Finerenone initiation linked to lower albuminuria in small retrospective diabetic kidney disease cohort

A small retrospective observational study found that finerenone initiation was associated with reduced urine albumin-to-creatinine ratio during six months of follow-up in adults with diabetic kidney disease. Among 15 patients with measurements at baseline, three months and six months, median UACR declined from 57.0 mg/g at baseline to 30.0 mg/g at three months and 18.0 mg/g at six months, with a statistically significant within-patient time effect.

The analysis used routinely collected data from a nephrology practice in Varna, Bulgaria. It included 16 adults with documented diabetic kidney disease who started finerenone 10 mg once daily and had baseline UACR, serum creatinine and serum potassium measurements. No additional eGFR or UACR eligibility restriction was applied, and one patient lacked laboratory measurements at six months. The available records did not systematically capture dose escalation, temporary interruption, permanent discontinuation or adherence.

The cohort had substantial kidney and cardiovascular risk. Mean age was 68.4 ± 9.6 years, 12 of 16 patients were women, and 11 had diabetes for more than 10 years. All 16 had documented hypertension and 13 had documented cardiovascular disease. Median baseline eGFR was 35.6 mL/min/1.73 m², with 12 patients below 60 and four below 30. Median baseline UACR was 57.0 mg/g: six patients were in the A1 albuminuria category and 10 in A2, with none in A3.

Most participants were already receiving multiple cardiorenal therapies. Fifteen received an ACE inhibitor or angiotensin receptor blocker, and 15 also received an SGLT2 inhibitor; 14 received both classes. Two patients used a GLP-1 receptor agonist and 10 used a diuretic. Mean office blood pressure was 127.8 ± 7.7/81.3 ± 6.2 mmHg, although six patients had a recorded pressure of at least 140/90 mmHg, and mean HbA1c was 6.88 ± 0.80%.

The primary endpoint was within-patient UACR change over six months. Secondary measures included UACR at three months, the proportions of patients with reductions of at least 30% and at least 50%, longitudinal eGFR and serum creatinine, and serum potassium. The investigators used Friedman tests for longitudinal changes, followed where appropriate by two-sided Wilcoxon signed-rank tests with Holm adjustment. Responder proportions were reported with 95% Wilson confidence intervals, and no multivariable model was fitted because of the small sample.

The median relative UACR change was −28.4% at three months and −54.1% at six months. Holm-adjusted comparisons with baseline were significant at both follow-ups. At three months, seven of 16 patients had a UACR reduction of at least 30% and four had a reduction of at least 50%. At six months, the corresponding available-case rates were 12 of 15 patients, or 80.0% (95% CI 54.8–93.0%), and nine of 15, or 60.0% (95% CI 35.7–80.2%). A sensitivity analysis that counted the patient with missing six-month data as a non-responder produced rates of 75.0% and 56.3%, respectively.

No statistically significant longitudinal change was detected for eGFR or serum creatinine. Median eGFR was 35.6 mL/min/1.73 m² at baseline, 40.2 at three months and 40.9 at six months, while median serum creatinine was 143.0, 135.5 and 130.0 µmol/L, respectively. Both Friedman tests produced p = 0.112. These descriptive findings did not establish equivalence, long-term kidney stability or an absence of clinically important change, and the short observation period could not assess kidney failure or sustained eGFR decline.

Mean serum potassium was 4.67 ± 0.45 mmol/L at baseline, 4.60 ± 0.41 at three months and 4.80 ± 0.45 at six months, with no significant longitudinal change detected by the Friedman test (p = 0.166). Two patients, or 12.5%, had at least one post-baseline value of at least 5.5 mmol/L, although one had already measured 5.5 mmol/L at baseline. No post-baseline value exceeded 5.5 mmol/L, and none reached 6.0 mmol/L. The findings do not demonstrate that hyperkalemia risk was absent, particularly because follow-up was short and information on dose modification, interruption and other adverse events was not systematically available.

The results describe temporal associations rather than causal treatment effects. The retrospective, single-center design, sample of 16, absence of a comparator and incomplete follow-up left the analysis vulnerable to regression to the mean, changes in background therapy, adherence and other unmeasured confounding. UACR was measured in routine spot urine samples without standardized repeated confirmation, and the small cohort could not support adjusted analyses or meaningful subgroup comparisons. Inclusion of patients with A1 albuminuria, eGFR below 25 mL/min/1.73 m² or baseline potassium of 5.5 mmol/L also limits direct comparison with pivotal trial populations. The authors characterized the findings as exploratory and called for confirmation in larger controlled real-world studies with standardized albuminuria assessment, more complete safety capture and longer follow-up.

Subject of Research: Six-month kidney-function, albuminuria and potassium outcomes after finerenone initiation in adults with diabetic kidney disease

Article Title: Finerenone in diabetic kidney disease: Findings from a single-center retrospective observational study

Article References: Shaleva, R., Danailova, S., Benkova-Petrova, M., Petrov, A., Petrov, P., Staykova, S., & Nanova, P. (2026). Finerenone in diabetic kidney disease: Findings from a single-center retrospective observational study. Pharmacia, 73, Article e189397. https://doi.org/10.3897/pharmacia.73.e189397

Image Credits: AI Generated

DOI: 10.3897/pharmacia.73.e189397

Keywords: finerenone, diabetic kidney disease, albuminuria, UACR, estimated GFR, serum potassium, hyperkalemia, retrospective cohort, SGLT2 inhibitors, ACE inhibitors, diabetic, kidney