Effectiveness and tolerability of pharmacological prophylaxis in migraine patients with prior preventive treatment failure: A systematic review and network meta-analysis of randomized controlled trials

Abstract

Background

Despite advances in migraine management, some patients fail to respond to preventive treatments for migraine. We aimed to assess the comparative effects of available pharmacological prophylaxis in adults with a treatment failure history.

Methods

We searched Medline, Embase, Cochrane Central, PsycINFO, Web of Science, and Scopus up to July 2025. Pairs of reviewers independently screened titles, abstracts, and full-text articles to identify randomized controlled trials of prophylactic pharmacological interventions that enrolled adults diagnosed with chronic or episodic migraine and a prior preventive treatment failure. We performed a frequentist random-effects network meta-analysis and used the GRADE approach to assess the certainty of evidence.

Results

We included 18 randomized trials (7281 participants). Compared to placebo, low certainty evidence suggest fremanezumab [mean difference (MD) −3.30 (95% CI: −4.11 to −2.49)], eptinezumab [MD −3.35 (95% CI: −4.38 to −2.32)], galcanezumab [MD −2.73 (95% CI: −3.43 to −2.03)], atogepant [MD −2.30 (95% CI: −3.47 to −1.13)], and erenumab [MD −2.20 (95% CI: −2.72 to −1.68)] may be among the most effective in reducing the monthly migraine headache days. Low to moderate certainty evidence suggests that, compared with placebo, galcanezumab [relative risk (RR) 1.94 (95% CI: 1.52 to 2.48)], fremanezumab [RR 3.98 (95% CI: 2.40 to 6.59)], atogepant [RR 2.80 (95% CI: 1.73 to 4.54)], erenumab [RR 2.56 (95% CI: 2.01 to 3.26)], and eptinezumab [RR 2.35 (95% CI: 1.61 to 3.42)] may increase the likelihood of achieving a 50% response rate.

Conclusion

Evidence for migraine patients with prior preventive treatment failure is limited. Low- to moderate-certainty data suggest that CGRP-targeted therapies may provide some benefit and are generally tolerable, but the available evidence is driven by a few industry-sponsored trials. Additional independent, well-powered studies with longer follow-up are needed to strengthen the evidence base.

Registration number

PROSPERO (CRD42024547860).

Graphical abstract

This is a visual representation of the abstract.

Keywords

migraine treatment failure refractory migraine network meta-analysis

Introduction

Migraine affects over one billion people worldwide and is the leading cause of disability, particularly among young adults and women, ranking eighth among all disorders across all ages and genders.^1,2 There are various medications to prevent migraine attacks, ranging from non-migraine-specific options, such as blood pressure medications, antiepileptics, and antidepressants, to recently introduced agents targeting the calcitonin gene-related peptide (CGRP) pathway, including CGRP monoclonal antibodies and CGRP receptor antagonists (gepants).³

While resistant and refractory migraine are presumed to be rare in general population, preventive treatment failures are highly prevalent among patients visiting specialist headache clinics, with approximately 60% experiencing one or more failures and 15% reporting more than four.^4,5 Failure of migraine prophylaxis can be due to lack of efficacy, high risk of adverse effects, and loss of treatment benefits over time, particularly with oral medications.⁶ This significantly impacts their quality of life, patient-reported outcomes, healthcare resource utilization, and costs.^5,7

Management of treatment failure in prevention of migraine can be challenging. Triptans, ditans and gepants have been suggested as suitable acute treatment options,⁸ with pharmacological and psychosocial interventions as preventive treatments. Recent systematic reviews have rarely focused on pharmacological prophylaxis in patients with a history of preventive-treatment failure^9,10 A few reviews examined this subgroup, but they focused on the effectiveness of CGRP monoclonal antibodies rather than the comparative effectiveness of all available management strategies.^11,12 In the current systematic review and network meta-analysis (NMA), we review randomized controlled trials (RCTs) involving people with migraine who have experienced prior preventive treatment failure, comparing the effectiveness and safety of different treatments.

Methods

We registered our review with PROSPERO (CRD42024547860) and followed PRISMA-NMA guideline statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses – extension for NMA) to report our findings.¹³

Data sources and searches

A medical librarian (RJC) developed database-specific search strategies for Medline, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), PsycINFO, Web of Science, and Scopus. Electronic databases were searched without language or publication status restriction from inception to January 2023, and we then updated our searches in July 2025 (eMethod 1). We reviewed reference lists of included trials and relevant reviews to identify additional eligible trials.

Study selection

Eligible studies were trials that (1) enrolled adults (18 years or older) diagnosed with migraine (episodic and/or chronic) who had experienced prior preventive treatment failure (due to inadequate efficacy, tolerability reasons or contraindications) with one or more preventive treatments for migraine,^14,15 (2) randomized them to any pharmacological interventions aimed at preventing migraine compared to an alternative intervention, or usual care, placebo, waitlist control, or no treatment, and (3) reported at least one effectiveness outcome. eMethod 2 provides the list of eligible interventions.

We excluded trials or trial arms that compared different doses or frequencies of the same intervention or trials of nonpharmacologic supplements and herbal medicines (e.g., vitamins, magnesium citrate, coenzyme Q10). Additionally, we excluded trials that exclusively enrolled migraine patients with medication overuse headaches.

Our outcomes of interest included, (1) monthly migraine headache days, (2) monthly migraine headache attacks, (3) pain free response (i.e., 100% response rate defined as 100% reduction in monthly migraine headache days or attacks), (4) 50% response rate (defined as 50% or more reduction in monthly migraine headache days or attacks), (5) analgesic medication use, (6) pain relief, (7) health-related quality of life, (8) migraine-related disability, (9) all-cause drop-out (withdrawal), (10) drop out (withdrawal) due to adverse events, (11) dizziness, (12) nausea, and (13) constipation.

Pairs of reviewers independently screened titles and abstracts of records identified through our searches. Full-texts of potentially eligible studies were reviewed by the same pairs of reviewers to identify eligible RCTs. Discrepancies were resolved through discussion and, if needed, involvement of a third reviewer (MK). Literature screening was performed using online systematic review software (DistillerSR Inc., Ottawa, ON, Canada http://systematic-review.net).

Data extraction and risk of bias assessment

Pairs of reviewers independently extracted data from eligible RCTs and assessed their risk of bias. We used pilot-tested Excel spreadsheets for data extractions and performed calibration exercises to ascertain the consistency and accuracy of data abstraction. We extracted the following information: (1) study characteristics, (2) participant and trial characteristics, (3) details of interventions and comparators, and (4) outcomes of interest. For outcomes that were reported at multiple follow-up times, we used data from the longest follow-up.

