This article is more than 5 years old. Information may no longer be current.
FDA advisory verdict: Continue rosiglitazone
For several months since the publication of the meta-analysis of rosiglitazone and ischemic disease by Nissen and Wolski in The New England Journal of Medicine, we have watched with growing concern the considerable debate regarding the methodology of the analysis as well as the questioning regarding ischemic risk and rosiglitazone. Clearly, there had been evidence of volume expansion with this class of antidiabetic agents with an attendant increase in edema and, in rare cases, a worsening of congestive heart failure. But there had been no evidence of increased risk of ischemic disease via increases in surrogates for atherogenesis, such as biomarkers, or architectural studies, such as carotid artery wall thickness studies.
Indeed, the entire class of PPAR-gamma agonists has been associated with improvements in these surrogates for atherosclerosis. Furthermore, the data from two published randomized multi-center controlled trials, DREAM and ADOPT, have suggested little adverse cardiovascular effect other than small increases in edema and in congestive HF.
Because of the wide usage of this class of the PPAR-gamma oral antidiabetic agents, especially rosiglitazone, or Avandia, these newly published findings gave rise to considerable public and governmental concern. This led to a public, FDA-sponsored joint meeting on July 30 of the Endocrinologic and Metabolic Drugs and Drug Safety and Risk Management advisory panels.
|
|
Arriving at a decision
At this lengthy meeting, medical safety officers from the FDA as well as the drug sponsor (GlaxoSmithKline) and selected members of the public made detailed presentations regarding the available databases, the published analyses, and recently performed but as-yet unpublished analyses regarding ischemic risk in rosiglitazone-treated patients. Previously unpublished analyses regarding pioglitazone were also presented.
A wide range of opinions and concerns soon developed in the hearing. Although several of the medical safety officers had considerable concern regarding rosiglitazone, this was not completely shared by members of the advisory committee, who seemed much more skeptical regarding the analytical tool of meta-analyses, especially in this setting, than did the medical safety officers. Committee members also seemed much more disturbed by the relative lack of any safety signal in the randomized control trials and the failure of these studies to confirm the findings of the meta-analyses. The findings of both the GlaxoSmithKline and Nissen meta-analyses seemed to be driven by substudies using rosiglitazone with insulin and in the elderly — populations at increased risk for ischemia even without the issue of rosiglitazone therapy.
In view of the basis for some degree of apparent ischemic risk associated with rosiglitazone therapy, no matter how poorly characterized, the advisory board voted that there seemed to be evidence for such risk. However, the board voted overwhelmingly in favor of continuing rosiglitazone on the market, while further studies, especially the randomized controlled trials designed to define risks and benefits accruing from rosiglitazone therapy, should continue to completion. The completion of these trials will definitively define the nature and amount of risk with regard to cardiovascular endpoints, and also provide an estimate of offsetting clinical benefits not available from the short-term trials already analyzed.
Thus, the committee seemed to strike a carefully reasoned compromise expressing considerable doubt regarding the meta-analysis of small short-term trials using ill-defined endpoints, while smoothing the path for the ongoing randomized and controlled safety and efficacy trials. This result should aid the FDA considerably in their ongoing effort to understand what, if anything, is the impact of rosiglitazone on the vasculature of patients with type 2 diabetes. As was carefully pointed out during the meeting by an National Heart, Lung, and Blood Institute representative, diabetes is a complex chronic illness lasting for many years. Different agents may have differing effects upon this illness at different stages of the disease, and only long-term trials can best define risks and benefits in such clinical circumstances.
So what are clinicians to do? In the likely absence of any drastic action by the FDA, patients on rosiglitazone without complications from that agent should continue on that agent. After all, diabetes is a serious, life-threatening illness whose treatment is an inherent benefit. No agent is free of adverse effect. Indeed, even in the meta-analyses, the differences noted were between rosiglitazone and placebo; no differences were noted between rosiglitazone and other alternative oral treatments for type 2 diabetes.
It is therefore clear that in the absence of definitive guidance to the contrary, rosiglitazone should continue to be used except in patients with unstable or NYHA class III-IV HF or with unstable angina. Rosiglitazone use in patients on high-dose insulin therapy should also be carefully monitored. In addition, a considerable degree of patience and calm will be needed while we await more definite outcome studies. Since type 2 diabetes is a progressive disease of worsening beta cell function, rosiglitazone offers a clear benefit with respect to highly durable glucose control, which is unavailable with alternative therapies. We should continue in our efforts to satisfactorily treat diabetes with all the agents at our disposal.
Alan J. Garber, MD, PhD, is Professor, Departments of Medicine, Biochemistry and Molecular Biology and Cellular and Molecular Biology at Baylor College of Medicine. He is a member of the Cardiology Today Editorial Board.