FDA advisory panel votes to keep rosiglitazone on market, but not without caveats

Key Points

An FDA advisory panel voted nearly unanimously to recommend that rosiglitazone (Avandia, GlaxoSmithKline) should remain on the market despite being in overwhelming agreement that the drug increases the risk of ischemic heart disease.

This advisory meeting was originally scheduled to take place in Gaithersburg, Maryland, in November 2007 but was moved up to July 30 because the results from 3 meta-analyses became available; these results demonstrated possible increased cardiovascular (CV) risks with rosiglitazone use.

LABELING CHANGE LIKELY

The recommendation for continued availability came even though the advisory panel voted 20 to 3 that the data made available to them supported the conclusion that rosiglitazone did indeed increase the risk of cardiac ischemic events.

FDA officials did not immediately indicate what action would be taken based on the advice they received at the meeting, but they did say that a labeling change was probably warranted. "We heard fairly clear recommendations that some additional labeling was needed, and I often heard the word 'warning,' but I did not uniformly hear the words 'boxed warning,'" said Robert J. Meyer, MD, FDA, Center for Drug Evaluation and Research (CDER), director of the Office of Drug Evaluation II. "I don't think that we can say without rigorous internal discussion whether it will be a boxed warning, contraindications, or what," Dr Meyer said.

RELIABILITY OF SHORT-TERM TRIALS

Representatives from GlaxoSmithKline spoke at the meeting and emphasized the short duration of the trials included in the 3 meta-analyses in which rosiglitazone was demonstrated to increase the risk of myocardial infarction (MI) by approximately 50% compared with placebo or other oral antidiabetic drugs. GlaxoSmithKline officials noted that these other oral agents, specifically metformin, glimepiride, and pioglitazone, were all associated with early CV risk, but longer-term follow-up revealed either CV benefit or no excess CV risk.

GlaxoSmithKline representatives also pointed to data from the company's controlled clinical trials-RECORD (Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes), ADOPT (A Diabetes Outcome Progression Trial), and DREAM (Diabetes Reduction Assessment with Ramipril and Rosiglitazone Medication)-and the company's epidemiologic studies, which cast significant doubt on the possibility of an increased CV risk with rosiglitazone use.

Specifically, it was GlaxoSmithKline's contention that any increased CV risk with rosiglitazone use is small in absolute terms; that the risk of MI is similar for rosiglitazone compared with other anti-diabetic agents; that the risk of MI is no different between rosiglitazone and pioglitazone when either agent is used as monotherapy, as part of combination therapy, or with insulin; and that fewer strokes are observed with rosiglitazone versus active comparators, said Ronald Krall, MD, senior vice president and chief medical officer at GlaxoSmithKline.

HIGHER RISK FOR SOME PATIENTS

FDA presented the results from its own meta-analysis of 42 controlled, randomized clinical trials of patients with type 2 diabetes; these results demonstrated a 40% increased risk (P=.02) of nonserious or serious myocardial ischemic events with rosiglitazone use. An FDA review of GlaxoSmithKline trials of rosiglitazone was especially noteworthy in that it identified 3 patient subgroups at higher risk of adverse CV events with rosiglitazone use: those also using insulin, those taking nitrates chronically, and those taking ACE inhibitors.

The interaction between rosiglitazone and ACE inhibitors was discovered in an analysis of the DREAM trial. In this trial, the outcome of congestive heart failure was 8 times more likely in patients receiving an ACE inhibitor plus rosiglitazone compared with those receiving an ACE inhibitor only; the risk was 4 times higher for patients receiving the combination compared with patients receiving rosiglitazone alone.

Clifford J. Rosen, MD, acting chair of the Endocrinologic and Metabolic Drugs Advisory Committee, made note of these findings when offering some guidance for the continued use of rosiglitazone. "There are reasons not to use this drug in certain type 2 diabetic patients," Dr Rosen said. "Those indications will likely fall under people who have congestive heart failure or are prone to congestive heart failure, those who have significant cardiovascular disease, particularly if they have been hospitalized previously for myocardial infarction, and I would be concerned about 2 other groups: the groups constantly using nitrates... and long-term users of insulin, who probably have advanced disease."

