Efficacy and safety of metformin versus empagliflozin on chronic kidney disease progression (MET-EMPA-CKD): a randomized controlled trial

Abstract

Background: Chronic kidney disease (CKD) is a devastating progressive condition accompanied with high morbidity and mortality rates. Sodium-glucose cotransporter-2 (SGLT2) inhibitors have recently proven their renoprotective effects, whereas evidence for metformin remains limited but suggestive of potential benefit. This study aimed at comparing the efficacy and safety of metformin versus empagliflozin, a SGLT2 inhibitor, on retarding CKD progression with exploring supposed mechanistic pathways in clinical settings.

Methods: In this 12-month randomized controlled trial, 120 moderate CKD patients were randomized into three groups: metformin 1000 mg/day (n = 40) or empagliflozin 10 mg/day (n = 40), both added orally to standard treatment, or control who continued standard of care (n = 40). The primary outcome was changes in estimated glomerular filtration rate (eGFR). Secondary analyses assessed percent changes of urinary albumin-to-creatinine ratio (uACR), transforming growth factor-β1 (TGF-β1), kidney injury molecule (KIM)-1, and beclin-1 (an autophagy biomarker). Other metabolic and safety issues were also assessed.

Results: 118 patients completed the study with comparable baseline data. Metformin and empagliflozin halted the decline in eGFR at study end with adjusted mean difference ± SE: 8.91 ± 1.92 (p˂0.001) and 5.1 ± 1.89 (p = 0.03), respectively, compared to control group. Metformin preserved its effect in diabetics and non-diabetics, with superiority than empagliflozin in non-diabetics. uACR was lowered by metformin and empagliflozin than control. Both of them tended to halt the deterioration of intermediates with %relative change of -28.8% (95% CI, -44.4 to -9, p = 0.003) and 179.3% (95% CI, 32.2 to 490, p = 0.003), for metformin versus control in TGF-β1 and beclin-1 levels, respectively. Empagliflozin reduced KIM-1 compared to control [-29% (95% CI, -49.3 to -0.5, p = 0.045)]. Study treatments showed benefits on lipid profile without changing urate levels significantly compared to the control arm. No significant changes were found between metformin and empagliflozin. Adverse effects were comparable across groups with tolerable increased urination frequency by empagliflozin.

Conclusion: 12-month metformin therapy demonstrated renoprotective effects comparable to empagliflozin, with a greater effect observed among non-diabetics as an exploratory insight. Metformin's renal actions were linked to antifibrotic and favorable autophagy effects while, empagliflozin preserved mainly tubular injury. Safety issues were generally comparable.

Gov identifier: NCT05373680, registered on 13/5/2022 "retrospectively".

Keywords: Autophagy; CKD progression; Empagliflozin; KIM-1; Metformin; Renoprotection; TGF-β.

Fig. 1

CONSORT flow diagram of the study participants. CKD: chronic kidney disease. *Referred to the evidence-based management of CKD, tailored according to relevant guideline recommendations, including blood pressure, glycemic, and lipid control with other supportive care as appropriate

Fig. 2

Trend changes in eGFR over 12-month follow-up period in pooled studied groups. Data are expressed as mean ± SE. eGFR: estimated glomerular filtration rate.** Independent p-value ˂ 0.01, ***Independent p-value ˂ 0.001, eGFR mean is adjusted to baseline eGFR and uACR levels

Fig. 3

Effects of study groups on uACR, markers of fibrosis, kidney injury, and autophagy with their % relative changes. a: Geometric mean of uACR with % relative changes, b: Geometric mean of TGF-β1with % relative changes, c: Geometric mean of KIM-1with % relative changes, d: Geometric mean of beclin-1 with % relative changes. uACR: urinary albumin-to-creatinine ratio; TGF-β1: transforming growth factor-β1; KIM-1: kidney injury molecule-1. Bars represent geometric means and error bars represent their 95% confidence interval, *Significant difference within group (p ˂ 0.05), **Significant difference within group (p ˂ 0.01), ***Significant difference within group (p ˂ 0.001)

Fig. 4

Significant pooled correlation for studied biomarkers after 12-month study period (n = 118). a: Correlation between eGFR and uACR; b: Correlation between eGFR and TGF-β1; c: Correlation between eGFR and KIM-1; d: Correlation between uACR and TGF-β1; e: Correlation between uACR and KIM-1; f: Correlation between KIM-1 and beclin-1; r: spearman’s correlation coefficient; eGFR: estimated glomerular filtration rate; uACR: urinary albumin-to-creatinine ratio; TGF-β1: transforming growth factor-β1; KIM-1: kidney injury molecule-1, *Statistically significant (p<0.05)

Fig. 5

Recorded adverse events in studied groups. GI: gastrointestinal; UTI: urinary tract infection; AKI: acute kidney injury, *Significant difference among groups (p<0.05)