Health Letter, October 2018
By Michael Carome, M.D.
On June 21, 2018, Public Citizen petitioned the Food and Drug Administration (FDA) to immediately remove from the market the widely used gout medication febuxostat because it poses unique, serious risks – some fatal – but provides no unique benefit.
In 2009, the FDA approved febuxostat, which is sold by Takeda Pharmaceuticals America under the brand name Uloric, for treating high uric acid blood levels in patients with gout. The approval came despite serious safety concerns about the drug — particularly its cardiovascular risks — as well as clinical trial data showing that it did not offer any major clinical advantage over the much older, clinically effective drug allopurinol, which has been the first-line medication for preventing gout attacks for decades.
In the decade since its approval, febuxostat has been aggressively marketed and prescribed to at least hundreds of thousands of patients. Takeda reported $1.9 billion in U.S. sales of febuxostat from fiscal years 2012 through 2017. For the one-year period ending in June 2015, there were 1.3 million U.S. prescriptions for Uloric, making it the 46th most prescribed brand-name medication in the U.S at the time.
Results of a large randomized clinical trial recently published in the New England Journal of Medicine (NEJM) showed that febuxostat significantly increased the risk of death from any cause and heart-related death compared with allopurinol. These findings provided definitive evidence that the risks of febuxostat outweigh any benefit. These new data prompted Public Citizen to petition the FDA to remove the drug from the U.S. market.
Gout develops in some patients who have high blood levels of the bodily waste product uric acid, which can be caused by the body producing too much uric acid or by the kidneys removing too little of it. Gout affects more than 8 million adults in the U.S. and occurs more commonly in men and in older people.
Acute gout attacks occur when uric acid crystals form in joints and surrounding tissues, resulting in severe joint pain, swelling and redness. Gout attacks usually affect a single joint, such as the big toe, ankle or knee; last several days; and occur intermittently. Standard treatments for acute gout attacks include nonsteroidal anti-inflammatory drugs (for example, ibuprofen, colchicine and injection of steroids into the affected joint.
Some gout patients develop large uric acid deposits, called tophi, over joints, tendons and bones. Tophi appear as hard bumps under the skin. Patients also may experience kidney damage and kidney stones caused by uric acid crystals forming in the kidneys and urine.
Drugs that lower uric acid levels are used to prevent acute gout attacks, tophi formation and kidney damage. The xanthine oxidase inhibitors allopurinol and febuxostat work by blocking the body’s production of uric acid.
For patients who experience an inadequate response to a xanthine oxidase inhibitor, a second drug that increases the removal of uric acid from the body by the kidneys — known as a uricosuric — can be added to further decrease uric acid levels. Probenecid (initially approved by the FDA in 1951) and lesinurad (initially approved by the FDA in 2015) are the only uricosurics currently available in the U.S.
Febuxostat no better than allopurinol
The FDA’s 2009 conclusion about febuxostat’s effectiveness was based on the results of three randomized clinical trials — called the APEX, FACT and CONFIRMS trials — that involved gout patients who had high blood uric acid levels. Altogether, these trials enrolled more than 4,000 subjects who were randomly assigned to receive febuxostat (at daily doses ranging from 40 to 240 milligrams [mg]) or allopurinol once daily for six months to one year. The APEX trial also included a small group of subjects who received only a placebo.
The FDA approved febuxostat at doses of 40 mg and 80 mg once daily. All three randomized clinical trials showed that febuxostat at doses of 80 mg or higher were more effective than allopurinol in lowering blood uric acid levels, a so-called “surrogate endpoint.” However, the CONFIRMS trial, which was the only trial that tested febuxostat at the 40-mg dose, revealed that the drug was no more effective at this dose than allopurinol in lowering blood uric acid levels.
Importantly, febuxostat at all of the tested doses was not more effective than allopurinol in preventing acute gout attacks or reducing tophi, which are clinically important outcomes in gout treatment.
In addition to not offering any major clinical advantage over allopurinol, serious concerns about the safety of febuxostat were raised before the drug was approved by the FDA.
