Cost effectiveness of liraglutide in type II diabetes: a systematic review

As novel treatments for type II diabetes enter the market, there is a need to assess their long-term clinical and economic outcomes against currently available treatment alternatives. Objective compilation and evaluation of current pharmacoeconomic evidence can assist payers and decision makers in determining the appropriate place in therapy of a new medication.

Our objective was to review the existing pharmacoeconomic literature evaluating the cost effectiveness and overall costs of treatment associated with liraglutide in type II diabetes.

Medical literature indexed in MEDLINE, EMBASE, PsycINFO, CINAHL, and EconLit through 1 June 2014 was searched.

Full-text, English-language cost-effectiveness, cost-utility, and other cost analyses in type II diabetes that compared liraglutide to one or more anti-diabetic agents were included. Initial screening was based on relevance of titles and abstracts followed by examination of the study methods of each remaining manuscript. Studies conducting original pharmacoeconomic analyses were chosen for inclusion.

Articles meeting the inclusion criteria were retrieved, and information on the study design and results was abstracted. Abstracted data elements were chosen and assessed based on the authors' experience as well as criteria set forth by the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Health Economic Evaluation Publication Guidelines Task Force. Additionally, reported incremental cost-effectiveness ratios (ICERs) and selected sensitivity analysis results were converted to $US, year 2012 values, in order to facilitate comparison across studies.

A total of six cost studies and seven cost-utility studies were identified for inclusion. Across cost studies, liraglutide treatment resulted in costs ranging from a loss of $US2,730 (liraglutide 1.8 mg vs. sitagliptin; pharmacy costs only) over a 1-year time horizon to a savings of $US9,367 (liraglutide 1.8 mg vs. glimepiride; diabetes-related complication costs only) over a 30-year time horizon. Cost-utility analysis results reported base-case ICERs ranging from $US15,774 (vs. glimepiride) to $US40,128 (vs. rosiglitazone) per quality-adjusted life-year (QALY) ($US, year 2012) for liraglutide 1.2 mg and $US8,497 (vs. exenatide) to $US66,031 (vs. rosiglitazone)/QALY ($US, year 2012) for liraglutide 1.8 mg. Estimates were most sensitive to variations in time horizon and cardiovascular complication rates. Based on frequently cited, country-specific cost-utility thresholds, liraglutide was determined to have a probability of being cost effective of between 58 % (liraglutide 1.8 mg vs. sitagliptin) and 93 % (liraglutide 1.2 mg vs. glimepiride).

Weaknesses of included studies related primarily to study model inputs that assumed long-term morbidity and mortality benefits in favor of liraglutide based on improvements in clinical biomarkers observed in short-term clinical trials. The exclusion of drug acquisition costs in two identified cost studies as well as the assumed lifetime duration of treatment with liraglutide in several cost-utility studies were also identified as weaknesses. The authors' review was limited by the possibility of incomplete literature retrieval, unintended omission of relevant data elements, and comparison of costs and ICERs generated from healthcare systems from differing countries.

The current literature presents liraglutide as a cost-effective adjunct treatment for type II diabetes that may also be associated with a reduction in diabetes-related complication costs; however, ICER values are largely dependent on assumptions regarding the benefits of long-term liraglutide treatment and the time horizon of the analysis. Real-world use may make liraglutide unattractive from a payer and policy-maker perspective.