Introduction:
The development of new preventive treatments represents a new era in the treatment of
migraine. Different drugs have shown a reduction in the number of headache days, average
intensity of pain and the need for symptomatic treatment in randomized, placebo-controlled
trials; nonetheless, there are unanswered questions from clinical trials which are of a
particular interest to routine clinical practice.
The duration of preventive treatments is not fully established, in the different clinical
trials the duration of treatment ranged between 3, 6 or 12 months 3-12 . Additionally, it is
known that the benefit of a preventive drugs may persist once interrupted, while in other
cases, patients may return to the previous situation in a relatively short period of time. In
the case of other preventive treatments, the therapeutic effect seems to be maintained for a
longer period after the discontinuation of the treatment.
Preventive drugs targeting the Calcitonin gene-related peptide (CGRP) or its receptor have
been approved. In Spain three of them (erenumab, galcanezumab and fremanezumab) are
subsidized for the treatment of episodic migraine or chronic migraine in patients with at
least 8 days of migraine per month and who have previously failed at least 3 preventive
treatments. Still, the long-term duration of therapeutic effect once these treatments are
suspended has not been clarified yet.
As the duration of the effect after discontinuing treatment and the optimal duration of
treatment are not fully known, the present study aims to evaluate patients with migraine who
have been treated with monoclonal antibodies. Additionally, the needed time to return to the
situation prior to this treatment will be studied. The factors that may be associated with
the duration of the benefit after the treatment discontinuation will be explored by using
survival analysis methods.
Material and methods Observational study with a prospective cohort design. Patients with
diagnosis of episodic or chronic migraine who, under the criteria of their neurologist and
according to the clinical practice guidelines and standard of care, receive treatment with
monoclonal antibodies against CGRP (galcanezumab, fremanezumab or eptinezumab) or else its
receptor (erenumab) will be included.
Concomitant treatment:
Taking another drug with preventive purposes will be allowed as long as it is included in the
national clinical practice guidelines as a preventive treatment and the drug is in a stable
dose for at least 6 weeks prior to entry.
A non-probabilistic sampling will be carried out for opportunism. Patients receiving
monoclonal antibody treatment in the participating centers will be included.
Intervention:
The intervention will consist of determining the time elapsed until the patients return to
their pre-treatment situation once it has been suspended, at the discretion of their main
neurologist or that they require a preventive treatment once again despite not having reached
the pre-treatment situation. For this, the patient will be provided with an in-paper or
electronic calendar in which the number of headache days, that will be completed daily. The
event, also known as the month of return to the previous situation, will be defined as that
month in which the number of days of pain is equal or with a difference of ± 1 day to the
month prior to the start of treatment. In the event that the patient starts any preventive
treatment for his migraine according to the national clinical practice guidelines, without
having reached a headache frequency like the one he had before starting treatment with
monoclonal antibody, it will also be classified as the event of interest.
To determine the factors that could be associated with a long-term effect, a series of
variables will be assessed. Including demographic variables: sex, age, type of migraine
(episodic or chronic), duration of migraine disease (in years) and duration of chronic
migraine (in months). Clinical variables evaluated at the baseline level will also be
analyzed: such as the number of headache days, number of migraine days, number of days of
symptomatic medication, number of days of triptans, average intensity of headache on an
analog verbal scale 0- 10 (0: no pain, 10: worst possible pain), MIDAS score at baseline,
baseline HIT-6 score, predominance of holocranial or hemicranial, presence of periocular
pain, predominant quality of pain (throbbing, pressing, stabbing, electrical) and presence of
allodynia (as observed on examination or reported in the anamnesis). The number of preventive
drugs previously used will be considered, understanding these as those drugs with a possible
prophylactic effect for migraine, present in the local clinical practice guidelines, taken
for long enough at the discretion of the responsible neurologist and at a sufficient dose or
that has been suspended due to adverse effects. The dose of the monoclonal antibody used and
its treatment duration in months will be evaluated.
Prior to the discontinuation of mABs, the monthly number of days with headache, migraine,
symptomatic medication, triptans use, average intensity of episodes and total score of MIDAS
and HIT-6 scale per month will be determined, by comparing the results obtained in the month
prior to the mAbs start with the results observed in the month prior to discontinuation of
treatment. It will also be described whether any adverse effect has occurred during the
treatment at the discretion of the investigator.
Sample size calculation:
There is no formal sample size estimation, and all the available patients will be included.
Schedule:
The duration of the study will be of at least the next 12 months after stopping preventive
treatment.
Confidentiality / Ethical aspects The study will be carried out in accordance with the
precepts of the Helsinki declaration and have been approved by the Valladolid East Clinical
Research Ethics Committee (PI 20-1687). All participants of the study will sign an informed
consent form.
In accordance with Fundamental Law 3/2018 of December 5, Protection of Personal Data and
Guarantee of Digital Rights will be respected. Be that as it may, the patient can request
access, rectification, cancellation and opposition of his/her personal data. Participants
will be explicitly informed that they can withdraw their participation whenever they want and
reassured that they will have no disbenefit or negative effects on their future health care
service from withdrawing their participation. Also, participants will be given the option to
withdraw the data they have contributed with even if it is encrypted and without given any
reason to request it. These will not have any economic cost for the patient. Tthe data
obtained will be encoded. The password and access to it will be guarded by the researcher
responsible for each center.
Possible limitations and proposed mitigations:
Observational studies are prone to biases and bias recalls, particularly when comes to
filling in questionnaires. To mitigate this, participants will be trained at the beginning of
study and reinforced every time is needed. Patients will have no limitations of time for
completing questionaries, allowing them to complete them with the most truthful and objective
information as possible; these instruments will be completed under the supervision of the
research staff. Questionnaires will always be administered in the same order, beginning with
the hetero-administered ones, and concluding with the self-administered ones, to avoid
changes due to the fatigue effect.
Statistical analysis:
Qualitative data will be presented as frequency and percentage and in the case of
quantitative data as mean and standard deviation or median and interquartile range according
to the type of distribution.
To determine the principal objective, a survival analysis and determination of the
Kaplan-Meier curve will be carried out to determine the probability of persistence of the
effect on a monthly basis up to 12 months and its 95% confidence interval.
An analysis will be performed using the Log-Rank Test (Mantel-Cox regression) to evaluate
which variables are associated with a more prolonged duration of the effect.
An alpha error of 5% will be considered and statistical compensation will be made for
multiple comparisons.