All-Cause Death Not Increased With Febuxostat Among Medicare Patients With Gout

In the wake of the CARES trial, data on nearly 100,000 US patients reassure but do not entirely erase concerns, experts say.

All-Cause Death Not Increased With Febuxostat Among Medicare Patients With Gout

In a cohort of nearly 100,000 people, researchers have found no uptick in the risk of MI, stroke, new-onset heart failure, coronary revascularization, or all-cause death for elderly patients who start taking febuxostat versus allopurinol to alleviate gout symptoms.

The observational analysis of US Medicare claims data from 2008 to 2013 comes in the wake of the randomized CARES trial. In that study, presented at the American College of Cardiology Scientific Session in March and simultaneously published in the New England Journal of Medicine, patients with gout and coexisting cardiovascular disease saw no increase in major cardiovascular events while taking febuxostat (Uloric; Takeda) versus allopurinol. Yet there were increases in both all-cause and CV death risk with the newer drug compared with its predecessor.

Seoyoung C. Kim, MD, ScD (Brigham and Women’s Hospital, Boston, MA), senior author of the study published online last week in Circulation, pointed out to TCTMD that while she and her colleagues found no difference in all-cause death between the two drugs, there was a nonsignificant trend toward a higher death risk with febuxostat in long-term users. “In our Medicare data, unfortunately, we did not have info on cause of death,” she said via email.

That rise in risk seen among people taking febuxostat more than 3 years could be genuine or it could be due to chance, Kim noted, adding, “Given the CARES data and our results, a larger study with longer duration of follow-up may be able to determine the long-term effect of febuxostat versus allopurinol on the risk of all-cause death or cardiovascular death.”

Allopurinol earned approval from the US Food and Drug Administration (FDA) in 1966 and febuxostat in 2009. CARES, a noninferiority trial, was conducted at the behest of the FDA as a postmarketing requirement to address an imbalance in CV events that had, without explanation, been seen during febuxostat’s clinical development. Based on preliminary results of CARES, the agency issued a safety communication in November 2017.

As to what clinicians should be doing right now, Kim advised that “patients with gout are at an increased risk of CVD, CV mortality, and all-cause mortality in general. Therefore, I believe we should be better in managing patients’ overall cardiovascular risk (eg, BMI, blood pressure, lipids, smoking) regardless of which gout drug that they are going to start with.”

CARES investigator William B. White, MD (University of Connecticut School of Medicine, Farmington), who was not involved in the new study, said its approach is sound.

“These investigators are experts at what they’re doing. They’re professionals at looking into the Medicare database,” he said, adding, “I can’t criticize or fault anything they’ve done methodologically, because they did the proper propensity scoring to try to match these treatment groups as best they could. But the reality is that, no matter what, any observational study has limitations because it’s not randomized. And you’ve got the possibility that there might be an underlying bias as to why a patient was started on one drug versus another—there’s some reason for that, some clinical rationale.”

For instance, patients with less insurance coverage might be more likely to take allopurinol, which has long been generic, or a patient who can’t tolerate allopurinol or who has chronic kidney disease may be more apt to take febuxostat, he said.

CARES and the Medicare analysis dovetail in that neither found an increase in CV events, White commented. It’s hard to know what to make of the similarity in all-cause mortality seen here, given that in the trial, the disparity in death was “pretty much completely driven by increasing cardiovascular mortality.” The imbalance was largely seen in sudden cardiac death and to a lesser degree in MI and heart failure, White noted. “They were not able to study that.”

Largely Similar Outcomes

Kim, lead author MaryAnn Zhang, MD (Brigham and Women’s Hospital), and colleagues used propensity-score matching to compare 24,936 patients who initiated febuxostat with 74,808 who started taking allopurinol over a 6-year period. Median age was 76 years, and 52% of patients were men. Twelve percent had been diagnosed with CVD at baseline, while 95% had hypertension, 58% chronic kidney disease, 55% diabetes, and 36% heart failure. Use of colchicine, nonsteroidal anti-inflammatory drugs, and steroids reached around 40%.

