SILVER SPRING, MD. – A reanalysis of data from a controversial cardiovascular safety study of rosiglitazone was well-conducted and provided results that were "highly similar" to the original results of the study, according to Food and Drug Administration officials who spoke during the first day of a 2-day advisory panel meeting focusing on the results of the reanalysis.
The meeting – of the FDA’s Endocrinologic Drugs and Drug Safety and Risk Management advisory committees – is being held to discuss the results of the independent readjudication of the results of the RECORD (Rosiglitazone Evaluated for Cardiovascular Outcomes in Oral Agent Combination Therapy for Type 2 Diabetes) trial, a prospective, open-label, noninferiority study conducted by rosiglitazone manufacturer GlaxoSmithKline in response to a request by European drug regulators to address concerns over possible CV risks of the drug.
RECORD is the only large randomized controlled CV outcomes study of rosiglitazone, and overall, the readjudication of the study "appears to support the previous observation that, in this trial, rosiglitazone was not associated with an increased risk of death or major adverse cardiovascular events," Dr. Karen Mahoney, diabetes team leader in the FDA’s division of metabolism and endocrinology products said at the meeting.
How the panel interprets these data could impact the availability of rosiglitazone, which is used to treat about 3,400 patients in the United States, down from about 117,000 patients the year before the FDA instituted a risk evaluation and mitigation strategy (REMS) to limit its use because of cardiovascular safety concerns. On the second day of the meeting, June 6, the panel will discuss these data and will vote on whether the drug should remain on the market with the same restrictions, whether the REMs and other restrictions should be lifted or modified, or whether it should be taken off the market altogether based on the readjudicated results and other available data. In Europe, rosiglitazone is no longer marketed because of concerns over CV risks.
There have been longstanding concerns that rosiglitazone may be associated with an increased risk of ischemic CV events, but the data have been mixed and limited by study designs and other issues.
But in 2007, a meta-analysis of 42 studies showed that rosiglitazone increased the risk of MI by up to 43% and the risk of CV death by up to 64% (N. Engl. J. Med. 2007;356:2457-71); the lead author was Dr. Steven Nissen, a vocal advocate for removing rosiglitazone from the market.
Considering these and other data, at a meeting of the same panels in July 2010, the majority of the panelists agreed that the data raised significant safety concerns regarding the risk of rosiglitazone and most recommended that the use of the drug be limited, but a significant proportion recommended that it be taken off the U.S. market. In September 2010, the FDA announced that rosiglitazone would remain available, with the REMS, limiting the drug to patients who were already on rosiglitazone, or for patients whose blood glucose could not be managed adequately on other medications and, could not take the only other approved thiazolidinedione (TZD) pioglitazone, "for medical reasons."
At that time, the FDA asked GSK to commission an independent readjudication of the RECORD data, which found a nonsignificant increase in MI in the rosiglitazone arm, along with a nonsignificant reduction in the risk of stroke and all-cause mortality. But the reliability of the study has been questioned because of the study design and other issues such as missing data, and in September 2010, when the FDA announced that rosiglitazone would be marketed under a risk evaluation and mitigation strategy, the agency requested that the company commission an independent readjudication of the data to determine if the results could reliably be used to evaluate the drug’s CV safety.
RECORD enrolled about 4,500 patients with type 2 diabetes in 364 centers in 25 countries outside of the United States, comparing CV outcomes among those on rosiglitazone with metformin or a sulfonylurea and metformin. Because of limitations of the study, the reliability of these results has been questioned by experts within and outside the FDA. GSK commissioned the Duke Clinical Research Institute (DCRI) to conduct the reanalysis, which involved blinded adjudication of more than 2,200 deaths, MIs, and strokes. The results, presented at the meeting on June 5, were consistent with the original RECORD results, with no significant differences in the composite of CV death, MI, or stroke; numerically fewer CV deaths and numerically fewer strokes in the rosiglitazone arm; and a lower risk of all cause death. There were more MIs in the rosiglitazone arm, but the difference was not significant.