Real-life data on the efficacy and safety of erenumab in the Abruzzo region, central Italy

Our study included patients aged 18 to 65 years consecutively treated with erenumab in the Headache Centers of Avezzano, L’Aquila, Sulmona, Teramo, Chieti, Lanciano, and Vasto, all located in the Abruzzo region, central Italy, from January to December 2019. The Abruzzo region hosts a population of 1,311,580 inhabitants according to the most recent data [23]. The Headache Centers of Avezzano-L’Aquila, Teramo, and Chieti offer a level 3 care, while the other offer a level 2 care according to the European Headache Federation/ Lifting the Burden (EHF/LTB) proposed classification [24]. The study was approved by the Internal review Board of the University of L’Aquila with the number 44/2019. All patients signed an informed consent.

The Avezzano-L’Aquila center started treating patients with erenumab in January 2019 and therefore recruited more patients than the remaining centers, which started adopting erenumab treatment from April 2019 onwards. In the absence of established reimbursement criteria from the Italian Agency for Drug administration (AIFA), erenumab was provided to patients from the producing company upon reasonable request from the Headache Centers. The drug was provided for patients with migraine with or without aura diagnosed by expert physicians according to the International Classification of Headache Disorders (ICHD) criteria [2]. All the study patients had to have > 4 monthly migraine days and failure of ≥2 prior preventive treatments for migraine, according to the eligibility criteria established by the European Headache Federation [14] and the American Headache Society [25]. Due to the limited availability of the drug, the study centers focused on the most difficult-to-treat patients, i.e. those with a long history of disabling migraine and/or previous treatment failure. Following the exclusion criteria of the available trials [17‐19, 26], we excluded from treatment patients with major medical or psychiatric illnesses.

Erenumab was administered during in-person visits in a monthly subcutaneous dose of 70 mg, with the option of switching to 140 mg monthly (i.e., two 70 mg doses) in case of a < 30% decrease in monthly migraine days (MMDs) compared with baseline; the dose escalation could be done since Dose 2 or even at treatment start in patients with several prior preventive treatment failures, according to the results of the LIBERTY trial [19]. In all patients, erenumab treatment was intended to be continued at least until Dose 6, but we acknowledged the possibility of early withdrawal because of severe adverse events, lack of compliance, or ineffectiveness (< 30% reduction in MMDs and/or lack of satisfaction with treatment).

Patients were allowed to start or continue concurrent oral preventive treatments for migraine at physicians’ discretion while the concurrent administration of botulinum toxin A for CM was not allowed. Withdrawing or adding concurrent oral preventive treatments for migraine was allowed over the course of treatment according to physicians’ judgement. Patients with CM and medication overuse were not detoxified prior to erenumab treatment, according to current recommendations [14].

For each included patient, we recorded sex, age, current treatments, and comorbidities; we also recorded age at migraine onset, age at CM onset in patients with CM, migraine frequency and intensity, associated symptoms, acute and preventive treatments as reported in the patients’ headache diaries. In each study Center, patients were asked to differentiate between migraine days, with attacks fulfilling the ICHD-3 criteria for migraine [2], and non-migraine headache days, which were not considered in the present study. We assessed attack severity by the 0–10 Numerical Rating Scale (NRS), disability by the Migraine Impact and Disability Assessment Scale (MIDAS), impact by the Headache Impact Test, 6th edition (HIT-6), and allodynia by the Allodynia Symptom Checklist-12 (ASC-12); we also recorded the scores of scales for psychiatric symptoms, including the Beck Depression Inventory (BDI) and Generalized Anxiety Disorder (GAD-7) Questionnaire. Migraine characteristics and the scores of HIT-6, NRS, and ASC-12 were assessed monthly, while the scores of MIDAS, BDI, and GAD-7 were assessed quarterly. Data were collected with a clinical interview and then reported on a standardized form with pre-determined answers which was the same for all participating centers. All the recorded data were stored in an anonymized computerized database.

Baseline was defined as the monthly mean of the 3 months preceding erenumab treatment. Patients reporting a ≥ 50% reduction of MMDs compared with baseline to at least one dose were defined as ‘anytime responders’. The co-primary efficacy outcomes of our analyses included the decrease in monthly migraine days (MMDs), days of analgesic and triptan use, and the proportion of anytime 50% responders. Secondary efficacy outcomes included the proportions of dose-specific 50%, 75%, and 100% responders (defined according to the percent decrease in MMDs after each dose) and the decrease in mean MIDAS, HIT-6, BDI, GAD-7, and NRS scores from baseline to the month following Dose 6. The safety outcomes included adverse events, and especially serious adverse events, i.e. those leading to hospitalization, death, or treatment withdrawal. All outcome variables were assessed through headache diaries. The proportions of responders were calculated over the total of patients with complete follow-up, irrespective of treatment discontinuation, while the decrease in MMDs and days of analgesic and triptan use was calculated over the total of patients who received all the six doses.

