Migraine Patients Treated With Erenumab and the Risk of Hypertension: A Registry-Based Real-World Study.

Christian Lampl: conceptualisation, writing – original draft, methodology, visualisation, writing – review and editing. Reem Suliman: conceptualisation, investigation, methodology, data curation, writing – original draft, methodology, visualisation, writing – review and editing. Viktoria Tischler‐Strasser: writing – original draft, writing – review and editing. Antoinette MaassenVanDenBrink: conceptualisation, writing – original draft, methodology, visualisation, writing – review and editing. Taoufik Alsaadi: conceptualisation, investigation, methodology, data curation, writing – original draft, methodology, visualisation, writing – review and editing, formal analysis, supervision.

The authors have nothing to report.

Participants in this retrospective, exploratory RWD study were adults aged 18 years or older with a confirmed diagnosis of migraine according to the ICHD‐3 criteria [ 18 ]. Data were extracted from the electronic health records (EHRs) of the American Centre for Psychiatry and Neurology in Abu Dhabi, UAE, a secondary headache clinic. From 2020 to 2024, 347 patients who received erenumab were identified and screened for inclusion. We included only those who completed at least 12 months of treatment. Inclusion was restricted to individuals with complete clinical records and regular follow‐up, including at least one visit every 3 months, and with data available to assess treatment outcomes. Patients were excluded if they had previously received anti‐CGRP mAbs or gepants for migraine prevention before starting the current therapy, were on any concomitant baseline migraine preventive treatment, did not complete the 12‐month follow‐up, or had medical records lacking sufficient data for complete analysis. The sample size was determined from the available data. Baseline characteristics, including sex, age, headache diagnosis and baseline headache and migraine days, were summarised using means, standard deviations (SDs), frequencies and proportions. A Chi‐square test of independence was conducted to assess the possible relationship between weight and BP elevation by comparing patients with elevated BP in the obese and overweight groups with those with normal BP. To evaluate BP trends and fluctuations, linear regression was used to determine the slopes of SBP and DBP throughout the treatment period. SBP and DBP were compared between the baseline visit before starting erenumab and the follow‐up visits at 3, 6, 9, 12 and 24 months while on treatment. In addition, paired‐sample t ‐tests were run for both systolic and diastolic blood pressures, comparing values obtained at 12 and 24 months with those at baseline. Patients treated with erenumab initially received 70 mg subcutaneously every 4 weeks. Based on shared decision‐making between patients and physicians, most agreed to increase the dose to 140 mg due to a lack of efficacy after at least 3 months of treatment. This aligns with the anti‐CGRP‐mAbs practice protocol at the American Centre for Psychiatry and Neurology (ACPN). The clinic's electronic patient records system documented BP measurements (in mmHg). Patients attended consultations at the ACPN at baseline and at visit 1 (V1), and at least one visit every 3 months for 12–24 months of follow‐up. Blood pressure measurements were obtained during routine clinical visits using automated, validated devices in the seated position. For measurements, we followed our standardised clinical routine: blood pressure was measured by trained nursing staff using an automated device with the patient seated after a minimum rest period of 5 min. The cuff size was adjusted according to the patient's weight. If an elevated BP reading was detected, the measurement was repeated. In such cases, confirmation was performed using both automated and manual devices to ensure accuracy. Any detected elevated BP prompted clinical action. However, as this was a retrospective real‐world study, fixed time‐of‐day measurements were not protocolised. Any sustained or clinically relevant increase would have prompted repeat measurements and clinical action. Data were collected at baseline and at each subsequent follow‐up visit, with a maximum follow‐up period of 24 months. Pre‐diagnosed HT was not an exclusion criterion for starting treatment with erenumab, provided there was no change in dose or medication in their antihypertensive treatment and BP was regularly controlled.

