GAITHERSBURG, MD. — Panelists at a joint Food and Drug Administration advisory committee meeting voted 20-12 with 1 abstention that the diabetes drug rosiglitazone can continue to stay on the market, but they urged the agency to impose new restrictions and warnings on use of the drug.
They further voted that if rosiglitazone remains on the market, that the FDA-mandated TIDE trial, examining the comparative cardiovascular safety of rosiglitazone, pioglitazone, and standard-of-care management of type 2 diabetes, be continued.
The Endocrinologic and Metabolic Drugs Advisory Committee and Drug Safety and Risk Management Advisory Committee were also asked by the FDA to consider whether new safety data showed that rosiglitazone (Avandia) increased the risk of cardiovascular events and death.
FDA Commissioner Margaret A. Hamburg addressed the two committees at the outset, requesting that they “cut through” the vast amounts of noise and “focus on the evidence.” She urged them to “follow your science wherever it leads and the rest will fall into place.”
Dr. Hamburg noted that usually, FDA opinions “line up nicely” when they are presented to advisory committees. In this case, they do not, she said, and advised panelists to factor all the divergent perspectives into their thinking.
Rosiglitazone maker GlaxoSmithKline presented a review of clinical and observational data, including the Rosiglitazone Evaluated for Cardiovascular Outcomes in Oral Agent Combination Therapy for Type 2 Diabetes (RECORD) trial, which was undertaken to further explore cardiovascular safety signals seen in earlier trials (Lancet 2009;373:3125-35). Dr. Philip Home of Newcastle University, London, chair of the RECORD steering committee, said that according to GSK's analysis, the cardiovascular events were similar between rosiglitazone and the controls, metformin and sulfonylureas.
Dr. Murray Stewart, vice president of clinical development for GSK, said that the drug was a much-needed option for patients because it offered durable, long-lasting glucose control.
But Dr. Steven Nissen, who is chair of cardiovascular medicine at the Cleveland Clinic and has conducted several meta-analyses of the rosiglitazone data, said that the RECORD study was biased and flawed and that it should not be used to judge the drug's suitability for marketing.
An FDA reviewer who delved into the RECORD study, Dr. Thomas A. Marciniak, seemed to agree. Dr. Marciniak of the FDA's Division of Cardiovascular and Renal Drug Products said that his review of a random sample of case reports from RECORD submitted by GSK were sloppy. Therefore, the rosy safety estimates could not be trusted, he said.
“From the study conduct biases, any estimates of risks you get from RECORD should not be viewed as precise statistical estimates, but as lower bounds; they could actually be higher,” Dr. Marciniak said.
Dr. Ellis F. Unger, deputy director of the Office of Drug Evaluation–I, said that, although Dr. Marciniak had uncovered a number of concerns about the quality of RECORD's data, his approach to evaluating the data was not typical FDA procedure. Dr. Unger's reading of the data was that RECORD data was not as definitive as it could have been, but that it did not show an increased risk of cardiovascular death or heart attack, he said.
Keep Avandia Available, Within Limits
Of the 20 panel members who voted to keep rosiglitazone on the market, 10 voted for the option that marketing could continue if more warnings were added to the label and restrictions were placed on the use of the drug. Only three panelists voted for the option to continue marketing with no changes to the current label, which includes a boxed warning about an increased risk of myocardial ischemic events associated with rosiglitazone in a meta-analysis of 42 clinical trials, but adds that the data on this risk are “inconclusive.” The remaining seven panelists voted for the option to allow marketing of the drug but with additional warnings in the label.
Dr. Sanjay Kaul, a panel member and cardiologist from the University of California, Los Angeles, was among those who voted for continued marketing but with some new labeling. “Was there clear and convincing evidence that incriminated this drug? No,” he commented. But Dr. Kaul said that rosiglitazone should not be available as a first-line therapy and that it should be contraindicated in patients with any stage of heart failure, the elderly, and those taking nitrates.
The 12 panelists who voted to withdraw rosiglitazone from the market said that there was not enough evidence proving absolute harm but that the data that had been presented was still concerning. “I don't see where this drug needs to be on the market anymore,” said panelist Dr. Morris Schambelan of the University of California, San Francisco. Dr. Schambelan added that pioglitazone is an acceptable alternative; he was a member of the Endocrinologic and Metabolic Drugs Advisory Committee in 2007 that voted to keep rosiglitazone on the market.
Dr. Peter Savage, of the Division of Diabetes, Endocrinology and Metabolic Diseases at the National Institute of Diabetes, Digestive, and Kidney Diseases, was among those who oscillated between voting to take it off the market or to keep it on the market with restrictions. He ultimately went with the latter option. “I think the evidence of potential harm associated with rosiglitazone is stronger now” than in 2007, and “importantly, the evidence about pioglitazone is substantially greater,” he explained.
