Different Doses of Erenumab for Preventive Treatment of Episodic Migraine: A Systematic Review and Network Meta-Analysis

Background Erenumab is a novel monoclonal calcitonin gene–related peptide receptor antibody that is used for the preventive treatment of migraine. Objectives To evaluate overall safety and ecacy and dose-response relationship of erenumab in patients with episodic migraine and patients with prior migraine treatment failures. Methods We searched randomized clinical trials on PUBMED, EMBASE database, and Cochrane Library database. A pair-wise meta-analysis and Bayesian network analysis were performed. Results For ecacy outcomes, the network meta-analysis suggests that compared with erenumab 70 mg, participants received erenumab 140 mg reported signicantly decreased MSMD and increased 50% response rate, and erenumab was most likely to be ranked rst for MMD, MSMD and 50% response rate. For safety outcomes, the network meta-analysis has found no signicant difference between the 70 mg group and the 140 mg group measured by AE and SAE. For patients with ≥ 2 treatment failures, 140mg erenumab group, patients with ≥ 2 treatment failures reported signicantly reduced MMD and MSMD, increased 50%, and 75% response rate, compared with placebo. For safety outcomes, no signicant difference was found between 140 mg erenumab group and the placebo group. Conclusion Erenumab was effective in patients with episodic migraine. 140 mg erenumab was associated with better ecacy outcomes without increased risk for developing adverse events compared with 70 mg erenumab, and 140 mg erenumab was effective in patients with prior migraine treatment failures. screened omissions. Disagreements were resolved through discussion. using of 1.005. The node-split method was used for assessing inconsistency. Log odds ratio (LOR) and mean difference with their 95% CI were calculated for binary variables and continuous variables, respectively. Furthermore, our study pooled both direct and indirect evidence with the Bayesian network meta-analysis method and demonstrated 140 mg erenumab outperforms 70 mg erenumab in multiple ecacy endpoints while related to the same risk for developing adverse events. In addition, our study demonstrated that 70 mg and 140 mg erenumab were associated with signicantly reduced migraine frequency and improved functional outcome in patients with prior migraine treatment failures. The heterogeneity was low across all the primary outcomes, indicating a high level of clinical evidence. It is the rst meta-analysis comparing the safety and ecacy of different doses of erenumab and the rst meta-analysis to demonstrate the safety and ecacy of erenumab in patients with ≥ 2 prior migraine treatment failures.


Introduction
Migraine is one of the most common neurological disorders, with an estimated global prevalence of 15% [1]. It is manifested by recurrent attacks of unilateral, pulsatile headache that could be triggered by daily activities. A migraine attack usually lasts for 4 to 72 hours and is frequently accompanied by nausea, vomiting, photophobia, phonophobia. Thus, patients may suffer from signi cant disabling pain and impaired quality of life.
As an expanding understanding of the migraine pathogenesis in the past decades, multiple targets have been identi ed for migraine therapy.
Among them, calcitonin gene-related peptide (CGRP) is considered one of the most promising targets [2]. Small molecule CGRP antagonists have been mainly used for acute treatment of migraine, while monoclonal antibodies (mAb) against CGRP or its receptor are used for preventive migraine treatment [3]. Erenumab, an IgG2 human mAb selectively binds to the CGRP receptor, was approved by the Food and Drug Administration for the prevention of migraine [4]. Previous trials showed that erenumab given as a 70 mg or 140 mg subcutaneous injection once per month was associated with a signi cant reduction of monthly migraine days (MMD) and monthly acute migraine-speci c medication days (MSMD) from baseline compared with the placebo, and is not related to an increased risk of developing adverse events [1]. Besides, the unique pharmacological mechanism suggests the potential advantages of erenumab in migraineurs with prior preventive treatment failures.
So far, prior studies have demonstrated that erenumab use was associated with reduced migraine frequency and improved functional outcome compared with placebo; however, no study has been performed to evaluate the dose-response relationship of erenumab for the preventive treatment of migraine. Furthermore, although the effectiveness of erenumab in patients with prior treatment failure has been demonstrated in multiple trials, no systematic approach has been used. The purpose of this study is to conduct a meta-analysis to evaluate the effectiveness and safety of different doses of erenumab in preventive treatment in migraine patients and in patients with prior preventive treatment failures.

