{"paper_id":"b19dfee2-7ddb-422f-a1e5-aff0774fdede","body_text":"Aimovig® (erenumab-aooe) is a first in class Food and Drug \nAdministration (FDA)-approved human monoclonal antibody for the prevention of \nmigraine in adults. It selectively targets and blocks the calcitonin gene-related \npeptide (CGRP) receptor, disrupting a key component of migraine pathophysiology \n[ 1 ]. Several studies have provided evidence of the safety and efficacy of \nerenumab in reducing the frequency of migraine compared to placebo [ 2 ,  3 ]. \nFurthermore, an open-label longer-term study found that erenumab was safe and \nwell-tolerated with a safety profile consistent with shorter-term \nplacebo-controlled studies through 5-years of treatment [ 4 ]. Erenumab has rapidly \nbecome a widely accepted prescription drug for the prevention of migraine, \nincluding episodic migraine and chronic migraine along with other anti-CGRP \nmonoclonal antibody-based therapies [ 5 ,  6 ].\nTemporomandibular disorder (TMD) is known to be co-morbid with the medical \ndiagnosis of chronic migraine [ 7 ,  8 ]. Temporomandibular disorder (TMD) is a \ncommon condition that may affect up to a third of the general population [ 9 ]. TMD \nis the most common orofacial pain condition of non-dental origin [ 10 ]. \nAdditionally, TMD has a major adverse impact on health-related quality of life \n[ 11 ,  12 ] as well as health care costs [ 13 ]. There is increasing interest in the \nconcept of Chronic Overlapping Pain Conditions (COPCs), which include TMD, \nfibromyalgia, irritable bowel syndrome, vulvodynia, chronic fatigue syndrome, \ninterstitial cystitis/painful bladder syndrome, endometriosis, chronic \ntension-type headache, migraine headache, and chronic lower back pain that may \nhave increased pain sensitivity as well as common genetic and biopsychosocial \nfactors [ 14 ].\nDuring 2018–2019, shortly after the commercial release of erenumab, one of the \nauthors (HCA) used erenumab to treat 5 patients with chronic severe TMD pain and \na history of migraine headaches. Four of these patients had substantial \nreductions in pain following this off-label administration of erenumab. These \npromising treatment results were the impetus for the pilot trial reported in this \npaper.\nChronic migraine is thought to originate within the trigeminovascular pathway \n(TGV) [ 15 ,  16 ]. TMD is also considered to originate within the TGV [ 17 ]. Thus, \nour working hypothesis is that a CGRP receptor antagonist for treatment of \nchronic migraine will also be effective in reducing TMD pain and related \nsymptoms. The purpose of this proof-of-concept study was to evaluate the safety \nand efficacy of erenumab in reducing Temporomandibular Disorder (TMD) pain \ncompared to placebo. The study design was a phase II randomized \nplacebo-controlled clinical trial. We postulated that erenumab would be superior \nto placebo in reducing pain intensity/severity over 20 weeks. Secondary outcomes \nincluded depression, anxiety and somatic symptoms; jaw function; and percentage \nof days taking pain medication.\n\nEligible participants included adults (age 18 to 59 years) who were diagnosed as \nhaving pain-related TMD using the diagnostic criteria (DC/TMD) for “myalgia”, \nrecommended by the International RDC/TMD Consortium Network and Orofacial Pain \nSpecial Interest Group [ 18 ]. Eligible participants also had to have a history of \nhead, face, neck, and/or shoulder pain for longer than 3 months; a good knowledge \nof the English language; and if taking prescription pain medication, a stable \ndose regiment for at least 2 months prior to the screening visit.\nThe Diagnostic Criteria for Temporomandibular Disorders Symptom Questionnaire \nand DC/TMD Examination Form was used to confirm the TMD diagnosis [ 19 ]. To meet \nthe diagnostic criteria for TMD Myalgia (IDC-9 729; ICD-10 M79.1), subjects must \nhave had a history of pain of muscle origin that was affected by jaw movement, \nfunction or parafunction, and demonstrated replication of this pain with \nprovocation testing of the masticatory muscles [ 18 ]. No minimum pain threshold \nwas used for eligibility because the DC/TMD criteria require documentation of \npain but no specific level of pain severity. The criteria included having a \npositive history for both pain in the jaw, temple, in the ear, or in front of ear \nand pain modified with jaw movement, function or parafunction. During clinical \nexamination, subjects must have had a confirmation of pain in the temporalis or \nmasseter muscle(s) and report familiar pain in the temporalis or masseter \nmuscle(s) with at least one of the following provocation tests: palpation of the \ntemporalis or masseter muscle(s) or maximum unassisted or assisted opening \nmovement(s).