A once-weekly injection of exenatide to improve blood sugar control in people with type 2 diabetes produces a significant improvement in blood sugar control, as well as inducing a mean weight loss of 2.6kg per patient. The current standard second-line therapy of insulin glargine (dosed to a specific target) also produces improved blood sugar control, but to a lesser extent than exenatide. Insulin glargine, which has to be injected daily, also produced a mean weight gain of 1.4kg per patient .The conclusions of the DURATION-3 study are reported in an Article in this week's American Diabetes Association meeting Special Issue of The Lancet.
Diabetes treatments are needed that are convenient, provide effective glycaemic control, and do not cause weight gain. While exenatide has been found to improve blood glucose control and induce weight loss in other studies, those studies addressed the twice-daily formulation of exenatide. A new formulation has been developed which only requires weekly injection. In this study, Professor Michaela Diamant, Diabetes Centre, VU University Medical Centre (VUMC), Amsterdam, The Netherlands, and colleagues tested the hypothesis that improvement in blood sugar control (measured by HbA1c concentration*) achieved with once-weekly exenatide is better than that achieved with the existing standard second-line treatment for patients not responding to oral blood-glucose-lowering agents—insulin glargine titrated to glucose targets.
In this 26-week randomised study, the authors compared once-weekly exenatide with insulin glargine in adults with type 2 diabetes who had inadequate blood sugar control despite use of maximum tolerated doses of first-line blood-glucose lowering drugs for 3 months or longer (these first-line drugs were either metformin or metformin plus sulphonylurea).
Patients were randomly assigned to add exenatide (2 mg, once-a-week injection) or insulin glargine (once-daily injection, starting dose 10 IU, target glucose range 4•0–5•5 mmol/L) to their blood glucose-lowering regimens. Randomisation was with a one-to-one allocation and stratified according to country and concomitant treatment (70% metformin only; 30% metformin plus sulphonylurea). Participants and clinical investigators were not blinded to treatment, but investigators analysing data were. The primary endpoint was change in HbA1c from baseline.
456 patients were randomly allocated to treatment and were included in the analysis (233 once-weekly exenatide, 223 insulin glargine). Participants who received at least one dose of study drug and for whom baseline and at least one post-baseline measurement of HbA1c were available were included in the primary efficacy analysis. Change in HbA1c at 26 weeks (from a mean level of 8.3% for both groups) was greater in patients taking exenatide (228 patients; –1•5%) than in those taking insulin glargine (220 patients; –1•3%). But the authors point out the clinical importance of this modest improvement is uncertain. A total of 12 (5%) of 233 patients allocated to once-weekly exenatide and two (1%) of 223 taking insulin glargine discontinued participation because of adverse events . Rates of hypoglycaemia (dangerously low blood sugar levels) were lower with exenatide than insulin glargine. There was a small but significant increase in mean heart rate with exenatide treatment, but the authors say the clinical significance of this is unclear.
Once-weekly exenatide was associated with a progressive decrease in bodyweight, but insulin glargine with a progressive increase (figure 3). Mean change in bodyweight at week 26 compared with baseline was –2•6 kg (SE 0•2) for patients allocated to exenatide (n=233), and 1•4 kg (0•2) for those taking insulin glargine (n=223; p<0•0001).
The authors note that their study consisted of mostly white participants, "and would be strengthened by replication in other populations." They add that a planned extension to DURATION-3 of 2.5 years is in progress to monitor long-term outcomes of this treatment strategy.
They say: "Diabetes treatments that lower HbA1c substantially, promote weight loss, and provide convenient dosing without the need for dose titration could improve adherence to treatment and lessen the burden of hyperglycaemia."
They conclude: "Especially important in our study was the finding that baseline bodyweight was not predictive of glycaemic response. Furthermore, weight loss was not strongly associated with nausea and vomiting—a finding that is supported by results of previous clinical trials. Taken together, these findings showed that once-weekly exenatide resulted in greater reduction in HbA1c than did insulin glargine titrated to target, and is an important therapeutic consideration for patients for whom convenience, weight loss, and risk of hypoglycaemia are particular concerns."
In an accompanying Comment, Dr Anoop Misra, Fortis Hospital, New Delhi, India, and Dr Shashank Joshi Lilavati and Bhatia Hospital, Mumbai, India, agree exenatide could be used in patients "with obesity and those in whom hypoglycaemia is a clinical concern."
They conclude: "Currently, there is more promise, few disadvantages, and some unknowns about treatment with long-acting exenatide for diabetes. Despite advances in antihyperglycaemic therapy, a drug which would lead to substantial prevention of macrovascular and microvascular complications, decreases mortality, and is convenient and affordable, remains the undiscovered Holy Grail of diabetes management."
Professor Michaela Diamant, Diabetes Centre, VU University Medical Centre (VUMC), Amsterdam, The Netherlands. T) +31-20-4440534/ +31-20-4440533 E) m.diamant@vumc.nl
Dr Anoop Misra, Fortis Hospital, New Delhi, India. T) 91-11-41759672 / +91 (0) 9811153997 (mobile) E) anoopmisra@metabolicresearchindia.com
For full Article and Comment see: http://press.thelancet.com/adaduration3.pdf
Note to editors: * HbA1c is used to indicate the average plasma glucose concentration of the preceding two to three months. In general, the reference range (that found in healthy persons who do not have diabetes), is about 4%—5.9%. Patients with diabetes usually have HbA1c levels above 6.5%
Journal
The Lancet