NPC Archive Item: DURATION-3; when added to metformin or metformin plus a sulphonylurea, weekly exenatide improves disease-orientated outcomes compared with insulin glargine

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
On The Horizon Rapid Review NPC Logo

18 November 2010

Long-acting exenatide 2mg once weekly was more effective at reducing HbA1c than insulin glargine (titrated to glucose targets), when added to metformin or metformin plus sulphonylurea, in the 26 week DURATION-3 trial (n=456). Exenatide was associated with a mean reduction in body weight of 2.6kg compared to a mean increase of 1.4kg in those participants on insulin glargine.

Level of evidence:
Level 3 (other evidence) according to the SORT criteria.

As yet, there are no data that shows exenatide reduces the risk of macrovascular or microvascular events. A market application has been made for this new long-acting formulation of exenatide and a NICE appraisal of this formulation as either dual or triple therapy is planned. No timelines are available currently for this guidance. Local decision making bodies need to plan for the managed introduction of the product; they may wish to take into account that the existing NICE Clinical Guideline advises a third-line role for exenatide twice daily for people who meet specific criteria.

What is the background to this?
Exenatide, an incretin-based therapy for T2DM, is currently available as a twice daily subcutaneous injection (Byetta®). A long-acting once weekly subcutaneous formulation is now in development and has been submitted to the European Medicines Agency (EMA).

NICE guidance on exenatide once weekly is in progress, for second-line (dual therapy) or third-line (triple therapy), but no timeline for final guidance is available currently. The NICE Clinical Guideline on newer agents for T2DM recommends the twice-daily formulation of exenatide as a third-line option in patients who remain poorly controlled despite metformin and a sulphonylurea and have:

  • a body mass index (BMI) > 35.0kg/m2 in those of European descent (with appropriate adjustment for other ethnic groups) and specific problems associated with high body weight, or
  • a BMI < 35.0kg/m2, and where therapy with insulin would have significant occupational implications or weight loss would benefit other significant obesity-related co-morbidities.

Treatment with the drug should only continue if a beneficial response occurs as measured by a reduction of at least 1.0 percentage point in HbA1c and a weight loss of at least 3% of initial body weight at 6 months.

A Drug Safety Update issued in March 2009 by the MHRA and CHM highlights that suspected adverse reaction reports of necrotising and haemorrhagic pancreatitis have been received in association with the twice daily formulation of exenatide. Some of these have had a fatal outcome. If pancreatitis is suspected, treatment with exenatide should be suspended immediately. If pancreatitis is diagnosed, exenatide should be permanently discontinued. Reports of renal impairment, including acute renal failure and worsened chronic renal failure have also been received. Therefore, exenatide is not recommended for use in patients with end-stage renal disease or severe renal impairment.

Further information on the management of T2DM is available on the diabetes section of NPC and a recent MeReC Extra has looked at blood glucose lowering and mortality in T2DM.

What does this study claim?
The primary outcome in this study was change in HbA1c at week 26 compared to baseline. In the exenatide group the change was –1.5% (standard error [SE] 0.05) and –1.3% (0.06) in the glargine group (treatment difference –0.16% (0.07), 95% confidence interval [CI] –0.29 to –0.03, P=0.017).

Secondary outcomes included body weight and the proportion of patients achieving HbA1c targets (<7.0% and <6.5%). 35% of exenatide patients (80/227) and 23% of insulin glargine patients (50/218) achieved a target HbA1c of < 6.5% at 26 weeks (P=0.004). It should be noted that reducing HbA1c to <6.5% may be associated with no benefits over less intensive control and significant harms including possibly an increase in total mortality. We have discussed these issues here. Exenatide was associated with a reduction of 2.6kg in body weight compared to an increase of 1.4kg in insulin glargine patients, a treatment difference of –4.0kg (95%CI –4.6 to –3.5, P<0.0001).

Minor hypoglycaemia occurred in 8% of exenatide patients and in 26% of those receiving insulin glargine. Gastrointestinal effects were more common with exenatide than insulin glargine; nausea was reported by 13% of exenatide patients compared to 1% of those receiving glargine. Diarrhoea occurred in 9% of the exenatide group vs. 4% glargine. Injection site reactions were noted in 13% and 2% of patients, respectively.

How does this relate to other studies?
The twice weekly and weekly formulations of exenatide have been compared in the DURATION-1 study.  We have discussed the results from the first 30 weeks of DURATION-1 previously. There is a review of the DURATION 1 extension study here together with description of the on-going DURATION 4 and 5 trials and there is a Rapid Review of DURATION 2, here.

A further GLP-1 analogue, liraglutide (Victoza®) is administered once daily as a subcutaneous injection. NICE guidance on this drug has been published in October 2010. We have previously discussed trial data for daily liraglutide compared to twice daily exenatide here.