\nSubjects that met any of the following exclusion criteria were not eligible: (1) \nlacking stable bilateral posterior occlusion; (2) currently using a complete \nmaxillary or mandibular prosthetic denture; (3) currently undergoing TMD \ntreatment elsewhere (exception is subjects undergoing TMD treatment involving the \nuse of oral orthotics for a minimum of 3 months prior to screening can be \nconsidered eligible for the study); (4) started orthodontic treatment during the \n3 months prior to Screening; (5) currently included in other experimental \nprotocols within the last 30 days or 5 half-lives before enrollment; (6) \ncurrently pregnant, planning to become pregnant or breastfeeding; (7) allergic to \nerenumab or any of the ingredients in Aimovig® (acetate, \npolysorbate 80 and sucrose); (8) allergic to rubber or latex; (9) having 8 or \nmore migraine days during the past 4 weeks; (10) started receiving massage, \nacupuncture or physical therapy treatment of the head, neck or shoulders during \nthe previous 3 months prior to Screening; (11) history of unstable or acute \nsevere non-head, neck or shoulder pain; (12) history of traumatic brain injury; \n(13) history of surgical treatment or recommended surgical treatment for TMD; \n(14) history of ongoing, unresolved disability litigation; (15) history of drug \nabuse; (16) started treatment for moderate to severe sleep apnea requiring \ncontinuous positive airway pressure (CPAP) or oral mandibular repositioning \nappliance during the previous 3 months prior to Screening; (17) history of \npreviously receiving erenumab-aooe or other anti-CGRP pathway therapies, \nincluding anti-CGRP pathway treatments; (18) history of chronic constipation \nand/or using medication associated with decreased gastrointestinal motility; (19) \nhistory of uncontrolled hypertension or risk factors for hypertension; (20) \nanything that would place the subject at increased risk or preclude the \nindividual’s full compliance with or completion of the study ( e.g. , \nmedical condition, laboratory finding, physical exam finding logistical \ncomplication).\nParticipants were recruited from November 2021 through July 2023 using fliers \nand advertisements placed in the Indiana University (IU) School of Dentistry \n(IUSD) and other locations on the Indiana University Purdue University \nIndianapolis (IUPUI) campus including IU Health facilities. We also used social \nmedia advertisements. Persons responding to an advertisement were given a brief \ndescription of the study and asked a series of questions related to the \ninclusion/exclusion criteria using an institutional review board (IRB) approved \nphone script. Those who were and appeared to meet the study requirements were \nscheduled for a screening visit at the Oral Health Research Institute (OHRI). A \nStudy Dentist qualified to diagnosis TMD reviewed the potential subject’s health \nhistory, medications and TMD history for the inclusion and exclusion criteria.\nPatients were randomized at baseline to one of the two treatment arms:\n● Arm A: erenumab 140 mg subcutaneous, administered every four weeks \nfor a total of five treatments.\n● Arm B: placebo subcutaneous, administered every four weeks for a \ntotal of five treatments.\nRandomization was stratified based on sex into two groups using block \nrandomization based on a schedule provided by the study statistician. Subjects, \ninvestigators, and study staff remained blinded to the identity of the treatment \nfrom the time of randomization until database lock. The randomization code was \nkept strictly confidential, and the identity of the study drug treatments \nconcealed using identical packaging and labeling.