The primary composite endpoint of hospitalization for MI or stroke occurred at an incidence rate of 3.43 per 100 person-years in the febuxostat group and 3.36 per 100 person-years in the allopurinol group (HR 1.01; 95% CI 0.94-1.08). All-cause mortality risk was similar in both groups (HR 0.95: 95% CI 0.89-1.02), as were the risks of MI, stroke, and coronary revascularization. Hospitalization for new-onset heart failure also did not differ, though among patients who already had heart failure, those in the febuxostat group were slightly less likely to be hospitalized for that condition than those in the allopurinol group (HR 0.94; 95% CI 0.91-0.99).

Stratifying all-cause mortality by follow-up time showed a reduced risk with febuxostat in the first year (HR 0.75; 95% CI 0.66-0.86) and then no statistically significant difference between drugs at 1 to 2 years (HR 0.85; 95% CI 0.63-1.15), at 2 to 3 years (HR 0.72; 95% CI 0.53-1.54), or at more than 3 years (HR 1.25; 95% CI 0.56-2.80).

For White, the mortality difference beyond 3 years is not convincing. “I thought that was a very weak finding. . . . It’s not robust. The confidence intervals for the third time period were wide, and it was statistically nonsignificant,” he emphasized. And in an observational study, it is hard to know whether patients that started on a drug actually kept on taking it over the long haul, White added.

Moving Forward From CARES

There’s no clear mechanism why long-term use of febuxostat would change outcomes, he observed. That said, White continued, there’s also “no mechanistic explanation for why we found what we did in the CARES trial. When I first saw the data, I’m going to be very honest with you . . . I actually didn’t believe it. I said, this cannot be right. Why would a drug that increased cardiovascular death have absolutely no effect on any of the nonfatal cardiovascular events?”

Why would a drug that increased cardiovascular death have absolutely no effect on any of the nonfatal cardiovascular events? William B. White

It’s still worth asking whether the CARES finding is “real,” he stressed. “And why would the finding not be real? Because we had a lot of loss to follow-up in the CARES trial. This was one of our major concerns with the trial, although drop-out from long-term trials is not uncommon when you’re studying patients with symptomatic diseases. They may drop out and return to ‘standard of care.’ What that standard is, however, whether febuxostat, allopurinol, a uricosuric drug, or nothing at all, isn’t known.

Both analyses of the CARES data and the FDA’s own process are ongoing, White said. But importantly, he added, outside the United States the drug is sold by various manufacturers that are conducting their own studies, such as FAST in Europe.

Already, though, the still unconfirmed concerns over febuxostat are having an effect. White pointed to a review article recently published online in Arthritis & Rheumatology that says febuxostat should, at least for now, be used as a second-line drug.

“CARES does not prove that febuxostat raises CV mortality risk, but suggests greater risk with febuxostat than allopurinol,” Hyon Choi MD, DrPH (Massachusetts General Hospital, Boston), and colleagues write in their review.

The trial’s “results do not support first-line use of febuxostat,” they conclude, adding, “In this era of shared medical decision-making, it is advisable to discuss with patients the comparative CV mortality risks of allopurinol and febuxostat, and other safety risks of allopurinol, febuxostat, and uricosurics, such as risks of side effects from severe drug hypersensitivity and drug-drug interactions, and renal adverse events including urolithiasis. The decisions made are based upon the best available information and the personal values and preferences of each patient.”

On June 21, the advocacy group Public Citizen called on the FDA to immediately pull febuxostat off the market based on the CARES trial and earlier studies. "There is overwhelming evidence that the serious cardiovascular harms of febuxostat outweigh any purported clinical benefit," they assert in their 23-page "citizen petition."

The study by Zhang et al is limited in various ways, they add, and as such "offers little meaningful data" for comparing the two drugs.

  • Zhang reports no relevant conflicts of interest.
  • Kim reports receiving research grants to Brigham and Women’s Hospital from Roche/Genentech, Pfizer, Bristol-Myers Squibb, Merck, and AstraZeneca for unrelated studies.
  • White reports receiving personal fees from Takeda USA during the conduct of the CARES trial as well as personal fees from AstraZeneca, Novartis, Abbvie, Sanofi-Aventis, GSK, and Pfizer Consumer Health Care, and nonfinancial support/other from Roche and Wolter Kluwers.
  • Choi reports receiving research support from the National Institutes of Health and AstraZeneca as well as consulting fees from Takeda, Selecta, Kowa, and Horizon (all less than $10,000). He is also a member of the Data Safety Monitoring Committee for the STOP GOUT clinical trial.