04 March 2010
Obese, non-diabetic patients received one of four doses of liraglutide▼ subcutaneously once a day versus open-label orlistat orally three times a day in a European, phase II, randomised, blinded, placebo-controlled trial. A dose-related reduction in bodyweight was seen with liraglutide ranging from 4.8 to 7.2 kg, compared to 2.8 kg with placebo and 4.1 kg in the orlistat group over a 20-week period. All participants were on a calorie-controlled diet and encouraged to participate in physical activity.
Level of Evidence: Level 3 (other evidence) according to the SORT criteria.
Practitioners should continue to follow the NICE clinical guideline for the treatment of obesity. Licensed drug treatments should only be considered after dietary and behavioural approaches and exercise have been started and evaluated.
Further studies of liraglutide, looking at potential benefits in terms of cardiovascular risk factors, mortality and prevention of diabetes, would need to take place and an extension of licence obtained before any changes to prescribing practice occur.
What is the background to this?
Liraglutide has been on the UK market since July 2009 as a treatment for type II diabetes (T2DM) in combination with other antidiabetic agents. It is a glucagon-like peptide-1 (GLP-1) analogue given as a once-daily subcutaneous injection. It is almost identical to human GLP-1 but has a longer half-life. Liraglutide is thought to work by suppressing appetite and by delaying stomach emptying. We previously blogged the LEAD trials studying liraglutide in diabetes, and in which patients on the drug lost weight.
What does this study claim?
Participants on all doses of liraglutide lost more weight than those on placebo or orlistat. Mean weight loss with liraglutide 1·2, 1.8, 2.4 and 3·0mg subcutaneously daily was 4·8kg, 5·5kg, 6·3kg, and 7·2kg, respectively, compared with 2·8kg with placebo and 4·1kg with orlistat 120mg orally three times a day. Nausea increased with dose for those on liraglutide (from 24.2% up to 47.3% of participants, compared to 5.1% on placebo and 4.2% on orlistat). 80% of nausea events and 50% of vomiting events occurred within the first 4 weeks, during dose titration. Diarrhoea was more common in participants taking orlistat: 25.3% compared to between 8.4 and 12.9% on liraglutide.
How does this relate to other studies?
Other studies of liraglutide have been in patients with diabetes. This trial used higher doses than are licensed for the treatment of T2DM. A phase III, year-long North American study of liraglutide 3·0mg versus placebo in obese patients is currently in progress.
The NICE clinical guideline recommends increasing physical activity levels or decreasing inactivity, improving the quality of the person’s diet and reducing energy intake. Orlistat should be prescribed only as part of the overall plan for managing obesity. In adults the criteria are a BMI of 28.0 kg/m2 or more with associated risk factors; or a BMI of 30.0 kg/m2 or more. Therapy should be continued beyond 3 months only if the person has lost at least 5% of their initial body weight. The guideline also advises on management of children.
Drug treatments available for obesity are currently limited. Marketing of rimonabant was suspended by the European Medicines Agency in 2008, see NPC blog. More recently the Agency has recommended suspension of the marketing authorisation for sibutramine. Doctors have been advised to no longer prescribe, and pharmacists should no longer dispense this medication. The review was initiated because data from the Sibutramine Cardiovascular Outcome Trial (SCOUT) showed an increased risk of serious, non-fatal cardiovascular events compared with placebo. In addition, the FDA is currently reviewing the safety of orlistat following reports of serious liver injury and this is discussed in detail in a blog.
An accompanying editorial expresses concern about the effect size of orlistat in this study, which is one of the lowest reported for individuals without diabetes. The 1·3kg net weight loss with orlistat is about half the 2·8kg (95% confidence interval [CI] 2·5 to 3·2) reported in a meta-analysis of other randomised trials of this medicine. One possible explanation is that the orlistat was given open-label thus introducing the potential for bias. The editorial also discusses the lower weight loss with liraglutide seen in diabetic patients in the LEAD-6 trial of liraglutide versus exenatide. A weight loss of 3.2 kg was seen with liraglutide 1.8mg in that study, as compared to 5.5 kg with the same dose in the study we discuss here. Therefore, it is hypothesized that non-diabetic patients may be more responsive to GLP-1 agonists.