Risk of bias among eligible RCTs were assessed using the modified Cochrane risk-of-bias tool (RoB 2.0).¹⁶ We used a macro-enabled Microsoft Excel tool to implement RoB 2.0 using signaling questions for the following five domains: (1) bias arising from the randomization process, (2) bias due to deviations from intended interventions, (3) bias due to missing outcome data, (4) bias in measurement of the outcome, and (5) bias in selection of the reported results. Disagreements in data extraction and risk of bias assessments were resolved through discussion or involving a third reviewer if needed (MK or BS).

Data synthesis and analysis

For continuous outcomes, we calculated mean difference (MD) and its 95% confidence intervals (CI) using change scores from baseline to the end of the follow-up to account for interpatient variability.^17,18 When standard deviation (SD) for the change score was not reported, it was imputed using baseline and end-of-study SDs and a median correlation coefficient derived from the trials at low risk of bias. We used the methods described by Weir et al.¹⁸ to impute mean and standard deviation when median, (interquartile) range, and sample size were reported. When pain intensity, migraine-related quality of life, and migraine-related disability was reported using different instruments across RCTs, we used linear transformation¹⁹ assuming instruments reporting shared domains have similar measurement properties, to convert pain intensity to 10-cm visual analogue scale (VAS), quality of life to 0–100 Migraine-Specific Quality-of-Life Questionnaire (MSQ), and disability to 36–78 Headache Impact Test (HIT6) and performed NMA using mean difference. Due to the differences in measurement of analgesic medication use across included trials, we pooled effects as the standardized mean difference (SMD). We used a minimally important difference (MID) of 1.5 cm reduction on a 10-cm VAS for pain, a 1.5 reduction of the HIT-6 score for disability,²⁰ and a 0.2 score for analgesic medication use as small effect size for SMD.²¹ We calculated the relative risk (RR) and the associated 95% CI for dichotomous outcomes.

NMA feasibility for each outcome was assessed by ensuring network connectivity and the availability of more RCTs than intervention nodes, performing NMA when at least 10 studies reported an outcome. We conducted NMA for monthly migraine headache days, 50% response rate, migraine related disability, analgesic medication usage reduction, all-cause dropout and dropout due to adverse events. We also performed conventional meta-analyses of two trials comparing botulinum toxin type A (onabotulinumtoxinA) with placebo for outcomes including monthly migraine headache days, migraine-related disability, and dropout due to adverse events. We used DerSimonian–Laird random-effects model for all direct comparisons and performed random-effects network meta-analysis assuming a common heterogeneity parameter using a frequentist approach.^22,23 Transitivity and coherence (a.k.a. consistency) are key assumptions for NMA. We ensured that all interventions among included trials were jointly randomizable to construct a network and that distribution of potential effect modifiers (e.g., age, proportion of female participants, diagnostic criteria) were similar across trials and comparisons. We assessed the transitivity assumption using NMA-studio web application (https://www.nmastudioapp.com).²⁴ For all direct comparisons, if at least 10 RCTs contributed to a meta-analysis, we evaluated small-study effect using the Harbord's test for binary outcomes and the Egger's test for continuous outcomes.²⁵

The ‘design-by-treatment’ model (global test) was used to assess the coherence assumption for each network and the side-splitting method was used to evaluate local (loop-specific) incoherence in each closed loop of the network as the difference between direct and indirect evidence.^26,27 We estimated the ranking probabilities among individual interventions and reported the Surface Under the Cumulative Ranking Curves (SUCRA) values, mean ranks, and rankograms.^28,29 We used Stata 17.0 (StataCorp, College Station, Texas, USA) for all statistical analyses.

Subgroup analysis and sensitivity analysis

We explored the effect of prespecified subgroups on treatment effects to explain the variability between trials. We considered mean age, proportion of female participants, type of migraine (i.e., episodic vs. chronic), diagnostic criteria, risk of bias (low vs. high), the number of prior preventive treatment failures, and duration of follow-up as potential effect modifiers. We used random-effects network meta-regression. We perform sensitivity analysis excluding studies with botulinum toxin A as a treatment arm. This was done because we observed considerable heterogeneity and statistical incoherence in evidence loops involving botulinum toxin A. Previous studies have shown similar patterns for botulinum toxin A, which may be due to differences in its effects based on whether the migraine is chronic or episodic.^10,30

Assessment of certainty of the evidence

We used the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach to assess the certainty of evidence for direct, indirect, and network estimates across all outcomes.^31,32 First, we assessed the certainty of direct estimates based on considerations of risk of bias, indirectness, publication bias, and inconsistency.³³ Then, the certainty of indirect estimates was assessed, with a focus on the dominant lowest-order loops. Finally, we rated the certainty of network estimates, considering further limitations due to incoherence and imprecision. We judged imprecision using the network estimate and rated down for imprecision if the 95% CI included half of the MID for pain intensity and headache-related disability, 0.2 for the pooled SMD in analgesic medication use,²¹ or the null effect for other outcomes.³⁴

Treatment hierarchy

We used the GRADE minimally contextualized approach to develop a hierarchy of interventions across all outcomes of interest.^35,36 In this approach interventions are categorized from the most effective to the least effective based on their relative effectiveness and safety and their associated certainty of evidence. We considered the null effect and MIDs as the decision threshold and placebo as the reference intervention. Interventions with no evidence of difference compared to placebo (i.e., 95% CI includes null value) are categorized as “among the worst”; interventions superior to placebo, but not superior to any other intervention superior to placebo, categorized as “among the intermediate”; and interventions superior to at least one of “among the intermediate” interventions then categorized as “among the best”. We subsequently classified interventions into two groups of “high to moderate certainty” or “low to very low certainty”.

Results

Search results and characteristics of included studies

We retrieved 43,719 unique records from our searches, of which full-texts of 413 studies were reviewed for eligibility. A total of 18 published RCTs were included in our systematic review, with 13 trials contributing to the NMA^{14,15,37–50} and 2 trials included in the conventional meta-analysis,^51,52 involving 7281 participants. Figure 1 provides details of the study selection process and excluded studies and reasons for exclusion are presented in eTable 1. The median of the mean age for participants among included trials was 42.8 years (interquartile range [IQR] 42.2–45.0 years), and on average, 87.8% of the enrolled participants were female (IQR 84.5–89.6). Eight trials (44.4%) enrolled only patients with episodic migraine, five trials (27.8%) enrolled only patients with chronic migraine, and five trials (27.8%) included both episodic and chronic migraine. The median intervention duration was 12 weeks of treatment (range: 12–24 weeks), and the median follow-up duration was 12 weeks (range: 12–52 weeks). Supplementary eTables 2 and 3 summarize the characteristics of the trials included in the NMA and systematic review, respectively.

