Merck said it’s stopping one of two big clinical trials of an experimental blood thinner that some viewed as the firm’s highest profile pipeline drug. The move marks a setback, analysts say, but nearly 20,000 patients will still be eligible to receive it, keeping alive hopes the drug may get back on track for approval.

The agent, vorapaxar, was being studied in two major clinical trials to evaluate its ability to prevent heart attacks, and Merck had expected to submit it to the FDA this year for use in patients with acute coronary syndromes (ACS).

“We’re obviously disappointed,” Merck’s head of R&D Peter Kim told analysts this morning about its decision to stop the drug in just those patients in the trial called TRA-2P who had a stroke before entry into the study, “but are pleased that 75% of patients will continue in this study.”

Kim said about 6,000 patients, out of 26,500, have a history of stroke or had one during TRA-2P. The trial is evaluating the drug in secondary prevention of heart attacks. The other trial, TRACER, a 13,000-subject study testing vorapaxar in patients with ACS, is being stopped altogether.

The removal of those who had a prior stroke from TRA-2P led to speculation that excessive bleeding could be to blame for TRACER’s discontinuation. Patients with ACS also tend to bleed more, a constant fear when evaluating new anti-clotting agents.

But the Data and Safety Monitoring Board (DSMB), which recommended making the study changes, did not give a reason, and Kim refused to provide additional details of the DSMB findings during the analyst call. “Some will attempt to draw conclusions, but keep in mind we don’t yet know the specific safety and efficacy results,” Kim said. “Detailed data will be available after review and analysis.”

In a statement, Merck said it will update its projections for vorapaxar regulatory filings once it receives efficacy and safety data from TRACER and can determine an updated completion date for TRA-2P.

Vorapaxar, which came to Merck from the Schering-Plough acquisition, has a different mechanism of action and more selectivity than Bristol-Myers Squibb/Sanofi-Aventis’ Plavix and Eli Lilly/Daiichi-Sankyo’s Effient and, in theory, could be used in combination with them or with AstraZeneca’s experimental antiplatelet agent Brilinta. “What we’re looking for is an antiplatelet agent we can add on top of the standard of care,” Kim said.