A total of 125 participants were eligible and included in the study, of whom 54 completed 24 months of treatment with erenumab. The cohort comprised 50 individuals with chronic migraine (40.0%) and 75 with episodic migraine (60.0%). The patient population was predominantly female, comprising 80.8% of the total ( n  = 101), while males accounted for 19.2% ( n  = 24). Ethnically, most patients were of Emirati origin, representing 82.4% ( n  = 103), followed by other Arab ethnicities at 10.4% ( n  = 13). Patients of white, Asian and South American origins made up 5.6% ( n  = 7), 0.8% ( n  = 1) and 0.8% ( n  = 1), respectively (Table  1 ). A positive family history of migraine was documented in only 8.0% ( n  = 10) of patients, with 92.0% ( n  = 115) reporting no familial history. Comorbidities assessed before initiating erenumab treatment were common across the cohort, particularly psychiatric, neurological and metabolic conditions (Table  2 ). Insomnia was the most prevalent comorbidity, affecting 29.6% of patients ( n  = 37), followed by major depressive disorder (19.2%) and generalised anxiety disorder (7.2%). Overall, psychiatric disorders were present in 62.4% of the cohort ( n  = 78). Seventeen point 3% ( n  = 19) had neurological disorders, including epilepsy and fibromyalgia/chronic pain syndrome. Metabolic and endocrine disorders were observed in 16.8% ( n  = 21), specifically type 2 diabetes (6.4%) and hypothyroidism (4.8%). Body Mass Index (BMI) values were analysed to determine their distribution across weight categories. Based on BMI classifications, 39 patients (31.2%) were classified as obese, 45 (36%) as overweight and two patients (1.6%) as underweight. The mean BMI across the sample was 28.20. Patient characteristics. Patient comorbidities. Abbreviations: BPD, borderline personality disorder; GAD, generalised anxiety disorder; GERD, gastroesophageal reflux disease; IBS, irritable bowel syndrome; MDD, major depressive disorder; OCD, obsessive compulsive disorder; PTSD, post traumatic stress disorder. For the interpretation of our data, we used the guidelines of the European Society of Cardiology and the European Society of Hypertension. Baseline BP measurements were available for all 125 patients before initiating erenumab. The overall mean SBP over 12 months was 120 mmHg, and the mean DBP was 79.5 mmHg, indicating that most patients had normal BP at baseline. When stratified by age group, both mean SBP and DBP remained stable across all cohorts (Figure  2 ). Regarding variability, the standard deviation (SD) across the 12‐month treatment period was low, with a coefficient of variation of 1.4%. SBP SD ranged from 10.1 to 13.1. In contrast, the DBP SD ranged from 8.4 to 10.6, indicating low variance and minimal fluctuation in BP over time (Figure  3 ). Throughout the 12‐month treatment period, BP measurements remained stable and within normal physiological ranges, suggesting no significant hypertensive effect. Across multiple visits, SBP ranged from 118 to 121 mmHg, and DBP remained between 78 and 80 mmHg. Also applying the paired sample t ‐test, no significant differences were found between the baseline BP and that obtained at 12 months for either SBP ( p : 0.206) or DBP ( p : 0.979). Average systolic and diastolic blood pressure across different age groups for participants in the 12‐month (left panel) and 24‐month (right panel) follow‐up groups. Values represent averages recorded over the course of follow‐up visits during each respective time period. Error bars represent ±1 standard deviation (SD). A plot showing the 12‐dose group, where the mean values of systolic and diastolic blood pressure across doses. The green lines represent the mean of systolic blood pressure, while the red lines represent the mean of diastolic blood pressure. The error bars in the plot illustrate the spread of values around the mean (±SD). Erenumab treatment over 24 months in 54 patients demonstrated stable BP, with no significant hypertensive effects observed. SD across the 24‐month treatment period showed low variability around the mean, with a coefficient of variation of 2.25%. For SBP, the SD ranged from 8 to 21 mmHg, indicating relatively low variability and consistent SBP control during treatment. Similarly, the DBP standard deviation ranged from 8 to 11 mmHg, indicating minimal fluctuation (Figure  4 ). As for the 12‐month period, the paired sample t ‐test revealed no significant differences between the baseline BP and that obtained at 24 months for either SBP ( p : 0.351) or DBP ( p : 0.336). A plot showing the 24‐dose group, where the mean values of systolic and diastolic blood pressure across doses. The green lines represent the mean of systolic blood pressure, while the red lines represent the mean of diastolic blood pressure The error bars in the plot illustrate the spread of values around the mean (±SD). Of the 125 patients, seven are of particular interest because their BP reached clinically meaningful levels (SBP ≥ 140 mmHg and DBP ≥ 90 mmHg) after erenumab use (Table  3 ). The baseline SBP of these participants ranged from 111 to 140 mmHg, while the baseline DBP ranged from 66 to 98 mmHg. Analysis showed that all seven patients experienced an increase in SBP, ranging from 4 to 30 mmHg. DBP also increased in all but one patient, with changes from 0 to +29 mmHg. Among these seven, three were overweight and four were obese according to their BMI. One of the seven individuals (pat 23) was pre‐diagnosed with HT. For the patient with pre‐existing HT, there was an increase in both systolic (+17 mmHg) and diastolic (+10 mmHg) BP. Vascular and/or metabolic comorbidities were found in two individuals (patients 9 and 23). 