Pre-approval cardiovascular safety concerns
Takeda Pharmaceuticals originally applied for FDA approval of the drug in 2004 after completion of the APEX and FACT trials., , The FDA rejected the application primarily because data from these randomized trials (and others) showed that a greater number of adverse cardiovascular events — such as heart attacks, strokes and heart failure — had occurred in subjects who had received febuxostat than in those who had received allopurinol or a placebo,. , Furthermore, across all studies of febuxostat at the time, eight subjects receiving febuxostat had died, whereas no subjects receiving allopurinol had died.
Takeda again applied for FDA approval of febuxostat in 2006 after reanalyzing data from the same clinical trials that had been submitted with the original application, as well as analyzing new data from two then-ongoing long-term non-randomized trials of febuxostat. The FDA again rejected approval of the drug because the company’s new analyses did not allay concerns about the possible cardiovascular risks of febuxostat. With this second rejection, the FDA recommended that the company conduct a new clinical trial assessing the cardiovascular risks of febuxostat at lower doses.
Takeda subsequently conducted the CONFIRMS trial and applied for FDA approval a third time in 2008. In contrast to earlier studies, the CONFIRMS trial did not show a greater number of adverse cardiovascular events with febuxostat compared with allopurinol. However, data combined from the APEX, FACT and CONFIRMS trials did show a slightly greater risk of adverse cardiovascular events in subjects who had received febuxostat than in those who had received allopurinol, but this finding was not statistically significant. Commenting on these results, one FDA medical officer noted that it was not possible to exclude either a greater or a lower risk of adverse cardiovascular events with febuxostat with much confidence.
Despite the uncertainty about febuxostat’s cardiovascular safety, the FDA in 2009 approved the drug but required that the company conduct a large post-approval randomized clinical trial to further assess the drug’s cardiovascular risks compared with allopurinol. This trial was called the CARES trial.
CARES trial confirms risk
The results of the CARES trial were published by the NEJM on March 29, 2018. From April 2010 to May 2017, the trial’s researchers randomly assigned nearly 6,200 subjects who had gout and cardiovascular disease to receive either febuxostat or allopurinol. The researchers followed the subjects for a median of 32 months.
The researchers found that subjects in the febuxostat group had a 22-percent higher risk of death from any cause and a 34-percent higher risk of heart-related death (for example, sudden cardiac death) than subjects in the allopurinol group. Both findings were statistically significant. These results confirm the worrisome cardiovascular safety concerns that were identified before the drug was approved.
As with the prior clinical trials, there was no evidence from the CARES trial that febuxostat is more effective than allopurinol in preventing acute gout attacks or other clinically important complications of gout, the most crucial reasons for using such treatments.
The results of the FDA-mandated CARES trial now provide high-quality evidence of a causal link between treatment with febuxostat and increased risk of death from any cause and death from cardiovascular causes. The FDA almost certainly would have denied approval of febuxostat if data from the CARES trial had been available before it approved the drug. The appropriate course of action now is obvious: Febuxostat should immediately be withdrawn from the U.S. market to avoid further preventable harm to patients.
 Food and Drug Administration. Medical Reviews for NDA 21-856. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2009/021856s000_MedR_P1.pdf. Accessed August 27, 2018. PDF page 24.
 Food and Drug Administration. Medical Reviews for NDA 21-856. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2009/021856s000_MedR_P4.pdf. Accessed August 27, 2018. PDF pages 46, 68.
 Food and Drug Administration. Medical Reviews for NDA 21-856. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2009/021856s000_MedR_P1.pdf. Accessed August 27, 2018. PDF page 25.
 Food and Drug Administration. Office director memo for NDA 21-856. February 13, 2009. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2009/021856s000_ODMemo.pdf. Accessed August 27, 2018. PDF pages 4-5.
 Food and Drug Administration. Medical Reviews for NDA 21-856. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2009/021856s000_MedR_P1.pdf. Accessed August 27, 2018. PDF page 56.