LACK OF BENEFIT

David Graham, MD, MPH, FDA, CDER, associate director for science and medicine, Office of Surveillance and Epidemiology (OSE), was the drug's most vocal critic. In examining only the placebo-controlled trials, he discovered a 70% increased risk of serious ischemic heart disease with rosiglitazone use. The DREAM trial alone demonstrated an approximate 40% increase in CV risk in a relatively low-risk population, Dr Graham said.

Dr Graham argued that because there is no evidence to support a benefit with rosiglitazone with respect to CV disease, diabetic retinopathy, nephropathy, or neuropathy, and because the drug has no unique short-term benefits related to glycemic control, its continued marketing is not justified. Countering those who argued against withdrawing the agent based on data from a meta-analysis, he said that this thinking "ignores that rosiglitazone has no known health benefit."

David S. Schade, MD, a temporary voting member of the Endocrinologic and Metabolic Drugs Advisory Committee and vice chairman for research at the University of New Mexico, Albuquerque, moved the discussion away from the unknown microvascular benefits provided by thiazolidinediones. "It should not be part of our argument; we know that glucose control lowers microvascular complications." Dr Schade said that a placebo-controlled trial to prove this assertion will never be conducted.

NEW DATA WON'T PROVIDE ANSWERS

Many panel members expressed frustration over the lack of quality data on which to base their decisions; however, Dr Graham pointed out that ongoing randomized, controlled clinical trials will probably fail to deliver the answers needed regarding the CV safety profile of rosiglitazone.

Dr Graham said that trials in progress "won't give us the clarity" regarding the CV safety of rosiglitazone because of design shortcomings in the studies. An interim analysis of RECORD (published in N Engl J Med. 2007;357:28–38), which was designed to assess whether rosiglitazone is noninferior to placebo and comparator drugs with respect to CV safety in patients with type 2 diabetes, did not demonstrate an increased risk of CV events with rosiglitazone use; however, RECORD has been criticized for being underpowered to answer this question, and the trial did not exclude the possibility of a 25% increased CV risk associated with rosiglitazone use.

Two other trials in progress, ACCORD (Action to Control Cardiovascular Risk in Diabetes), a study sponsored by the National Heart, Lung, and Blood Institute (NHLBI) to assess the effect of rosiglitazone on major adverse coronary events in 10,251 patients with type 2 diabetes, and BARI 2D (Bypass Angioplasty Revascularization Investigation 2 Diabetes), which is designed to assess all-cause mortality in patients with type 2 diabetes and stable coronary artery disease (CAD), have similar limitations. For instance, the BARI 2D trial is not randomized or blinded and is studying patients who are receiving several oral antidiabetes medications.

PIOGLITAZONE MAY OFFER CV BENEFITS

Kate Gelperin, MD, MPH, FDA, CDER, OSE, Division of Drug Risk Evaluation, presented preliminary findings from a study conducted by Takeda Pharmaceuticals that demonstrated a significant reduction in the adjusted risk of MI with pioglitazone (Actos, Takeda) relative to rosiglitazone (HR=0.78; 95% CI, 0.63–0.96).

Dr Graham also presented results from a Takeda meta-analysis that demonstrated a strong trend (HR=0.75; 95% CI, 0.55–1.02) towards a reduction in all-cause death, nonfatal MI, and nonfatal stroke favoring pioglitazone over placebo. He called this study "the single most important additional source of data for this hearing." These data support those from PROactive (Prospective Pioglitazone Clinical Trial in Macrovascular Events), a randomized, controlled trial that also demonstrated trends towards a reduction in adverse CV outcomes with pioglitazone use. Dr Graham said that rosiglitazone should be viewed in the context of the pioglitazone findings and called rosiglitazone "an outmoded drug compared with pioglitazone."

Although the FDA advisory panel members weren't able to review the newly presented pioglitazone data prior to the meeting, Gerald Dal Pan, MD, MPH, FDA, CDER, director of OSE, said that "the pioglitazone data were clinically relevant to the issue, so we presented it, with the caveat that they have to undergo FDA review."