US FDA review also shows Glaxo diabetes drug risk

WASHINGTON, May 26 (Reuters) A US Food and Drug Administration preliminary analysis found a level of heart-attack risk with GlaxoSmithKline Plc's diabetes drug Avandia similar to what was reported in a study earlier this week, the agency confirmed.

FDA spokeswoman Julie Zawisza stressed that other data conflicted with that finding, and officials still were weighing whether to take any action.

A Cleveland Clinic analysis released Monday said the drug increased the chances of having a heart attack by 43 per cent.

Speaking on the Senate floor late on Thursday, Republican Sen Charles Grassley of Iowa said preliminary findings from an FDA analysis of Glaxo's clinical trials found a 40 per cent increased risk, while the company puts the figure at about 30 per cent.

''Those numbers seem like a high enough threshold to me for the FDA to warn the American people of the possibility of a problem,'' Grassley said.

The FDA's Zawisza confirmed that the agency's evaluation of Glaxo's clinical trials found a higher risk of about 40 per cent.

But she said ''this alone is not conclusive'' and officials are trying to reconcile that with clinical trial data that did not show an elevated risk.

''We don't have the scientific data upon which to make any conclusions or decisions yet,'' she said via e-mail.

If a 40 percent risk were confirmed, it would mean that a person with diabetes who does not take the drug has about a 1 per cent risk of having a heart attack or other, similar problem, while someone who takes the drug would have a 1.4 percent risk, Zawisza said.

Grassley also said some FDA scientists recommended months ago that Avandia carry the strongest possible warning, a ''black-box,'' about heart-attack risk.

Zawisza said she could not comment on ''ongoing regulatory matters'' such as whether staff had recommended a black box.

Grassley also said it had been reported to him that the incidence of heart attacks with Avandia could be as many as 60,000 to 100,000 from 1999 to 2006. He did not cite the source of the numbers, and experts say it is difficult to make such estimates across a broad population.

The FDA's Zawisza said a relationship between Avandia and that number of events had not been established. ''We don't have data to support such a conclusion,'' she said.

GlaxoSmithKline spokeswoman Mary Anne Rhyne said Grassley's use of the estimate was ''irresponsible.'' ''They are extrapolations of extrapolations, and do not reflect reality. If these numbers were true, we would have seen them clearly'' in long-term studies and databases of real-world use, she said.

Glaxo has said Avandia's cardiovascular safety is similar to other oral diabetic drugs, and the benefits of treatment outweigh the risks.

The FDA is continuing to review data on Avandia and plans to ask an advisory panel of outside experts for input.

The Union of Concerned Scientists and several consumer groups wrote the FDA yesterday urging officials to chose advisers with no financial ties to Glaxo for the past 36 months.

''By staffing the Avandia advisory committee with scientists from diverse specialties who are non-conflicted, you will send a message that the FDA cares about avoiding conflicts of interest and evaluating drugs on the basis of unbiased scientific findings,'' the letter said.

Reuters DKS GC0440