\nThe study sponsor provided the active and placebo free of charge through the \nInvestigator Sponsored Studies (ISS) Program (CAMG334AUS01T). The investigational \nproducts (erenumab and placebo) were supplied in prefilled syringes, using \nidentical packaging and labeling and shipped through ISS to the unblinded IU \nHealth Investigational Drug Services Pharmacy. The IU Health Pharmacy dispensed \nthe investigational products according to a randomization schedule provided by \nthe study statistician. Doses were administered in the upper arm, thigh, or \nabdomen by a study dentist qualified in subcutaneous drug administration.\nStudy products were stored and handled according to labeling instructions and \nstored in a secure area of the IU Health Investigation Drug Services Pharmacy to \nwhich only the pharmacy staff had access. The IU Health Pharmacy maintained \nrecords documenting the receipt, use, loss or other disposition of the products \non the electronic Investigational Agent Accountability Record. The clinical site, \nOHRI, working with the pharmacy also maintained a Drug Administration Form \ndocumenting the date and time of transport to the blinded site staff, the date \nand signature of the blinded site staff receiving the study drug, the date and \ntime the study drug was received from the pharmacy, the date and time the study \ndrug was administered, and the randomized injection site body location noted by \nthe study dentist. These procedures coupled with the use of identical prefilled \nsyringes for the erenumab, and placebo groups assured blinding of the study \nsubjects, investigators, research staff and outcome assessors.\nAt Baseline and Weeks 4, 8, 12, 16, 20 and 24 subjects were instructed to \ncomplete patient-reported outcomes regarding pain and other TMD-relevant \nsymptoms. The outcome measures were based on consensus recommendations for \nresearch assessments in chronic pain and TMD research [ 19 ,  20 ].\nThe primary study outcome was the Brief Pain Inventory (BPI) pain severity scale \nwhich rates the severity of pain on 4 items (current, worst, least and average \npain in past week) [ 21 ,  22 ,  23 ]. Each item is rated on a 0 (no pain) to 10 (pain as \nbad as you can image) scale. The BPI pain severity score is the average of the \nitems and ranges from 0 to 10, with higher scores representing greater pain \ninterference.\nThree other pain outcomes were assessed. The Brief Pain Inventory (BPI) pain \ninterference scale rates pain-related interference in 7 areas (mood, physical \nactivity, work, social activity, relations with others, sleep, and enjoyment of \nlife). Each item is rated on a 0 (does not interfere) to 10 (completely \ninterference) scale [ 21 ]. The BPI pain interference score is the average of the \nitems and ranges from 0 to 10, with higher scores representing greater pain \ninterference. The Patient Global Impression of Change (PGIC) assesses change in \npain on a 7-item Likert scale where 1 = much better; 2 = moderately better; 3 = a \nlittle better; 4 = no change; 5 = a little worse; 6 = moderately worse; 7 = much \nworse [ 24 ]. Daily use of pain medications was tracked each month by asking how \nmany days medications were taken for TMD-related pain.\nDepressive and anxiety symptoms were assessed by the Patient Health \nQuestionnaire (PHQ-4) which comprises 2 depression items and 2 anxiety items \n[ 25 ,  26 ,  27 ]. Individuals are asked how much they have been bothered by each of the \nsymptoms during the past 2 weeks on a scale of 0 (not at all) to 3 (nearly every \nday). The PHQ-4 total score ranges from 0 to 12 with higher scores representing \nmore severe symptoms. Jaw function was assessed with the Jaw Function Limitation \nScale (JFLS-8) which asks the level of limitation during the past month in 8 \nactivities (chew tough food; chew chicken; eat soft food requiring no chewing; \nopen wide enough to drink from a cup; swallow; yawn; talk; smile) [ 28 ,  29 ]. Each \nitem is scored from 0 (no limitation) to 10 (severe limitation). The JFLS-8 score \nis the average of the 8 items and ranges from 0 to 10, with higher scores \nrepresenting greater jaw functional impairment. The Somatic Symptom Scale (SSS-8) \nasks how much each of 8 common physical symptoms have bothered the individual \nduring the past 7 days on a 5-point Likert scale ranging from 0 (not at all) to 4 \n(very much) [ 30 ]. Total scores range from 0 to 32 with higher scores representing \nhigher somatic symptom burden.