Abstract
Background
Ictal cutaneous allodynia, common in chronic migraine, is associated with reduced responses to acute treatment with triptans. Allodynia’s impact on the efficacy of newer preventive treatments such as erenumab is unknown.
Methods
Post-hoc subgroup analysis of a double-blind, randomized, placebo-controlled 12-week study of erenumab in chronic migraine, contrasting those with no allodynia with those with moderate-severe allodynia assessed with the Allodynia Symptom Checklist-12, was undertaken.
Results
Of 648 randomized individuals with baseline Allodynia Symptom Checklist-12 scores, 386 (59.6%) had no allodynia and 153 (23.6%) had moderate-to-severe allodynia. Mean (standard deviation) baseline monthly migraine days were 17.6 (4.8) and 18.9 (4.3), respectively. Compared to placebo, the erenumab group had greater reductions in monthly migraine days and monthly acute migraine-specific medication days in both no allodynia and allodynia subgroups. Mean (95% confidence interval) treatment differences in change from baseline for monthly migraine days at week 12 were −2.5 (−3.7, −1.4) in the no allodynia subgroup and −3.3 (−5.3, −1.3) in the moderate-severe allodynia subgroup. Change in acute migraine-specific medication days were −3.3 (−4.3, −2.3) and −2.5 (−4.3, −0.8), respectively.
Conclusions
Erenumab’s efficacy in reducing monthly migraine days and acute migraine-specific medication days in chronic migraine was not impacted by the presence of moderate-severe ictal allodynia.
Trial registration: ClinicalTrials.gov NCT02066415
Introduction
An estimated two-thirds of people with migraine in the US population experience ictal cutaneous allodynia (i.e., during migraine attacks) (1–3). Allodynia may also be present between attacks (i.e. interictally) (3). Allodynia is characterized by the sensation of pain provoked by ordinarily non-noxious stimuli (3). Allodynia of the head and face arises from sensitization of nociceptive neurons in the trigeminal nucleus caudalis, which receives convergent afferent inputs from first division trigeminal (V1) and upper cervical afferents (4,5). Extracephalic allodynia may arise from sensitization of thalamic neurons (2).
Allodynia is of clinical importance for several reasons. First, it is a clinical reflection of central sensitization and provides clues to the mechanisms of pain evolution during migraine attacks (6). Second, the presence of allodynia has been associated with migraine progression or the new onset of chronic migraine (CM) among people with episodic migraine (EM) and can have implications for treatment and prognosis (3,6). Third, there are important clinical differences among people with migraine who do and do not have allodynia. The prevalence of cutaneous allodynia is higher in individuals who have had a long duration of illness, high frequency of attacks, more severe symptoms, acute medication overuse, depression and anxiety, and greater disability (7). The presence and severity of allodynia has been associated with several aspects of migraine, including nausea, photophobia, and phonophobia. Finally, the presence of allodynia may be associated with a reduced response to acute treatment (8–10). However, relatively little is known regarding the impact that allodynia has, if any, on the efficacy of preventive migraine treatments.
Erenumab (erenumab-aooe in the US) is a fully human monoclonal antibody (mAb) that selectively targets and blocks the canonical calcitonin gene-related peptide (CGRP) receptor and is approved for the prevention of migraine in adults (11). At present, the effect of allodynia on the efficacy of erenumab is unknown; therefore, we conducted a post-hoc subgroup analysis contrasting subgroups of individuals with extremes of allodynia from a pivotal study of erenumab in individuals with CM (12).
Methods
Study design
A post-hoc subgroup analysis of a multicenter, randomized, double-blind, placebo-controlled study (NCT02066415) was performed for subgroups of individuals with CM and extremes of ictal cutaneous allodynia. Following an initial screening phase of up to 3 weeks and a subsequent 4-week baseline phase, those eligible were randomized in the parent study in a 3:2:2 fashion to receive either subcutaneous placebo (n = 286), erenumab 70 mg (n = 191), or erenumab 140 mg (n = 190), every 4 weeks during the 12-week double-blind treatment phase. Safety follow-up lasted 12 weeks after the treatment (12). Adults (≥ 18 to ≤ 65 years of age) with a diagnosis of migraine based on the International Classification of Headache Disorders, third edition (beta version) (ICHD-3 beta) (13) and ≥ 15 headache days per month, of which ≥ 8 were migraine days, were included. Individuals who had no therapeutic response to > 3 classes of prophylactic treatment were excluded. The primary endpoint for the parent study was change from baseline in monthly migraine days (MMD) at week 12. Subgroups were analyzed according to the severity of allodynia at baseline, and subgroups with extremes of allodynia (defined as no allodynia vs. moderate-severe) were compared in this analysis. We defined allodynia using the Allodynia Symptom Checklist-12 (ASC-12), a recall-based measure that captures only ictal allodynia (3). Subgroup analysis endpoints at week 12 included change from baseline in MMD, monthly acute migraine-specific medication (AMSM) days in patients with baseline AMSM use, and proportion of patients achieving ≥ 50% reduction in MMD from baseline.
