Abstract
Background and objective
We aimed to provide real-world data on the effectiveness of an anti-calcitonin gene-related peptide monoclonal antibody administered for treating migraine in Korean patients.
Methods
We prospectively recruited patients with migraine who received galcanezumab treatment at a single university hospital from June 2020 to April 2021. The treatment response was assessed after three consecutive monthly injections. A 50% responder rate was evaluated based on ≥50% reduction in the number of moderate/severe headache days.
Results
Overall, 87 patients were included in the analysis. Most patients were women (83.9%). They had a mean age of 41.7 ± 12.3 years (range 17–72). Sixty-five patients (74.7%) had chronic migraine, 35 patients (40.2%) had a history of medication-overuse headache, and 32 patients (36.8%) were previously unresponsive to or found intolerable five classes of preventive medication. After three months of treatment, mean changes in numbers of monthly headache days, moderate/severe headache days, crystal clear days, and days of acute medication use were −7.2 ± 8.43, −4.3 ± 9.76, 7.3 ± 8.50, and −4.1 ± 7.93, respectively. The 50% responder rates were 58.3%, 44.2%, and 40.6% for patients with unsuccessful previous use of 0–1, 2–4, and 5 preventive medication classes, respectively. Headache Impact Test-6 and Migraine Disability Assessment Test scores also decreased (−4.4 ± 8.09 and −32.9 ± 77.04, respectively).
Conclusion
In our cohort, the effectiveness and safety of galcanezumab were comparable with those reported in clinical trials, whereas a higher response rate was observed in the difficult-to-treat patient subset than that reported in trials. We provide real-world evidence of galcanezumab treatment benefits in Asian patients with migraine.
Keywords
Introduction
Anti-calcitonin gene-related peptide (CGRP) monoclonal antibody (mAb) is a new disease-specific drug for the preventive treatment of migraine (1). All four anti-CGRP mAbs (erenumab, fremanezumab, galcanezumab, and eptinezumab) have been approved by the FDA. Among them, the efficacies of galcanezumab, erenumab, and fremanezumab have been proven for the treatment of patients with two or more previous oral preventive medication failures for migraine (2–4).
Galcanezumab was approved in June 2019, and it is currently the only available anti-CGRP mAb in Korea. There have been four randomized controlled studies of galcanezumab for the treatment of migraine: EVOLVE-1 for episodic migraine (EM) (5), EVOLVE-2 for EM (6), REGAIN for chronic migraine (CM) (7), and CONQUER for treatment-resistant migraine (8). In these studies, the 50% responder rate was approximately 60% for EM, approximately 30% for CM, and lower for patients with a history of failures of 2–4 preventive medication categories. However, these randomized control trials (RCTs) have certain limitations applicable to real-world settings of the Asian population. Asians comprised only a small proportion of the total population, and patients with extremely difficult-to-treat migraine (daily headaches, failures of >4 categories of preventive medications, and significant comorbidities) were excluded from the trial. Recently, Yang et al. published a post-hoc analysis of phase 3 studies on EM and CM with respect to the safety and efficacy of galcanezumab in Taiwanese patients (9). Taiwanese patients who participated in the EVOLVE-2 and REGAIN trials were included in the study. The percentage of patients with ≥50% reduction in migraine headache days during the double-blinded treatment period was higher in the galcanezumab group in each study. However, this study had the following limitations: 1) the number of patients included was small (15 patients with EM each in galcanezumab 120 mg and 240 mg groups, and 12 patients with CM each in galcanezumab 120 mg and 240 mg groups), 2) the ‘real-world’ dose was not administered (240 mg loading and 120 mg monthly), and 3) Asian populations other than Taiwanese were included in the comparison group for the Taiwanese group. In addition, the EVOLVE-2 and REGAIN studies excluded those who failed more than one preventive medication before the study. Consequently, the result of this study might not be enough to apply to real-world clinical setting where patients who failed multiple drugs are considered candidates for CGRP-targeting monoclonal antibody treatment.