Methods
A meta-analysis was performed following the Preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines [5]. The study was registered at PROSPERO (http://www.crd.york.ac.uk/PROSPERO; registry number: CRD42020198985).

Search strategy
Two researchers independently conducted the online search procedure. The related articles published before April 1, 2020, were searched through PUBMED, EMBASE database, and Cochrane Library database. The search terms include migraine, erenumab, AMG334, and randomized clinical trials. The full electronic search strategy was listed in Supplement S1. The references listed in the relevant articles were screened to avoid omissions. Disagreements were resolved through discussion.

Study Selection
The study included in the meta-analysis all met the following criteria: 1) randomized clinical trials; 2) patients with a history of episode migraine for ≥ 12 months according to the International Classi cation of Headache Disorders; 3) using erenumab as an experimental group intervention.

Data extraction
The extraction of data was carried out by two investigators independently. The extracted data include (1) literature information: title, author, publication time, sample size; (2) characteristic of the object of the study including region, age, sex ratio, BMI, age at migraine onset, MMD, MSMD, number of patients who have failed previous preventive migraine medication due to insu cient e cacy or (and) unacceptable tolerability; (3) interventions: the dose of erenumab included in any phase III trials; (4) outcome data: the change in MMD, MSMD at month 3, 50% response rate (the proportion of patients achieved a 50% or greater reduction in MMD from baseline) and 75% response rate (the proportion of patients achieved a 75% or greater reduction in MMD from baseline) at month 3, the change in Headache Impact Test (Hit-6) at month 3, Migraine Physical Function Impact Diary-Everyday Activities and Physical Impairment (MPFID-EA and MPFID-PI) at month 3, number of patients with adverse events (AE) and serious adverse events (SAE) throughout the study.

Pair-wise meta-analysis
We conducted a pair-wise meta-analysis to evaluate the overall safety and e cacy of erenumab compared with the placebo and the safety and e cacy of different doses of erenumab in patients with prior treatment failures. Dosages were included if they were accessed in any phase III trials.
The random-effects model with the Hartung-Knapp-Sidik-Jonkman method was used. Mean difference (MD) and risk ratio (RR) with their 95% con dence intervals (CI) were used to evaluate continuous and categorical variables, respectively. The heterogeneity between the included studies was evaluated with I 2 . R Software version 3.6.2 (R Foundation for Statistical Computing) was used to conduct the analysis.

Network meta-analysis
As studies with direct and indirect comparison exist across different dosage subgroups, we conducted a Bayesian network meta-analysis to compare safety and e cacy outcomes between different dosages using the gemtc package in R software (version 3.6.2). Both xed-effects and random-effects models were generated with Markov Chain Monte-Carlo simulation for each outcome using default priors with 5000 adaptation iterations with 100,000 iterations of 4 chains. The model with the lowest deviance information criterion was used for further analysis. The potential scale reduction factor was used to assess the model convergence, using a cut-off value of 1.005. The node-split method was used for assessing inconsistency. Log odds ratio (LOR) and mean difference with their 95% CI were calculated for binary variables and continuous variables, respectively.

Risk of Bias
For assessing the risk of bias of RCTs, we used the standard approach developed by the Cochrane collaboration, which included: selection bias, performance bias, detection bias, attrition bias, reporting bias, and other potential biases. The risk of bias plot in individual studies was created using the Review Manager 5.3 software.
Safety and e cacy of erenumab in patients with prior treatment failures

