JAMA: Docs should consider switching patients off Avandia
Switching prescription drugs within the same therapeutic class has become increasingly common with tiered drug plan formularies. Therefore, patients and physicians may see the switch from rosiglitazone (Avandia, GlaxoSmithKline) to pioglitazone (Actos, Takeda Pharmaceuticals) as “simple, straightforward, and appropriate, because patients are receiving a similar drug without the increased risk of MI,” according to commentary published Feb. 8 in the Journal of the American Medical Association.

At present, there is insufficient evidence from clinical trials to support the choice of one specific class of glucose-lowering agents over another with respect to long-term outcomes, wrote Kasia J. Lipska, MD, and Jonathan S. Ross, MD, both from Yale University School of Medicine in New Haven, Conn. “The prevailing assumption has been that the degree to which these agents reduce glucose levels will determine how well they reduce the risk of complications, including major cardiovascular events, regardless of the medication strategy chosen. Recent experience with rosiglitazone suggests otherwise.”

As a result, the FDA recently began to require data on cardiovascular outcomes for approval of diabetes agents. Last July, a contentious meeting of the FDA’s Endocrinologic and Metabolic Drugs Advisory Committee and the Drug Safety and Risk Management Advisory Committee voted to keep Avandia on the U.S. market—most with further restrictions and warnings. Later, in September 2010, the agency decided to keep GSK’s controversial diabetes drug on the market, but with significant restrictions associated with its use.

In lieu of long-term clinical outcomes data for most glucose-lowering agents, the current consensus statement of the American Diabetes Association and European Association for the Study of Diabetes makes therapeutic recommendations based on surrogate outcomes, such as effectiveness of glucose lowering, safety, tolerability and expense.

According to the commentary authors, these comparisons tend to favor metformin; therefore, metformin is recommended as the first-line agent for treatment of diabetes along with lifestyle modification. “Metformin is safe and well tolerated and reduces blood glucose levels effectively without causing hypoglycemia or weight gain. Moreover, in a small subgroup of the U.K. Prospective Diabetes Study, metformin reduced the risk of cardiovascular events,” they wrote.

The consensus statement recommends either insulin or sulfonylureas (depending on glucose level) as preferred second-line medications with the strongest supportive evidence if metformin is contraindicated or does not achieve desired glucose control. Sulfonylureas are as effective as metformin for glucose lowering and insulin can be used to manage glucose to achieve a more tightly controlled target level, wrote Lipska and Ross; however, both sulfonylureas and insulin have potential adverse effects, including weight gain and hypoglycemia.

Thiazolidinediones (TZDs), the class to which Avandia and Actos belong, are not recommended as part of the well-validated core therapy for diabetes because of their adverse effect profile, the authors pointed out. Both rosiglitazone and pioglitazone increase the risk of significant weight gain and lower-extremity edema and have been associated with accelerated bone loss and fractures among women. Moreover, both medications now carry an FDA black box warning after multiple studies have confirmed that they increase the risk of heart failure and related adverse events.

“Although the magnitude of the risk associated with each agent varies by degree among different studies, the most recent FDA meta-analysis confirmed that the risk was elevated for both agents—nearly 1.5 greater odds for pioglitazone and double for rosiglitazone,” Lipska and Ross wrote.

The authors also pointed to evidence of increased heart failure events with TZDs.

Yet, “physicians rapidly adopted TZDs for the treatment of diabetes,” the commentary stated. Although first introduced to the market in 1997, by 2005 TZDs were prescribed during 34 percent (estimated at 11.2 million) of treatment visits for type 2 diabetes in the U.S. (Diabetes Care 2010;33(4):823–825). Even after the class-wide FDA black box warning for congestive heart failure was added in 2007, pioglitazone use was largely unchanged and it was prescribed during an estimated 5.8 million treatment visits for type 2 diabetes between 2008 and 2009 (Diabetes Care 2010;33(4):823–825).

The use of TZDs as monotherapy, which should have a very limited role in diabetes treatment according to treatment guidelines, accounted for 25 percent of all TZD treatment visits (Arch Intern Med 2008;168(19):2088–2094).

“Given its substantive use, in the U.S., more money is spent on TZDs than on any other major class of oral glucose-lowering agents. Thiazolidinediones are brand medications and as such are quite expensive,” the authors wrote. In 2007 alone, an estimated $4.2 billion was spent on TZDs compared with an estimated $2.5 billion spent on the preferred generic choice, metformin (Arch Intern Med 2008;168(19):2088–2094).
“Given the increasing prevalence of diabetes and the rising costs of diabetes therapies, more comparative-effectiveness data are needed on glucose-lowering agents and their ability to improve both patient-important outcomes (such as microvascular and macrovascular complications and mortality) as well as patient-reported outcomes,” Lipska and Ross concluded. “Effects on inflammation, lipid profiles and other surrogate outcomes may provide insights into mechanisms of action, but should not serve as the basis for therapeutic decisions. Ideally, physicians and patients should be guided by definitive evidence when choosing a specific medication.”