Sodium-glucose cotransporter-2 (SGLT2) inhibitors lower risk of cardiovascular and kidney events in people with type 2 diabetes. In the empagliflozin in type 1 diabetes clinical program (Empagliflozin as Adjunctive to inSulin thErapy [EASE]), glycemic control, body weight, and BP improved with empagliflozin as adjunct to insulin, although diabetic ketoacidosis risk was higher with the 10- and 25-mg doses versus the 2.5-mg dose. The kidney effects of empagliflozin in type 1 diabetes remain incompletely understood. Here, we report changes in kidney parameters in two phase 3 placebo-controlled trials, Empagliflozin as Adjunctive to inSulin thErapy over 52 Weeks in Patients with Type 1 Diabetes Mellitus (EASE-2; empagliflozin 10/25 mg; 52 weeks; n=730) and Empagliflozin as Adjunctive to Insulin Therapy over 26 Weeks in Patients with T1DM (EASE-3; empagliflozin 2.5/10/25 mg; 26 weeks; n=975).
Baseline characteristics of the EASE cohort have been reported previously (1). Briefly, mean ± SD baseline eGFRs (milliliters per minute per 1.73 m2) in EASE-2/EASE-3 were 97±18 and 99±19, respectively, and median (interquartile range) baseline urine albumin-creatinine ratio (UACR; milligrams per gram creatinine) was 6.2 (2.7–14.1) in both studies. After 26 weeks of treatment in EASE-3, mean placebo-corrected eGFR changes with empagliflozin 2.5, 10, and 25 mg were −0.1 (P=0.87), −3 (P=0.004), and −4 (P<0.001), respectively. After 52 weeks in EASE-2, mean placebo-corrected eGFR changes with empagliflozin 10 and 25 mg were −2 (P=0.02) and −3 (P=0.002), respectively. eGFR changes 3 weeks post-treatment (follow-up) returned to above baseline (Figure 1A).
;)
Figure 1-_A.: Effects of empagliflozin versus placebo on kidney parameters, hematocrit (HCT), and serum albumin and uric acid levels in the EASE-2 and EASE-3 trials. (A) eGFR change from baseline (BL). (Left panel) Empagliflozin as Adjunctive to inSulin thErapy over 52 Weeks in Patients with Type 1 Diabetes Mellitus (EASE-2). The model includes BL eGFR (P<0.001) as the linear covariate and treatment (P=0.001), visit (P<0.001), visit by treatment interaction (P<0.001), and BL eGFR by visit interaction (P=0.004) as fixed effects. Week 52 difference versus placebo: *empagliflozin 10 mg, –2.0; 95% confidence interval (95% CI), –4.0 to –0.1; P=0.04; †empagliflozin 25 mg, –2.0; 95% CI, –4.0 to –0.4; P=0.02. (Right panel) Empagliflozin as Adjunctive to Insulin Therapy over 26 Weeks in Patients with T1DM (EASE-3). The model includes BL eGFR (P<0.001) as the linear covariate and treatment (P=0.13), visit (P=0.003), visit by treatment interaction (P<0.001), and BL eGFR by visit interaction (P=0.14) as fixed effects. Week 26 difference versus placebo: ‡empagliflozin 10 mg, –1.0; 95% CI, –3.0 to 0.6; P=0.17); §empagliflozin 25 mg, –2.0; 95% CI, –4.0 to –0.1; P=0.04). (B and C) Change from BL in urine albumin-creatinine ratio (UACR) by albuminuric status at 26 weeks. The model includes BL eGFR and BL glycated hemoglobin as linear covariates and BL preexisting insulin therapy, treatment, visit, BL UACR, treatment by visit interaction, visit by BL UACR interaction, treatment by BL UACR interaction, and treatment by visit by BL UACR interaction as fixed effects. (D) Change from BL in HCT. (Left panel) EASE-2. The model includes BL HCT (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.03), visit by treatment interaction (P<0.001), and BL HCT by visit interaction (P=0.004) as fixed effects. (Right panel) EASE-3. The model includes BL HCT (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.01), visit by treatment interaction (P<0.001), and BL HCT by visit interaction (P=0.15) as fixed effects. (E) Change from BL serum albumin. (Left panel) EASE-2. The model includes BL albumin (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.03), visit by treatment interaction (P=0.001), and BL albumin by visit interaction (P=0.02) as fixed effects. (Right panel) EASE-3. The model includes BL albumin (P<0.001) as the linear covariate and treatment (P=0.04), visit (P=0.46), visit by treatment interaction (P<0.001), and BL albumin by visit interaction (P=0.29) as fixed effects. (F) Change from BL in uric acid. (Left panel) EASE-2. The model includes BL uric acid (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.001), visit by treatment interaction (P=0.002), and BL uric acid by visit interaction (P=0.01) as fixed effects. (Right panel) EASE-3. The model includes BL uric acid (P<0.001) as the linear covariate and treatment (P<0.001), visit (P<0.001), visit by treatment interaction (P=0.07), and BL uric acid by visit interaction (P<0.001) as fixed effects. EASE, Empagliflozin as Adjunctive to inSulin thErapy in type 1 diabetes; FU, follow-up; N/A, not analyzed.