Evaluating the Risks
Before voting on whether to withdraw the drug, 18 panel members said they found the data sufficient to suggest that rosiglitazone increased the risk of ischemic cardiovascular events relative to non-thiazolidinedione diabetic agents. Six panelists said it did not increase that risk, and nine said that they were not able to make a finding. Next, 21 committee members voted that rosiglitazone's risks were increased relative to competitor pioglitazone (Actos, Takeda Pharmaceuticals). Three said there was no increased risk, and nine said they could not make a determination.
Panelists were not as convinced that rosiglitazone increased the risk of death from any cause, whether compared with older antidiabetic agents or with pioglitazone. The majority said the evidence was not sufficient to make a determination.
Some panelists said they were still concerned, however, largely because of an analysis of Medicare claims data conducted by the FDA's associate director for science and medicine, Dr. David J. Graham, which found an increased risk of stroke (hazard ratio 1.27), heart failure (1.25), and death (1.14), but not of acute MI (1.06). The risk for the composite end point of all these outcomes combined was significantly higher in the rosiglitazone patients (HR 1.18), compared with pioglitazone patients, especially as patients aged. The results of that study were published in June (JAMA 2010;304 [doi:10.1001/jama.2010.920]).
And panelist Dr. Curt Furberg of Wake Forest University Health Sciences, Winston-Salem, N.C., said, “I do have significant safety concerns about cardiovascular mortality.” Dr. Furberg voted to take rosiglitazone off the market because of what he said were serious adverse effects that overwhelmed any benefit, which “is modest at best and there are other treatment alternatives.” Indeed, several panel members voiced that they voted for withdrawal because although the signal for cardiovascular harm was weak, so too was the evidence of any unique benefits of rosiglitazone.
TIDE's Fate Tied to Rosiglitazone's
Regarding whether GlaxoSmithKline should proceed with enrollment in the TIDE (Thiazolidinedione Intervention With Vitamin D Evaluation) trial, 20 panelists voted yes, 10 voted no, 2 abstained, and 1 did not vote. Most of the panelists who voted yes said that if rosiglitazone remains on the market, TIDE must proceed in order to give answers.
Dr. Hertzel Gerstein of McMaster University in Hamilton, Ontario, who helped design the TIDE study, said in a presentation that the trial was necessary and completely ethical, despite what critics alleged previously during the meeting.
The TIDE trial was mandated by the FDA in the wake of earlier safety concerns. But the FDA's Dr. Graham offered his personal opinion that the TIDE trial was unethical, because it did not properly inform patients that it was seeking to determine if there was an increased risk of cardiovascular events or death with rosiglitazone. “The presence of informed consent does not render an unethical trial ethical. Consent does not make exploitation ethical,” he said.
In the wake of these allegations, Dr. Gerstein vigorously defended TIDE. First, he said, it is a randomized controlled trial that will compare rosiglitazone to pioglitazone to placebo. As an additional benefit, it will answer a question that no one would pay to study—whether vitamin D has any effect on malignancies, Dr. Gerstein said.
The TIDE study aims to enroll 16,000 patients worldwide; that is large enough to accurately determine an event rate, he said. The primary outcome will be whether addition of a thiazolidinedione reduces myocardial infarction, stroke, or cardiovascular death when compared with placebo.
Dr. Gerstein criticized the meta-analyses that have largely informed the debate over rosiglitazone safety so far and were debated by the panel the day before. “History has shown that we can be seriously, seriously misled by the types of data that were discussed yesterday,” he said.
After the committee voted, Dr. Gerstein said in an interview that the panel “clearly voted that this study should be done, and we [TIDE investigators] agree with them. This is a superiority study to see whether these drugs can prevent or reduce cardiovascular events or what the risk vs. benefit is. We will discuss this among the TIDE leadership and barring some other decision we will continue doing the study.”
Expert Groups Urge Caution
The Endocrine Society, American Diabetes Association and American Association of Clinical Endocrinologists rushed a statement out after the votes, urging patients not to stop taking rosiglitazone without consulting their health care providers.
“Stopping diabetes medications can cause significant harm and result in higher levels of blood glucose that may cause severe short-term health problems and could increase the risk of diabetes-related complications in the long term,” wrote Dr. Robert A. Vigersky, immediate past president of the Endocrine Society.
Dr. Daniel Einhorn, president of AACE, said “[t]he worst outcome would be to not treat diabetes properly, thereby risking its complications.”
The FDA usually follows its panels' advice, but it does not always hew to those opinions.
Dr. Janet Woodcock, director of the FDA's Center for Drug Evaluation and Research, said she and the FDA will evaluate the proceedings and the regulatory options available to them, and will come to a decision “as soon as possible and will announce that publicly.” At press time, no decision had been made.
Disclosures: Members of FDA advisory panels have been cleared of conflicts related to the topic under discussion. Dr. Gerstein stated that he has served as a consultant, given scientific talks, and/or his institution has received research, educational and CME grants from the following companies: Sanofi-Aventis, GlaxoSmithKline, Novo Nordisk, Eli Lilly, Bristol-Myers Squibb, AstraZeneca, Roche, Boehringer Ingelheim, and Bayer.
Elizabeth Mechcatie contributed to this article.
FULL TEXT