Discussion
Our study pooled ve randomized clinical trials evaluating the safety and e cacy of erenumab for the treatment of episodic migraine. Our analysis showed that erenumab was e cacious and safe for the treatment of migraine. Furthermore, our study pooled both direct and indirect evidence with the Bayesian network meta-analysis method and demonstrated 140 mg erenumab outperforms 70 mg erenumab in multiple e cacy endpoints while related to the same risk for developing adverse events. In addition, our study demonstrated that 70 mg and 140 mg erenumab were associated with signi cantly reduced migraine frequency and improved functional outcome in patients with prior migraine treatment failures. The heterogeneity was low across all the primary outcomes, indicating a high level of clinical evidence. It is the rst meta-analysis comparing the safety and e cacy of different doses of erenumab and the rst meta-analysis to demonstrate the safety and e cacy of erenumab in patients with ≥ 2 prior migraine treatment failures.
Targeting CGRP is considered a breakthrough in migraine therapy in the past decades. Serum CGRP levels are elevated during migraine attacks [12].
Meanwhile, the infusion of human CGRP could induce headaches and migraine in migraineurs [13]. These evidences showed that CGPR might play an essential role in migraine attacks. CGRP receptors are widely distributed in both the central and peripheral nervous systems [14]. Since CGRP antagonists have a limited ability to cross the blood-brain barrier (BBB) and there is no increase in BBB permeability during a migraine attack [15], the brainstem, which has a high level of CGRP receptor expression and defective BBB, was considered the target of action of CGRP receptor antibodies [16]. Drugs targeting CGRP were developed based on the role of CGRP in the pathogenesis of migraine, which is highly selective for migraine treatment. CGRP antagonists and CGRP receptor monoclonal antibodies are mainly used as migraine prevention, and small-molecule CGRP antagonists are mainly used for acute management of migraine.
Although current evidence suggests the effectiveness of CGRP monoclonal antibodies in migraine treatment is not dramatically increased compared with traditional therapy, the monoclonal antibodies targeting CGRP and its receptors have their unique advantages in migraine treatment.
Low compliance in patients with migraine with chronic prophylactic medication use is frequently reported [17]. Erenumab has been designed and modi ed to extend its circulating half-lives [18]. In clinical practice, it could be given once per month, which could potentially increase compliance among migraineurs. A cost-analysis based on US societal perspective Markov health state transition model has demonstrated the use of erenumab was associated with reduced migraine-related direct and indirect costs compared with supportive care [19].
As CGRP ligand could dilate blood vessels, the potential risk on the cardiovascular system caused by CGRP targeted therapies attracts much attention. A study conducted on human isolated cranial arteries has found that the use of erenumab was not associated with direct vasoconstriction and did not in uence the effect of endogenous or exogenous vasoactive compounds [20]. A 12-week trial conducted by Tepper et al. con rmed that erenumab did not affect the blood pressure and 24-hour blood pressure changes in healthy volunteers [21]. No signi cant difference was found in the emergence of vascular adverse events between the erenumab group and the placebo group in the short-term migraine treatment phase [22]. In addition, a study conducted in patients with stable angina has found that the use of erenumab had no effect on exercise time [23]. Although these studies have demonstrated CGRP targeted therapies are not likely to induce severe adverse events caused by vasoconstriction, CGRP may play a more critical role in the change of vascular tension in hypertensive rats than in normotensive rats [24]. Hence, future researches are still needed to evaluate the long-term use of erenumab in patients with cardiovascular risk or hypertension.
Our study mainly pooled short-term placebo-controlled trials for episodic migraineurs. Erenumab also showed its effect on chronic migraine patients [25]. So far, a handful of studies have been conducted to evaluate the long-term use of erenumab. An open-label study evaluating long-term safety and e cacy of erenumab in patients with episodic migraine found that erenumab use is related to improved function and a favorable safety and tolerability pro le [26]. The long-term adverse effects of erenumab include injection-site reactions, constipation, and muscle spasm [27]. In another similar long-term study, erenumab was found to be safe and well-tolerated, and the adverse events were considered consistent with shorterterm placebo treatment [28]. Furthermore, evidence suggests that the long-term use of erenumab increased the conversion from chronic migraine to episode migraine [29], especially 140 mg dose [30]. Besides, a previous study reported that erenumab use was associated with a higher response rate in patients who had a high susceptibility to migraine induction by CGRP [31].
