Cardiologist questions safety of Lilly's prasugrel

By Ransdell Pierson

NEW ORLEANS (Reuters) - Researchers have overestimated the ability to administer Eli Lilly and Co's experimental prasugrel blood thinner without causing dangerous bleeding, a prominent cardiologist said on Monday.

U.S. regulators in recent months have twice delayed a decision on whether to approve prasugrel, a blood clot preventer being developed in partnership with Daiichi Sankyo that would compete with Bristol-Myers Squibb Co's Plavix.

Prasugrel proved far better able than Plavix to prevent dangerous blood clots in a large trial called Triton whose results were unveiled last year

But it was associated with a far higher degree of serious bleeding than Plavix, raising concern whether the U.S. Food and Drug Administration would deem its potential benefits worth the risks.

The Triton study involved patients who underwent angioplasties to clear clogged coronary arteries.

Researchers of the Lilly-sponsored trial had theorized that prasugrel could be safely given, as long as it was not taken by three groups shown to have the highest bleeding risk in the trial: patients weighing less than 130 pounds, those age 75 or older, or those who had previous strokes or "mini-strokes."

But Dr. Sanjay Kaul, a cardiologist with Cedars-Sinai Heart Institute in Los Angeles, said on Monday that theory was at odds with scientific facts.

"The investigator analysis is off target because they could not draw the conclusion, based on the data, that bleeding risk is higher in these (supposed) hi-risk patients," said Kaul, who is making several presentations on prasugrel at the annual scientific meeting here of the American Heart Association.

Kaul specializes in analyzing results of published trials.

Kaul said the risk of serious bleeding in the supposedly high-risk patients was 42 percent higher than with Plavix, compared with 24 percent higher risk among other patients receiving prasugrel.

Although the bleeding risk for the "high-risk" groups was numerically higher than for the "non-high risk" patients taking prasugrel, it was not statistically significant, he said in an interview.

"So you cannot identify which patients should be excluded," he said, or which ones are at greater or lesser bleeding risk.

Anthony Ware, head of Lilly's cardiovascular program, said Kaul's statistical method was different from the pre-specified method used by the Triton researchers.

"The highest-risk groups were identified with statistical significance," using Triton's own statistical formula," Ware said. "We would argue that the overall study shows an exciting benefit to risk, including all these groups."

He noted Lilly was in "productive and collegial" discussions with U.S. and European regulators as they continue to weigh prasugrel's approval.