A trial comparing weekly exenatide to daily liraglutide is expected to finish in April 2011. Several other trials of the long-acting exenatide formulation are being carried out. For example, a 26-week trial in comparison to insulin detemir is examining change in HbA1c and weight loss.

So what?
The primary outcome of the trial was the disease-oriented measure of HbA1c, a surrogate outcome, rather than a patient-oriented outcome such as the effect on macrovascular or microvascular events, diabetes-related mortality or total mortality. The FDA in the United States has issued guidance for the pharmaceutical industry concerning the development of new agents for T2DM; sponsors are advised to undertake studies to demonstrate that the therapy is not associated with an unacceptable cardiovascular risk.  Similar guidance is being developed in Europe.

The marketing of the weekly formulation in the US has been delayed due to a FDA request for a study to investigate if exposures to higher concentrations of exenatide than typical therapeutic levels are associated with a prolongation of the QT interval in electrocardiograms, a condition that could predispose to potential cardiac arrhythmias.

A long-term clinical trial (EXSCEL) of once weekly exenatide looking at cardiovascular outcomes is currently recruiting patients (a similar study for liraglutide has recently started). Until such data are available, local decision makers may wish to follow NICE guidance on the place in therapy of GLP-1 analogues.

Study details
Diamant M, Van Gaal L, Stranks S et al. Once weekly exenatide compared with insulin glargine titrated to target in patients with type 2 diabetes (DURATION-3): an open-label randomised trial. Lancet 2010; 375:2234-43

Design: Open-label, randomised 26-week study with concealed allocation. The investigators analysing the data were blinded. Patients are continuing to receive open-label treatment in an extension period (for up to 2.5 years).

Patients: Adult T2DM patients with sub-optimal glycaemic control despite maximum tolerated doses of metformin or metformin plus sulphonylurea for at least three months were included. 84% of the patients were white; patients had had diabetes for a mean of 8 years and had a mean HbA1c at baseline of 8.3%.

Intervention and comparison: 233 patients received weekly 2mg exenatide subcutaneous injections and 223 patients received insulin glargine daily. The initial dose was 10 international units daily subsequently adjusted to achieve target fasting blood glucose concentrations of 4.0 to 5.5 mmol/L (mean 31 I.U per day at end point).

These treatments were added to the patients’ previous metformin or metformin plus sulphonylurea regimens. The dose of sulphonylurea could be reduced if hypoglycaemia was confirmed.

Outcomes and results: The primary outcome was change in HbA1c at week 26 compared to baseline. Secondary outcomes included body weight and percentage of patients achieving HbA1c targets. Results are for the modified intention-to-treat population, defined as randomised patients who received at least one dose of study drug and had both baseline and at least one subsequent measurement of HbA1c. In the exenatide group (n=228) the change in HbA1c from baseline (least squares mean) was –1.5% (SE 0.05) and –1.3% (0.06) in the glargine group (n=220). This gives a treatment difference of –0.16% ([0.07], 95%CI –0.29 to –0.03, P=0.017).

60% of exenatide patients (129/216) and 48% of insulin glargine patients (101/212) achieved a target HbA1c of < 7.0% at 26 weeks (P=0.010). 35% of exenatide patients (80/227) and 23% of insulin glargine patients (50/218) achieved a target of < 6.5% (P=0.004). These included only those patients who had not reached the target before baseline.

Exenatide was associated with a reduction in body weight: mean change of –2.6kg (SE 0.2) compared to an increase of 1.4kg (0.2) in insulin glargine patients, a treatment difference of –4.0kg ([95%CI –4.6 to –3.5], P<0.0001).

Adverse events:  Minor hypoglycaemia occurred in 8% of exenatide patients and in 26% of those receiving insulin glargine. Three patients (one taking exenatide plus metformin, one receiving glargine plus metformin and one taking glargine, metformin and a sulphonylurea) reported an episode of hypoglycaemia, which needed the assistance of another person, but did not lead to withdrawal from the study.

More withdrawals due to adverse events were reported in exenatide patients than in patients receiving insulin glargine: 5% (12/233) vs. 1% (2/223). Nausea was reported by 13% of exenatide patients compared to 1% of glargine patients and diarrhoea occurred in 9% receiving exenatide vs. 4% of those receiving glargine.  Injection site reactions were noted in 13% and 2% of patients, respectively; 2% of exenatide patients stopped treatment due to injection site reactions. Mean heart rate increased by 4 beats per minute in exenatide, but not glargine, patients (P<0.0001), but the authors state that the clinical significance of this remains unclear.

One patient in the exenatide group experienced oedematous pancreatitis but was fully recovered at a follow-up visit about two months later. Anti-exenatide antibodies were detected in some patients but the presence of antibodies was not predictive of treatment response.

Sponsorship: Amylin Pharmaceuticals Inc, Eli Lilly and Company.

Please comment on this blog in the NPC discussion rooms, or using our feedback form.

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.