\nAdverse events were assessed and documented at each follow-up visit. This study \nwas conducted in compliance with the US Code of Federal Regulations (CRF) \ngoverning informed consent, the IRB, and Investigator conduct. This study was \nperformed according to Good Clinical Practice for research. Standard operating \nprocedures for the trial were on file with the Quality Assurance staff of the \nOral Health Research Institute. All study staff who had direct contact with \nsubjects were required to review the WARNINGS AND PRECAUTIONS for \nHypersensitivity Reactions, Constipation with Serious Complications and \nHypertension found in Section 5 of the US Prescribing Information (USPI) for \nAimovig.\nA sample size of 12 subjects per group has been suggested for pilot studies to \nevaluate feasibility and to estimate group means and standard deviations (SD) for \nfuture study planning [ 31 ,  32 ]. Based on two-sided paired  t -tests and \ntwo-sample  t -tests, all conducted at a 5% significance level, this \npilot study had 80% power to detect effect sizes of 0.9 for changes over time \nwithin groups and effect sizes of 1.2 for differences between groups. To account \nfor dropout, the study enrolled 15 subjects per group, for a total of 30 \nsubjects.\nMixed model repeated measures (MMRM) analysis was used to evaluate changes over \ntime in the BPI pain and BPI interference scores, JFLS-8, PHQ-4 total score, \nPHQ-4 anxiety, and depression scores, SSS-8 Scale, and PGIC pain change within \nand between treatment groups. The MMRM included factors for treatment group, \ntime, and their interaction. The MMRM also included sex as a covariate due to \nstratification by sex in the randomization. A two-sided 5% significance level \nwas used for all tests without multiplicity adjustment between multiple \nendpoints.\n\nThirty-two individuals met the inclusion/exclusion criteria. Two individuals who \nqualified for the study were never randomized to study treatment; one decided not \nto continue and the other due to scheduling issues. A total of 30 subjects, 26 \nfemales and 4 males (equally balanced between erenumab and Placebo groups), \nmedian age of 34 years old (range 21 to 58 years old) were randomized into the \nstudy.  Table 1  summarizes demographic characteristics of the sample. The mean BPI \ninterference and severity scores were in the mild range (2.19 and 2.95, \nrespectively).\nBaseline characteristics of study participants.\nSD: standard deviations.\nTwenty-two subjects completed the study (10 Erenumab; 12 Placebo). Fig.  1  shows \nthe participant flow in this randomized trial. Four subjects in the erenumab \ngroup withdrew from the study for the following reasons: constipation, which was \nconsidered by the PI to be definitely associated with the investigational \nproduct; concern about hypertension; family issues; and desire to donate plasma. \nOne subject in the Placebo group withdrew from the study because the treatment \nwas not improving pain. Three subjects were lost to follow-up after repeated \nattempts to contact them. There were no gender differences between treatment \ngroups and no differences in the percentage of subjects seen at each follow-up \nvisit between groups.\nParticipant flow in the randomized trial.\nThe study findings are presented for the primary outcome of pain interference \nand the secondary pain outcomes of pain severity and global change in pain in \nFig.  2 . There were no significant differences in pain outcomes between treatment \narms. In the small subset of 7 subjects (4 on erenumab, 3 on placebo) who had \nmore than mild pain ( i.e. , BPI ≥4), pain outcomes were similar. \n Table 2  summarizes findings for all study outcomes. Overall, the outcomes were \nsimilar between erenumab and placebo for all outcomes except the PHQ-4 which \nshowed that depression/anxiety symptoms were modestly worse in the erenumab \ngroup. Although between-group PHQ-4 differences were not significant at all 7 \ntimepoints, the overall effect using repeated measures analysis was significant \n( p  = 0.032).\nMean (adjusted for sex) Brief Pain Inventory (BPI) Interference \nand Severity scores and Patient Global Impression of Change (PGIC) scores with \n95% CI by treatment group.\nOutcome comparisons by treatment arm (adjusted for sex) * .\n* Higher score is worse for all scales. Difference = placebo minus \nerenumab score. BPI: Brief Pain Inventory; CI: confidence interval; PHQ: Patient \nHealth Questionnaire; SSS: Somatic Symptom Scale.