Therefore, the aim of this study was to provide real-world data on the preventive treatment of migraine using galcanezumab in the Asian region.
Methods
Patients
We prospectively recruited patients who started galcanezumab treatment for migraine and were registered in the prospective headache registry at the Samsung Medical Center headache clinic between June 2020 and April 2021. The Samsung Medical Center is a tertiary university hospital that requires a referral letter. However, patients can easily obtain a referral letter directly to the tertiary hospital if they want under the Korean national healthcare system. Therefore, our headache clinic has a wide spectrum of patients from treatment-naïve patients to those who failed several treatments received in other hospitals. Among the patients recruited from the prospective registry, patients with migraine who completed the follow-up evaluation at 3 months were included in this analysis, regardless of the completion of three consecutive monthly injections. Age range, disease duration, or age at onset was not considered for inclusion or exclusion. Patients who had other concomitant primary or secondary headache diagnosis were excluded, and no one received any other CGRP mAb or gepants as these drugs were not available in Korea during the study period. Exceptionally, one elderly patient with elderly onset of severe headache was excluded because the diagnosis could not be confirmed as the patient was not able to describe her headache well. The diagnosis of migraine was based on the International Classification of Headache Disorders (ICHD)-3 (10) by a headache specialist (MJL): 1.1, migraine without aura and 1.2, migraine with aura. Patients were further classified into the EM or CM groups according to the ICHD-3 criteria. Medication-overuse headache (MOH) was also diagnosed when appropriate using ICHD-3 criteria.
Intervention
All patients were administered monthly galcanezumab injection to prevent migraine attacks. A loading dose of 240 mg of galcanezumab was injected subcutaneously for the first month, and 120 mg of galcanezumab was injected monthly thereafter. Other oral preventive medications, except botulinum toxin A (BTX), were allowed, and triptans, ergotamine/caffeine, NSAIDs, and rarely acetaminophen/tramadol were used as acute headache medication. The data collected included demographic data, headache characteristics, baseline headache days, moderate/severe headache days, crystal clear days, and days of acute headache medication based on patients’ headache diary during the 30 calendar days immediately before the first injection. Current and previous use of migraine preventive medications and maximal doses of each drug were also reviewed. At the baseline visit, the patients completed self-reported questionnaires including the Headache Impact Test-6 (HIT-6) (11) and Migraine Disability Assessment Test (MIDAS) (12) for life impact assessment due to headaches and Generalized Anxiety Disorder-7 (GAD-7) (13), and the Patient Health Questionnaire-9 (PHQ-9) (14) were used to determine the presence of anxiety and depression, respectively.
Follow-up evaluation
The patients were scheduled for follow-up three months after the consecutive monthly treatments. The patients were instructed to complete a structured prospective headache diary. At the follow-up visit, we collected information on headache days, moderate/severe headache days, crystal clear days, and days of acute headache medication use during the previous month using the prospective diary and administering a structured follow-up questionnaire. Clinical interviews were conducted to ensure the accuracy and reliability of patients’ reports. HIT-6 and MIDAS scores were also assessed.
Outcome
We investigated changes in the number of moderate/severe headache days, headache days, crystal clear days, and days of acute headache medication use. The primary outcome was the 50% responder rate, which was based on ≥50% reduction in the number of moderate/severe headache days compared between that at baseline and the third month of treatment (15,16). The secondary outcomes were the 30%, 75%, and 100% responder rates, measured by ≥30%, ≥75%, and 100% reduction in moderate/severe headache days, respectively, and changes in total headache days, moderate/severe headache days, days of acute headache medicine, crystal clear days, and HIT-6 and MIDAS scores. Data on any adverse effect reported by patients and reasons for discontinuation were also collected.