In this phase II trial, the proportion of patients losing 5 and 10% of their body weight was assessed:76.1% of patients receiving liraglutide 3.0 mg lost more than 5% of baseline weight at week 20, P<0.0001 versus placebo or orlistat. However, none of the results for 10% loss of body weight were statistically significant compared to that for placebo or orlistat.
The placebo-subtracted weight loss in the lowest dose liraglutide groups was modest (2 to 3kg for the 1.2 and 1.8mg doses). There was a high incidence of nausea in the liraglutide arms of the trial (between 24.2 and 47.3% of participants, with increasing dosage). Although this was mainly transient, it is possible that weight loss was related to nausea.
Whether obese patients will find the administration of a daily injection off putting, or if it will prove more of an incentive than just taking a tablet, is yet to be established. In addition, the long-term risk–benefit profile of high doses of liraglutide, as well as its ability to maintain weight loss is unknown. Eligible patients from this trial were able to continue in an 84-week extension study. The results have not yet been published so we do not know if weight loss was maintained. It is also not known if weight is regained once treatment is stopped.
Astrup A, Rössner S, Van Gaal L et al on behalf of the NN8022-1807 Study Group. Effects of liraglutide in the treatment of obesity: a randomised, double-blind, placebo-controlled study. Lancet 2009;374:1606–16
Design: Phase II, randomised, placebo-controlled 20-week trial. Blinding was achieved for whether patients received liraglutide or placebo (but not dose), and orlistat was given on an open-label basis. Allocation was concealed. The trial started with a two-week single-blind, placebo run-in period, followed by a four-week titration and 16-week constant-dose periods.
Patients: 564 obese, non-diabetic patients (average age 45.9 years, average BMI 34.8kg/m²) with fasting plasma glucose of less than 7mmol/L at run-in were included. All individuals had a 500kcal per day energy-deficit diet and were encouraged to increase or maintain their physical activity. Participants completing the study could enrol in an 84-week open-label extension period, if eligible.
Intervention: Liraglutide 1·2mg, 1·8mg, 2·4mg or 3·0mg (n=90 to 95) administered subcutaneously once a day.
Comparison: Placebo (n=98) administered subcutaneously once a day or open label orlistat (n=95) 120mg orally three times a day.
Outcomes and results: The primary endpoint was change in bodyweight during the 20 weeks of the study in the modified intention-to-treat population (defined as all randomised patients who took at least one dose of medication and had at least one post-baseline bodyweight assessment).
Mean weight loss with liraglutide 1·2, 1.8, 2.4 and 3·0mg was 4·8kg, 5·5kg, 6·3kg, and 7·2kg, respectively, compared to 2·8kg with placebo and 4·1kg with orlistat. The mean difference (95%CI) for each dose of liraglutide compared to placebo was -2·1kg (–3.6 to –0.6, P=0·003), -2.8 (–4.3 to –1.3, P<0.0001), -3.5 (–5.0 to –2.0, P<0.0001) and –4.4 (–6.0 to –2.9, P<0.0001), respectively. 76.1% of individuals on the highest liraglutide dose lost more than 5% of baseline weight compared to 29.6% of those on placebo and 44.2% of the orlistat group, P<0.0001.
Adverse events were mainly transient and rarely led to discontinuation of treatment. More than twice as many people reported gastrointestinal events with liraglutide than with placebo. Nausea increased with dose for those on liraglutide (from 24.2% up to 47.3% of participants, compared to 5.1% on placebo and 4.2% on orlistat). 80% of nausea events and 50% of vomiting events occurred within the first 4 weeks, during dose titration. No pancreatitis was reported. Diarrhoea was more common in participants taking orlistat: 25.3% compared to 7.1% on placebo and between 8.4 and 12.9% on liraglutide.
Sponsorship: Novo Nordisk.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you