* Bleeding 2 to 2.5 times higher than standard clot drug
* Effectiveness matches or beats Sanofi's Lovenox
* Doctors cast doubt on approval for acute illness group
* Bayer shares down 3.6 pct; J&J shares fall 0.6 pct (Adds details on overall bleeding risk for drugs)
By Ransdell Pierson and Bill Berkrot
NEW ORLEANS, April 5 A blood clot preventer from Johnson & Johnson and Bayer caused a surprisingly high rate of bleeding in a trial of patients with acute illnesses, representing a setback for the drugmakers.
Bayer AG (BAYGn.DE) shares fell 3.6 percent, while J&J (JNJ.N) slipped 0.6 percent after the data were presented at the annual scientific meeting of the American College of Cardiology in New Orleans on Tuesday.
Industry analysts had predicted the trial of the drug rivaroxaban, if successful, would create a potential $2.8 billion annual market among the study's population of patients who are susceptible to blood clots while hospitalized for illnesses such as cancer and pneumonia.
"This will surely impact the chances of admission of the drug and is a serious disappointment," said Markus Huber, a senior trader at ETX Capital, adding that it could have financial repercussions for Bayer.
A lead investigator for earlier North American trials of rivaroxaban in patients getting knee replacements also took a pessimistic view of the new data.
"The data today would not be approvable .... Why approve something with no overall benefit" for patients like those in the trial, said Alexander Turpie, professor of medicine at McMaster University in Hamilton, Ontario.
Turpie and industry analysts said the results do not preclude other big opportunities for the drug, which is already sold in Europe by Bayer under the brand name Xarelto to prevent blood clots in patients undergoing hip and knee surgery.
The drugmakers are developing the pill to prevent stroke in patients with an irregular heartbeat called atrial fibrillation. Analysts see that market having the potential for $3 billion in annual sales.
Development of new blood clot preventers has been one of the hottest areas in cardiology. Several pharmaceutical companies are vying to come up with a drug of choice to displace decades-old warfarin and other medicines.
Potential rivals to rivaroxaban include apixaban by Pfizer Inc (PFE.N) and Bristol-Myers Squibb (BMY.N), edoxaban from Japan's Daiichi Sankyo (4568.T) and Pradaxa, already being sold by privately held Boehringer Ingelheim.
The 8,101-patient study released on Tuesday compared the bleeding risk and effectiveness of rivaroxaban with that of the standard injectable treatment enoxaparin in patients hospitalized for acute medical conditions, including heart failure, infectious disease and breathing difficulty.
Injections of enoxaparin, sold by Sanofi-Aventis (SASY.PA) under the brand name Lovenox, are typically given in the hospital, with treatment lasting no more than two weeks.
After 10 days of treatment in the study, rivaroxaban and enoxaparin were deemed equally effective in preventing dangerous blood clots in the extremities and in the lungs.
Although enoxaparin is not typically used for long periods, J&J wanted to assess its long-term effectiveness compared with rivaroxaban. To set up a comparison, patients took enoxaparin for 10 days, and then took placebos for the subsequent 25 days, while rivaroxaban was given for 35 days.
By that measure, rivaroxaban proved significantly more effective than enoxaparin.
But patients taking the J&J/Bayer drug had a significantly higher rate of bleeding. Some 2.8 percent of patients taking rivaroxaban had clinically relevant bleeding at 10 days, compared with 1.2 percent of those receiving enoxaparin.
At 35 days, 4.1 percent of the rivaroxaban group experienced bleeding, versus 1.7 percent taking enoxaparin. Although there was a disparity in risk, researchers said the overall rate of serious bleeding was small for both drugs.
"This is a setback for rivaroxaban because its efficacy is counterbalanced by the risk," said Dr. Catalin Balan of University Emergency Hospital in Sibiu, Romania, who had no role in the trial. "It worsens the overall view of the drug."
Lead researcher Alexander Cohen of King's College Hospital in London said the drug fell short because safety was assessed not only by major or fatal bleeding, but also by minor bleeding such as nosebleeds, big bruises and blood in the urine.
Had the trial only included serious bleeding, he said the benefit of rivaroxaban would have outweighed its risks. Seven patients taking rivaroxaban died from major bleeding, versus one for enoxaparin. But 30 patients taking enoxaparin died from lung clots, versus 19 taking the J&J drug.
Peter DiBattiste, vice president of cardiovascular development for J&J, said the company will conduct additional analyses to see if rivaroxaban can be used more selectively in patients hospitalized with acute medical illness.
DiBattiste said that compared to enoxaparin, a higher bleeding risk was not seen for his drug in earlier successful trials among patients undergoing orthopedic surgery.
"This is a different population than we've seen before, with different organ systems involved," he said. "And they're probably more predisposed to bleeding because of multiple illnesses and multiple medicines."
Wall Street expects the drug to be approved this summer to prevent clots in patients undergoing orthopedic surgery, with approval by Fall for stroke prevention in atrial fibrillation patients.
Under their marketing agreement, J&J will market the pill in the United States and pay royalties of up to 30 percent of its revenue to Bayer. Deutsche Bank analysts estimate that Bayer stands to retain 61 percent of worldwide sales in 2015. (Additional reporting by Bill Berkrot in New Orleans, Josie Cox in Frankfurt and Ben Hirschler in London; Editing by Michele Gershberg, Dave Zimmerman, Richard Chang and Carol Bishopric)