Among 132 patients who started erenumab treatment during the observation period, 43 (32.6%) did not yet complete a 6-month follow-up. Among the remaining 89 patients, 13 (14.6%) discontinued treatment for ineffectiveness (12 patients) or adverse events (1 patient), while the remaining 76 (85.4%) continued treatment during all the study period.

Patient characteristics are reported in Table 1. Most patients (84; 94.4%) had CM, while 5 (5.6%) had EM; 64 patients (71.9%) had medication overuse. All patients had multiple preventive treatment failures (Table 1). In detail, 53 patients (59.6%) reported failures of antiepileptics, 50 (56.2%) of antidepressants, 42 (47.2%) of botulinum toxin A, 35 (39.3%) of calcium antagonists, 22 (24.7%) of beta blockers, and 6 (6.7%) of other preventive treatments. Failures were reportedly due to ineffectiveness in 57 patients (64.1%), adverse events in 6 (6.7%), and both in 26 (29.2%).

In 37 (41.6%) patients, erenumab was added to an ongoing oral preventative at baseline. During the treatment period, 10 patients (11.2%) withdrew the prior oral treatment, while 8 (9.0%) started a new one. Forty-three patients (48.3%) escalated the erenumab dose from 70 to 140 mg monthly across the study period. Figure 1 reports the proportions of patients on treatment with 70 mg or 140 mg across each dose. Table 2 reports the comparison between patients escalating and not escalating erenumab dose. Patients escalating erenumab dose had a higher median number of MMDs at baseline compared with those not escalating the dose (25 vs 17; P = 0.002).

Over the study period, 64 patients (71.9%) were classified as anytime responders, i.e. had a ≥ 50% reduction of MMDs after at least one dose. Rates of dose-specific responders increased over time, ranging from 31.5% to 57.3% for ≥50% responders, from 15.8% to 46.0% for ≥75% responders, and from 1.1% to 11.2% for 100% responders (Fig. 2).

In the 76 patients who completed a 6-dose treatment, we observed a decrease in the median number of MMDs from 19 to 4 (P < 0.001). The mean number of analgesic use days decreased from 10 to 2 (P < 0.001), while the mean number of triptan use days decreased from 5 to 1 (P < 0.001); significant decreases of MMDs and days of analgesic or triptan use were observed since the first month of treatment (Fig. 3a). We also found that erenumab treatment improved intensity of attacks, disability, impact, allodynia as well as depressive and anxiety symptoms over the 6-month treatment period (Fig. 3b).

Forty-six (71.9%) of the 64 patients with medication overuse withdrew the overused medication during the follow-up (Fig. 4).

Among the 44 patients who had failed treatment with botulinum toxin A, 31 (70.5%) were anytime responders; the proportion of dose-specific ≥50% responders increased from 25.0% after Dose 1 to 56.8% after Dose 6 (Fig. 5).

Twenty-eight patients (31.5%) had ≥50% response after Dose 1; 9 patients (10.1%) not responding to Dose 1 had ≥50% response after Dose 2, and 16 (18.0%) not responding to the first two doses had ≥50% response after Dose 3; 12 (13.5%) additional patients had ≥50% response after Dose 4 to Dose 6.

Among the 53 patients having a ≥ 50% response to at least one of the first 3 doses, 36 (67.9%) maintained the response to all subsequent doses, 5 (9.4%) to two doses, and 3 (5.7%) to one dose.

Compared with the 25 non-responders, the 64 anytime responders had a lower median number of MMDs at baseline (18 vs 26.5; P = 0.026) and a lower median number of monthly analgesic use days (8 vs 20; P = 0.008); the remaining baseline characteristics did not differ between responders and non-responders (Table 3).

During the 6-month follow-up, over the 76 patients completing the 6-dose follow-up and the 13 patients who discontinued the treatment, we recorded 24 adverse events in 20 (22.5%) patients. The most common adverse event was constipation, which was observed in 12 (13.5%) patients (Table 4). One adverse event led to treatment discontinuation, namely allergic reaction (Table 4). The event resolved after treatment discontinuation.