Evaluating Mean Level and Within‐Person Consistency in Migraine Pain Intensity and Migraine‐Related Disability for AVP‐825 vs Oral Sumatriptan: Results from the COMPASS Study, A Randomized Trial

• Published in: Headache Vol. 59; no. 7; pp. 1002 - 1013
• Main Authors: McGinley, James S., Buse, Dawn C., Shulman, Kenneth J., Wirth, R. J., Hugentobler, Emmanuelle, Lipton, Richard B.
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01-07-2019
• Subjects: Consistency; Crossovers; Demographics; Demography; Headache; Migraine; Pain; Powder; Sumatriptan; Tablets
• ISSN: 0017-8748; 1526-4610
• DOI: 10.1111/head.13530

Background
Consistency of response across multiple attacks is typically measured as the proportion of study participants who achieve a categorical endpoint over a specified number of attacks (ie, 2‐hour pain‐free response in 2 of 3 attacks). We applied a novel analytic approach for measuring consistency of response in the acute treatment of episodic migraine using data from the COMPASS study.
Methods
The COMPASS study (NCT01667679) was a multiple attack crossover study which compared AVP‐825, a Breath Powered® intranasal delivery system for low‐dose sumatriptan powder (22 mg), with 100‐mg oral sumatriptan tablets in the acute treatment of migraine. Participants were 18–65 years old, met ICHD‐2 criteria for migraine with or without aura, and had migraine for ≥1 year prior to screening. They were instructed to treat up to 5 migraine attacks with each treatment and recorded migraine pain intensity and disability data at pre‐dose and 10, 15, 30, 45, 60, 90, and 120 minutes post‐dose for each attack. We explored the mean level and within‐person variability (WPV; a measure of consistency) in migraine pain intensity and migraine‐related disability across multiple attacks after treatment with AVP‐825 (22 mg) vs oral sumatriptan (100 mg) using location scale mixed‐effects models (LSMEMs). LSMEMs controlled for pre‐dose pain/disability, demographics, treatment sequence, and treatment period.
Results
The mean age was 40 and the sample was 84.6% women. Participants (N = 259) treated an average of 6.8 attacks each during the course of the study. Attacks treated with AVP‐825 showed significantly lower mean pain intensity and mean disability from 10 to 90 minutes post‐dose (effect sizes ranged from −0.09 to −0.29 and P values ranged from P < .0001 to P = .01). WPV was significantly greater at 10–15 minutes (WPV ratios ranged from 1.20 to 1.58 and P values ranged from P < .0001 to P = .04) but significantly reduced from 45 to 120 minutes for attacks treated with AVP‐825 compared to oral sumatriptan (WPV ratios ranged from 0.67 to 0.81 and P values ranged from P < .0001 to P = .03).
Conclusions
LSMEMs demonstrate that treatment with AVP‐825 is associated with lower average migraine pain intensity and disability from 10 to 90 minutes and greater within‐person consistency across multiple migraine attacks (reduced WPV) from 45 to 120 minutes post‐dose compared to oral sumatriptan. These findings may reflect the more rapid and consistent absorption of sumatriptan using AVP‐825. Increased WPV with AVP‐825 in the first 15 minutes likely reflects the earlier onset of treatment effects with the device compared to oral sumatriptan. LSMEMs show promise as a novel approach for assessing and comparing consistency of treatment response in migraine trials.