Statistical analysis
The data are shown as numbers (percentages) and means ± standard deviations, depending on the data distribution. The subgroups based on CM, MOH, previous treatment failure, and prior response to BTX were analyzed with regard to treatment outcomes (50% response) by using logistic regression. Previous medication failures were further classified into 0–1, 2–4, and 5 classes. A two-tailed P-value < 0.05 denoted statistical significance in all analyses. Statistical analyses were performed using SPSS version 26.0 software (SPSS Inc., Chicago, IL, USA) and Stata 15.0 (Stata Statistical Software release 15, StataCorp LLC, College Station, TX).
Results
A flowchart of the study is shown in Figure 1. Among the 92 patients who received galcanezumab for the treatment of migraine, five were excluded because they declined follow-up evaluations or were lost to follow-up unexpectedly. Finally, 87 patients were included in the analysis. Most patients were women (83.9%). The mean age was 41.7 ± 12.3 (range 17–72) years. All patients developed migraine at the age of <50 (range 8–53) years, except for one whose age at onset was 53 years. All had a disease duration >1 (range 1.2–41) years. Sixty-five (74.7%) patients had CM, 35 (40.2%) had MOH, and 32 (36.8%) had a history of failure of five classes of preventive medications (Table 1). Forty-three (49.4%) patients received BTX treatment previously; 17 (19.5%) showed no response, 25 (28.7%) could not sustain their response, and one (1.1%) sustained the response.
Flow chart of patient inclusion and exclusion.
Demographics and characteristics of patients.
Data are presented as n (%) or mean ± standard deviation. MWA, migraine with aura; MOH, medication-overuse headache; BTX, botulinum toxin A; PHQ-9, 9-item Patient Health Questionnaire; GAD-7, General Anxiety Disorder-7.
*Non-sustained response was defined as the decrease in injection effect as the number of injections increased. Sustained response was defined as reproducible sufficient effect every time the injection was administered.
At baseline, 22.2 ± 8.44 headache days, 12.1 ± 8.52 moderate/severe headache days, 11.9 ± 9.30 days of acute headache medicine use, and 7.9 ± 8.63 crystal clear days were recorded (Table 2). After three months of treatment, the numbers of monthly headache days, moderate/severe headache days, and days of acute medication use decreased to 14.8 ± 10.31, 7.8 ± 8.87, and 7.8 ± 8.26 days (−7.2 ± 8.43, −4.3 ± 9.76, and −4.1 ± 7.93 days compared with those at baseline), respectively. The number of clear crystal days increased to 15.2 ± 10.29 (7.3 ± 8.63 compared with that at baseline).
Changes in the number of headache days and patient-reported outcomes after three month galcanezumab treatment.
Data are shown as mean ± standard deviation. HIT-6, Headache Impact Test-6; MIDAS, Migraine Disability Assessment Test.
The primary outcome: the overall 50% responder rate was 44.8% (54.5% for EM and 41.5% for CM) (Figure 2). The 30% responder rates for EM, CM, and overall were 59.1%, 55.4%, and 56.%, respectively. About one-fourth (27.7%) of the patients with CM and 27.3% of the patients with EM had a 75% response. A total of 12 patients (13.8%; 22.7% of EM; 10.8% of CM) showed a 100% response in moderate/severe headache days.
Responder rates of 30%, 50%, 75%, and 100%.
Subgroup analysis showed that the treatment response was not significantly affected by MOH (p = 0.250) or prior failure to BTX (p = 0.064, 0.843 compared with the never used group with non-effective and effective group) (Table 3 and Figure 3). In the subgroups based on previous failures of multiple drug classes, the 50% responder rates were 58.3%, 44.2%, and 40.6% for patients with failures of 0–1, 2–4, and 5 preventive medication classes, respectively; however, this trend was not significant (p = 0.384 using the linear-by-linear association test).
Subgroup analysis of MOH, CM, and prior BTX response.
MOH, medication-overuse headache; BTX, botulinum toxin A.
Subgroup analysis of MOH, previous medication failures, and prior BTX response.
