Linagliptin non-inferior to glimepiride in two-year trial

Linagliptin was non-inferior to glimepiride in lowering hemoglobin A1c (HbA1c) in type 2 diabetes patients, according to a study designed, conducted and analyzed by linagliptin's manufacturer, Boehringer Ingelheim.

Researchers conducted a randomized, double-blind, parallel-group, active controlled, non-inferiority trial at 209 sites in 16 countries. Results appeared online June 28 at The Lancet. Study participants were aged 18 to 80 years, had type 2 diabetes and a body mass index less than 40 kg/m², were receiving metformin at a stable dose of at least 1,500 mg/d (alone or with one other oral antidiabetic drug), and had an HbA1c of 6.5% to 10% on metformin alone or 6% to 9% on metformin and one additional oral antidiabetic drug.

After a washout period of the patients' second diabetes drugs, linagliptin (5 mg once daily) or glimepiride (initially 1 mg once daily) was added to metformin. The dose of glimepiride was increased in 1-mg increments up to a maximum of 4 mg once daily, at four-week intervals during the first 12 weeks of treatment. The dose of glimepiride was increased if the patients' self-monitored fasting plasma glucose values were greater than 6.1 mmol/L (110 mg/dL). At any time, the dose of glimepiride could be decreased to prevent hypoglycemia.

The primary efficacy endpoint was change in HbA1c from baseline to week 104. The two key secondary endpoints were occurrence of hypoglycemic episodes and change in body weight. After two years of treatment, linagliptin was non-inferior to glimepiride in reducing HbA1c. At week 104, adjusted mean changes in HbA1c from a baseline of 7.7% were −0.16% with linagliptin and −0.36% with glimepiride; the difference between treatment groups was 0.20% (97.5% CI, 0.09 to 0.30; P=0.0004).

An HbA1c of less than 7% at week 104 was achieved by 232 (30%) of 764 patients on linagliptin and 263 (35%) of 755 patients on glimepiride. An HbA1c less than 6.5% was achieved by 92 (12%) of 764 patients in the linagliptin group and 120 (16%) of 755 patients in the glimepiride group. Overall, 200 (26%) of 764 patients on linagliptin and 253 (34%) of 755 patients on glimepiride achieved an HbA1c reduction of 0.5% or greater.

There were 4.8 times fewer hypoglycemic events with linagliptin than with glimepiride (58 [7%] of 776 patients vs. 280 [36%] of 775 patients; P<0.0001). Severe hypoglycemia occurred in one patient receiving linagliptin compared with 12 patients receiving glimepiride. The proportion of patients who had an A1c less than 7% and at least one hypoglycemic event was four times lower with linagliptin than with glimepiride (31 [4%] of 776 vs. 152 [20%] of 775 patients). Body weight decreased with linagliptin (−1.4 [SE, 0.2] kg) but increased with glimepiride (1.3 [SE, 0.2] kg) from similar mean baseline values (86.0 [SE, 0.7] kg vs. 87.0 [SE, 0.6] kg).

An accompanying editorial commented that linagliptin was non-inferior to glimepiride after two years, although the non-inferiority of 0.20% was lower than the predefined criterion of 0.35%.

“Only a longer follow-up (presumably at least 4 years) would allow investigators to show whether a more durable glucose-lowering effect can be achieved with a DPP-4 inhibitor than with a sulphonylurea,” the editorialists stated. “If so, stabilization of metabolic control would be of great interest by decreasing the future treatment burden for patients with diabetes, thus justifying the higher cost of DPP-4 inhibitors.”