Figure 1.

PRISMA flow diagram for study selection.

Risk of bias

Of the 18 trials, we judged 9 (50.0%) to be at low risk of bias, and 3 (16.7%) to have some concerns. Most studies (17 RCTs) were judged to have a low risk of bias or raised some concerns due to issues related to the randomization process. Four studies (22.2%) were affected by high missing outcome data. We judged 12 (66.7%) trials to be at low risk of bias in the measurement of outcomes. Supplementary eFigures 1 and 2 provide RoB assessment details.

Monthly migraine headache days

The 12 RCTs involving 5337 participants reported monthly migraine headache days (Figure 2(a)). Of the 5 direct comparisons, 3 comparisons were informed by at least two trials (eTable 4). Compared to placebo, low certainty evidence suggests that fremanezumab [MD −3.30 (95% CI: −4.11 to −2.49)], eptinezumab [MD −3.35 (95% CI: −4.38 to −2.32)], galcanezumab [MD −2.73 (95% CI: −3.43 to −2.03)], atogepant [MD −2.30 (95% CI: −3.47 to −1.13)], and erenumab [MD −2.20 (95% CI: −2.72 to −1.68)] may be among the most effective interventions in reducing monthly migraine headache days (Table 1 , eTable 5).

Figure 2.

Network of effectiveness outcomes: monthly migraine headache days (a), 50% response rate (b), migraine related disability (c), analgesic medication uses (d).

Table 1.

GRADE summary of findings for the comparisons of active treatments and placebo for each outcome.

	Effectiveness outcomes				Tolerability outcomes
Interventions	Monthly migraine headache daysMD (95% CI)	50% Response rateRR (95% CI)	Migraine related disabilityMD (95% CI)	Analgesic medication useSMD (95% CI)	Drop out due to AERR (CI 95%)	All-cause drop outRR (CI 95%)
Erenumab	−2.20 (−2.72, −1.68)	2.56 (2.01, 3.26)	−7.05 (−10.87, −3.22)	−0.68 (−1.10, −0.26)	0.88 (0.36, 2.10)	0.56 (0.24, 1.28)
Galcanezumab	−2.73 (−3.43, −2.03)	1.94 (1.52, 2.48)	−3.56 (−5.08, −2.05)	−0.59 (−1.18, −0.00)	2.97 (0.12, 72.63)	1.73 (0.27, 11.19)
Eptinezumab	−3.35 (−4.38, −2.32)	2.35 (1.61, 3.42)	−5.51 (−8.02, −3.00)	−0.53 (−1.39, 0.33)	3.52 (0.43, 28.46)	2.21 (0.40, 12.21)
Fremanezumab	−3.30 (−4.11, −2.49)	3.98 (2.40, 6.59)	−2.35 (−4.79, 0.09)	−0.64 (−1.36, 0.08)	0.83 (0.20, 3.46)	0.58 (0.12, 2.82)
Atogepant	−2.30 (−3.47, −1.13)	2.80 (1.73, 4.54)	–	−0.53 (−1.63, 0.58)	2.01 (0.37, 10.83)	2.01 (0.37, 11.08)
Gabapentin	–	1.64 (0.61, 4.44)	–		4.89(0.60, 40.12)	1.40(0.29, 6.74)

Certainty evidence	Statistically significant difference with placebo and at least one other treatment	Statistically significant difference with placebo	Statistically no difference with placebo
High or moderate	Among the most effective	Inferior to the most effective, but superior to placebo	No more effective than placebo
High or moderate	Among the less tolerable	more tolerable than some alternatives, but less tolerable than placebo	No more tolerable than placebo
Low or very low	May be among the most effective (or less tolerable)	May be inferior to the most effective, but superior to placebo (or more tolerable than some alternatives, but less tolerable than placebo	May be no more effective than placebo
Low or very low	May be among the most tolerable	May be inferior to the most tolerable, but superior to placebo	May be no tolerable than placebo

Notes: The number is the point estimate with 95% confidence interval (CI). Results are the mean difference (MD) for frequency of headache days per month; lower is better (minimally important difference [MID] is null value). For migraine-related disability, scores range from 36 to 78; lower MD is better (MID is −1.5). Results are Standardized mean differences (SMD) for analgesic medication uses; lower is better (MID is considered as small effect size of 0.2). A relative risk (RR) > 1 indicates the treatment is superior to placebo for 50% response rate, for drop out due to AE and all-cause drop out, it states the treatment is inferior to placebo. Numbers in bold represent statistically significant results at p-value < 0.05. There were not enough trials to conduct a NMA for monthly migraine headache attacks, 100% response rate, pain relief, and health related quality of life. GRADE: Grading of Recommendations Assessment, Development and Evaluation.

The effects of botulinum toxin A on monthly migraine headache days were reported in two studies involving 200 participants. The evidence is very uncertain about the effect of botulinum toxin A on monthly migraine headache days compared to placebo [MD −5.90 (95% CI: −17.74 to 5.93), very low certainty] (eFigure 3, eTable 6).

Monthly migraine headache attacks

Changes in monthly migraine headache attacks were reported in six RCTs (1245 participants); however, the intervention network was not connected (eTable 7). Compared to placebo, eptinezumab probably improves the monthly migraine headache attacks [MD −2.15 (95% CI: −2.89 to −1.41), moderate certainty].¹⁴ The remaining comparisons, which were supported by low- to moderate-certainty evidence, showed little to no important difference in changes in monthly migraine headache attacks.^39,52–55

Response rate

The 12 RCTs that reported 50% response rate enrolled 5360 participants (Figure 2(b)). Of the 6 direct comparisons, 2 comparisons were informed by at least two trials (eTable 8). Compared to placebo, moderate certainty evidence suggests erenumab [RR 2.56 (95% CI: 2.01 to 3.26)], eptinezumab [RR 2.35 (95% CI: 1.61 to 3.42)], and galcanezumab [RR 1.94 (95% CI: 1.52 to 2.48)] were probably among the most effective in improving the likelihood of 50% response. Low certainty of evidence suggests fremanezumab [RR 3.98 (95% CI: 2.40 to 6.59)] and atogepant [RR 2.80 (95% CI: 1.73 to 4.54)] may increase the likelihood of 50% response (Table 1, eTable 9).

