Invokana reduces HbA1c, body weight in adults with type 1 diabetes

Issue: November 2015

Adults with type 1 diabetes randomly assigned the sodium-glucose cotransporter 2 inhibitor Invokana saw reductions in HbA1c, body weight and insulin doses, with no increase in hypoglycemia, compared with those randomly assigned placebo, according to study findings presented at the 51st European Association for the Study of Diabetes Annual Meeting.

In an 18-week, double blind, parallel-group, phase 2 study analyzing the efficacy and safety of Invokana (canagliflozin, Janssen) in 351 patients with type 1 diabetes, researchers also found an increase in diabetic ketoacidosis among participants assigned canagliflozin compared with those assigned placebo.

“[Canagliflozin is] approved for the treatment of type 2 [diabetes], but not in type 1 [diabetes],” Robert Henry, MD, professor of medicine at University of San Diego, said while presenting the study findings. “Canagliflozin works by lowering the renal threshold for glucose and increasing urinary glucose excretion, and these mechanisms are independent of insulin secretion, so it makes reasonably good sense to be able to use it in type 1 [adults], and that was the basis for this study.”

Robert Henry

Henry and colleagues analyzed data from 351 adults with type 1 diabetes aged 25 to 65 years on a stable insulin regimen for at least 8 weeks (56% men; 91% white; mean age, 41 years; mean HbA1c, 7.9%; mean body weight, 83 kg; mean BMI, 28 kg/m²). The average duration of type 1 diabetes within the cohort was 22 years. Those with a history of diabetic ketoacidosis or a severe hypoglycemic event within 6 months of randomization, or a history of cardiovascular events or uncontrolled hypertension, were excluded from the study.

Patients were randomly assigned 100 mg (n = 110) or 300 mg (n = 110) canagliflozin or matching placebo (n = 110) for 18 weeks. Before randomization, study participants were advised to reduce insulin doses to reduce hypoglycemic risk, Henry said. Adults with an HbA1c of 8% or less were advised to reduce basal insulin doses by 20%; those with an HbA1c greater than 8% were advised to reduce insulin doses by 10%.

“It was not mandated, it was suggested,” he said. “So, many of the investigators did it; some did not.”

Researchers used a mixed model for repeated measures to analyze changes from baseline in HbA1c, fasting plasma glucose, body weight and insulin dose. The primary endpoint — the proportion of patients with an HbA1c reduction of up to 0.4% with no increase in body weight — was analyzed longitudinally with a generalized linear mixed model.

Patients also wore continuous glucose monitors for 1 week before randomization and again at week 18.

“So there is [continuous glucose monitoring] data which has not been fully analyzed,” Henry said.

Researchers found that both doses of canagliflozin increased the proportion of adults who met the primary endpoint, when compared with placebo.

“The change in [HbA1c] was about 45% in both the 100-mg and 300-mg arms of canagliflozin vs. 23% in the placebo,” Henry said. “The change in body weight was higher ... and the combined endpoint ... was significantly better at 37% with 100 mg and 41% with 300 mg of canagliflozin.”

After 18 weeks of treatment, adults assigned 100 mg canagliflozin saw a mean HbA1c reduction of 0.27% and an average 3.1% reduction in body weight; adults assigned 300 mg saw a mean HbA1c reduction of 0.24% and an average 5.1% reduction in body weight. Canagliflozin was also associated with reduced FPG, but results were not statistically significant, Henry said.

Adults assigned both 100 mg and 300 mg canagliflozin also reduced their insulin doses.

There were no differences across groups for hypoglycemia events; there was a slight increase in urinary tract infections and mycotic infections, particularly in the 300-mg dose in women, Henry said.

Within the canagliflozin arms, five patients assigned 100 mg canagliflozin and seven patients assigned 300 mg experienced diabetic ketoacidosis. There were no ketone-related events in the placebo group. Henry noted that all patients experiencing diabetic ketoacidosis had factors contributing to the event, including pump failure, skipped insulin doses or illness.

“In type 1, the canagliflozin treatment may make patients more susceptible when there are precipitating factors,” Henry said. “It may be related to a combination of reducing the insulin doses excessively, reduced carbohydrate intake, or other effects of SGLT inhibition.”

Mitigation strategies, including more monitoring of ketones or lower doses of canagliflozin, are being considered for future clinical trials in adults with type 1 diabetes, he said.

“Clearly, from this data, it appears that lower doses of canagliflozin will be necessary,” Henry said. – by Regina Schaffer

Reference:

Henry, R, et al. Efficacy and safety of canagliflozin: results of an 18-week phase 2 study in patients with type 1 diabetes. Presented at: 51st EASD Annual Meeting; Sept. 14-18, 2015; Stockholm.

Disclosure: Henry reports serving as a board member for or receiving consultant fees or grants from Abbott, Alere, Amgen, AstraZeneca, Boehringer Ingelheim, Elcelyx, Gilead, Isis Pharmaceuticals, Intarca, Janssen, Ligand, Eli Lilly, Merck, Novo Nordisk, Roche-Genentech, Sanofi and ViaCyte.