To date there are no direct comparative studies of these agents within the same cohort. Accordingly, caution must be exercised when comparing results of the available data, as they are subject to bias effects of different study populations.
Monotherapy: In two placebo-controlled studies of rosiglitazone monotherapy, HbA1c was 1.5% lower in the treatment vs. the placebo group after 26 weeks on maximal dose rosiglitazone monotherapy (4 mg BID), and was 1.6% lower than placebo in a series of three placebo-controlled studies of pioglitazone monotherapy of 26 weeks duration. As monotherapy for type 2 diabetes, rosiglitazone and pioglitazone thus appear to be roughly equivalent in glucose-lowering potential.
The primary failure rate of the thiazolidinediones is approximately 25%-30%, and has been correlated with low levels of endogenous insulin secretion and C-peptide levels <1.5 ng/mL in nonresponders.The efficacy of thiazolidinediones as monotherapy may be compared to that of the sulfonylurea-metformin combination, although the blood glucose-lowering effects of the thiazolidinediones are more gradual (requiring about 2 weeks for initial effects and up to 3 months for complete effects). An additional important feature of monotherapy with thiazolidinediones is a virtual lack of hypoglycemic potential. As they do not raise (and in fact they often lower) endogenous insulin levels, hypoglycemia with monotherapy is exceedingly rare.
Combination Therapy: There are data indicating the efficacy of combination therapy with sulfonylurea, metformin and as add-on to insulin therapy.
Sulfonylureas: In a multicenter, placebo-controlled study of subjects failing sulfonylurea monotherapy, addition of pioglitazone 30 mg for 16 weeks reduced HbA1c 1.2% (p < 0.05) relative to baseline. In a similar multicenter placebo-controlled study involving addition of rosiglitazone 4 mg/day for 24 weeks to sulfonylurea monotherapy, HbA1c was reduced by 0.9% (p < 0.001) relative to baseline.
Metformin: The effects of thiazolidinediones in combination with metformin are similar. The addition of pioglitazone 30 mg/day to metformin therapy resulted in reduction in HbA1c of 0.6% relative to baseline (p < 0.05 significant only relative to placebo). Similarly, when rosiglitazone 4 mg/day or 8 mg/day was added to metformin therapy (2,500 mg/day) for 26 weeks, HbA1c decreased by 0.56% and 0.78% relative to baseline (p < 0.001). While the magnitude of HbA1c reduction with this combination is generally less than with sulfonylurea combination, the lack of hypoglycemic potential, and the potential for cardiovascular risk reduction and beta cell function preservation make this combination particularly attractive.
Insulin: When 15-mg or 30-mg pioglitazone was added for 16 weeks to patients with type 2 diabetes uncontrolled on insulin monotherapy (median insulin dose 60.5 units daily), HbA1c decreased 1.0% and 1.26%, respectively, relative to baseline (p < 0.05 significant only relative to placebo). Additionally, 16% of subjects on the 30-mg dose had a >25% reduction in their average daily insulin dose at the end of the study. When rosiglitazone 4 mg BID was added to insulin monotherapy for 26 weeks, HbA1c was similarly reduced by 1.2% relative to baseline (p < 0.006 significant only relative to placebo), and average daily insulin dose was reduced by 9 units; however, as of the February 2001 rosiglitazone package insert, insulin combination therapy is not indicated due to an increased incidence of congestive heart failure.