Three RCTs reporting a pain-free response (100% response rate) enrolled 1542 participants, and the results are provided in eTable 10. nty], Compared to placebo, the evidence is very uncertain about the effect of galcanezumab, erenumab and fremanezumab on achieving a 100% response.^38,44,46

Pain relief

Four RCTs reporting a reduction in pain intensity enrolled 405 participants, and the results are provided in eTable 11. Compared to placebo, erenumab may show a little to a non-important impact on the improvement of pain intensity.⁴⁰ Botulinum toxin A in comparison with placebo may suggests a little to a non-important impact on the changes in pain intensity.⁵¹

Health related quality of life

Three RCTs reporting health-related quality of life enrolled 2111 participants, and the results are provided in eTable 12. Compared to placebo, moderate certainty evidence suggests that eptinezumab [MD 14.26; (95% CI: 13.48 to 15.04)], galcanezumab [MD 11.58; 95% (CI: 8.16 to 15.00)], and fremanezumab [MD 9.71; (95% CI: 6.11 to 13.31)] may improve quality of life.^14,38,44

Migraine related disability

The 7 trials reporting migraine related disability enrolled 3025 participants (Figure 2(c)). Of the available 4 direct comparisons, 2 comparisons were informed by two or more trials (eTable 13). Compared to placebo, low certainty evidence suggests that erenumab [MD −7.05 (95% CI: −10.87 to −3.22)], eptinezumab [MD −5.51 (95% CI: −8.02 to −3.00)], and galcanezumab [MD −3.56 (95% CI: −5.08 to −2.05)] may be among the most effective in improving migraine-related disability (Table 1, eTable 14).

The effects of botulinum toxin A on migraine-related disability were reported in two studies involving 200 participants. The evidence suggests that botulinum toxin A [compared to placebo MD −3.11 (95% CI: −5.75 to −0.48), low certainty] may improve migraine-related disability (eFigure 4, eTable 6).

Analgesic medication usage reduction

The reduction in monthly analgesic medication usage was reported in 10 trials involving, 4684 participants (Figure 2(d)). Of 5 direct comparisons, 2 comparisons were informed by at least two trials (eTable 15). Moderate certainty evidence suggests that erenumab [compared to placebo SMD −0.68 (95% CI: −1.10 to −0.26)] was probably among the most effective in reducing the analgesic medication use. Compared to placebo, low certainty evidence suggests galcanezumab [compared to placebo SMD −0.59 (95% CI: −1.18 to −0.00)] may reduce the analgesic medication use (Table 1, eTable 16).

All-cause drop-out

The 10 RCTs reporting all-cause drop-out (tolerability) enrolled 4339 participants (Figure 3(a)). Of the 6 direct comparisons, one comparison was informed by at least two RCTs and showed significant heterogeneity (I² = 79.2%; eTable 17). Compared to placebo, no intervention results in a statistically significant increase in dropout rates (Table 1, eTable 18).

Figure 3.

Network of tolerability outcomes: all-cause drop out (a), drop out due to adverse events (b).

Drop out due to adverse events

The 10 trials reporting drop out due to adverse events enrolled 4124 participants (Figure 3(b)). Of the 6 direct comparisons, one was informed by two RCTs (eTable 19). Compared to placebo, low and very low certainty of evidence indicates that no intervention results in a statistically significant increase in dropout due to adverse events (Table 1, eTable 20). eTables 21 to 23 provide comparative effects for three adverse effects, including dizziness, nausea, and constipation respectively.

The effects of botulinum toxin A on drop out due to adverse events were reported in two studies involving 200 participants. The evidence is very uncertain about the effect of botulinum toxin A on drop out due to adverse events compared to placebo [RR 0.55 (95% CI: 0.03 to 8.60), very low certainty]. (eFigure 5, eTable 6).

Additional analysis

The transitivity assessment suggested no critical imbalances in potential effect modifiers including mean age, proportion of female participants, risk of bias, type of migraine, diagnostic criteria, follow-up duration, or number of prior preventive treatment failures across treatment comparisons (eFigures 6–39). The eTables 24 to 29 provide ranking probabilities and SUCRA values for all outcomes. We performed subgroup analysis and network meta-regression to explore the impact of mean age, proportion of female participants, risk of bias, type of migraine, diagnostic criteria, and longest follow-up duration (eTables 30–41). Network meta-regression analyses suggested that follow-up duration may modify the effect of erenumab on the 50% response rate [RR 1.90 (95% CI: 1.27 to 2.85), p-value for test of interaction = 0.002] (eTable 31). Subgroup analyses suggested that risk of bias may modify the effect of erenumab, with an estimated RR of 2.20 (95% CI: 1.70 to 2.86) in trials at low risk of bias and 3.39 (95% CI: 2.41 to 4.80) in trials at high risk (p-value for test of interaction = 0.047). Subgroup analyses showed that migraine type may modify the effect of galcanezumab on the 50% response rate. The estimated RR was 2.20 [(95% CI: 1.80 to 2.69), p-value for test of interaction = 0.029] in chronic migraine, compared with 1.80 (95% CI: 1.27 to 2.56) in episodic migraine and 1.82 (95% CI: 1.34 to 2.51) in mixed episodic and chronic populations (eTable 37).

The proportion of female participants may modify the effect of galcanezumab on migraine-related disability score [MD 46.45 (95% CI: 4.12 to 88.79), p-value for test of interaction = 0.032] (eTable 32). Subgroup analyses also suggested modification by risk of bias, with an estimated MD of −2.77 (95% CI: −3.89 to −1.66) in trials at low risk of bias and −5.30 (95% CI: −7.12 to −3.47) in trials at high risk (p-value for test of interaction = 0.021). Subgroup analyses showed that migraine type may modify the effect of galcanezumab on the improvement of disability. The estimated MD was −2.77 [(95% CI: −3.89 to −1.66), p-value for test of interaction = 0.021] in episodic migraine, compared with −3.93 (95% CI: −6.14 to −1.73) in chronic migraine and −4.06 (95% CI: −6.20 to −1.91) in mixed episodic and chronic populations (eTable 38). We did not find evidence of effect modification for the remaining comparisons/factors. There was insufficient and inconsistent data to investigate the impact of the number of prior treatment failures, and insufficient variability to assess the effect of diagnostic criteria. eTable 42 provides definitions of prior treatment failure across the included studies.