7 of 125 patients exhibited an increase in blood pressure throughout their 12‐month treatment period. A Chi‐square test of independence was conducted to examine the relationship between weight and BP elevation, comparing patients with elevated BP in the obese and overweight groups with those with normal BP. The findings indicate no statistically significant association between the variables and BP, χ 2 (1, N  = 125) = 3.62, p  = 0.057. To evaluate BP trends and fluctuations, linear regression was performed for each of the 7 patients to determine the slopes of SBP and DBP throughout their treatment period. SD was also calculated to assess temporal variability. The systolic slopes ranged from −0.60 to +2.72, reflecting diverse BP responses among the patients. Patients 9, 87 and 83 showed positive systolic slopes, indicating a consistent rise in SBP. The other four exhibited flat or negative slopes, suggesting stable or slightly decreasing trends. For DBP, slopes varied from −0.57 to +2.05, with five patients displaying positive diastolic slopes, indicating a moderate upward trend, while the remaining two showed minimal or negative diastolic changes. SD in SBP and DBP ranged from 7.15 to 13.06 mmHg and 4.88 to 11.08 mmHg, respectively. Patient 9 exhibited the greatest SBP variability (SD = 13.06 mmHg), whereas patient 93 showed the greatest DBP variability (SD = 11.08 mmHg). To assess whether BMI influenced changes in BP after treatment, patients were categorised into weight groups (Normal, Obese, Overweight and Underweight). Obese patients experienced an average increase of +1.70 mmHg in SBP and +3.38 mmHg in DBP. The Overweight group showed a slight decrease in SBP (−0.78 mmHg) and a slight increase in DBP (+0.98 mmHg). In September 2025, we conducted a follow‐up visit with the seven patients, during which an increase in BP was observed while they were using erenumab. None of them was subsequently diagnosed with HT, as the last BP measurement remained within normal ranges, except for patient 23, who had a pre‐existing diagnosis of HT (last measurement: 158/98). Four of the seven patients continued treatment with erenumab, one switched to eptinezumab (with no change in BP), and two stopped taking anti‐CGRP‐mAbs. Patients were also classified according to whether they had pre‐existing HT. In the group with pre‐existing HT ( n  = 5), the mean changes were +0.80 mmHg for SBP and −1.20 mmHg for DBP. For the remaining 120 patients without a diagnosis of HT, the mean changes were −1.99 mmHg for DBP and +0.18 mmHg for SBP.

The findings of this RWD study suggest that, in individuals with migraine, treatment with erenumab at doses of 70 and 140 mg was associated with BP remaining within normal ranges across visits, doses and age groups. However, in 7 out of 125 individuals, we observed increases in BP at multiple time points. This may indicate that further evaluation could lead to a diagnosis of HT. Interestingly, all seven individuals were overweight or obese. Although our study was not statistically powered to perform subgroup analyses, we conducted a post hoc chi‐square test to examine whether overweight patients were overrepresented among those who experienced an increase in BP. While this test did not reach statistical significance, the borderline p ‐value (0.057) warrants further investigation in future studies. Our findings align with most current publications on this topic. In a recent retrospective analysis of data from 259 patients with migraine, prophylactic treatment with three different monoclonal antibodies did not significantly raise arterial BP during follow‐up. The analyses also included patients who switched to a second, third, or fourth antibody, with no impact on BP [ 19 ]. An Italian group recently published results from a multicentre, prospective, observational study focusing solely on patients aged 60 years or older who initiated treatment with anti‐CGRP mAbs for migraine prevention [ 20 ]. In this population, which has a higher risk of HT and cardiovascular events, the estimated effects on SBP and DBP were negligible at all time points. These findings align with data from large‐scale randomised clinical trials and early real‐world studies, which have shown no evidence of cardiovascular risk in migraine patients treated with erenumab [ 21 , 22 , 23 , 24 ]. In another retrospective cohort study analysing data from 69,589 patients in the Veterans Health Administration, no association was found between initiating anti‐CGRP‐mAbs and an increase in BP or a diagnosis of HT among non‐hypertensive patients [ 25 ]. However, among those with pre‐existing