BACKGROUND: Galcanezumab is a humanised monoclonal antibody against human calcitonin gene-related peptide (CGRP) and is indicated for the prophylaxis of migraine in adults who have at least 4 migraine days per month and is currently under evaluation for the preventive treatment of episodic cluster headache. As migraine and cluster headache (CH) are chronic conditions, continuous or intermittent long-term treatment is anticipated in routine clinical practice. Exposure to galcanezumab was limited in clinical trials. Therefore, infrequent adverse effects and effects which have a longer latency period, such as serious hypersensitivity and/or malignancy, could occur in real-world practice. Additionally, patients with recent cardiovascular (CV) events and/or serious CV risk as well as patients over the age of 65 years, who were excluded from the clinical trial population, might be part of the user population in everyday clinical practice. The implications of long-term inhibition of CGRP are unknown, including the impact on long-term safety. As a result, the long-term safety of galcanezumab in larger patient populations requires further characterisation.
Research question and objectives: The purpose of this study is to evaluate the utilisation and safety of galcanezumab with respect to serious hypersensitivity events and long-term safety (up to 5 years), including malignancy and cardiovascular events in routine clinical practice in Europe. The primary objective is to characterise the utilisation of galcanezumab for the treatment of migraine or CH and to assess the incidence of serious CV events, serious hypersensitivity reactions, and malignancies in real world clinical practice in Europe.
This objective is primarily descriptive and aims to assess:
? the utilisation of galcanezumab overall and in special populations of interest, particularly: o patients with recent (within the last 6 months) serious CV events and/or serious CV risk prior to initiating galcanezumab o patients ?65 years of age
? the incidence of serious CV events in patients who receive galcanezumab treatment overall, by duration of galcanezumab treatment and in the special populations of interest including:
– patients with recent (within the last 6 months) CV events and/or serious CV risk prior to initiating galcanezumab
– patients ?65 years of age
? the incidence of serious hypersensitivity reactions such as anaphylaxis in patients who receive galcanezumab
? the incidence and distribution of time to malignancies (excluding non-melanoma skin cancers) in patients exposed to galcanezumab
The secondary objective is to provide context for incidence rates of safety events seen in the galcanezumab-exposed migraine cohort by describing the incidence rates observed in a comparator cohort of migraine patients that initiated another (prophylactic) migraine medication and, as feasible, to conduct comparative safety analyses of serious CV events, and malignancies. Non CGRP prophylactic migraine medications are small molecules that are typically less immunogenic than monoclonal antibodies. There is also uncertainty around the ability to accurately identify serious hypersensitivity events. As such, formal comparative analyses are not proposed; however, rates in the comparator cohort will be provided for context.
STUDY DESIGN: An observational cohort study using data from various population-based healthcare databases from seven different European countries.
POPULATION: The study population for the primary objective will include all patients with a prescription or dispensing of galcanezumab in the population-based healthcare databases from seven different European countries including France (FR), Germany (DE), Italy (IT), the Netherlands (NL), Spain (ES), Sweden (SE) and the United Kingdom (UK). This cohort will be divided into three sub-cohorts based on the diagnosis and/or treatment of migraine or CH. In addition, sub-cohorts of patients with recent (within the last 6 months) serious CV events and/or serious CV risk prior to initiating galcanezumab and patients ?65 years of age will also be created. The study population for the secondary objective will include all migraine patients initiating treatment with galcanezumab in the population-based healthcare databases from the seven different European countries. To provide context for incidence rates of the safety events seen in the galcanezumab-exposed migraine cohort, migraine patients who initiate treatment with comparator medication topiramate, will be used as a comparator cohort for analyses of serious cardiovascular events and malignancies. A formal comparative analysis for serious hypersensitivity reactions will not be undertaken due to the lack of an appropriate comparator and uncertainty around the ability to accurately identify these events. Rates in the comarator cohort will be provided for context. For each of the databases, the study period starts at the time galcanezumab was approved by the European Commission (i.e. 14 November 2018). The end of the study period will depend on the available data at the end of data collection (Q3 2025) for each country. Assuming an average data lag time of approximately 1.5 years, the end of the study period will be Q4 2025 at maximum.