As a preventive therapy, erenumab is likely to be prescribed for long term use. Thus, the cross-talk of erenumab with other commonly-used drugs is another concern for physicians. From the pharmacokinetics perspective, as mAbs are eliminated via catabolic pathways, they may not compete directly with other small molecule drugs that are mainly eliminated via hepatic, renal, or biliary processes [32]. A study conducted in 2017 found that subcutaneous erenumab did not in uence the effect of estrogen/progestin combination oral contraceptives among healthy females [33]. A placebocontrolled trial investigating co-administration of erenumab 140 mg and sumatriptan 12 mg have found no additional effect on averages of mean arterial pressure or on the pharmacokinetics of sumatriptan [34].
Optimal management of chronic migraine patients with concurrent overuse of acute medications poses a challenge for clinicians [35]. Results from clinical trials have demonstrated that erenumab is effective in patients with medication overuse [36][37][38]. In a post-hoc analysis, erenumab showed comparable e cacy between medication over-users and the non-medication overuse group [39].
Preventive treatment of migraine is an essential component of migraine management. To date, multiple drugs have been applied to the preventive treatment of migraine, including beta-blockers, topiramate, amitriptyline, and onabotulinumtoxinA [40]-however, none of these drugs highly selective to pharmacological targets on the migraine pathogenesis pathway which would lead to disturbing adverse events. Propranolol and timolol are the only beta-blockers approved by the FDA for migraine prevention [41]. Frequently reported adverse events for beta-blockers include fatigue, exercise intolerance, and orthostatic hypotension. Besides, beta-blockers are contraindicated in patients with decompensated heart failures and patients with asthma. Valproate and topiramate are frequently used as migraine preventive therapy. Previous evidence suggests about 50% of patients who received valproate and topiramate achieve a 50% reduction in headache frequency [42]. However, for the safety outcomes, the use of valproate and topiramate is related to increased adverse events. Commonly reported adverse events for valproate including nausea, fatigue, tremor, weight gain, and hair loss, and topiramate commonly causes paresthesias, fatigue, cognitive impairment, taste perversion, weight loss, and nephrolithiasis [43]. When compared with valproate and topiramate, erenumab is related to comparable e cacy outcomes and is not associated with increased risk for developing adverse effects. Amitriptyline is the only tricyclic antidepressant shown to be effective for migraine prevention in clinical trials, but it often causes sedation, dry mouth, and weight gain [44]. To sum up, although no clinical trial has been conducted to evaluate the safety and e cacy of erenumab versus other traditional methods, current evidence indirect comparison suggests erenumab outperform these traditional migraine preventive drugs [45]. Pericranial intramuscular injections of onabotulinumtoxinA (Botox) is another FDA-approved novel therapy for the prevention of headaches in adults with chronic migraine [46]. However, until now, the effectiveness of Botox in episodic migraineurs remains uncertain on account of the low quality of limited evidence [47].
Our study had a few limitations. First, although our study included ve multicenter randomized trials that have a low risk for bias, the heterogeneity of overall functional improvement in the pair-wise meta-analysis was considerably high, indicating a low level of clinical evidence. Second, all the included clinical trials were limited to short-term use; further pooled analysis based on long-term, or real-world studies are needed. Third, our analysis of 70 mg erenumab in patients with previous treatment failure only includes one trial with an insu cient sample size. Further studies are still needed to provide more robust evidence.

Conclusion
In this systematic review and meta-analysis, the use of erenumab given subcutaneously once per month as a preventive treatment for episodic migraine in adults led to a signi cantly decreased MMD and MSMD and increased 50% response and was not associated with increased risk for developing adverse events compared with placebo. Our results from network meta-analysis suggest that 140 mg erenumab was related to better safety outcomes compared with 70 mg. Furthermore, current evidence suggests that 140 mg erenumab was effective in patients with previous migraine treatment failure.

Declarations
Ethics Approval and consent to participate Not Applicable

Consent for Publication
Not Applicable Data Availability All the additional data involved in the study was provided in the supplement le.

Conflict of interest statement
The authors declare that they have no con ict of interest.