Febuxostat Emerges Relatively Unscathed After FDA Advisory Committee Review

Concerns about CV risk are prompting calls to update the label, but most panel members said the drug has a favorable risk-benefit balance.

Febuxostat Emerges Relatively Unscathed After FDA Advisory Committee Vote


(UPDATED) Despite worries over a potential CV risk with febuxostat, there is indeed a population of patients with gout for whom the risk-benefit profile is favorable for the treatment of hyperuricemia, agreed 19 out of 22 panel members at a joint meeting of two US Food and Drug Administration advisory committees held today.

There were two “no” votes against that idea and one abstention among members of the Arthritis Advisory Committee and the Drug Safety and Risk Management Advisory Committee.

The agency is seeking expert advice on what new action, if any, to take following a signal of increased cardiovascular deaths among patients taking febuxostat (Uloric; Takeda) in the CARES postmarketing study. Those data were released last spring at the 2018 American College of Cardiology Scientific Session in Orlando, FL, and published in the New England Journal of Medicine, as reported by TCTMD.

C. Michael Gibson, MD (Beth Israel Deaconess Medical Center, Boston, MA), voted yes. Voicing support for a Dear Healthcare Provider letter, Gibson said, “I would suggest that patients and doctors make informed choices, and I don’t know that it’s the job of the FDA to enforce those conversations. That’s what physicians should be doing. Part of the conversation could be that if you’re on aspirin, that would minimize the risk.”

Gibson also suggested reaching out to the data safety monitoring board of the ongoing European FAST study, to see if unblinded data would reveal any clues.

Also voting yes, Laurel A. Habel, MPH, PhD (Kaiser Permanente Northern California, Oakland), an epidemiologist, said that at the very least febuxostat’s label should “include the CARES results, even though they’re not definitive.” This might come in the form of a black-box warning or some other phrasing to “result in more restrictive use and encourage it as a second-line therapy,” she commented. “I like the idea of a letter or some sort of material that doctors can use [to] improve risk communication with patients, so that they really can make an informed decision.”

Further, professional societies should also frame febuxostat as a second-line therapy, Habel asserted, though there shouldn’t be barriers that make the drug hard to obtain for patients who really need it to treat debilitating gout.

John J. Cush, MD (Baylor University Medical Center, Dallas, TX), a rheumatologist, said he voted yes because he’d “like to see the drug continue to be used but used with greater intelligence and with restriction. I would recommend strong if not radical changes to the label including a boxed warning.

“I don’t think the problem is that people don’t read boxed warnings,” Cush continued. “I think the problem is that people don’t read the package insert until they have a question about something inconsequential and they go and read three lines and they’re done. [So] I think there has to be a strong educational commitment by the manufacturer to right this wrong.”

Steven Nissen, MD (Cleveland Clinic, OH), and Martin Kulldorff, PhD (Harvard Medical School, Brigham and Women’s Hospital, Boston, MA), both voted against febuxostat having a favorable risk-benefit balance.

“I wanted to vote and I could vote yes, but I believe patients need to be informed of these risks. And I think the idea of informed consent is critical,” Nissen stressed, expressing frustration that REMS didn’t get more support as a way to mitigate risk. “If a REMS could be constructed, to ensure that patients that are truly intolerant to allopurinol are the ones that get febuxostat, then I would be okay with keeping the drug on the market. But in the absence of that, I do not think that warning labels work, and I think there’s actually quite ample literature to suggest that even boxed warnings on the utilization of drugs this far into the market is very limited.”

Kulldorff, for his part, cited the argument that “there is evidence that the drug causes death in some people.” Moreover, it isn’t known whether patients who are unable to take allopurinol would in fact do well on febuxostat, he said. “I think we basically don’t know. I think the future of this drug for the moment is to withdraw it from the market and encourage prospective clinical trials to study the drug among those who do not tolerate allopurinol.”

Febuxostat was approved in 2009 and allopurinol in 1966. The two drugs are the only xanthine oxidase inhibitors (XOIs) on the market for the treatment of gout, which affects approximately 9 million people in the United States and is growing in prevalence.

Febuxostat is the only first-line alternative for patients who cannot take allopurinol, due to being genetically predisposed to a hypersensitivity reaction, having chronic kidney disease, or developing a skin reaction or other side effects on the drug. In addition to the XOIs, there are four second-line options. But of the six approved urate-lowering drugs currently available in the US, two are slated to go off the market this year.


