Rosiglitazone monotherapy for type 2 diabetes mellitus--too soon to ADOPT?

BACKGROUND Type 2 diabetes mellitus is a progressive disease, which requires intensification of treatment regimens over time in order to maintain target glucose levels. OBJECTIVE To compare rosiglitazone, metformin and glyburide monotherapies for the treatment of type 2 diabetes mellitus. DESIGN AND INTERVENTION A Diabetes Outcome Progression Trial (ADOPT) was an international, multicenter, randomized, double-blind, head-to-head study of patients with newly diagnosed type 2 diabetes mellitus. Eligible patients were aged 30-75 years, with a fasting plasma glucose (FPG) level 7-10 mmol/I, and no previous pharmacotherapy for type 2 diabetes mellitus. Patients with congestive heart failure, uncontrolled hypertension, angina, renal impairment, or liver disease were excluded. Patients were randomly allocated to receive one of three monotherapies: 4 mg rosigiitazone daily, 500 mg metformin daily, or 2.5 mg glyburide daily. Doses were adjusted to maintain an FPG level <7.8 mmol/l. OUTCOME MEASURES The primary outcome measure was the interval from baseline to treatment failure (defined as an FPG level >10 mmol/I). Secondary outcome measures included HbA1c levels, insulin sensitivity, β-cell function, weight, and adverse events. RESULTS The study enrolled 1,456 patients in the rosiglitazone group, 1,454 in the metformin group and 1,441 in the glyburide group. Median treatment duration was 3.3 years in the glyburide group and 4.0 years in the rosiglitazone and metformin groups. The study completion rates for rosiglitazone, metformin and glyburide were 63%, 62% and 56%, respectively. The 5-year Kaplan-Meier cumulative incidence of treatment failure was 15% with rosiglitazone, 21% with metformin and 34% with glyburide. The risk of treatment failure was reduced by 32% and 63% when rosiglitazone was compared with metformin and glyburide, respectively. Progression to an FPG level >7.8mmol/I occurred in 15.4% of patients treated with rosiglitazone, 24.4% of patients treated with metformin and 33.3% of patients treated with glyburide. Annual rates of increase in the FPG levels were 0.04 mmol/I (rosiglitazone), 0.15mmol/I (metformin) and 0.31mmol/I (glyburide). Annual rates of increase in the HbA1c levels were 0.07% (rosiglitazone), 0.14% (metformin) and 0.24% (glyburide). At study end, 40% of patients in the rosiglitazone group had HbAlc levels <7%, compared with 36% and 26% of patients in the metformin and glyburide groups, respectively. Insulin sensitivity improved in the rosiglitazone and metformin groups, but not in the glyburide group. β-Cell function declined by 2.0% (rosiglitazone), 3.1% (metformin) and 6.1% (glyburide). Treatment with rosiglitazone was more frequently associated with weight gain, congestive heart failure, edema, high LDL cholesterol levels, and non-vertebral fractures than treatment with metformin or glyburide. Metformin treatment had the highest rate of gastrointestinal events, whereas glyburide treatment had the highest rate of hypoglycemia but the lowest risk of cardiovascular events. CONCLUSION Although rosigiitazone was the most effective monotherapy in this study, cost and adverse events could limit its use as an initial treatment for type 2 diabetes mellitus. [ABSTRACT FROM AUTHOR]