology development. He cited an instance where a digital consent platform was developed for a trial, but sites refused to use it.
"Sponsors need to receive feedback from sites on what they need before applying any future innovation," Owen urged. He also highlighted AI's significant potential for patient screening and protocol inquiries.
Balancing Digital Innovation with Personal Contact
Despite technological advancements, Warner emphasized that personal and in-person contact with sites remains irreplaceable. "In-person contact needs to be incorporated to augment any technologically advanced way to engage with sites," he said.
The panelists noted that successful implementation isn't about sponsors dictating solutions, but creating adaptable options. Owen agreed that involving patient advocacy groups to ensure patient voices are considered is crucial, with protocol flexibility allowing patients to choose between remote or in-person visits.
Addressing Implementation Challenges
Willert raised ethical considerations regarding wearables, including privacy and data collection concerns. She advocated for transparency and boundaries in data handling, noting that new regulations now require patient letters explaining trial results in accessible language at study completion.
Warner stressed that in indications focused on symptom tracking, wearable devices can relieve the burden of frequent questionnaires and may improve compliance, though concerns around data volume and privacy remain valid.
The conference highlighted that while digital solutions offer significant promise for improving patient-centric approaches in oncology trials, successful implementation requires careful consideration of site needs, patient preferences, and privacy concerns, combined with maintaining essential human connections in the clinical research process.