Patient-reported outcomes also improved. The HIT-6 and MIDAS scores decreased after treatment (63.4 ± 7.08 vs 59.0 ± 9.26, p = 0.001 and 107.3 ± 101.18 vs 74.4 ± 74.04, p = 0.009, respectively) (Table 2).
During the study period, three patients discontinued galcanezumab. Reasons for discontinuation included inefficacy (n = 1), high cost (n = 1), and adverse events (n = 1). A total of 17 patients (19.5%) reported adverse events (Table 4). Seven patients (8.0%) experienced an injection site reaction. Other adverse effects were dizziness (seven patients, 8.0%), gastric or chest pain (two patients, 2.3%), severe headache (one patient, 1.1%), somnolence (one patient, 1.1%), weight loss (one patient, 1.1%), constipation (one patient, 1.1%), sweating (one patient, 1.1%), and transient sense of depersonalization (one patient, 1.1%).
Adverse events during study period*.
*One patient discontinued galcanezumab due to adverse event.
Discussion
This is the first set of real-world data on the effectiveness and tolerability of anti-CGRP monoclonal antibody therapy in the Korean population. Our study showed that galcanezumab is an effective preventive treatment for difficult-to-treat migraine in Asian populations. Both EM and CM patients responded well to galcanezumab with a comparable or even better response rate than the RCT results (5,6,8,17). The efficacies of patients with and without MOH, previous medication failure, and prior BTX response were similar.
Our study provides an Asian-specific treatment outcome following galcanezumab treatment. Of the galcanezumab RCTs, the EVOLVE-2 and CONQUER studies included 29.5% of the non-white population and 15.6% of the Asian population, respectively (6,8). Recently, results of a Taiwanese cohort from the EVOLVE-2 and REGAIN studies were reported (9). The number of patients was 30 and 24, mean age was 41.4 and 43.2 years, and female ratio was 86.7% and 75% in EVOLVE-2 and REGAIN, respectively. In EVOLVE-2, monthly migraine headache days showed no significant improvement when compared with that in the control group owing to the small number of patients, but in REGAIN, the proportion of patients achieving ≥50% response during the double-blinded period improved or was maintained during the open-label period.
In Japan, an Asian-specific outcome of anti-CGRP monoclonal antibody treatment of migraine was reported, as part of an open-label extension of the Japanese erenumab trial for the prevention of EM (18). In this trial, patients with no therapeutic response to >2 migraine-preventive treatment categories were not included, and the 50% and 75% responder rates were 44.5% and 23.6%, respectively. Compared with EVOLVE-2, CONQUER, and the Erenumab open-label extension study from Japan, our real-world study showed a responder rate comparable with or even better than those reported by RCTs involving Asian patients, despite the high disease severity in our patients (74.7% had CM, and 36.8% had histories of five classes of preventive medication failure). The effectiveness of galcanezumab regardless of the presence of CM or previous preventive medication may be attributed to the fact that oral preventive medications were maintained in most patients in our study, providing a possibility of additive or synergistic effect between oral preventive medications and galcanezumab: oral preventive medications act on the central nervous system, while the major site of action of CGRP-targeting monoclonal antibodies is outside the brain (19,20). In addition, the placebo response may be greater in real-world settings than in clinical trials, as physicians were more convinced of the effect of galcanezumab.
To date, there have been two real-world studies of galcanezumab: one is a multicenter cohort study from Italy and the other is a single clinic-based study from Spain (21,22). These two studies recruited difficult-to-treat patients with migraine, although the patient profiles differed slightly. In the Italian study, all patients who was enrolled had failed at least three preventive medications, and the Spanish group reported that nearly 90% of the patients had a history of failure of four or more classes of migraine prophylactic agents. When comparing these two studies, the 50% responder rate at the third month of treatment was numerically higher (66.7% for CM and 67.6% for high frequency EM) in the Italian study than in the Spanish study (51.6%). Our patient profile and responder rates were between those of the two studies. When restricted to patients who had a history of failure of four or more classes of preventive treatments, our study results were comparable with those of the Spanish study. This indicates that the effectiveness of anti-CGRP monoclonal antibodies differs not by ethnic group but by the number of previous treatment failures (Table 5).
