Background/Purpose: The cardiovascular (CV) safety study CARES has the longest study duration of any randomized controlled trial in patients (pts) with gout and CV disease. In CARES, the proportion of pts with the primary endpoint (composite of CV death, nonfatal myocardial infarction, nonfatal stroke, and unstable angina with urgent revascularization) was noninferior between febuxostat (feb) and allopurinol (allo), but there was an imbalance in rates of CV mortality (4.3% and 3.2%, respectively).¹ We evaluated the general safety of feb and allo in CARES and the relationship between CV mortality and serum urate levels (sUA) or gout flares.

Methods: Pts were randomly assigned to once-daily feb (40 or 80 mg, based on sUA at Week 2) or allo (dose titrated in 100 mg increments from 200‒400 mg or 300‒600 mg, based on kidney function). The modified intention-to-treat population (mITT) comprised pts who were randomized and received treatment. Safety endpoints included treatment-emergent adverse events (TEAEs); key efficacy parameters were changes from baseline in sUA and gout flares. The sUA and flares (while on study drug) were evaluated in those pts with CV mortality versus those in the overall mITT.

Results: Randomized pts were treated with feb (n=3098) or allo (n=3092) over a median follow-up of 32 months (max 85 months); 57.3% and 55.9% of patients randomized to feb and allo, respectively, discontinued treatment early. In the feb and allo treatment groups (mITT), similar proportions of pts had TEAEs (83% vs 82%, respectively) or serious TEAEs (34% vs 32%); most common TEAEs were diarrhea (10% vs 9%) and arthralgia (8% vs 10%). Total incidences of rashes, eruptions, or exanthemas for feb and allo were 4% and 5%, respectively. A total of 13% pts in each group discontinued study drug due to TEAEs. Baseline sUA were higher in pts with CV mortality (feb: 9.3 mg/dL [n=134]; allo: 9.8 mg/dL [n=100]) versus the overall mITT (8.7 mg/dL with feb and allo), but comparable between treatment groups. In the overall mITT, sUA with feb were lower than with allo at Week 2 and Months 3, 6, 12, 24, 36, 48, 60 and 72; however, sUA were not consistently lower with feb in the CV mortality cohort (Figure 1). Gout flare rates were similar with feb and allo and decreased across the study period (Figure 2); flare rates within 3 months of CV mortality were low in both the feb (6%) and allo (2%) groups.

Conclusion: Feb and allo treatments had comparable safety and tolerability findings in pts with gout and CV disease. There were no distinct relationships between CV mortality and sUA or gout flares.

1. White WB et al. N Engl J Med 2018;378:1200‒10