Login to Your Account

CRDAC can’t overlook trial flaws, votes against The Medicine Co.’s cangrelor


By Mari Serebrov
Washington Editor

A flawed pivotal trial coming on the heels of two well-designed trials that failed wasn’t enough to convince an FDA advisory committee of the benefits of The Medicines Co.’s cangrelor in reducing thrombotic cardiovascular events.

The Cardiovascular and Renal Drugs Advisory Committee (CRDAC) voted 7-2 Wednesday against approving the intravenous antiplatelet drug for use in percutaneous coronary intervention (PCI) and voted unanimously against its use as a bridge from oral antiplatelet therapy to surgery.

The PCI indication wasn’t an easy vote for some of the panelists. Although he voted against PCI approval, Milton Packer, a cardiology professor at the University of Texas Southern Medical School, wanted to vote yes. “The concept behind this drug is intuitively appealing,” he said, referring to cangrelor’s fast action. But he recognized that drugs can’t be approved based on what they ought to do.

Cangrelor’s fast action has been touted as one of its key advantages over clopidogrel (Plavix, Bristol-Myers Squibb Co. and Sanofi SA), the current standard of care in PCI. Another advantage is that its P2Y12 inhibitory activity disappears within an hour or so, whereas clopidogrel’s activity lasts for days after the drug is stopped. The fact that cangrelor can be used and stopped quickly gives doctors flexibility until they figure out how to treat a given patient. (See BioWorld Today, March 12, 2013.)

Much of CRDAC’s debate over the drug focused on problems with the PHOENIX study, an adaptive randomized, blinded trial that enrolled more than 11,000 subjects globally. Approved by regulators in 12 countries and numerous institutional review boards, the pivotal trial followed the CHAMPION PCI and PLATFORM studies, both of which were large phase III trials halted in 2009 for lack of efficacy. (See BioWorld Today, May 14, 2009.)

Packer said even though Medicines, of Parsippany, N.J., did a good job of reassuring him, he still was uncomfortable with some aspects of PHOENIX, especially since the well-designed PCI and PLATFORM trials failed.

Thomas Fleming, a professor of biostatistics at the University of Washington, agreed that the other two trials can’t be discounted, especially since PHOENIX was a “marginal” trial.

The two yes votes on the PCI indication came from CRDAC Chairman Philip Sager, a professor at Stanford University School of Medicine, and Jennifer Li, director of pediatric research at the Duke Translational Medicine Institute. While both recognized problems with the trial design, they voted yes on the basis of the clinically significant primary endpoint and the lack of major risk.

In the PHOENIX trial, subjects undergoing PCI who were given cangrelor had their odds reduced by 22 percent of experiencing the primary endpoint, which was a composite incidence of death, heart attack, stent thrombosis or ischemia-driven revascularization at the 48-hour mark after they were randomized. Cangrelor also showed a 38 percent drop in the chances of the key secondary endpoint, incidence of stent thrombosis at 48 hours.

The efficacy data were called into question by Thomas Marciniak, the FDA’s medical team leader on the drug’s review, because of delays in administering clopidogrel and differences in the dosing of the standard of care. He also raised ethics issues, noting that the informed consent forms for PHOENIX didn’t tell subjects of the advantages of earlier use of clopidogrel or inform them of alternative treatments such as prasugrel (Effient, Eli Lilly and Co.) and ticagrelor (Brilinta, Astrazeneca plc), both of which have demonstrated superiority to clopidogrel.

While some CRDAC members were concerned about the variations in dosing and timing of clopidogrel and how they impacted the efficacy data, they didn’t seem overly bothered about the lack of information about the other drugs since they aren’t as commonly used in PCI. Effient, which was approved in 2009, was just coming on the scene when PHOENIX started, and Brilinta wasn’t approved until 2011.


What tripped cangrelor up in the bridge indication is that the need hasn’t been clearly defined. Packer acknowledged that this is a gap that’s seen all the time clinically, but there’s no clear thought on how that gap should be filled.

James DeLemos, the cardiology service chief at Parkland Memorial Hospital and a professor at the University of Texas Southwestern Medical Center, said he thought cangrelor might be able to fill the gap if the bridge “space turns out to be important.”

The FDA doesn’t have to follow the committee’s lead as it approaches the April 30 PDUFA date. While Marciniak is strongly opposed to approving cangrelor for any indication, two other agency reviewers supported the drug’s approval for PCI but not as a bridge to surgery. (See BioWorld Today, Feb. 11, 2014.)

Medicines (NASDAQ:MDCO) halted trading during the meeting Wednesday. It closed at $33.10 Tuesday, regaining much of the ground it lost Monday when the FDA’s briefing documents were released. With its top-selling drug, Angiomax (bivalirudin), facing a patent challenge from Hospira Inc. and patent expiration next year, the company would welcome a cangrelor approval, but the drug is not critical to the company’s future.

If approved, cangrelor, licensed from Astrazeneca plc in 2003, could take advantage of the infrastructure already in place for Angiomax. And there’s the possibility of combining it with Angiomax to potentially treat high-risk PCI patients.

Although disappointed by CRDAC’s decision, Medicines Chairman and CEO Clive Meanwell said the company appreciated the dialogue and looks forward to subsequent discussion with the FDA as the agency completes its review.