Design of a Phase 2b Dose-finding Trial to Evaluate Safety and Efficacy of the CNS-penetrant BTK Inhibitor SAR442168 in Patients with Relapsing Forms of Multiple Sclerosis
Anthony Traboulsee1, Christopher LaGanke2, Xavier Montalban3,4, Sibyl Wray5, Xinyan Zhang6, Andre Matta6, Timothy Turner6, Erik Wallstrom6
1University of British Columbia, 2North Central Neurology Associates, 3Division of Neurology, St Michael’s Hospital, University of Toronto, 4Department of Neurology-Neuroimmunology, Multiple Sclerosis Center of Catalonia, Vall d’Hebron University Hospital, 5Hope Neurology, 6Sanofi
Objective:
To describe the design of a phase 2b, proof-of-concept, dose-finding trial of the CNS-penetrant Bruton’s tyrosine kinase inhibitor SAR442168 in patients with relapsing MS, which implements important design elements intended to improve efficiency and reduce exposure to placebo.
Background:
With an increasing number of MS therapies available, to justify placebo arms in clinical trials is becoming difficult, and using alternative approaches is necessary. MRI outcomes in phase 2 trials are well-established predictive biomarkers for clinical outcomes in phase 3 MS drug development.
Design/Methods:
The phase 2b trial (DRI15928, NCT03889639) was a 16-week, placebo-controlled, crossover trial testing four SAR442168 doses (5, 15, 30, or 60 mg q.d., administered orally) in patients with relapsing MS. Patients were assigned to receive placebo for 4 weeks (weeks 1–4) then one of four doses of SAR442168 (weeks 5–16), or to receive SAR442168 (weeks 1–12) then placebo (weeks 13–16), in a double-blinded fashion. The Multiple Comparison Procedure with Modeling was selected to assess the dose-response relationship based on the reduction in the number of new gadolinium (Gd)-enhancing T1-hyperintense lesions; this method will greatly increase the power of dose-finding exercises based on the a priori assertion that the dose-response relationship is best described by one of six prespecified models (2 Emax, linear, quadratic, logistic, exponential).
Results:
Our modeling approach could provide ≥89% power to detect a maximum 85% risk reduction in the number of new Gd-enhancing T1-hyperintense lesions using a total of 120 patients (26 per dose, assuming 15% drop-out rate) with a 2-sided α=0.05.
Conclusions:
By applying modern trial design elements, we developed a compact, MRI-based trial for proof-of-concept and dose-finding for SAR442168 in patients with relapsing MS without the need for a placebo-only arm.