Allodynia assessment
The validated, ASC-12 questionnaire was used to assess ictal cutaneous allodynia associated with migraine attack severity. It included 12 questions about the frequency of various allodynia symptoms in association with headache attacks, encompassing the three domains of allodynia (thermal, mechanical dynamic, and mechanical static). Response options were “never,” “rarely,” “less than half the time,” and “half the time or more,” or “does not apply to me.” Following the validated scoring system, ASC-12 items were scored as 0 (never, rarely, or does not apply to me), 1 (less than half the time), and 2 (half the time or more), yielding total scores ranging from 0 to 24. “No allodynia” (none or minimal) was defined as ASC-12 score < 3. Mild allodynia was defined as ASC-12 score ≥ 3 to < 6. Moderate-to-severe allodynia was defined as ASC-12 score ≥ 6 (3,14). Individuals with mild allodynia were not included in analyses of the extreme subgroups of allodynia; we hoped to increase our ability to see an effect of allodynia by focusing on the moderate-to-severe allodynia group. In addition, for the group with mild allodynia there were only 42 placebo treated participants, which might have limited power to identify the effects of allodynia on treatment. For the mild allodynia subgroup proportions were imbalanced both for prior preventive treatment failure (placebo 54.8% versus erenumab 71.6%) and for use of migraine specific acute treatment use (placebo 61.9% and erenumab 76.1%) (Table 1). Of note, ASC-12 assesses the frequency of 12 allodynia-related symptoms during headaches and was not designed to distinguish phases of the migraine attack or to measure interictal allodynia.
Baseline characteristics.
Note: Data represent mean (SD) unless otherwise indicated.
Statistical analysis
Owing to sample size considerations, results were pooled for the two erenumab dose groups. For the purposes of these analyses, the efficacy analysis set included those randomized with baseline ASC-12 scores within category extremes (no/minimal allodynia and moderate-to-severe allodynia) and who completed ≥ 1 post-baseline monthly electronic diary measurement. Adjusted analysis used a generalized linear mixed model, which included treatment, visit, treatment by visit interaction, stratification factors region and medication overuse, and baseline value as covariates and assuming a first-order autoregressive covariance structure. Nominal p-values are presented without multiplicity adjustment.
Results
Baseline characteristics
Of the 648 randomized individuals with baseline ASC-12 scores, 386 (59.6%) had no or only minimal symptoms of allodynia, 109 (16.8%) had mild allodynia, and 153 (23.6%) had moderate-to-severe allodynia. Of the 386 individuals with no allodynia, 170 were treated with placebo and 216 were treated with erenumab (70/140mg). Of the 153 individuals with moderate-to-severe allodynia, 61 were treated with placebo and 91 were treated with erenumab (70/140mg). Of the 109 individuals with mild allodynia, 42 were treated with placebo and 67 were treated with erenumab (70/140mg).
Baseline characteristics were generally well balanced across the treatment groups, with a few exceptions (Table 1). Most of the study population was female (72.9 − 91.9%); the average age was between 41 and 42 years; most (61.3 − 83.5%) had previous use of preventive treatment, and most (61.9 − 80.6%) were using acute migraine-specific medication. Prior preventive treatment use was highest in the placebo non-allodynia subgroup and lowest in the placebo moderate-to-severe allodynia subgroup. Baseline MMD was similar across treatment groups ranging from 17.3 to 19.0 days, but slightly higher for the moderate-to-severe subgroup. Baseline monthly AMSM days were similar across treatment groups ranging from 9.9 to 12.6 days among the 76% of patients who used AMSM at baseline (n = 492; Table 1).
