Authors' objectives: Background Calcitonin gene-related peptide (CGRP) antagonists are monoclonal antibodies indicated for the preventative treatment of migraine in adults; erenumab (Aimovig®), fremanezumab (Ajovy®), galcanezumab (Emgality®) and eptinezumab (Vyepti®) are provisionally listed on the Spezialitätenliste until February 2024, February 2024, April 2024 and April 2024, respectively. The Spezialitätenliste listing for each drug has limitations relating to the effectiveness of treatment after 3, 6 and 12 months, and requires patients to have failed at least 2 prior prophylactic therapies. Objective This HTA evaluates the clinical effectiveness and safety, costs, cost-effectiveness and budget impact of CGRP antagonists (erenumab, fremanezumab, galcanezumab, eptinezumab) for migraine prophylaxis compared to the current standard of care (beta blockers [propranolol, metoprolol], calcium antagonists [flunarizine], anticonvulsants [topiramate], antidepressants [amitriptyline]), other CGRP antagonists and placebo in patients diagnosed with episodic and chronic migraine.

Authors' results and conclusions: Results Overall, 27 RCTs were included: • 8 RCTs reported data comparing erenumab to placebo and 1 RCT compared erenumab to topiramate. The median sample size was 577 (range 246–955), with 5,057 participants included across all 9 independent trials. The duration of treatment ranged from 3–6 months. • 3 RCTs reported data comparing eptinezumab with placebo. The median sample size was 665 (range 364–1072), with 2,101 participants included across all 3 independent trials. The duration of treatment ranged from 3–9 months. • 7 RCTs reported data comparing fremanezumab to placebo. The median sample size was 571 (range 177–1,130), with 4,245 participants included across all 7 independent trials. The duration of treatment ranged from 2–3 months. • 7 RCTs reported data comparing galcanezumab to placebo and 1 RCT compared galcanezumab 120 mg to 240 mg doses. The median sample size was 459 (range 207– 1,113), with 4,501 participants included across all 8 included trials. The duration of treatment ranged from 3–12 months. Selected key outcomes are summarised as follows. Almost all the included studies reported significantly fewer monthly migraine days (MMDs), significantly more patients with a response rate of >50% and significant improvements in the Migraine-Specific Quality of Life questionnaire (MSQ) for all CGRP antagonists compared to placebo, irrespective of dose. There was more RCT evidence for patients with episodic migraine than for chronic migraine, and a greater number of trials conducted for erenumab and galcanezumab compared to fremanezumab or eptinezumab. Subgroup analyses of patients with >2 prior treatment failures were reported for studies of erenumab, with one each conducted for fremanezumab and galcanezumab. While almost all trials of CGRP antagonists reported significantly fewer MMDs, the evidence was strongest for erenumab, followed by galcanezumab. Adverse events were not well reported in the included trials. Where reported, most trials showed no significant differences in adverse events between CGRP antagonists and placebo. Costs and cost-effectiveness CGRP antagonists are reimbursed for patients in Switzerland who have failed at least 2 prior prophylactic therapies. The clinical results from trials that specifically included this patient population, or presented subgroup analyses, were used as assumptions in the modelling. The clinical evidence section of the report refers to the comparator in key trials as the placebo arm. This arm is also used in the economic model, but is referred to as best supportive care (BSC). Patients in both trials were generally allowed concomitant medication, which varies by migraine frequency. The cost-effectiveness of CGRP antagonists versus BSC ranged from Swiss francs (CHF)134,152 to CHF318,982 per QALY gained over an analysis period of 1 year among episodic migraine patients, and CHF53,067 to CHF84,033 per QALY gained among chronic migraine patients. CGRP antagonists appear to be more cost-effective among chronic migraine patients. Analyses were also conducted at 5 and 10 years. These results are similar to existing analyses of models submitted to the Canadian Agency for Drugs and Technologies in Health (CADTH) for reimbursement. Univariate, probabilistic and scenario sensitivity analyses were used to explore different model assumptions. Specifically, differing doses, medicines cost, Swiss-Diagnosis-related group (DRG) cost weights for health states, structural assumptions and estimated health state utilities were included in sensitivity analyses. The analyses indicated the ICUR was most sensitive to medicines cost assumptions used in the model. Scenario analysis included the rate of MMDs experienced by those discontinuing treatment; response rates and estimated utilities were the most important assumptions driving modelling results. A budget impact analysis was undertaken to determine the additional cost of CGRP antagonists. The cost of CGRP antagonists was estimated to be CHF19.3 million in 2021 and CHF25.5 million in 2022. Given the high uncertainties associated with uptake and the sensitivity of economic modelling results to medicines prices, a range of hypothetical uptake and price scenarios were included in the budget impact analysis. The net cost of CGRP antagonists increases to CHF79.9, CHF199.8 and CHF400.9 million by 2026 at current prices assuming 10%, 25% and 50% uptake. Conclusions CGRP antagonists showed significantly fewer MMDs, significantly more patients with a response rate of >50% and significant improvements in the MSQ compared to placebo, irrespective of dose and with minimal side effects. Most of the evidence was for erenumab, followed by galcanezumab. CGRP antagonists appear to be most cost-effective among chronic migraine patients compared with episodic migraine patients. Changes in unit costs have the largest impact on estimated costeffectiveness, so strategies to reduce prices would significantly enhance economic attractiveness and reduce the budget impact of these medicines. Trials were limited by relatively short follow-up compared to modelling horizons, along with the absence of a preventive comparator. The placebo arms of clinical trials were used as a comparator, as acute medicine use was allowed. Sensitivity analyses show that results vary when longer-term effectiveness assumptions are changed.

