PHILADELPHIA – Daily glargine use among patients with type 2 diabetes produced similar clinical outcomes to those of patients using liraglutide, but with lower associated costs, according to a retrospective analysis of real-world use that was presented at the annual scientific sessions of the American Diabetes Association.
When liraglutide or glargine was added to a combined metformin-plus-sulfonylurea therapy in a previous phase III trial, liraglutide demonstrated a greater reduction in hemoglobin A1c than did glargine (Diabetologia 2009;52:2046-55).
To date, however, no real-world comparative data have been published comparing use of injectable therapy with insulin glargine vs. the glucagonlike peptide–1 agonist liraglutide among patients with type 2 diabetes, said Dr. Philip Levin, director of the diabetes center at Mercy Medical Center, Baltimore.
Administrative claims data were analyzed from a managed care database comprising about 50 U.S. health care plans and 107 million patients. Of those, investigators identified 967 adults who had type 2 diabetes and had initiated injectable pen therapy with either glargine (n = 557) or liraglutide (n = 410) between January and June 2010. All patients who had baseline HbA1c levels less than 7% were excluded, because it’s possible that they were started on liraglutide for weight loss rather than glucose control, Dr. Levin noted.
Because there were significant differences between the two groups at baseline – glargine initiators were older, sicker, less obese, and more likely to be male, and had higher HbA1c levels, among other differences – the 336 eligible patients were propensity matched to make the two groups more comparable. After that was done, the 168 glargine and 168 liraglutide patients did not differ significantly in sex (46% of glargine and 44% of liraglutide patients were women), age (53 years for both groups), mean Charlson comorbidity index (0.29 and 0.36, respectively), or other baseline characteristics.
The group taking glargine stayed on therapy significantly longer than did those on liraglutide (279 vs. 257 day).
Reductions in HbA1c from baseline did not differ significantly between the glargine and liraglutide groups, decreasing from 8.96% to 7.94% at 1 year with glargine, and from 8.76% to 7.81% with liraglutide (reductions of 1.02 and 0.95 percentage points, respectively). The average dose for glargine was 27.4 U/day, and for liraglutide was 1.16 mg/day.
Hypoglycemia was more common with glargine (7.7% vs. 4.7%), but this difference did not reach statistical significance. Hypoglycemia requiring emergency department or inpatient treatment occurred in just 1.1% of both groups.
Health care costs were significantly lower with glargine. Study drug costs were $1,198 for glargine vs. $2,784 for liraglutide; diabetes-related drug costs were $2,958 vs. $3,988, respectively; and total diabetes-related costs were $5,653 vs. $7,976. Each of those differences was statistically significant, Dr. Levin said.
These findings will be further explored by the planned INITIATOR study, a large-scale hybrid prospective/retrospective real-world study, he noted.
The study was funded by Sanofi-Aventis. Dr. Levin disclosed that he is on the advisory panel for that company, as well as for Novo Nordisk, the maker of liraglutide. He also consults for, receives research support from, or is on the speakers bureau for a long list of other manufacturers of diabetes-related products.