Discussion

We included 18 RCTs involving 7281 participants with prior preventive treatment failure—mostly middle-aged women—and a median study duration of 12 weeks. Compared to placebo, moderate-certainty evidence identified erenumab, eptinezumab, and galcanezumab as probably among the most effective for achieving a 50% reduction in migraine frequency, while low-certainty evidence suggested fremanezumab and atogepant may increase the likelihood of response. Low certainty evidence further indicated benefits of CGRP-targeted therapies in reducing monthly migraine headache days and migraine-related disability. Our findings suggested that botulinum toxin A that botulinum toxin A may also improve migraine-related disability. Moderate certainty evidence showed that erenumab reduced analgesic medication use compared with placebo, while low certainty evidence suggested that galcanezumab may also reduce analgesic use. The evidence is very uncertain about the effects of interventions on all-cause dropout and discontinuation due to adverse events. We found evidence of effect modification for certain factors (e.g., duration of follow-up, sex, type of migraine), but these seems to be of low-to-very low credibility and should be interpreted cautiously due to the observational nature of these comparisons, possible aggregation bias, and small number of observations.

Effective migraine management begins with evaluating factors that may reduce treatment success, such as incorrect diagnosis, medication overuse headache, overlooked comorbidities, or lifestyle issues like inadequate sleep, diet, anxiety, and chronic stress.⁵⁶ Real-world data show that resistant and especially refractory patients have higher rates of psychiatric comorbidities, thyroid and cerebrovascular disorders, asthma/rhinitis, obesity, and musculoskeletal trigger points compared with non-resistant/non-refractory groups.⁵⁷ This clinical complexity contributes to a substantially greater disease burden, highlighting the need for early recognition of emerging resistance and timely identification of refractory cases.^57,58 After ensuring that treatment failure is not due to these issues, potential reasons for treatment failure include lack of effectiveness, side effects, poor tolerability, or non-compliance.^56,59 Choosing the appropriate medication with the correct dose, duration, and delivery route is crucial. Symptoms such as nausea and vomiting may necessitate non-oral administration, and factors like patient preference, daily compliance, and affordability should also be considered.⁵⁶ Many migraine preventatives have significant side effects and contraindications. For instance, medications causing weight gain should be avoided in obese patients, those lowering blood pressure in hypotensive individuals, and beta-blockers in people with asthma.^60,61

Treatment failure and labeling patients as refractory can erode trust in the healthcare system, increase frustration, and add to stigma, potentially straining the patient-doctor relationship and complicating the management of migraine patients.⁵⁹ To improve outcomes and reduce the likelihood of treatment failure, treatment should start early with options that have a higher probability of success and proven efficacy, even in those with prior treatment failures. Alternative treatments can be considered for non-responders to initial treatments.

For migraine patients who require preventive treatments, beta blockers such as propranolol and metoprolol, candesartan, and topiramate are still considered first-line treatments in many countries.⁶⁰ Medications such as flunarizine and amitriptyline are then used as second-line options, followed by third-line treatments targeting CGRP.⁶⁰ In a previous network meta-analysis, we demonstrated that CGRP inhibitors, along with beta blockers and botulinum toxin A, were the most effective and safe options for migraine prevention among all available medications.⁶² Our findings suggest CGRP-targeting therapies may be beneficial in those with a history of migraine treatment failure.

Questions about treatment duration and long-term safety are central to the use of anti-CGRP therapies. These medications generally act reasonably quick, suggested to be well tolerated, and show acceptable safety signals based on short-term follow-ups, which may be associated with better adherence—an area where many traditional preventives fall short.^63–65 Mechanistically, there are concerns over long-term use of CGRP blockers suggesting they may cause cardiovascular and gastrointestinal system-related side effects; however, their long-term effects and safety is yet not well-studied.⁶⁶ While evidence seems to point to benefit of certain medications in this class for management of episodic and chronic migraine headaches compared to established migraine treatments,^65,67,68 their access and reimbursement remain key constraints in many regions.⁶³ Additionally, guidance on how long to continue therapy varies partly because current evidence from RCTs and observational studies have short-term follow-ups. Some recommendations suggest assessing benefit only after a six-month trial,⁶⁹ and current European guidelines advise continuing treatment for roughly 12–18 months before considering interruption.⁷⁰

This study has some limitations. Firstly, we did not identify sufficient studies on conventional migraine preventive treatments in individuals with prior treatment failure, which prevented us from making comparisons. Secondly, the number of available studies was very limited, and none included head-to-head comparisons; the only common intervention across studies was placebo, which may limit the generalizability to real-world scenarios. Although individuals with a greater number of prior treatment failures are likely to have more refractory disease,⁷¹ the available data were insufficient and inconsistently reported across studies to reliably incorporate the number of failed preventive treatments as an effect modifier in meta-regression analyses. There was variability in how monthly migraine days and analgesic medication use were defined across studies, which may have contributed to variability in effect estimates.⁷² Furthermore, the analysis did not distinguish between low-frequency episodic migraine and high-frequency episodic migraine, despite accumulating evidence suggesting that these populations may be biologically distinct and may respond differently to preventive treatments.⁷³ The relatively short follow-up duration in most studies also limits the assessment of long-term safety and sustained effectiveness.

Another limitation of our network meta-analysis is the significant inconsistency introduced by botulinum toxin A, which precluded its inclusion in the network. Therefore, we evaluated its effects separately using a conventional meta-analysis. All trials of anti-CGRP monoclonal antibodies were industry-funded, and we therefore downgraded the certainty of evidence for potential risk of bias related to sponsorship. Ultimately, this review underscores substantial evidence gaps in the comparative effectiveness of preventive treatments, highlighting the need for well-designed comparative trials to strengthen the evidence base.

Nevertheless, this study contributes to filling the gap by comparing different interventions through an NMA, providing valuable insights for clinicians, decision-makers, and patients in improving migraine management. We also evaluated multiple outcomes and included all available trials, thereby providing a comprehensive perspective. While future studies are necessary to assess more treatment options and compare them within a single trial, our findings can help pave the way for rethinking guidelines, influencing policies, and refining the selection of the best preventive treatments for migraine, especially in patients who have experienced prior treatment failure.

In conclusion, this network meta-analysis of randomized trials found that, in patients with prior preventive treatment failure, moderate certainty evidence suggests erenumab, eptinezumab, and galcanezumab as probably among the most effective for achieving a 50% response rate, while low-certainty evidence indicates fremanezumab and atogepant may also provide some benefit. Low certainty evidence further suggests that CGRP-targeted therapies may reduce monthly migraine days, migraine-related disability, and analgesic medication use, whereas the evidence for botulinum toxin A remains sparse and uncertain. These treatments showed little and uncertain impact on dropout rates compared to placebo. Given the limited number of available trials, many of which are industry-sponsored, further independent and well-powered studies are needed to better characterize comparative effectiveness. Efforts to expand access to effective migraine-specific treatments in more countries may help improve patient outcomes.

