Drucker DJ, Buse JB, Taylor K et al for the DURATION-1 Study Group. Exenatide once weekly versus twice daily for the treatment of type 2 diabetes: a randomised, open-label, non-inferiority study. Lancet DOI:10.1016/S0140-6736(08)61206-4
In terms of effect on glycated haemoglobin (HbA1c) once weekly exenatide▼ appears to be non-inferior to the twice-daily formulation. However, a per-protocol analysis is needed to confirm non-inferiority: unfortunately this was not reported.
Area Prescribing Committees will need to consider this study alongside other evidence for the management of type 2 diabetes. The new once weekly formulation of exenatide may be similar in effectiveness to the twice daily preparation but we still do not know if this drug improves patient oriented outcomes compared with other interventions. Data from long-term studies comparing once weekly exenatide with insulin and other antidiabetic treatments will be necessary before the place in therapy of this new formulation can be defined.
What is the background to this?
Exenatide (Byetta▼) is a recently marketed diabetes agent given via sub-cutaneous (s/c) injection twice a day. A long-acting version is the subject of on-going research. We have also blogged about safety warnings issued by MHRA and FDA about this agent.
Further information about the management of diabetes is available on the cardiovascular floors of NPC.
What does this study claim?
In this 30 week open-label, non-inferiority trial, long-acting exenatide was non-inferior to the twice-daily formulation on changes to HbA1c levels in the intention-to-treat population. However, no per-protocol analysis has been quoted, which you would expect in a non-inferiority study. The researchers showed that the long-acting formulation was associated with greater changes in HbA1c than the twice-daily formulation (–1.9 [standard error 0.1%] vs. –1.5 [0.1%], mean difference –0.33, 95% CI –0.54 to –0.12%, P=0.0023).
How does this relate to other studies?
A recent NICE clinical guideline on the management of type 2 diabetes states that exenatide is not recommended for routine use. However, it does give some indication of situations in which it could be considered, for example: patients with a body mass index over 35.0 kg/m2, or when inadequate blood glucose control (HbA1c ≥ 7.5%) has been obtained with conventional oral agents, or when other high-cost medication, such as a thiazolidinedione or insulin injection therapy, would otherwise be started. This section of the guideline is expected to be updated in a further NICE clinical guideline due to be published in 2009 called ‘Newer agents for blood glucose in type 2 diabetes’.
This study used a surrogate outcome, HbA1c, rather than a patient oriented outcome such as the effect on cardiovascular events and mortality. It has been reported that a US advisory committee has recommended that long-term studies for new drugs for diabetes should measure cardiovascular events. This conclusion was reached after a meta-analysis suggested that patients on rosiglitazone were at increased risk of cardiovascular events, despite known favourable effects on blood glucose.
Only results from the intention-to treat analysis are reported in this study. However, because an intention-to-treat analysis includes patients who did not actually take the treatment as intended, any differences between the two groups may be blurred and the chance of falsely finding non-inferiority is increased. As explained in a recent Drug and Therapeutics Bulletin, a per-protocol analysis, which uses data only from patients who satisfied the entry criteria, actually received treatment and properly followed the protocol, should also have been undertaken. Equivalence or non-inferiority can only be established if both analyses support the finding.
As the long-acting formulation of exenatide was not compared to insulin, it is difficult to comment on the place of this formulation in therapy. Longer-term studies are necessary to confirm efficacy and safety. The DURATION-3 trial of once weekly exenatide compared to insulin glargine is currently recruiting.
The existing formulation of exenatide costs £64 for 28 days, although pricing details of a long-acting formulation are not yet available.
Design: open-label, randomised, non-inferiority trial.
Patients: adults (n=303) with type 2 diabetes treated with diet/exercise or oral drugs. Baseline HbA1c of 7.1 – 11.0%.
Intervention: exenatide long-acting formulation (n=148) 2mg s/c once weekly for 30 weeks.
Comparison: exenatide 5 micrograms s/c twice daily for 28 days, 10 microgram twice a day thereafter (n=147).
Outcomes and Results: primary, surrogate, endpoint of change in HbA1c levels. At 30 weeks results were assessed on basis of anonymised data.
Pre-specified secondary endpoints included changes to body weight and proportion of patients reaching target HbA1c levels. Patients in both groups lost weight, in the region of 4kg. Once weekly exenatide resulted in a greater proportion of patients achieving HbA1c levels of 7% or less (77% vs. 61%, P=0.0039).
Adverse effects: no patients experienced major hypoglycaemia. Anti-exenatide antibodies were higher with the once weekly rather than twice daily dosing (P=0.0002). The significance of this finding is not known. Cases of acute pancreatitis have been reported with the twice daily formulation, particularly in the US. In this trial, no cases of pancreatitis were reported.
Injection site pruritis was more common with the once weekly regimen (17.6% vs. 1.4%). The twice-daily formulation was associated with more nausea and vomiting (nausea 26.4% vs. 34.5%, vomiting 10.8% vs. 18.6%, once weekly vs. twice daily, respectively). P values were not quoted.
Withdrawals due to adverse events occurred in 6.1% of the once weekly group and 4.8% of the twice-daily patients.
Sponsorship: Amylin Pharmaceuticals, Eli Lilly and Company.