In CARES, a multicenter, double-blind trial, more than 6,000 patients with gout and coexisting cardiovascular disease were randomized to febuxostat or allopurinol. While febuxostat was noninferior to allopurinol in terms of overall MACE risk, the drug was also linked to a 1.1% absolute increase in CV mortality (HR 1.34; 95% CI 1.03-1.73), mainly sudden cardiac death.

Yet nonfatal CV events were similar between the two drugs, and most MACE occurred not while patients were taking their medications, but in the first 30 days after they’d stopped.

MACE Incidence Rate per 100 Person-Years










After Treatment Discontinuation

    1-30 Days

    31+ Days










Numerous subgroup analyses and other investigations have revealed no clues as to mechanism, said Beth Anne Knapp, VP of Global Regulator Affairs, Marketed Products, at Takeda. “We could not find an explanation within CARES . . . or identify a biologically plausible cause" for higher CV death with febuxostat, she noted in her presentation to the advisory committees.

It’s also a mystery why patients taking NSAIDs at baseline had a lower CV death risk with allopurinol versus febuxostat. For patients on low-dose aspirin, febuxostat came out ahead. But William B. White, MD (University of Connecticut School of Medicine, Farmington), principal investigator for the trial, speaking today at the panel, cautioned that the number of events was small and that the observed associations could be due to confounding.

Some theories tossed about in the discussion were that febuxostat might affect platelet activation, plaque rupture, or heart rhythm.

CARES was conducted at the behest of the FDA as a condition of the drug’s approval 10 years ago, to monitor for signs of CV risk that had emerged earlier in the drug’s development. Based on preliminary results from CARES, the agency had issued a safety communication in November 2017. Seven months later, the nonprofit consumer advocacy group Public Citizen called for febuxostat to be pulled from the market. In August 2018, the FDA announced that it was conducting a comprehensive review of the data, with the plan of meeting in early 2019.

‘Magnitude of Missingness’

Amid specific questions about the trial’s methodology, several participants in today’s meeting voiced uncertainty about how much insight CARE could provide given its shortcomings—fully 57% of patients discontinued their assigned study medication and 45% lacked complete follow-up over the course of the 7-year study. However, one-third of patients had some level of follow-up.

“You’ve said the noninferiority criteria were met [for MACE], but when you have this magnitude of missingness it drives the hypothesis towards a noninferiority finding,” Gibson pointed out.

According to calculations from Takeda, an estimated 6.5% of that missing one-third would have had events captured by complete follow-up. To undermine the noninferiority of CARE’s primary endpoint, their risk ratio would have to be 1.65, or 22 times higher than the current hazard ratio of 1.03 for febuxostat versus allopurinol, something that a biostatistician presenting the numbers on behalf of Takeda framed as unlikely.

Pressed to answer why so many patients either stopped taking their drug or dropped out, White described the gout population as challenging to study. Much like in the PRECISION trial of patients with arthritis, where 68.8% stopped taking the study drug, the CARES population is in pain, he said.

As White explained, “This particular patient population is a very difficult one. They have a lot of other problems: there’s a lot of alcoholism, there’s a lot of obesity, there’s a lot of other cardiovascular disease. And they’re ornery, they’re unhappy, they’re in pain all the time from their disease. So the first thing they’re going to do when they get a gout attack is say, ‘This doctor’s not doing something good for me. I’m leaving the study.’

“We didn’t know this was going to happen—no one would have predicted that half the patients in the trial were going to discontinue study drug,” he stressed. “But at least we can say one thing: that it was equal in each treatment group.”

At the end of the day, despite much debate, the majority of panel members agreed that the drug plays an important role in the management of this condition.

Takeda announced the outcome of the meeting in a press release that expressed optimism but offered few specifics on how the company plans to move forward.

“Patient safety has always been our first priority at Takeda, and we appreciate that common purpose and the thoughtful discussion today with the Advisory Committee members about CARES,” said Tom Harris, Senior VP and Head of Global Regulatory Affairs for Takeda Pharmaceuticals. “We have studied the safety of Uloric for more than 15 years, and remain confident in Uloric as an important option for the chronic management of hyperuricemia in gout. We look forward to additional discussions with the FDA regarding these CARES data.”

As pointed out in the press release, though, the Advisory Committee vote “is nonbinding and will be taken into consideration by the FDA.”

We Recommend