\nEleven potential study-related adverse events (AEs) were reported in 6 subjects \n(5 in erenumab arm and 1 in placebo arm). These 11 AEs included irritation at \ninjection site (2 erenumab, 2 placebo), myalgia (1 erenumab,1 placebo), nausea (2 \nerenumab), constipation (1 erenumab), drowsiness (1 erenumab) and COVID-19 \nsymptoms (1 placebo).\n\nThe findings of this randomized controlled pilot study do not support the \npremise that erenumab is beneficial in reducing facial, jaw or TMD pain \nintensity/severity in individuals with pain using the diagnostic criteria \n(DC/TMD) for “myalgia”. This finding was consistent for the primary outcome \nmeasure pain interference using the Brief Pain Inventory (BPI) as well the \nsecondary pain-related measures BPI pain severity, global improvement in pain \n(PGIC), jaw function limitation (JFLS-8), somatic symptom severity (SSS-8), and \ndays of use of TMD pain-specific medication per month.\nThere are several possible reasons for the lack of benefit of erenumab compared \nto placebo for pain. Firstly, study participants had only relatively low levels \nof pain at baseline; both groups had a mean pain score <3 which, on a 0 to 10 \nnumeric rating scale, indicates mild pain [ 33 ]. This could have created a floor \neffect in our study’s ability to show a reduction in pain. It is possible that \nerenumab’s separation from placebo might differ in patients with more severe TMD \nmyalgic pain. Notably, all secondary scales had relatively low scores at baseline \nsuggesting a sample with mild overall symptoms and good jaw function. Secondly, \nthe placebo response may be particularly high in some TMD patients and, in a \nsmall sample, might have contributed to our null findings [ 34 ]. Thirdly, it may \nbe that our hypothesis was incorrect that erenumab would be beneficial in TMD \npain because of its comorbidity with migraine and potential shared pathways.\nUnexpectantly, there were worse results for erenumab compared to placebo for \ndepression and anxiety based on the Patient Health Questionnaire (PHQ-4). This \nisolated finding should be put in the context of what is known about the \npsychological effects of erenumab in previous trials for migraine as well as \npost-marketing data. Firstly, the severity of depression and anxiety symptoms was \nrelatively low. PHQ-4 scores of 3–5 are considered mild [ 25 ], and scores were \n<3 in both groups at baseline and never rose above 4.4 in the erenumab group at \nany assessment. Secondly, the between-group differences at most follow-up time \npoints were only mildly statistically significant which is important because \nmultiple secondary outcomes were tested. Thus, it is possible the single \nsecondary outcome differing between groups represents a chance finding. Thirdly, \ndata regarding the psychological effects of erenumab are inconclusive. Data from \ntrials have not found it to be a common adverse event, and there is some evidence \nfrom other studies that erenumab might even be beneficial for depression and \nanxiety [ 35 ]. In the post-marketing setting, there is a \nslight increase in the reporting of depression and anxiety with erenumab compared \nto other acute or preventive migraine treatments [ 36 ]. It should be noted, \nhowever, that findings from disproportionality analyses do not confirm causality. \nImportantly, 3 systematic reviews have not found psychological symptoms to differ \nbetween erenumab and placebo [ 37 ,  38 ,  39 ]. Thus, a large body of evidence including \nPhase III trial data coupled with extensive post-marketing surveillance do not \nindicate that erenumab has psychoactive effects.\nThe most important study limitation is the generally mild level of pain and \nother secondary outcomes at baseline which reduced the amount of improvement that \ncould be detected ( i.e. , a floor effect). Second, the primary outcome \nwas assessed with one of the most commonly recommended general pain measures. \nMeasuring TMD-specific pain may have also been informative.\n\nIn conclusion, erenumab compared to placebo was not effective in reducing pain \nin a small pilot trial of patients with TMD with low pain intensity. Whether the \nmedication might be beneficial in patients with more severe pain requires further \nresearch. For now, the use of erenumab in treating TMD-related pain in the \nabsence of comorbid chronic migraine cannot be recommended.","source_license":"CC-BY-4.0","license_restricted":false}