UPDATE 2-Bayer to advance clot drug despite mixed results

(Adds researcher comments, background on studies)

By Bill Berkrot and Ransdell Pierson

NEW ORLEANS, Nov 10 (Reuters) - An experimental blood clot preventer being developed by Bayer AG (BAYG.DE) and Johnson & Johnson (JNJ.N) missed its primary efficacy goal in a mid-stage study in patients with acute coronary syndrome and led to more bleeding than a placebo, according to data released on Monday.

But the drug, rivaroxaban, did succeed in a secondary goal of showing combined risk reduction in death, heart attack and stroke compared with a placebo -- giving the drugmakers enough confidence to move the medicine into large and costly late-stage trials.

The companies selected two lower doses of rivaroxaban -- 2.5 milligrams or 5 milligrams given twice daily -- that will be moved forward into late-stage testing among more than 13,000 patients. Selecting the doses was one of the main reasons for the study.

The primary efficacy goal was to achieve statistically significant combined risk reduction in death, heart attack, stroke and severe recurrent chest pain requiring angioplasty or bypass surgery.

J&J officials said there were not enough patients in the study to prove effectiveness, a common limitation of mid-stage studies.

"The overall results of this trial were persuasive enough that we will be conducting a 13,000-patient trial to evaluate the drug at low doses," said Dr. Michael Gibson, a Harvard cardiologist who helped lead the trial.

The companies said up to 16,000 patients may be enrolled in the new Phase III program.

The recent trial involved about 3,500 patients with acute coronary syndrome, a chronic life-threatening heart condition caused by artery blockages that can often lead to serious chest pains and heart attacks.

Patients were typically enrolled in the study soon after undergoing angioplasty procedures that clear plaque from clogged coronary arteries.

All patients received low-dose aspirin, which helps to prevent blood clots, while many also received another blood-thinning agent, such as Bristol-Myers Squibb Co's (BMY.N) widely used Plavix (clopidogrel). Patients were then divided into groups that additionally received either rivaroxaban or placebo.

Bleeding ranged from 6.1 percent of patients receiving 5 milligrams of rivaroxaban once a day, to 15.3 percent of patients who received 20 mg daily, compared with 3.3 percent in the placebo group. Bleeding is typically a major concern with drugs designed to prevent blood clots.

Researchers who presented the data at the American Heart Association scientific meeting in New Orleans said 82 percent of the bleeding was considered mild.

"This was a Phase II trial to find the right doses for the right patients with the right frequency, and it looks like a low dose gives just as good effectiveness with lower bleeding" risk than higher doses, Gibson said.

"Doctors must weigh the relative benefits of preventing death, heart attacks and stroke against the risk of causing a major bleed. This will vary with the age of the patient" and the patient's other medical conditions, Gibson said.