HT, a slight but consistent increase (3.7%) in the number of antihypertensive medications was observed after starting anti‐CGRP(r)‐mAbs treatment, suggesting a possible worsening of HT. Key limitations of this study include the absence of a comparison between follow‐up and baseline BP readings. Additionally, the wide variety of anti‐CGRP‐mAbs treatments was analysed collectively, preventing a direct assessment of how initiation of anti‐CGRP‐mAbs affects BP. A recent review of the impact of anti‐CGRP‐mAbs on BP in individuals with migraine [ 26 ] identified four eligible studies [ 17 , 27 , 28 , 29 ] that met predefined inclusion criteria (randomised controlled trials and observational cohort or case–control studies) and compared outcomes with a control group. They also used the Newcastle‐Ottawa Scale for risk of bias assessment and customised this evaluation by adding fields not explicitly covered in the NOS scale. These additional criteria included ‘sufficient sample size’, ‘distinction between anti‐CGRP treatment’, ‘number of dropouts specified’, ‘stability on anti‐hypertensive medication’ and ‘company funding’. This approach presents an intriguing method for reducing bias. Indeed, they found a high risk of bias, except for De Vries Lentsch et al. [ 17 ]. In that study, a consistent increase was observed in mean SBP and DBP at 3, 6, 9 and 12 months compared with baseline for patients on erenumab. The maximum estimated increase in mean SBP for erenumab was 9.1 mmHg (95% CI 6.2–12.0, p  < 0.001), and for DBP, it was 6.3 mmHg (95% CI 4.4–8.3, p  < 0.001). However, none of their measurements exceeded the SBP values of 130 mmHg or the DBP values of 90 mmHg, which are the key criteria for HT. The authors further concluded that blocking CGRP may be potentially problematic only for patients already at risk of developing HT. Kudrow et al. [ 27 ] found no difference between the placebo group ( n  = 1032) and the erenumab treatment groups (70 mg, n  = 885; 140 mg, n  = 504) in the proportion of patients with clinical worsening of BP category from baseline to months 1–3 (14% [143/1032] placebo vs. 13% [114/885] and 14% [71/504] for erenumab 70 and 140 mg, respectively), regardless of baseline BP category. Similar conclusions were drawn from pooled results with eptinezumab [ 29 ] and galcanezumab [ 28 ]. As previously mentioned, only the Leiden group's results [ 17 ] reported a modest effect on mean BP and did not identify a significant risk of developing HT. In the control group (individuals with migraines with a similar distribution of sex, age and migraine diagnosis, and without any migraine prophylactic treatment or other medication that could influence their BP), the mean SBP from the first measurement was 121.3 ± 14.9 mmHg, and the mean DBP was 77.0 ± 9.9 mmHg. In comparison, the mean follow‐up SBP was 120.9 ± 14.8 mmHg, and the mean follow‐up DBP was 77.7 ± 8.7 mmHg. In the erenumab‐treated group, the mean SBP from the first measurement was 118.8 ± 13.8 mmHg, and the mean DBP was 76.2 ± 8.4 mmHg, indicating slightly lower BP than in the control group. Unfortunately, the authors did not provide data on the mean follow‐up for the erenumab‐treated group. CGRP may be involved in the regulation of blood pressure and thus in the pathophysiology of hypertension; indeed, stable CGRP analogues have been suggested as potential antihypertensive treatments [ 30 ]. However, this does not imply that blocking CGRP actions will induce hypertension. Indeed, studies in αCGRP‐knockout mice suggest that the sole absence of αCGRP is insufficient to induce hypertension, and that other factors, such as an activated renin‐angiotensin system, are required for its development [ 31 ]. An ongoing role of α‐calcitonin gene‐related peptide as part of a protective network against hypertension, vascular hypertrophy and oxidative stress [ 31 ]. Therefore, blocking CGRP might not be harmful in normotensive patients. However, in hypertensive patients, particularly those not adequately controlled by medication, the loss of a compensatory mechanism such as CGRP could increase SBP. We must recognize that the retrospective design and the absence of a control group are limitations of our study. Furthermore, changes in adherence to antihypertensive drugs and lifestyle factors during follow‐up cannot be excluded and could have acted as possible confounders . However, as explicitly criticised in the systematic review by van der Arend et al. [ 32 ], having multiple local principal investigators can lead to greater variability in BP labelling; therefore, our trial was conducted at a single centre with a single local principal investigator. Additionally, our study was not industry‐driven, did not overlook potential side effects, and imposed no administrative burden in identifying and reporting adverse events. Our EHR‐generated RWD provides access to large, clinically relevant patient populations, including individuals of diverse ethnicities and those who may be excluded from RCTs. Moreover, our strict inclusion and exclusion criteria enabled us to assess the risk of HT in patients treated with erenumab. Unlike other studies in the