VARIABLES:
– Exposure
Primary objective: The exposure of interest will include any treatment with galcanezumab. Secondary objective: The exposure of interest will include new users of galcanezumab or comparator medication topiramate. Patients will be considered to be new users if they have not been prescribed/dispensed their index drug within the 12 months prior to the index date (i.e first prescription/dispensing of galcanezumab or comparator medication within the study period).
– Outcomes
The outcomes of interest are serious CV events, serious hypersensitivity reactions and malignant neoplasms. Serious CV events under consideration include: hospitalisation for: myocardial infarction (MI), transient ischaemic stroke (TIA), ischaemic stroke, ischaemic heart disease, unstable angina pectoris (AP), percutaneous coronary intervention (PCI), coronary revascularisation and CV death. The final selection of CV outcomes will depend on the accuracy of the diagnosis within in each data source. Serious CV events will be reported as composite and individual outcomes. Serious hypersensitivity reactions will be identified by a hospitalisation or emergency care visit for anaphylaxis or allergy. As feasible, hospitalisations or emergency care visits for angioedema, acute asthma or acute bronchospasm, acute upper airway obstruction, epinephrine administration and death from serious hypersensitivity reactions (i.e. death after any of these events, to be detailed in the SAP) will additionally be assessed as part of this outcome. Malignant neoplasms will be identified as a diagnosis of any cancer, excluding non-melanoma skin cancer (NMSC). Malignancies will be reported as a composite outcome (all malignancies, excluding NMSC), and by type of the first malignancy (i.e. per primary cancer site). All outcomes of interest will be identified by using the database specific coding systems (e.g. READ codes in the UK, International Classification of Diseases (ICD) 9 th or 10th revision, or International Classification of Primary Care (ICPC) in the Netherlands), as well as the specific provenance of the data (e.g. hospital admission, primary care or emergency care visits). The specific diagnostic codes to identify the outcomes of interest will be included in the statistical analysis plan (SAP). Furthermore, table shells will be provided in this document to show which results will be delivered.
– Covariates
Demographic and clinical covariates will be considered for descriptive purposes and as confounders in the comparative study. All available data prior to start of the treatment will be used to assess baseline demographics and clinical covariates.
DATA SOURCES: All data will be obtained from seven European databases. The following databases will be used: ? SNDS (Système National des Données de Santé) from France
? GePaRD (German Pharmacoepidemiological Research Database) from Germany
? ARS (Agenzia regionale di sanità della Toscana database) from Italy
? PHARMO (The PHARMO Database Network) from the Netherlands
? SIDIAP (Sistema d’Informació per al Desenvolupament de la Investigació en Atenció Primària) from Catalonia, Spain
? SHR (Swedish Health Registers) from Sweden
? CPRD (The Clinical Practice Research Data Link Gold plus Hospital Episodes Statistics (HES) Data) from the UK
STUDY SIZE: The number of galcanezumab-exposed patients will depend on the uptake of galcanezumab in each country. The descriptive outcomes in the primary objective as well as the incidence rates as part of the secondary objective will be presented in the final report regardless of the size of the study population. The comparative analyses will only be performed if target sample sizes are attained that will allow us to detect a 2-fold difference in this primary outcome of serious CV events with 80% power for the composite endpoint of serious CV events (N = 7,245 person-years in galcanezumab-exposed patients per database). Pooled analysis or meta-analysis will be considered to help achieve sufficient power, depending on the appropriateness of combining data sources.
DATA ANALYSIS: Descriptive statistics for demographic and clinical characteristics at baseline will be provided. Incidence rates for serious CV events, serious hypersensitivity reactions and malignant neoplasms will be presented for the galcanezumab-exposed migraine and topiramate migraine cohort and will be estimated by dividing the number of events of interest by the person-time at risk. If feasible, based on accrual of target sample size, comparative analyses will be conducted to compare the incidence of serious CV and malignancy events among migraine patients that are new users of galcanezumab with propensity score-matched new users of topiramate. This will be further detailed in the SAP. Relative risks for the outcomes of interest will be estimated as hazard ratios (HRs). All analyses will be done for each country separately as well as pooled (if deemed appropriate based on the similarity of results and outcome definitions).