Key findings

Across 18 randomized trials with 7281 adults who had failed prior preventive treatments, low- to moderate-certainty evidence suggests that several CGRP-targeted therapies may reduce monthly migraine days and increase the likelihood of achieving a 50% response.

Evidence for botulinum toxin A in this population was sparse and inconsistent.

While CGRP-targeted therapies show the most promise for preventing migraine headaches, the existing evidence is limited in number and largely industry-sponsored, highlighting the need for independent, well-powered trials with longer follow-up to strengthen comparative effectiveness estimates

Supplemental Material

sj-docx-1-cep-10.1177_03331024261441287 - Supplemental material for Effectiveness and tolerability of pharmacological prophylaxis in migraine patients with prior preventive treatment failure: A systematic review and network meta-analysis of randomized controlled trials

Supplemental material, sj-docx-1-cep-10.1177_03331024261441287 for Effectiveness and tolerability of pharmacological prophylaxis in migraine patients with prior preventive treatment failure: A systematic review and network meta-analysis of randomized controlled trials by Malahat Khalili, Faraidoon Haghdoost, Amin Liaghatdar, Kian Torabiardakani, Fatemeh Mahdian, Tariq Atkin-Jones, Tal Levit, Sara Moradi, Ehsan Hedayati, Farzaneh Ahmadi, Sahar Khademioore, Ahmad Sofi-Mahmudi, Vivek Patil, Fatemeh Mirzayeh Fashami, Soheil Mehmandoost, Rachel J Couban, Kameshwar Prasad, Seyed-Mohammad Fereshtehnejad and Behnam Sadeghirad in Cephalalgia

Footnotes

Acknowledgements

The authors gratefully acknowledge the contributions of Dr Norman Buckley (Professor Emeritus, Anesthesiologist), Sairan Nili (patient partner), and Brandon Van Dam (patient partner) for sharing their invaluable insights in prioritizing patient-important outcomes.

Author Note

Kameshwar Prasad is affiliated with the Fortis Healthcare Research Foundation, New Delhi, India.

ORCID iDs

Malahat Khalili

Faraidoon Haghdoost

Seyed-Mohammad Fereshtehnejad

Ethical considerations

Given the secondary nature of data used in this systematic review and network meta-analysis, it did not require institutional ethics approval or consent to participate.

Consent to participate

Not applicable.

Consent for publishing

All authors reviewed the manuscript and provided their consent for publication.

Author contributions

Drs Malahat Khalili and Behnam Sadeghirad had full access to all of the data in the study and took responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Drs Behnam Sadeghirad, Malahat Khalili. Acquisition, analysis, or interpretation of data: Drs Khalili, Behnam Sadeghirad, Faraidoon Haghdoost. Drafting of the manuscript: Drs Malahat Khalili, Faraidoon Haghdoost, Behnam Sadeghirad. Critical revision of the manuscript for important intellectual content: All authors. Statistical analysis: Drs Malahat Khalili, Behnam Sadeghirad. Administrative, technical, or material support: Dr Behnam Sadeghirad. Study supervision: Drs Malahat Khalili, Behnam Sadeghirad.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Malahat Khalili is an awarded post-doctoral fellow at the Michael G. DeGroote Institute for Pain Research and Care, McMaster University.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

All data are presented within the paper. The datasets for the conventional meta-analysis and network meta-analysis are available from the corresponding author upon reasonable request (sadeghb@mcmaster.ca).

Open practices

Not applicable.

Supplemental material

Supplemental material for this article is available online.

References

Husøy

Steiner

. GBD 2023: Over twice the health loss from headache disorders among females compared with males, and one fifth of the loss is attributable to medication overuse. J Headache Pain 2025 Oct 21; 26: 27.

Husøy

Steinmetz

, et al. Global, regional, and national burden of headache disorders, 1990–2023: a systematic analysis for the Global Burden of Disease Study 2023. Lancet Neurol 2025 Dec; 24(12): 1005–1015. DOI: 10.1016/S1474-4422(25)00402-8

Ashina

Buse

Ashina

, et al. Migraine: integrated approaches to clinical management and emerging treatments. Lancet 2021 Apr 17; 397: 1505–1518.

Ornello

Andreou

De Matteis

, et al. Resistant and refractory migraine: clinical presentation, pathophysiology, and management. EBioMedicine 2024 Jan; 99: 104943.

Pozo-Rosich

Lucas

Watson

DPB

, et al. Burden of migraine in patients with preventive treatment failure attending European headache specialist centers: real-world evidence from the BECOME study. PainTherapy 2021 Dec; 10: 1691–1708.

Loder

Rizzoli

. Tolerance and loss of beneficial effect during migraine prophylaxis: clinical considerations. Headache 2011 Sep; 51: 1336–1345.

Newman

Zhou

, et al. Health care utilization and costs in patients with migraine who have failed previous preventive treatments. Neurol Clin Prac 2021 Jun; 11: 206–215.

Sacco

Lampl

Amin

, et al. European Headache Federation (EHF) consensus on the definition of effective treatment of a migraine attack and of triptan failure. J Headache Pain 2022 Oct 12; 23: 33.

Vélez-Jiménez

Martínez-Mayorga

Rodriguez-Leyva

, et al. Comprehensive preventive treatments for episodic migraine: a systematic review of randomized clinical trials. Front Neurol 2025 Aug 18; 16: 1611303.

10.

Lampl

MaassenVanDenBrink

Deligianni

, et al. The comparative effectiveness of migraine preventive drugs: a systematic review and network meta-analysis. J Headache Pain 2023 May 19; 24: 56.

11.

Wang

Wen

, et al. Efficacy and safety of monoclonal antibody against calcitonin gene-related peptide or its receptor for migraine patients with prior preventive treatment failure: a network meta-analysis. J Headache Pain 2022 Sep 8; 23: 05.

12.

Sevivas

Fresco

. Treatment of resistant chronic migraine with anti-CGRP monoclonal antibodies: a systematic review. Euro J Med Res 2022 Jun 4; 27: 86.

13.

Hutton

Salanti

Caldwell

, et al. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations. Ann Iintern Med 2015 Jun 2; 162: 777–784.

14.

Ashina

Lanteri-Minet

Pozo-Rosich

, et al. Safety and efficacy of eptinezumab for migraine prevention in patients with two-to-four previous preventive treatment failures (DELIVER): a multi-arm, randomised, double-blind, placebo-controlled, phase 3b trial. Lancet Neurol 2022 Jul; 21: 597–607.