Efficacy
MMD: In both the no allodynia and moderate-severe allodynia subgroups, erenumab resulted in a significantly greater mean reduction in MMD compared with placebo at week 12 (Figure 1). Least-square mean (95% CI) changes were −6.6 (−7.4, −5.8) versus −4.0 (−4.9, −3.1) in the no allodynia subgroup and −7.3 (−8.6, −6.1) versus −4.1 (−5.7, −2.5) in the moderate-to-severe allodynia subgroup. Respective treatment differences were −2.5 (−3.7, −1.4) (p < 0.001) and −3.3 (−5.3, −1.3) (p = 0.001), with a treatment-by-subgroup interaction p-value > 0.5. Furthermore, the interaction p-value between ASC-12 score and treatment group for the comparison between placebo and combined 70/140 mg group was p = 0.7906 and for the comparison between 70 mg and 140 mg groups was p = 0.9448. Given the higher proportion of those who failed prior preventive treatments in the no allodynia group receiving placebo, a sensitivity analysis was conducted on the primary endpoint. The change in MMD for those who failed prior preventive treatments was similar to the primary subgroup analysis; LSM (95% CI) difference from placebo for the combined 70/140 mg group was −3.6 (−5.8, −1.4) (p = 0.002) for the moderate-severe subgroup and −3.8 (−5.0, −2.6) (p < 0.001) for the no allodynia subgroup.
Efficacy over time in patients without allodynia (a) and with moderate-to-severe allodynia (b). Least-squares mean changes from baseline in monthly migraine days (MMD) among patients with chronic migraine who were assigned to receive erenumab (70 mg or 140 mg) or placebo every 4 weeks. The error bars represent 95% confidence intervals (95% CI).
At least 50% response: In both the no allodynia and moderate-to-severe allodynia subgroups, erenumab resulted in a significantly higher proportion of patients achieving ≥ 50% reduction from baseline in MMD compared with placebo at week 12 (Figure 2). Response rates were 41.5% versus 23.2% (adjusted odds ratio [95% CI] erenumab vs. placebo: 2.34 [1.49, 3.68]; p < 0.001) for no allodynia and 44.0% versus 23.0% (adjusted odds ratio: 2.69 [1.27, 5.70]; p = 0.008) for moderate-to-severe allodynia. The interaction p-value for between-group comparisons (“no allodynia group” and the “moderate to severe allodynia group”) of treatment efficacy for at least 50% Response was p = 0.145.
Responder rates over time in patients without allodynia (a) and with moderate-to-severe allodynia (b). Proportion of patients with chronic migraine who were assigned to receive erenumab (70 mg or 140 mg) or placebo every 4 weeks achieving ≥ 50% reduction from baseline in monthly migraine days.
Monthly AMSM days: Among patients who used AMSM at baseline in both the no allodynia and the moderate-to-severe allodynia subgroups, erenumab resulted in a significantly greater reduction in mean monthly AMSM days compared with placebo at week 12 (Figure 3). Least-square mean (95% CI) changes were −5.3 (−6.03, −4.63) versus −2.0 (−2.82, −1.25) days in the no allodynia subgroup and −4.9 (−6.00, −3.79) versus −2.4 (−3.69, −1.03) days in the moderate-to-severe allodynia subgroup. Respective treatment differences were −3.3 (−4.3, −2.3) (p < 0.001) and −2.5 (−4.3, −0.8) (p = 0.004) days. The interaction p-value for between-group comparisons (“no allodynia group” and the “moderate to severe allodynia group”) of treatment efficacy for monthly AMSM days was p = 0.0545.
Change from baseline in monthly acute migraine-specific medication days over time in patients who used baseline acute migraine-specific medication without allodynia (a) and with moderate-to-severe allodynia (b). Least-squares mean changes from baseline in monthly acute migraine-specific medication (AMSM) days among patients with chronic migraine who were assigned to receive erenumab (70 mg or 140 mg) or placebo every 4 weeks. The error bars represent 95% confidence intervals (95% CI).
Discussion
In this subgroup analysis of a double-blind placebo-controlled study of erenumab for the prevention of CM, erenumab led to similar improvements in patients with moderate-to-severe ictal allodynia and in those without ictal allodynia for each of the efficacy endpoints: change from baseline in MMD and monthly AMSM days, as well as the proportion of patients achieving ≥ 50% MMD reduction. In contrast to migraine-specific acute medications (e.g. triptans) where presence of allodynia has been reported to be associated with a reduced response (15), we did not observe such an association between the presence of ictal cutaneous allodynia and the treatment response to erenumab.
The possibility that ictal allodynia may not affect treatment response to preventive medications has been reported in two previous open-label studies of onabotulinumtoxinA. In one study, assessment of ictal allodynia (using Quantitative Sensory Testing), obtained before and 4–12 weeks after treatment of high-frequency episodic migraine patients with onabotulinumtoxinA showed no significant difference between responders and non-responders (16). In another open label study of onabotulinumtoxinA in CM, treatment was effective in subgroups defined by ASC-12 as with and without allodynia (17). Improvement in headache day frequency at week 108 was significantly greater in those without allodynia compared to those with allodynia. For most other endpoints (e.g. Headache Impact Test-6 (HIT-6) and Migraine-Specific Quality of Life questionnaire (MSQ) scores), differences were not significant between the with and without allodynia subgroups. To our knowledge, this is the first analysis looking at the impact of allodynia on response to a preventive therapy targeting the CGRP pathway.
