Short and Mid-Term Predictors of Response to OnabotulinumtoxinA

Real-Life Experience Observational Study

Alicia Alpuente, MD; Victor José Gallardo, MSc; Marta Torres-Ferrús, MD, PhD; José Álvarez-Sabin, MD, PhD; Patricia Pozo-Rosich, MD, PhD


Headache. 2020;60(4):677-685. 

In This Article

Abstract and Introduction


Objective: To identify clinical predictors of excellent response to OnabotulinumtoxinA in patients with chronic migraine (CM) at 6 and 12 months of follow-up.

Background: Clinical predictors of response to OnabotulinumtoxinA are scarce and have not been clearly reproduced and analyzed in detail. So far, predictors of response to OnabotulinumtoxinA assess response in general or good response, but not an excellent response.

Methods: Cohort study of patients attended in a specialized Headache Clinic in treatment with OnabotulinumtoxinA were classified according to their improvement in frequency: no-response (<25%) and excellent response (≥75%). A comparative analysis was carried out at 6 and 12 months identifying clinical predictors of excellent response to OnabotulinumtoxinA at each timepoint.

Results: Data were collected from 221 patients. After 6 and also 12 months, we observed a statistically significant mean reduction in frequency and analgesic intake. At month 6, independent variables associated with excellent response (OR[95%CI]) were daily headache frequency (0.32[0.14–0.74]; P = .005), medication overuse (MO) (2.28[1.19–4.37]; P = .013), and a higher ratio of migraine days/month (MDM) (1.20[1.10–1.45]; P = .018) at baseline. At month 12, independent predictors of excellent response were patients with less than 30 years of migraine evolution (0.43[0.23–0.82]; P = .011), presence of anxiety (0.44[0.23–0.85]; P = .018), and aura (0.48[0.25–0.92]; P = .037). Excellent responders showed a higher improvement rate in pain intensity at 6 and 12 months.

Conclusions: Patients with daily frequency and MO show a clinical improvement in short-term. Patients with comorbidities who start treatment earlier in the course of the disease need a longer duration of treatment. The profile of response to treatment with OnabotulinumtoxinA determines its minimum treatment duration and the timepoint of a meaningful response.


Chronic migraine (CM) is a disabling neurological disease which is estimated to affect 2% of the population.[1,2] Current preventative treatments in migraine are focused on reducing the frequency of headache. Nevertheless, they also aim to reduce other factors such as intensity of pain, acute medication intake, or disability with a final goal of improving the patients' quality of life (QoL).

Since its approval, treatment with OnabotulinumtoxinA in CM is widely used in the United States and most European countries. Real-life clinical studies[3–13] have demonstrated its efficacy during longer time periods than the year analyzed in the PREEMPT clinical trials[14,15] providing clarity to practical issues such as safety, maintenance, or withdrawal of treatment.

Seeking clinical biomarkers that may contribute to predict how our patients will evolve is an important goal for physicians since it helps to manage both patient and physician's expectations, better understand the mechanisms of CM, and optimize the therapeutic resources.[16] So far, most predictors of response to OnabotulinumtoxinA are not only based on clinical features,[17–21] but also in molecular biomarkers such as calcitonin gene-related peptide (CGRP).[22–24] However, results are limited or controversial, and are based on the presence of a good response (>50%).

In order to be able to clinically deep phenotype our patients and to better understand migraine and its process of chronification, we decided to identify the clinical predictors of excellent and no-response, being these opposite ends of the spectrum of response to preventive treatment with OnabotulinumtoxinA. So, we did a secondary analysis of the study done on the evaluation of the efficacy and late gain in CM in high-frequency episodic migraine (EM) with OnabotulinumtoxinA.[25]