Clinical Trials: ADOPT and CHICAGO
Two important studies on thiazolidinediones (TZDs), a class of insulin sensitizers, were published in November and December of 2006. The two currently available TZDs are rosiglitazone (Avandia) and pioglitazone (Actos). While it isn't correct to apply the study results of one drug to another drug, overall both TZDs seem to have the same mechanism of action and effects. The ADOPT trial was a five-year study that included 4,360 adults with newly diagnosed type 2 diabetes. The participants were randomized, or placed in separate groups, to receive either metformin, Avandia (rosiglitazone), or glyburide (a sulfonylurea). After five years, 40% in the Avandia group had an A1c <7%, compared to 36% in the metformin group and 26% in the glyburide group. This suggests that glyburide doesn't last as long compared to Avandia. Patients also maintained A1c <7% the longest with Avandia (60 months), followed by metformin (45 months) and glyburide (33 months). Thus, rosiglitazone provides more stable diabetes control than the other two drugs. However, Avandia also caused significant weight gain. Patients on the TZD gained 4.8 kg (10.5 lb) over five years, while those on metformin lost 2.9 kg (6.4 lb) and those on glyburide gained 1.6 kg (3.5 lb). Weight gain is clearly undesirable, and in previous studies Avandia has been linked to another cardiovascular complication, congestive heart failure, although in ADOPT, there was no difference shown. The CHICAGO study was a 72-week trial that included 462 adults with type 2 diabetes. The participants received either Actos (pioglitazone) or glimepiride, a sulfonylurea, to lower their fasting plasma glucose to <140 mg/dL (<7.8 mmol/L). The study was designed to look at the drugs' effect on a commonly used marker of atherosclerosis (plaque on the arteries) called carotid intima-media thickness (CIMT). Increases in CIMT are associated with higher risk for cardiovascular disease. At the end of the study, mean CIMT decreased by 0.001 mm in the Avandia group but increased by 0.012 mm in the glimepiride group. If we accept that CIMT is a valid surrogate measure of cardiovascular risk, then these results are highly favorable for Avandia. However, the study was not big enough to measure cardiovascular risk directly, so we don't actually know if risk was reduced. It's possible that Avandia improves CIMT but does not affect the more important endpoint: cardiovascular disease. Still, the TZD did produce more sustained improvements in A1c than glimepiride and raised levels of HDL (good cholesterol) while glimepiride did not.
The bottom line: TZDs produce more durable declines in A1c levels than metformin and sulfonylureas but cause significant weight gain and are more expensive. In their favor, they are also more beneficial to the vascular system than sulfonylureas and may reduce the risk of cardiovascular disease. Sulfonylureas are not a good choice for oral anti-diabetic therapy because they cause weight gain and do not provide sustained glycemic benefits. Overall, monotherapy does not keep patients at goal very well, so the biggest takeaway is that combination therapy is essential in type 2 diabetes.
(Kahn S. et al. "Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy." NEJM. 7 Dec 2006. 355(23):2427-2443. Mazzone T. et al. "Effect of Pioglitazone Compared With Glimepiride on Carotid Intima-Media Thickness in Type 2 Diabetes." JAMA. 6 Dec 2006. 296(21):2572-2581.)
