SAN FRANCISCO — A novel investigational agent that suppresses resorption of glucose in the kidney improved glycemic control in a 12-week trial of 389 treatment-naive patients with type 2 diabetes.

Bristol-Myers Squibb Co.'s dapagliflozin selectively inhibits the renal sodium-glucose cotransporter 2 (SGLT2), which is the primary transporter of glucose in the kidney, while avoiding the intestinal glucose transporter SGLT1. The inhibition of SGLT2 modulates resorption of glucose in the proximal tubule, which results in excretion of glucose into the urine.

Previous studies have shown that SGLT2 inhibition reduces blood glucose independently of insulin secretion or action and also results in urinary loss of calories, Dr. James F. List said at the annual scientific sessions of the American Diabetes Association.

Dr. List, associate director of global clinical research for Bristol-Myers Squibb Co., Princeton, N.J., presented results from the phase IIb dose-ranging study of dapagliflozin. After a 2-week placebo lead-in phase, the patients were randomized in equal ratios to either once-daily dapagliflozin in doses of 2.5, 5, 10, 20, and 50 mg, metformin (750 mg titrated to 1,500 mg), or placebo for 12 weeks. The 389 patients had a mean age of about 55 years, a mean weight of 89 kg, and a mean body mass index of 31 kg/m

Mean hemoglobin A_1c (HbA_1c) values ranged from 7.7% to 8.0% in all treatment arms and were similar to the median HbA_1c in all groups except the 20-mg arm, which had a median HbA_1c of 7.4%. Mean fasting plasma glucose (FPG) was about 150 mg/dL in all the treatment arms. At baseline, glycosuria levels ranged from 6 g in 24 hours to 11g/24 hours.

Reduction in HbA_1c at 12 weeks, the primary end point, was significant in all dapagliflozin dose groups, ranging from 0.55 percentage points in the 20-mg group to 0.90 in the 50-mg group. Reductions in HbA_1c were 0.71 with 2.5 mg and 0.72 with 5.0 mg, similar to the 0.73 seen with metformin. The placebo group, in contrast, had a reduction of only 0.18. Dose-dependent drops in FPG were similarly significant in all of the dapagliflozin treatment arms, from 16.2 mg/dL in the 2.5-mg treatment group, to 21.1 mg/dL in the 10-mg arm, to 30.5 mg/dL in the group receiving 50 mg dapagliflozin. Reductions in FPG averaged 18.0 mg/dL with metformin and just 5.8 mg/dL with placebo.

Changes in postprandial glucose, assessed by 75-g 3-hour oral glucose tolerance test, also were significant with all doses of dapagliflozin, compared with placebo. However, unlike FPG, those reductions showed no clear dose-dependency, Dr. List said.

The normalized 24-hour urinary glucose:creatinine ratio ranged from 31.72 g/g creatinine at the 2.5-mg dose to 64.75 g/g creatinine for the 20-mg dose, also significant and dose-dependent. There were no changes in those ratios for either metformin or placebo.

The nonnormalized values, expressed as total urinary glucose in 24 hours at week 12, ranged from 51.8 g/day with the lowest dose to a high of 85 g/day with the 20-mg dose (the 50-mg value was similar), compared with baseline values of 5.8–10.9 g/day. That degree of glycosuria translates to a loss of 208–340 calories/day, Dr. List noted.

Changes in weight ranged from 2.5 kg with the 5-mg dapagliflozin dose to 3.4 kg for both the 20-mg and 50-mg doses, compared with 1.7 kg and 1.2 kg with metformin and placebo, respectively. There were no significant changes in serum sodium, potassium, calcium, or urinary calcium, and no changes in renal function assessed by serum creatinine and by measured 24-hour creatinine clearance.

Adverse events were evenly distributed among the study arms, with 63% of the dapagliflozin groups reporting at least one event of any kind, compared with 68% of the metformin patients and 54% with placebo. There were no deaths, and a total of five serious events that were evenly distributed among the dose arms. Discontinuations due to adverse events were also similar between groups, with no particular event standing out, he said.

Hypoglycemia (defined as a confirmed glucose value of 50 mg/dL or below) was not increased in the dapagliflozin groups, in which it ranged from 6%–10%, compared with 9% in the metformin group and 4% with placebo. Rates of urinary tract infections—a concern because of the possibility that the excess urinary glucose could serve as a growth medium for microorganisms—were 5%–12% with dapagliflozin, compared with 9% and 6% for metformin and placebo, respectively.

Genital infections occurred in 2%–3% of the dapagliflozin groups, compared with 2% on metformin and none with placebo. Hypotension—a theoretical concern because of the osmotic diuretic effect of the urinary glucose—was seen in one patient on 50 mg dapagliflozin, two metformin patients, and one on placebo.

Dapagliflozin is now in phase III testing.

Dapagliflozin Lowers Glucose by Raising Glycosuria