Clinical Trial Design in Chronic Migraine: Lessons Learned from the NINDS CRC Chronic Migraine Treatment Trial (CMTT)

Publication Type
Conference Paper
Year of Publication
Dodick, D; Silberstein, S; Lindblad, A; Holroyd, K; Mathew, N; Cordell, J; Hirtz, D
Date Published
Washington, D.C.
Objectives: Design an efficient, robust randomized placebo controlled clinical trial in chronic migraine.Background: Open-label studies and conventional wisdom suggest that the combination of two migraine preventive agents may result in incremental benefit over a single agent when efficacy is suboptimal. Prior to the launch of the CMTT, no randomized trials had assessed the value of frequently used preventive agent combinations for chronic migraine (CM). CMTT was expected to randomize 250 CM subjects not adequately controlled (≥10 headaches/month) with topiramate (up to 100mg/day) to either the addition of propranolol LA (up to 240mg/day) or placebo. Methods: The study used the International Classification of Headache Disorders-II (ICHD-II) guidelines and diagnostic criteria. Up to 70 sites were planned for activation. Initial eligibility required subjects with CM already on preventive medication to undergo a two month washout period. Current use of topiramate, not exceeding 100mg, was permitted. Subjects not taking topiramate were required to participate in a 4 week topiramate titration-phase prior to beginning the 28-day baseline diary. Results: 70 sites were activated and 48 randomized at least one subject. Unwillingness to terminate other migraine preventatives for two months contributed to subject screening ineligibility. 68% initiating screening failed to randomize. The primary reasons for failure to randomize were loss/refusal of follow-up (30%), too few headaches to meet the ICHD-II CM definition (23%), and inability to tolerate at least 50 mg of topiramate (19%). The number of headaches required for eligibility after topiramate titration was modified from 15 to 10 days/month to minimize ineligibility rates and increase the number of subjects who could be randomized. Conclusions: Lessons learned from CMTT regarding trial design, implementation, and recruitment, will inform the design of future trials evaluating the efficacy of combination preventive medications in CM subjects who demonstrate a suboptimal response to one preventive medication.