OBJECTIVES : To compare the efficacy and safety of three new acute treatments for migraine (lasmiditan, rimegepant, ubrogepant) to placebo, triptans (sumatriptan and eletriptan) and each other.

METHODS : We systematically identified and reviewed randomized controlled trials of the interventions and comparators in adults with a diagnosis of migraine with or without aura as specified by ICHD diagnostic criteria. Study selection and data abstraction were completed independently by three reviewers. Study results were synthesized using a Bayesian network meta-analysis, and results were presented in terms of odds ratio (OR).

RESULTS : We identified 33 single migraine attack trials, of which 10 trials assessed the new treatments and 23 trials assessed one or more of the triptans. The NMA showed lasmiditan, rimegepant, and ubrogepant were statistically superior on the primary outcome of pain freedom at two hours compared to placebo (OR 2.12 -3.01), however, compared to each other there was no statistically significant difference. In contrast, all these new interventions showed lower odds of achieving pain freedom at two hours compared to eletriptan (OR 0.38-0.54) and sumatriptan (0.52-0.73). Similar results were observed for headache relief at two hours and sustained pain freedom at 24 hours. NMA results showed no differences in the odds of treatment emergent adverse events (TEAE) among rimegepant, ubrogepant, triptans, and placebo in the single-attack trials. However, lasmiditan had higher odds of causing TEAE compared to placebo (5.99, 95% Crl: 3.3, 12.52), rimegepant (4.00, 95% CrI: 1.38, 12.04), ubrogepant (5.10, 95% CrI: 2.31, 12.95), and the triptans (OR 2.57-3.27).

CONCLUSIONS : Lasmiditan, rimegepant and ubrogepant provide substantial benefit relative to placebo in acute treatment of migraine. They had similar benefit compared to each other, but lasmiditan had more side effects. All three appear less effective than triptans. Relative benefits and harms of these new treatments with repeated use over time remains to be determined.