Real-world study results of galcanezumab.
aThe data is based on third months visit.
bFailure of three or more preventive medications.
cOther patients received erenumab.
dReduction in monthly moderate/severe headache days.
CM, chronic migraine; HFEM, high-frequency episodic migraine (8–14 migraine days per month); MOH, medication-overuse headache.
The subgroup analysis showed that the responses of patients with and without MOH were comparable. This is consistent with previous RCT results (23–26), suggesting a new treatment paradigm for MOH. Previously, the mainstay of MOH treatment was withdrawal therapy, and the use of preventive treatment was considered an add-on to withdrawal. However, anti-CGRP monoclonal antibody trials showed promising results in the MOH subgroup without attempting withdrawal. In addition, our study showed similar effectiveness of galcanezumab in patients with and without prior BTX exposure. This finding is consistent with a recent report from a Spanish group (21,27). However, when restricted to patients who never responded to BTX, the 50% responder rate was low, though not significant. Likewise, our datashowed a numerical decrease in 50% responder rates when the number of previously failed preventive treatment classes increased, although the difference was not significant. Larger samples may be needed to determine the small difference in possible determinants of treatment response.
Our study had several strengths. This study is an Asian study to evaluate the effectiveness and safety of galcanezumab in a real-world setting. Real-world studies reflect the challenges in practice that were not included or analyzed in RCTs, such as previous failures of >4 preventive agents, concomitant use of multiple preventive drugs, and previous failures to BTX. In addition, we prospectively evaluated patients using a structured headache diary and questionnaires. The limitations of this study include the relatively small number of patients, single-center setting, and single ethnic group (i.e., Korean). The small sample size is a major limitation of our study, as it may not be representative of the entire migraine population. Further real-world studies from Asian regions are needed to generalize our study results. Furthermore, some reports regarding CGRP-targeting monoclonal antibodies have been recently published from an Asian perspective, limiting the novelty of our study (9,18,28). In addition, we considered both ineffectiveness and intolerability as failures of previous treatments. Although this is a commonly used concept in clinical trials and other real-world studies, failure might not be the most appropriate nomenclature for this, and it might be better to specify it as ineffectiveness or intolerability. However, these two factors were often mixed, and we did not classify the cause of failure in our study. Lastly, we only showed the treatment outcome at three months, and the long-term effectiveness and safety of galcanezumab were not determined in our study.
In conclusion, this study revealed that galcanezumab is effective for the prevention of difficult-to-treat migraine in an Asian population. Anti-CGRP monoclonal antibodies are highly promising treatment options for patients with history of failures of several preventive agents including BTX, and for patients with MOH.
Clinical implications
In this real-world study from Asia, the three month galcanezumab treatment yielded a high response rate with acceptable tolerability in Korean patients with difficult-to-treat migraine. The response rate was high regardless of migraine chronicity, comorbid medication overuse, and prior response to botulinum toxin treatment.
Footnotes
Acknowledgment
The authors thank Ms. Jinju Lee and Ms. Miran Jeong for data management.
Ethics approval
The study was approved by the Institutional Review Board of Samsung Medical Center (IRB #2018-10-029). All participants provided written informed consent.
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: SK and YEG report nothing to disclose. MJL has served as a consultant to Eli Lilly; has lectured for Eli Lilly, Sanofi-Aventis, and YuYu Pharma; and has been the PI in trials sponsored by Lundbeck, Novartis, Teva (Otsuka), Allergan, Biohaven Pharmaceuticals, Ildong Pharmaceutical Co., and Yuhan Company, none of which were related to this work.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. NRF-2020R1A2B5B01001826). Yuhan Company supported data management.