Authors' methods: Methods A systematic literature search was conducted from 1 January 2012 to 9 March 2022 in Medline, Embase, Cochrane Library, EconLit, INAHTA HTA database, Cost-Effectiveness Analysis Registry, and grey literature sources. Studies were prioritised for inclusion by study design using a hierarchical selection process, whereby meta-analyses were included preferentially, followed by randomised controlled trials (RCTs) and finally non-randomised studies of interventions; only the highest level of available evidence was included. Meta-analysis of RCTs was conducted using a random effects, inverse variance modelling approach. If meta-analysis could not be conducted, results were reported narratively. Risk of bias of the included RCTs was evaluated using the Cochrane Risk of Bias 2.0 tool, and the overall strength of evidence for key outcomes was evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. A Markov model was developed to quantify the cost-utility of CGRP antagonists using incremental quality-adjusted life years (QALY), with univariate, scenario and probabilistic sensitivity analyses evaluating uncertainties in the model. The results have been presented as incremental cost-utility ratios (ICUR) and as a series of cost-effectiveness acceptability curves to show the probability that a given intervention can be considered cost-effective under a range of willingness-to-pay thresholds (WTPs).

Project Status: Completed

URL for project: https://www.bag.admin.ch/bag/en/home/versicherungen/krankenversicherung/krankenversicherung-leistungen-tarife/hta/hta-projekte/cgrp-antagonists.html

URL for protocol: https://www.bag.admin.ch/dam/bag/en/dokumente/kuv-leistungen/leistungen-und-tarife/hta/berichte/h0055migr-hta-protocol.pdf.download.pdf/h0055migr-hta-protocol.pdf

Year Published: 2023

URL for published report: https://www.bag.admin.ch/dam/bag/en/dokumente/kuv-leistungen/leistungen-und-tarife/hta/berichte/h0055migr-hta-report.pdf.download.pdf/h0055migr-hta-report.pdf

English language abstract: An English language summary is available

Publication Type: Full HTA

Country: Switzerland

Organisation Name: Swiss Federal Office of Public Health (FOPH)

Contact Address: Federal Office of Public Health, Schwarzenburgstrasse 157, CH-3003 Berne, Switzerland

Contact Name: Stephanie Vollenweider

Contact Email: hta@bag.admin.ch

Copyright: Swiss Federal Office of Public Health