The objective of the present work was to use modeling and simulation to inform trial design of a proof-of-concept study for agents used in the treatment of hyperhidrosis. Data were available from 36 subjects who received the vehicle, 2% or 4% topical glycopyrrolate wipes daily for 4 weeks, with response (hyperhidrosis disease severity scale [HDSS] and sweat production [SP]) measured weekly. The HDSS and SP time courses were best described using a longitudinal model with maximum response achieved by 1 week. Glycopyrrolate 4% had a higher HDSS responder rate than 2% (50% vs 33%) and placebo (0%) at week 1. Mean change from baseline (mg/5 min [SD]) in SP at week 1 was -90 (220), -185 (214), and -271 (265) for placebo, 2%, and 4% glycopyrrolate, respectively. Subjects with higher baseline SP had higher sweat reduction from baseline. Virtual clinical trials were simulated and analyzed using conventional (at the end of the study) versus model-based methods to determine sample size for achieving 80% power to identify a dose-response relationship. Twenty-seven subjects compared with at least 120 subjects would be needed using model-based and conventional methods, respectively. Thus, the model-based method using longitudinal data required fewer subjects than the conventional single-point method.
Keywords: Hyperhidrosis Disease Severity Scale (HDSS); glycopyrrolate; modeling and simulation; sweat production (SP).
© 2015, The American College of Clinical Pharmacology.