FDA Advisers Call for Cardiovascular-Safety Assessment of New Diabetes Drugs

By Peggy Peck
ROCKVILLE, Md., 5 july2008 -- Type 2 diabetes drugs aiming to pass FDA muster should be required to consider long-term cardiovascular safety as part of the approval mix, according to agency advisers. The Endocrinologic and Metabolic Drugs Advisory Committee voted 14-2 to recommend that the FDA add long-term cardiovascular-safety trials to the regulatory burden. Most of the advisers said the risk assessment should be completed prior to drug approval, but the panel did not issue a clear-cut recommendation on that question. Currently, type 2 diabetes drugs are approved on the basis of a surrogate -- the ability to lower hemoglobin A1c, because it was believed that lowering glucose would translate into clinical benefit.
This notion has been challenged over the past year, first by a series of meta-analyses of rosiglitazone (Avandia), a drug that demonstrated significant efficacy in lowering glucose.
But beginning with a meta-analysis of published trial data by Steven Nissen, M.D., of the Cleveland Clinic and colleagues in May 2007, which revealed an increased risk of myocardial infarction among patients treated with rosiglitazone, a number of studies has linked the drug to increased risk of MI. As a result, the FDA added additional warnings to the rosiglitazone label. Pioglitazoane (Actos), a drug in the same class, already carried boxed warnings about congestive heart failure.
More recently, three trials reported at the American Diabetes Association meeting last month -- ACCORD, ADVANCE, and VADT -- found that aggressive glucose lowering did not reduce cardiovascular events.At a press briefing following the FDA advisory panel meeting, John Jenkins M.D., director of the FDA's Office of New Drugs, said the agency will consider several options for cardiovascular-risk assessment, including asking drug makers to continue patients as randomized for up to 18 months.
As it stands now, a typical trial design collects randomized data for a limited period -- 12 weeks is common -- to assess efficacy, and then crosses patients over to the investigational agent to gain long-term data.
Other possibilities would to require companies to include a single prospective cardiovascular-risk assessment trial as part of the NDA application process, or to permit a pooling of data from ongoing clinical trials.
But pooling was probably not a viable option for drugs in the approval pipeline because companies are not currently adjudicating cardiovascular risk, said Mary Parks, M.D., director of the agency's Division of Metabolism and Endocrinology Products.
Robert Temple, M.D., director of the FDA's Office of Medical Policy, said the same cardiovascular-risk assessment approach might be used to assess other drugs that are currently evaluated on the basis of a surrogate marker, such as hyperlipidemia agents and NSAIDs.
The panel considered, but rejected, the concept that a diabetes drug should be required to show cardiovascular benefit.