The notion that central sensitization and cutaneous allodynia impact the outcome of treating an acute migraine attack with triptans was established in pre-clinical studies in which triptans were found to reverse central sensitization if given immediately after induction of sensitization in the spinal trigeminal nucleus (part of the trigeminal cervical complex) but not when given 4 h later (1,18). Parallel clinical studies showed that migraine attacks associated with cutaneous allodynia were readily terminated by early (i.e. < 1 h after attack onset), but not late (i.e. ≥ 4 h after onset), triptan administration (15). The main conclusion of these series of studies was that triptans are effective when central sensitization is still developing (i.e. when the activity of sensitized neurons in the trigeminal cervical complex depends on incoming pain signals from the meninges) but not when it is fully established (i.e. when the activity of sensitized neurons in the trigeminal cervical complex is independent of incoming pain signals from the meninges). This may be due to the fact that although triptans cross the blood-brain barrier, they do not inhibit second-order trigeminovascular neurons in the spinal trigeminal nucleus as these neurons lack 5HT1b/1d receptors (19–21). Outside the acute attack phase, several studies have shown that individuals with chronic migraine are more likely to experience interictal allodynia than those with episodic migraine (22,23), suggesting that the lower efficacy of triptans in individuals with chronic migraine may be due to the ongoing headache being mediated by sustained activity independent of afferent input (15).
If the main site of action of monoclonal antibodies targeting the CGRP pathway (CGRP-mAbs) is outside the central nervous system, where they attenuate meningeal nociceptors activation (29) and dilatation of dural arteries (30,31), one would expect these drugs to have little effect on central sensitization and on migraine attacks whose pathophysiology include cutaneous allodynia. While we know that CGRP-mAbs can prevent central sensitization from developing (31), and effectively prevent migraine (12,24–26), including in those with ictal allodynia (current study), it is not known whether residual attacks while on treatment (among both responders and non-responders) are associated with ictal allodynia, or whether the prolonged elimination of nociceptive signals from the meninges to the spinal trigeminal nucleus alters the molecular mechanisms of central sensitization, to the extent that it becomes dependent on the input it receives from the peripheral nociceptors again. As CGRP-mAbs do not appear to access the spinal trigeminal nucleus or thalamic neurons that mediate ictal allodynia (27,28), presumably due to the blood-brain barrier, we should consider the possibility that ictal allodynia and central sensitization may switch from being independent of peripheral pain signals from meningeal nociceptors to being dependent on these peripheral signals due to their prolonged suppression by the CGRP-mAbs. In such a case, a peripherally-acting drug such as erenumab can be effective in preventing migraine attacks in patients with ictal allodynia.
There are several limitations to the current analysis. The modest number of study participants with allodynia necessitated the pooling of data from both erenumab dose groups and from the moderate and severe allodynia subgroups. Therefore, neither drug dose effects nor effects of gradations of allodynia severity could be assessed. We were unable to include individuals with mild allodynia in these analyses due to the small sample size and heterogeneity in baseline characteristics of the mild allodynia subgroup. This limits the analysis and conclusions to just the extremes of allodynia (none or moderate-to-severe) and further research would be required to determine any impact of mild allodynia on treatment response. The study was limited to a duration of 12 weeks. Although erenumab has demonstrated efficacy in clinical trials of up to 5 years’ duration (32), it is unknown whether longer follow-up would result in attenuation of the currently observed benefit in patients with allodynia. Although the ASC-12 is a standard and validated tool assessing cutaneous allodynia in migraine, it requires subjects to recall ictal allodynia symptoms and estimate their frequency during their most severe headaches, making this data collection method vulnerable to recall bias. Notably, ∼60% of individuals in this study had no or minimal symptoms of allodynia, in contrast to the reported two thirds of patients assessed ictally with quantitative sensory testing, reflecting potential differences in patient selection or sensitivity of the ASC-12. However, limiting the analysis to the extremes of ictal cutaneous allodynia may have helped minimize the impact of recall bias on categorizing patients within subgroups.