literature, which documented BP measurements every 3 months, we aimed to minimise bias by excluding participants who had previously used gepants or anti‐CGRP mAbs. The use of RWD can expand understanding of treatment effects across patient groups that are often underrepresented in clinical trials. Furthermore, RWD may be more suitable for constructing external controls when clinical trial experience with a disease or subtype is limited. In conclusion, none of the cited studies demonstrate that anti‐CGRP‐mAb treatment for migraine prevention causes HT. However, in one study [ 17 ], an increase in SBP of 5.8 mmHg was observed at T1 in the erenumab group, with a similar rise of 5.8 mmHg at T3 for DBP. Pharmacotherapy is generally used when patients have elevated BP, as seen in HT. This approach is typically reflected in clinical guidelines, which set the BP threshold at ≥ 140/90 mmHg to warrant pharmacological intervention to lower BP. Baseline BP, as a sole risk factor, does not accurately predict CVD risk. Risk‐based approaches for determining treatment strategies require further investigation. Some multivariable models for estimating an individual's risk incorporate information on personal characteristics, lifestyle and clinical factors—including age, sex, smoking status, BP, lipids and body size—to calculate the likelihood of developing a cardiovascular event. Our data suggest that although some patients do experience increases in blood pressure (this should be monitored when this medication is initiated), the risk may be higher in those with pre‐existing risk factors. However, the incidence of a significant increase in BP, leading to de novo hypertension, seems low. Additional research is necessary to fully understand the potential hypertensive risk associated with anti‐CGRP‐mAb treatment. The current understanding of CGRP's role in the pathophysiology of HT and its interaction with anti‐CGRP‐mAb therapies remains under development.

Migraine ranks among the top three causes of disability‐adjusted life years (DALYs) and is the second leading cause among adults aged 20–59 years. In this age group, the proportion of the overall neurological disease burden attributable to migraine is 21.8% [ 1 ]. Besides the ictal burden, the interictal burden also forms a significant part of the Total Migraine Burden [ 2 , 3 ] and is strongly associated with seeking care for migraine [ 4 ]. Monoclonal antibodies targeting the neuropeptide calcitonin gene‐related peptide (CGRP) are recommended as first‐line preventive treatments for migraine [ 5 , 6 ]. They may also reduce the overall pain burden [ 7 ] and the interictal burden [ 8 ]. Since CGRP is involved in the pathophysiology of migraine, blocking this peptide is an effective treatment strategy [ 9 , 10 ]. However, CGRP also plays an essential role in cardiovascular function under ischaemic or hypertensive (HT) conditions [ 11 , 12 ]. In fact, a stable CGRP analogue has been proposed as a potential therapeutic for cardiovascular diseases, including HT [ 13 ]. Recently, debates have emerged regarding the safety of anti‐CGRP monoclonal antibodies (anti‐CGRP‐mAbs) in individuals with HT and whether patients with migraine are at risk of developing HT when treated with these mAbs [ 14 ]. HT, as defined by the European Society of Cardiology (ESC) and the European Society of Hypertension (ESH), implies an office systolic blood pressure (SBP) of ≥ 140 mmHg and/or a diastolic blood pressure (DBP) of ≥ 90 mmHg. This classification is supported by evidence from multiple randomised controlled trials (RCTs) showing that treating patients with these BP values is beneficial [ 15 ]. A significant change in the 2017 ACC/AHA guidelines [ 16 ] was the adoption of a lower threshold (130/80 mmHg) for defining HT. SBP of 130–139 mmHg or DBP of 80–89 mmHg characterises stage 1 HT, while any values greater than 140/90 mmHg indicate stage 2 hypertension (Figure  1 ). Thresholds for hypertension definition, treatment initiation and treatment targets following different classifications. The US flag indicates guidelines by the American College of Cardiology and the American Heart Association (2017), the European flag indicates guidelines by the European Society of Cardiology and the European Society of Hypertension (2018), and the lower line represents guidelines by the International Society of Hypertension (2020). For references, please refer to [ 16 ]. Redrawn using Biorender after [ 16 ]. Erenumab is an anti‐CGRP (receptor) mAb and was approved by the FDA in 2018 as the first drug of its kind. Although preclinical models and clinical trials in patients treated with erenumab showed no evidence of HT, recent post‐marketing studies have suggested a link between erenumab and HT [ 17 ]. However, it remains unclear to what extent these results, obtained in a tertiary headache centre in the Netherlands, can be generalised to populations from other clinics and different ethnicities. This real‐world data (RWD) study aims to explore and assess the risk of HT in patients treated with erenumab attending a headache clinic in the United Arab Emirates.

The authors declare no conflicts of interest.