A Risk-Difference Meta-Analysis for the Prophylactic Treatments of Chronic Migraine

Chronic migraine (CM) imposes significant personal, societal, and financial burdens, historically lacking specific prophylactic treatments. Monoclonal antibodies (mAbs) targeting calcitonin gene-related peptide (CGRP) represent a novel, mechanism-based, and migraine-specific prophylactic approach. Four mAbs, namely, erenumab, fremanezumab, galcanezumab, and eptinezumab, have been marketed, although head-to-head trials with standard anti-migraine treatments are absent. This study aimed to compare the efficacy and safety of anti-CGRP mAbs with standard anti-migraine treatments using a cross-trial indirect model of the absolute risk difference (ARD) of a 50% responder rate, in order to express the final results in terms of the number needed to treat (NNT) and number needed to harm (NNH). Phase 3 and 2b randomized controlled trials (RCTs) for CM prophylaxis were searched in the MEDLINE and CENTRAL databases with specific inclusion and exclusion criteria. The ARD between groups for the percentage of trial participants who reported a 50% reduction in monthly migraine days and the differences in the number of adverse events (AEs), serious adverse events (SAEs), and participants who withdrew from each RCT were calculated, and subsequently, the NNT and NNH were calculated for each one of the outcome measures. In total, eight RCTs were considered eligible. A similar efficacy and safety have been demonstrated among CGRP mAbs and all standard CM treatments. The results of the ARD for the total number of studies concerning efficacy, total adverse events, serious adverse events, and dropout from the RCTs ranged from -0.688 (95% confidence interval (CI): -0.821-(-0.513)) to -0.018 (95% CI: -0.044-(0.007)), from 0.032 (95% CI: -0.041, 0.104) to -0.380 (95% CI: -0.589, -0.126), from -0.025 (95% CI: -0.046, -0.006) to 0.014 (95% CI: -0.015, 0.42), from 0.048 (95% CI: -0.112, 0.014) to 0.232 (95% CI: -0.016, 0.458) correspondingly. All anti-CGRP mAbs showed a roughly equal statistically significant ARD and similar NNTs, ranging from 5 to 8, while the ARD of onbotulinum toxin A (oBTA) was not significant with an NNT 56. The two studies of topiramate showed contradictory results, the one significant while the other not, with NNTs 2 and 22, respectively. All four anti-CGRP mAbs showed an invariably high efficacy among their studies, in terms of the ARD and its derivative measure of NNT, in contrast to oBTA, while in topiramate, the results are contradictory between the two studies.