15.

Tassorelli

Nagy

Pozo-Rosich

, et al. Safety and efficacy of atogepant for the preventive treatment of episodic migraine in adults for whom conventional oral preventive treatments have failed (ELEVATE): a randomised, placebo-controlled, phase 3b trial. Lancet Neurol 2024 Apr; 23: 382–392.

16.

RoB

. RoB 2: A revised Cochrane risk-of-bias tool for randomized trials. Available at: methods.cochrane.org/bias/resources/rob-2-revised-cochrane-risk-bias-tool-randomized-trials (2020, accessed 2022/07/10 2022).

17.

Higgins

Deeks

. Choosing effect measures and computing estimates of effectCochrane handbook for systematic reviews of interventions. 2nd Edition. Chichester (UK): John Wiley & Sons, 2019, pp.143–176.

18.

Weir

Butcher

Assi

, et al. Dealing with missing standard deviation and mean values in meta-analysis of continuous outcomes: a systematic review. BMC Med Res Methodol 2018 Mar 7; 18: 25.

19.

Thorlund

Walter

Johnston

, et al. Pooling health-related quality of life outcomes in meta-analysis—a tutorial and review of methods for enhancing interpretability. Res Synthesis Methods 2011 Sep; 2: 188–203. Epub 2011.

20.

Smelt

AFH

Assendelft

WJJ

Terwee

, et al. What is a clinically relevant change on the HIT-6 questionnaire? An estimation in a primary-care population of migraine patients. Cephalalgia 2013 Jan; 34: 29–36.

21.

Cohen

. Statistical power analysis for the behavioral sciences. New York, NY (USA): Academic Press, 2013.

22.

White

. Multivariate random-effects meta-analysis. Stata J 2009; 9: 40–56.

23.

Sadeghirad

Foroutan

Zoratti

, et al. Theory and practice of Bayesian and frequentist frameworks for network meta-analysis. BMJ Evid Based Med 2023; 28: 204–209.

24.

Metelli

Papakonstantinou

, et al. NMAstudio 2.0: An interactive tool for network meta-analysis to enhance understanding, interpretation, and communication of the findings. Res Synth Methods. 2026: 1–14. doi: 10.1017/rsm.2026.10074

25.

Harbord

Egger

Sterne

. A modified test for small-study effects in meta-analyses of controlled trials with binary endpoints. Stat Med 2006 Oct 30; 25: 3443–3457.

26.

Higgins

Jackson

Barrett

, et al. Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies. Res Synth Methods 2012; 3: 98–110.

27.

White

Barrett

Jackson

, et al. Consistency and inconsistency in network meta-analysis: model estimation using multivariate meta-regression. Res Synth Methods 2012; 3: 111–125.

28.

Chaimani

Salanti

. Visualizing assumptions and results in network meta-analysis: the network graphs package. Stata J 2015; 15: 905–950.

29.

Chaimani

Higgins

Mavridis

, et al. Graphical tools for network meta-analysis in STATA. PLoS One 2013; 8: e76654.

30.

Herd

Tomlinson

Rick

, et al. Cochrane systematic review and meta-analysis of botulinum toxin for the prevention of migraine. BMJ Open 2019; 9: e027953.

31.

Puhan

Schünemann

Murad

, et al. A GRADE Working Group approach for rating the quality of treatment effect estimates from network meta-analysis. Br Med J 2014 Sep 24; 349: g5630.

32.

Brignardello-Petersen

Bonner

Alexander

, et al. Advances in the GRADE approach to rate the certainty in estimates from a network meta-analysis. J Clin Epidemiol 2018 Jan; 93: 36–44.

33.

Guyatt

Oxman

Vist

, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Br Med J 2008 Apr 26; 336: 924–926.

34.

Zeng

Brignardello-Petersen

Hultcrantz

, et al. GRADE Guidelines 32: GRADE offers guidance on choosing targets of GRADE certainty of evidence ratings. J Clin Epidemiol 2021; 137: 163–175.

35.

Brignardello-Petersen

Florez

Izcovich

, et al. GRADE Approach to drawing conclusions from a network meta-analysis using a minimally contextualised framework. Br Med J 2020 Nov 11; 371: m3900.

36.

Phillips

Sadeghirad

Busse

, et al. Development and design validation of a novel network meta-analysis presentation tool for multiple outcomes: a qualitative descriptive study. BMJ Open 2022; 12: e056400.

37.

Ashina

Tepper

Brandes

, et al. Efficacy and safety of erenumab (AMG334) in chronic migraine patients with prior preventive treatment failure: a subgroup analysis of a randomized, double-blind, placebo-controlled study. Cephalalgia 2018; 38: 1611–1621.

38.

Ferrari

Diener

Ning

, et al. Fremanezumab versus placebo for migraine prevention in patients with documented failure to up to four migraine preventive medication classes (FOCUS): a randomised, double-blind, placebo-controlled, phase 3b trial. Lancet 2019; 394: 1030–1040.

39.

Feuerstein

Quebe-Fehling

. A double-blind, placebo-controlled, parallel-group, multicenter study of the safety and efficacy of gabapentin (CI-945) as a prophylactic interval therapy in patients with common migraine (Protocols 879-201, −205,-206,-207,-209). 1990. Research Report No. RR 4301-00066. Freiburg: Goedecke AG Research and Development.

40.

Filippi

Messina

Bartezaghi

, et al. The effect of erenumab on brain network function in episodic migraine patients: a randomized, placebo-controlled clinical trial (RESET BRAIN). J Neurol 2023; 270: 5600–5612.

41.

Goadsby

Barbanti

Lambru

, et al. Eptinezumab improved patient-reported outcomes and quality of life in patients with migraine and prior preventive treatment failures. Euro J Neurol 2023; 30: 1089–1098.

42.

Goadsby

Paemeleire

Broessner

, et al. Efficacy and safety of erenumab (AMG334) in episodic migraine patients with prior preventive treatment failure: a subgroup analysis of a randomized, double-blind, placebo-controlled study. Cephalalgia 2019; 39: 817–826.

43.

Hirata

Sakai

Takeshima

, et al. Efficacy and safety of erenumab in Japanese migraine patients with prior preventive treatment failure or concomitant preventive treatment: subgroup analyses of a phase 3, randomized trial. J Headache Pain 2021; 22: 20210918.

44.

