Duke backs safety findings from controversial Avandia trial

June 3 (Reuters) - An independent analysis of a controversial clinical trial appears to back the safety of GlaxoSmithKline Plc’s diabetes drug Avandia, U.S. regulators said on Monday, although scores of smaller trials still raise questions over whether the largely abandoned pill increases heart risks.

Backing for Avandia’s heart safety came from the Duke Clinical Research Institute, a group within Duke University School of Medicine that conducts and analyzes drug studies. The U.S. Food and Drug Administration had commissioned the group three years ago to analyze, or “readjudicate,” data from a large Glaxo-sponsored trial of Avandia called RECORD.

“Overall, the readjudication appears to support the previous observation that in this trial, (Avandia) was not associated with an increased incidence of major adverse cardiovascular events,” FDA staff members said in documents released by the agency on Monday.

The 500-page FDA briefing document was released ahead of a planned joint meeting on Wednesday and Thursday of two medical advisory panels to the FDA, where the Duke group’s findings and restrictions on sales of the medicine will be discussed.

The FDA in September 2010 placed severe restrictions on use of Avandia due in part to ambiguity about possible increased risk of heart attack and stroke seen in RECORD, as well as a review of dozens of other studies.

The drug was placed in a highly unfavorable light in 2007 after Dr. Steven Nissen, head of cardiology at the Cleveland Clinic, reported results of a so-called meta-analysis in which data from 42 studies was pooled and analyzed, showing a 43 percent increased risk of heart attack from Avandia.

Avandia had been one of Glaxo’s top-selling medicines with sales of $3.2 billion in 2006. But sales plunged following the negative publicity.

Avandia was withdrawn from the market in Europe in 2010 and is now taken by only 3,000 Americans, down from 120,000 before the restrictions were put in place.

The FDA commissioned the Duke group to analyze results of the RECORD trial to better assess Avandia’s safety and to examine criticisms that the trial was poorly designed and that its data was mishandled.

RECORD showed a non-statistical increase in the risk of heart attack, compared with two other types of widely used oral diabetes drugs - metformin and sulfonylureas. There was also a non-statistical reduction in risk of stroke and death among those taking Avandia.

But criticism at the time of RECORD’s design and allegations that its data was mishandled cast doubt on the trial’s general conclusion that Avandia did not significantly raise the risk of cardiovascular problems.

The drug, also known by its chemical name rosiglitazone, is available under tight restrictions of a so-called Risk Evaluation and Mitigation Strategies program.

During the two day meeting, the FDA advisers could recommend removal, continuation, or modification of the restrictions, or full withdrawal of Avandia from the U.S. market.

Even if the FDA committee declares the drug safe and lifts usage restrictions, the London-based drugmaker said it was not likely to put its marketing muscle behind Avandia again.

“I don’t think from a commercial prospect it will be promoted again in the way it was long ago,” Glaxo spokeswoman Mary Ann Rhyne said.

Murray Stewart, senior vice president in Glaxo’s clinical research group, said the drugmaker did not take part in re-evaluating data from the RECORD trial.

“Thousands of documents went to Duke,” he told Reuters. “Our role was to be the postman and send everything to Duke.”