P225: Successful Implementation of a Fully Remote Participation Option in the Phase 3 Pivotal GLISTEN Study
Poster Presenter
Vamsidhara Dhulipala
Senior Director, Global Regulatory Affairs at GSK
GSK United States
Objectives
To assess the impact of including a fully remote, direct-to-patient model for the first time in a published pivotal hepatology trial, the Phase 3 GLISTEN study, which investigated the efficacy and safety of linerixibat for the treatment of cholestatic pruritus in patients with PBC (NCT04950127).
Method
We report trial conduct and outcomes in GLISTEN between the fully remote participation model (eConsent, remote screening and clinical outcomes assessments, telemedicine, home health visits, direct-to-patient treatment shipping) implemented in the US and the traditional on-site participation.
Results
Of 238 patients in GLISTEN, 17 (7%) participated via the fully remote model, representing 47% (17/36) of US participants. The remaining 221 (93%) participated via physical sites globally, including 6 (3%) with hybrid participation (=1 remote visit). The proportion of patients completing the study was similar between fully remote (n=14; 82%) and physical sites (n=197; 89%). In the fully remote model, the most common reason for study withdrawal was physician decision (n=2; 12%); no patients withdrew due to burden of procedures. At physical sites, the main reason for withdrawal was patient decision (n=15; 7%); 4 patients (2%) cited burden of procedures. The most common important protocol deviations were missed or out of window/out of order assessments (fully remote, n=9 [53%]; physical, n=146 [66%]), and deviation from study procedures (fully remote, n=10 [59%]; physical, n=100 [45%]). Across visits most fully remote and physical site patients were highly compliant with data reporting requirements via their electronic diary (e-diary), enabling calculation of the monthly itch score (key efficacy data for GLISTEN). Pruritus improvement from baseline over 24 weeks (primary endpoint) was seen in both the fully remote model and physical sites, with no statistically significant interaction between visit model (fully remote or physical) and treatment (post-hoc; p=0.734). In the overall study population, pruritus improvement from baseline over 24 weeks was greater with linerixibat versus placebo (least squares mean change: -2.86 vs -2.15; adjusted mean difference -0.72 [95% CI -1.15, -0.28]; p=0.0013); these results were similar when the fully remote population was excluded in a sensitivity analysis at the request of a regulator (linerixibat: -2.76 vs placebo: -2.07; adjusted mean difference -0.68 [95% CI -1.14, -0.23]; p=0.003). Adverse events (AEs) were experienced by 94% (n=16) patients in the fully remote model and 82% (n=181) at physical sites.
Conclusion
To our knowledge, this piloted participation model was the first published pivotal hepatology trial where patients could join a study entirely remotely, without the need to visit a physical site. Efficacy data were collected in all patients via e-diary technology, facilitating fully remote participation. The fully remote model had the highest enrollment of all US sites, indicating strong patient interest in this option. The proportion of patients completing the trial was similar between the fully remote model and physical sites, with similar e-diary compliance regardless of the visit model, and comparable rates of protocol deviations, supporting the feasibility and compliance potential of fully remote participation. The incidence of AEs across the fully remote and physical site populations suggest that safety data can be captured adequately without relying on traditional site visits. Similar trends in primary efficacy data were observed between patients from physical sites and those in the fully remote model with no significant interaction between the treatment and visit model. A sensitivity analysis showed that the primary efficacy data remained consistent when the fully remote population was excluded, suggesting that data integrity was maintained in the remote model. The fully remote option enables individuals, particularly those in wide geographical areas or experiencing challenging symptoms (such as pruritus, sleep disturbances, and fatigue), to participate in clinical trials with a lower overall burden; this is valuable in rare diseases such as PBC, especially when patient-reported outcomes (key study endpoints) can be provided by patients in a remote setting. By reducing patient burden and eliminating geographic access as a barrier, this model significantly expands our ability to conduct clinical trials without compromising efficacy, data integrity, or safety oversight.
Funding: GSK (212620)