Mulleners

Kim

Láinez

MJA

, et al. Safety and efficacy of galcanezumab in patients for whom previous migraine preventive medication from two to four categories had failed (CONQUER): a multicentre, randomised, double-blind, placebo-controlled, phase 3b trial. Lancet Neurol 2020; 19: 814–825.

45.

Pozo-Rosich

Dolezil

Paemeleire

, et al. Early use of erenumab vs nonspecific oral migraine preventives: the APPRAISE randomized clinical trial. JAMA Neuro 2024; 81: 461–470.

46.

Reuter

Goadsby

Lanteri-Minet

, et al. Efficacy and tolerability of erenumab in patients with episodic migraine in whom two-to-four previous preventive treatments were unsuccessful: a randomised, double-blind, placebo-controlled, phase 3b study. Lancet 2018 Nov 24; 392: 2280–2287.

47.

Ruff

Ford

Tockhorn-Heidenreich

, et al. Efficacy of galcanezumab in patients with chronic migraine and a history of preventive treatment failure. Cephalalgia 2019; 39: 931–944.

48.

Ruff

Ford

Tockhorn-Heidenreich

, et al. Efficacy of galcanezumab in patients with episodic migraine and a history of preventive treatment failure: results from two global randomized clinical trials. Eur J Neurol 2020; 27: 609–618.

49.

Schwedt

Kuruppu

Dong

, et al. Early onset of effect following galcanezumab treatment in patients with previous preventive medication failures. J Headache Pain 2021; 22: 15.

50.

Tepper

Ailani

Ford

, et al. Effects of galcanezumab on health-related quality of life and disability in patients with previous failure of 2-4 migraine preventive medication categories: results from a phase IIIb randomized, placebo-controlled, multicenter clinical trial (CONQUER). Clin Drug Investigation 2022; 42: 263–275.

51.

Bono

Mazza

Magro

, et al. Regional targeted subcutaneous injection of botulinum neurotoxin type A in refractory chronic migraine: a randomized, double-blind, placebo-controlled study. Toxins (Basel) 2023 May 9; 15: 20230509.

52.

Cady

Schreiber

. Botulinum toxin type A as migraine preventive treatment in patients previously failing oral prophylactic treatment due to compliance issues. Headache 2008 Jun; 48: 900–913.

53.

Mansoureh

Rahmat Jirde

Nilavari

, et al. Cinnarizine in refractory migraine prophylaxis: efficacy and tolerability. A comparison with sodium valproate. J Headache Pain 2008 Apr; 9: 77–82.

54.

Mohammadianinejad

Abbasi

Sajedi

, et al. Zonisamide versus topiramate in migraine prophylaxis: a double-blind randomized clinical trial. Clin Neuropharmacol 2011; 34: 174–177.

55.

Bartolini

Silvestrini

Taffi

, et al. Efficacy of topiramate and valproate in chronic migraine. Clin Neuropharmacol 2005; 28: 277–279.

56.

Lipton

Silberstein

Saper

, et al. Why headache treatment fails. Neurology 2003; 60: 1064–1070.

57.

Rosignoli

Ornello

Caponnetto

, et al. Resistant and refractory migraine - two different entities with different comorbidities? Results from the REFINE study. J Headache Pain 2024; 25: 12.

58.

Pensato

Ornello

Rosignoli

, et al. Longitudinal assessment of migraine burden in resistant and refractory migraine - data from the prospective REFINE study. J Headache Pain 2025; 26: 20250815.

59.

Charles

. A new opportunity after migraine treatment failure? Lancet 2018 Nov 24; 392: 2241–2242.

60.

Eigenbrodt

Ashina

Khan

, et al. Diagnosis and management of migraine in ten steps. Nat Rev Neurol 2021; 17: 501–514.

61.

Jenkins

. Migraine management. Aust Prescr 2020; 43: 148–151.

62.

Khalili

Liaghatdar

Mahdian

, et al. P.067 pharmacological prophylaxis for chronic migraine: a systematic review and network meta-analysis of randomized controlled trials. Canadian J Neurol Sci 2024; 51: S34–S34.

63.

Waliszewska-Prosół

Raffaelli

Straburzyński

, et al. Understanding the efficacy and tolerability of migraine treatment: a deep dive into CGRP antagonists. Expert Rev Clin Pharmacol 2024; 17: 1039–1051.

64.

Messina

Huessler

E-M

Puledda

, et al. Safety and tolerability of monoclonal antibodies targeting the CGRP pathway and gepants in migraine prevention: a systematic review and network meta-analysis. Cephalalgia 2023; 43: 03331024231152169.

65.

Haghdoost

Puledda

Garcia-Azorin

, et al. Evaluating the efficacy of CGRP mAbs and gepants for the preventive treatment of migraine: a systematic review and network meta-analysis of phase 3 randomised controlled trials. Cephalalgia 2023; 43: 03331024231159366.

66.

Rivera-Mancilla

Villalón

MaassenVanDenBrink

. CGRP Inhibitors for migraine prophylaxis: a safety review. Expert Opin Drug Saf 2020; 19: 1237–1250.

67.

Drellia

Kokoti

Deligianni

, et al. Anti-CGRP monoclonal antibodies for migraine prevention: a systematic review and likelihood to help or harm analysis. Cephalalgia 2021; 41: 851–864.

68.

Damen

JAA

Yang

Idema

, et al. Comparative effectiveness of pharmacologic treatments for the prevention of episodic migraine headache: a systematic review and network meta-analysis for the American college of physicians. Ann Intern Med 2025; 178: 369–380.

69.

Puledda

Sacco

Diener

H-C

, et al. International headache society global practice recommendations for preventive pharmacological treatment of migraine. Cephalalgia 2024; 44: 03331024241269735.

70.

Sacco

Amin

Ashina

, et al. European headache federation guideline on the use of monoclonal antibodies targeting the calcitonin gene related peptide pathway for migraine prevention – 2022 update. J Headache Pain 2022; 23: 67.

71.

Buse

Pozo-Rosich

Dupont-Benjamin

, et al. Impact of headache frequency and preventive medication failure on quality of life, functioning, and costs among individuals with migraine across several European countries: need for effective preventive treatment. J Headache Pain 2023; 24: 15.

72.

Loose

Paemeleire

Goadsby

, et al. What's a migraine day? Analysis of the variety in the definition of a migraine day across phase III trials with drugs targeting the CGRP pathway. Cephalalgia 2025; 45: 3331024251393986.

73.

Cammarota

De Icco

Vaghi

, et al.

High-frequency episodic migraine: time for its recognition as a migraine subtype?

Cephalalgia 2024; 44: 3331024241291578.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.71 MB