IMPORTANCE: Steroidal mineralocorticoid receptor antagonists, when added to a renin-angiotensin system blocker, further reduce proteinuria in patients with chronic kidney disease but may be underused because of a high risk of adverse events. OBJECTIVE: To evaluate the safety and efficacy of different oral doses of the nonsteroidal mineralocorticoid receptor antagonist finerenone, given for 90 days to patients with diabetes and high or very high albuminuria who are receiving an angiotensin-converting enzyme inhibitor or an angiotensin receptor blocker. DESIGN, SETTING, AND PARTICIPANTS: Randomized, double-blind, placebo-controlled, parallel-group study conducted at 148 sites in 23 countries. Patients were recruited from June 2013 to February 2014 and the study was completed in August 2014. Of 1501 screened patients, 823 were randomized and 821 received study drug. INTERVENTIONS: Participants were randomly assigned to receive oral, once-daily finerenone (1.25 mg/d, n = 96; 2.5 mg/d, n = 92; 5 mg/d, n = 100; 7.5 mg/d, n = 97; 10 mg/d, n = 98; 15 mg/d, n = 125; and 25 mg/d, n = 119) or matching placebo (n = 94) for 90 days. MAIN OUTCOMES AND MEASURES: The primary outcome was the ratio of the urinary albumin-creatinine ratio (UACR) at day 90 vs at baseline. Safety end points were changes from baseline in serum potassium and estimated glomerular filtration rate. RESULTS: The mean age of the participants was 64.2 years; 78% were male. At baseline, 36.7% of patients treated had very high albuminuria (UACR ≥300 mg/g) and 40.0% had an estimated glomerular filtration rate of 60 mL/min/1.73 m2 or lower. Finerenone demonstrated a dose-dependent reduction in UACR. The primary outcome, the placebo-corrected mean ratio of the UACR at day 90 relative to baseline, was reduced in the finerenone 7.5-, 10-, 15-, and 20-mg/d groups (for 7.5 mg/d, 0.79 [90% CI, 0.68-0.91; P = .004]; for 10 mg/d, 0.76 [90% CI, 0.65-0.88; P = .001]; for 15 mg/d, 0.67 [90% CI, 0.58-0.77; P<.001]; for 20 mg/d, 0.62 [90% CI, 0.54-0.72; P < .001]). The prespecified secondary outcome of hyperkalemia leading to discontinuation was not observed in the placebo and finerenone 10-mg/d groups; incidences in the finerenone 7.5-, 15-, and 20-mg/d groups were 2.1%, 3.2%, and 1.7%, respectively. There were no differences in the incidence of the prespecified secondary outcome of an estimated glomerular filtration rate decrease of 30% or more or in incidences of adverse events and serious adverse events between the placebo and finerenone groups. CONCLUSIONS AND RELEVANCE: Among patients with diabetic nephropathy, most receiving an angiotensin-converting enzyme inhibitor or an angiotensin receptor blocker, the addition of finerenone compared with placebo resulted in improvement in the urinary albumin-creatinine ratio. Further trials are needed to compare finerenone with other active medications. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT1874431.