In addition, the current analysis was limited to the impact of ictal allodynia on erenumab’s efficacy rather than including a determination of possible effects of erenumab on allodynia symptoms, disability, or other migraine-associated symptoms (such as photophobia or nausea), which could possibly be assessed in future analyses. Subgroups were based on those who completed the ASC-12 questionnaire, which, although a well-validated measure, does not distinguish early and late ictal allodynia and does not measure interictal allodynia. Nonetheless, the current analyses show that erenumab is effective in the preventive treatment of CM in patients with and without ictal allodynia, as measured by the ASC-12. Further studies are needed to determine if treatment response is affected by interictal allodynia, which may represent individuals with sustained sensitization or those whose central sensitization is independent of incoming peripheral input from the dura.
Clinical implications
In patients with chronic migraine, erenumab was equally effective in reducing the number of monthly migraine days and acute migraine-specific medication days in patients with moderate-to-severe allodynia and in those without allodynia. The presence of ictal allodynia does not appear to negatively impact response to erenumab treatment for the prevention of chronic migraine.
Footnotes
Acknowledgements
Erenumab is co-developed in partnership with Amgen Inc. and Novartis. The authors acknowledge Shannon Rao and Jon Nilsen (both of Amgen Inc.) for medical writing assistance. The authors thank Farhad Sahebkar-Moghaddam, MD for his assistance in the early-stage planning and design of the analysis.
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: RBL is the Edwin S Lowe Professor of Neurology at the Albert Einstein College of Medicine in New York. He receives research support from the NIH, FDA as well as the National Headache Foundation and the Marx Foundation. He also receives research support from Allergan/Abbvie, Amgen, Eli Lilly and Electrocore. He receives personal fees as a consultant or advisor from Allergan/Abbvie, Amgen, Biohaven Holdings, Dr. Reddy’s, GlaxoSmithKline, Grifols, Lundbeck, Merck, Novartis and Teva Pharmaceuticals. He holds stock or options in Biohaven Holdings and CtrlM Health. In addition, he receives royalties for Wolff’s Headache 7th and 8th edition.
RB is the John Hedley-Whyte Professor of Anesthesia and Neuroscience at the Beth Israel Deaconess Medical Center and Harvard Medical School. He has received research support from the NIH: R01 NS094198-01A1, R37 NS079678, R01NS095655, R01 NS104296, R21 NS106345, Allerga, Teva, Dr. Reddy’s, Eli Lilly, Trigemina and the Migraine Research Foundation [‘Dr Ready’ changed to ‘Dr. Reddy’s’, please check.]. He is a reviewer for NINDS, holds stock options in AllayLamp, Theranica and Percept; serves as consultant, advisory board member, or has received honoraria from: Alder, Allergan, Amgen, Autonomic Technologies, Avanir, Biohaven, Dr. Reddy’s Laboratory, electroCore, Eli Lilly, GlaxoSmithKline, Merck, Pernix, Theranica, Teva, and Trigemina. He has received CME fees from Healthlogix, Medlogix, and WebMD/Medscape.
DCB, in the past 12 months, has received grant support and honoraria from Allergan, Amgen, Biohaven, Lilly, Promius, and Teva. She is on the editorial board of Current Pain and Headache Reports.
DWD reports the following conflicts within the past 12 months: Consulting: AEON, Amgen, Clexio, Cerecin, Cooltech, Ctrl M, Allergan, Alder, Biohaven, Linpharma, Lundbeck, Promius, Eli Lilly, eNeura, Novartis, Impel, Satsuma, Theranica, WL Gore, Nocira, XoC, Zosano, Upjohn (Division of Pfizer), Pieris, Praxis, Revance, and Equinox. Honoraria: CME Outfitters, Curry Rockefeller Group, DeepBench, Global Access Meetings, KLJ Associates, Academy for Continued Healthcare Learning, Majallin LLC, Medlogix Communications, MJH Lifesciences, Miller Medical Communications, Southern Headache Society (MAHEC), WebMD Health/Medscape, Wolters Kluwer, Oxford University Press, and Cambridge University Press. Research support: Department of Defense, National Institutes of Health, Henry Jackson Foundation, Sperling Foundation, American Migraine Foundation, and Patient Centered Outcomes Research Institute (PCORI). Stock options/shareholder/patents/board of directors: Ctrl M (options), Aural analytics (options), ExSano (options), Palion (options), Healint (options), Theranica (options), Second Opinion/Mobile Health (options), Epien (options/board), Nocira (options), Matterhorn (shares/board), Ontologics (shares/board), King-Devick Technologies (options/board), and Precon Health (options/board). Patent 17189376.1-1466:vTitle: Botulinum Toxin Dosage Regimen for Chronic Migraine Prophylaxis.
RK is an employee of and stockholder in Amgen Ltd.
JK is an employee of and stockholder in Novartis.
SC and DEC are employees of and stockholders in Amgen Inc.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study (NCT02066415) was funded by Amgen Inc. and Novartis.