You are here
On January 11, 2019 the FDA convened two advisory panels (the Arthritis Advisory Committee [AAC] and the Drug Safety and Risk Management Advisory Committee [DSaRM]) to consider the cardiovascular safety of febuxostat (Uloric) based on the 2018 “Cardiovascular Safety of Febuxostat and Allopurinol in Patients with Gout and Cardiovascular Morbidities (CARES) study.
The FDA posed the following voting question to the advisory panel: “Based upon the available data, is there a patient population in which the benefit-risk profile for febuxostat is favorable for the treatment of hyperuricemia in patients with gout?”
After a day of presentations by FDA reviewers and the sponsor (Takeda), and review of the data, the AAC and DSaRM advisory committee members voted as follows:
- 19 Yes
- 2 No
- 1 Abstention
The panel’s recommendation and discussion are advisory to the FDA, who will issue a final decision on the future of febuxostat at a later date.
The Panel Members
The panel included members of the AAC and DSaRM advisory committees, comprising a total of 22 voting members that included rheumatologists, cardiologists, internists, epidemiologists/statisticians, PharmD pharmacologists, patient representatives and a non-voting industry representative.
Febuxostat (a xanthine oxidase inhibitor [(XOI]) was FDA approved on February 13, 2009 for the chronic management of hyperuricemia in patients with gout. However, during development, febuxostat (FEB) was twice denied approval largely because of concerns over cardiovascular (CV) events and death compared to allopurinol in 6-12 mos. phase 2 trials (FACT, APEX) and their open label extensions. This led to the larger CONFIRMS trial designed to assess the CV safety and ultimately demonstrated no increase in APTC CV events for FEB versus allopurinol. Combining all 3 clinical trials (APEX, FACT, CONFIRMS) the risk of adjudicated CV events tended to be higher for FEB compared to allopurinol (APTC events 0.74/100PY vs 0.60/100PY), but was not significant. The CARES study was a post-marketing committment to further study the CV risks associated with FEB treatment. The results of the study were published earlier in 2018 in the NEJM.
Compared to population controls, gout patients have a significantly higher risk of mortality (28% higher) and CV death (38%) (Choi et al, Circulation 2007; 116:894). Thus, teasing out a low risk potentially imposed by a drug may be difficult.
The FDA reported that in 2017 there were 3.1 million prescriptions for allopurinol and 200,000 Rx for FEB, with the majority of prescribers for both being primary care physicians. Takeda reported that since FDA approval, FEB has amassed 15 million patient-years of exposure and thus the safety profile of this drug is well established. While FEB has certain advantages, including preferred use in patients with allopurinol hypersensitivity syndrome, use in patients with renal insufficiency and a two dose option, it does have unique risks including severe skin reactions and uncommon hepatotoxicity.
The CARES study was a double blind, active comparator trial of 6190 gout patients, wherein the primary objective was to rule out a hazard ratio greater than 1.3 for MACE (major adverse cardiovascular events) associated with FEB relative to allopurinol use. The primary endpoint was MACE (defined as CV death, non-fatal MI, non-fatal stroke, unstable angina requiring coronary revascularization) composite events. To be included in the trial patients had to have gout, hyperuricemia, >1 flare in the last 12 mos. and a history of major CV or cerebrovascular disease.
Patients were treated in a treat to target (SUA < 6.0; or SUA < 5.0 for tophaceous gout) fashion and nearly 70% achieved this target. However, only 55% of the patients completed the trial for a median of 2.6-2.7 years of follow-up and a median of 2 years of drug exposure. Non-completers were similar between groups, with 7% lost to follow-up, 19% withdrew consent and 19% dropped for other reasons. There was missing data on roughly 8%. Several panel members were quick to remind everyone that having such a large drop out and data deficit tends to favor the null hypothesis of finding no difference between FEB and allopurinol. This data gap takes on greater meaning given the mixed outcomes below.
Outcomes: while the primary endpoint (MACE) showed no significant difference in MACE events (335 vs 321 events; HR 1.03; 95%CI, 0.89, 1.21) comparing FEB and allopurinol, there was a significant difference in CV death (134 vs 100 events; HR 1.34; 95%CI 1.03, 1.73). The other MACE elements were not significantly different between groups, including non-fatal MI (HR 0.93), non-fatal stroke (HR 1.01) and unstable angina (HR 0.86) – all not significant. Most of the MACE events and CV deaths occurred after treatment discontinuation, especially in the first 30 days. Reasons for this observation were not elucidated.
Multiple subanalyses failed to show a risk differential between FEB and allopurinol exposed patients, with one exception. Not receiving aspirin (roughly 40%) appeared to increase the risk of CV death with FEB use (HR 2.05; 95%CI 1.42, 2.95) leading some panelist to postulate FEB having potential platelet effects and suggest that FEB should be taken with daily ASA.
The panel was uniform in their belief that there is an unmet need for gout education and additional therapies for gout and that another urate lowering therapy (e.g., febuxostat) has played an important role in gout management, especially with rheumatologists. Gout patients have an estimated 30% or higher overall cardiovascular risk that may be ameliorated by allopurinol, but (possibly) not as well with FEB therapy.
The advisory panels also struggled with the interpretation of this incomplete data and overall were concerned that incomplete data may obscure and even higher cardiovascular risk than that presented.
Lastly the recommendations of the panel waffled when considering stronger labeling guidance or the use of a “boxed” warning, as several members recalled that such changes often have little impact on future product use. It was also noted that it may not be that boxed warnings are ineffective, instead the problem could be that most physicians do not read the product label in its entirety or at all.
At the close of the meetings each panelist explained their vote and detailed their recommendations for an improved benefit: risk profile of gout patients who may be treated with FEB. Most called for education, some favored a REMS program, many favored an update to the label (including a boxed warning), some favored a change in indication that FEB only indicated after allopurinol use and a few recommended the product be removed from the market.
The AAC and DSaRM advisory committee members were overwhelmingly in favor of keeping febuxostat on the market but clearly expressed concerns that febuxostat use may pose a small CV risk to some gout patients, especially those with severe gout and those not receiving daily aspirin. The FDA decision on febuxostat is pending.
(Dr. Cush, [the author of this article] participated in this FDA Advisory panel as a SGE, invited guest and voting panel member)