The Effect of Atrasentan on Kidney and Heart Failure Outcomes by Baseline Albuminuria and Kidney Function: A Post Hoc Analysis of the SONAR Randomized Trial.

BACKGROUND AND OBJECTIVES: Atrasentan reduces the risk of kidney failure but increases the risk of edema and, possibly, heart failure. Patients with severe CKD may obtain greater absolute kidney benefits from atrasentan but may also be at higher risk of heart failure. We assessed relative and absolute effects of atrasentan on kidney and heart failure events according to baseline eGFR and urinary albumin-creatinine ratio (UACR) in a post hoc analysis of the Study of Diabetic Nephropathy with Atrasentan (SONAR) trial.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: The effect of atrasentan versus placebo in 3668 patients with type 2 diabetes and CKD with elevated albuminuria was examined in the SONAR trial. We used Cox proportional hazards regression analysis to study effects on the primary kidney outcome (composite of doubling of serum creatinine, kidney failure, or kidney death) and heart failure hospitalization across subgroups of eGFR (<30, ≥30-45, and ≥45 ml/min per 1.73 m2 ) and UACR (<1000, ≥1000-3000, and ≥3000 mg/g).

RESULTS: Atrasentan reduced the relative risk of the primary kidney outcome (hazard ratio, 0.71; 95% confidence interval, 0.58 to 0.88) consistently across all subgroups of baseline eGFR and UACR (all P interaction >0.21). Patients in the highest UACR and lowest eGFR subgroups, in whom rates of the primary kidney outcome were highest, showed the largest absolute benefit (all P interaction <0.01). The risk of heart failure hospitalization was higher in the atrasentan group (hazard ratio, 1.39; 95% confidence interval, 0.97 to 1.99) and was consistent across subgroups, with no evidence that relative or absolute risks differed across eGFR or UACR subgroups (all P interaction >0.09).

CONCLUSIONS: Atrasentan reduced the relative risk of the primary kidney outcome consistently across baseline UACR and eGFR subgroups. The absolute risk reduction was greater among patients in the lowest eGFR and highest albuminuria category who were at highest baseline risk. Conversely, the relative and absolute risks of heart failure hospitalization were similar across baseline UACR and eGFR subgroups. Clinical Trial registry name and registration number: Study of Diabetic Nephropathy with Atrasentan (SONAR), NCT01858532.