Figure 1-_B.: Effects of empagliflozin versus placebo on kidney parameters, hematocrit (HCT), and serum albumin and uric acid levels in the EASE-2 and EASE-3 trials. (A) eGFR change from baseline (BL). (Left panel) Empagliflozin as Adjunctive to inSulin thErapy over 52 Weeks in Patients with Type 1 Diabetes Mellitus (EASE-2). The model includes BL eGFR (P<0.001) as the linear covariate and treatment (P=0.001), visit (P<0.001), visit by treatment interaction (P<0.001), and BL eGFR by visit interaction (P=0.004) as fixed effects. Week 52 difference versus placebo: *empagliflozin 10 mg, –2.0; 95% confidence interval (95% CI), –4.0 to –0.1; P=0.04; †empagliflozin 25 mg, –2.0; 95% CI, –4.0 to –0.4; P=0.02. (Right panel) Empagliflozin as Adjunctive to Insulin Therapy over 26 Weeks in Patients with T1DM (EASE-3). The model includes BL eGFR (P<0.001) as the linear covariate and treatment (P=0.13), visit (P=0.003), visit by treatment interaction (P<0.001), and BL eGFR by visit interaction (P=0.14) as fixed effects. Week 26 difference versus placebo: ‡empagliflozin 10 mg, –1.0; 95% CI, –3.0 to 0.6; P=0.17); §empagliflozin 25 mg, –2.0; 95% CI, –4.0 to –0.1; P=0.04). (B and C) Change from BL in urine albumin-creatinine ratio (UACR) by albuminuric status at 26 weeks. The model includes BL eGFR and BL glycated hemoglobin as linear covariates and BL preexisting insulin therapy, treatment, visit, BL UACR, treatment by visit interaction, visit by BL UACR interaction, treatment by BL UACR interaction, and treatment by visit by BL UACR interaction as fixed effects. (D) Change from BL in HCT. (Left panel) EASE-2. The model includes BL HCT (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.03), visit by treatment interaction (P<0.001), and BL HCT by visit interaction (P=0.004) as fixed effects. (Right panel) EASE-3. The model includes BL HCT (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.01), visit by treatment interaction (P<0.001), and BL HCT by visit interaction (P=0.15) as fixed effects. (E) Change from BL serum albumin. (Left panel) EASE-2. The model includes BL albumin (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.03), visit by treatment interaction (P=0.001), and BL albumin by visit interaction (P=0.02) as fixed effects. (Right panel) EASE-3. The model includes BL albumin (P<0.001) as the linear covariate and treatment (P=0.04), visit (P=0.46), visit by treatment interaction (P<0.001), and BL albumin by visit interaction (P=0.29) as fixed effects. (F) Change from BL in uric acid. (Left panel) EASE-2. The model includes BL uric acid (P<0.001) as the linear covariate and treatment (P<0.001), visit (P=0.001), visit by treatment interaction (P=0.002), and BL uric acid by visit interaction (P=0.01) as fixed effects. (Right panel) EASE-3. The model includes BL uric acid (P<0.001) as the linear covariate and treatment (P<0.001), visit (P<0.001), visit by treatment interaction (P=0.07), and BL uric acid by visit interaction (P<0.001) as fixed effects. EASE, Empagliflozin as Adjunctive to inSulin thErapy in type 1 diabetes; FU, follow-up; N/A, not analyzed.
In participants with UACR<30 mg/g, no significant changes in UACR were observed. In EASE-3, in people with baseline UACR ≥30 mg/g, empagliflozin 2.5 mg (n=36) for 26 weeks did not attenuate UACR versus placebo (n=34) (Figure 1B). In a pooled analysis (EASE-2 + EASE-3), in participants with baseline UACR ≥30 mg/g, UACR decreased by 16% (P=0.27) and 30% (P=0.02) with empagliflozin 10 (n=71) and 25 mg (n=77), respectively, versus placebo (n=65) at 26 weeks (Figure 1C). Hematocrit and serum albumin increased in EASE-2 and EASE-3 with empagliflozin, whereas serum uric acid decreased; changes in these parameters returned toward baseline at follow-up after washout following 26 and 52 weeks of treatment (Figure 1, D–F).
Previous analyses with other SGLT2 inhibitors suggest that these therapies lead to similar changes in eGFR and UACR in people with type 1 diabetes versus type 2 diabetes. Importantly, the initial small eGFR dip is likely hemodynamic related (2); rapidly reversible after brief washout periods, irrespective of diabetes status; and not linked with kidney injury (3). Consistent with observations in people with type 2 diabetes, changes in eGFR 3 weeks post-treatment returned toward baseline levels. Importantly, empagliflozin 2.5 mg daily resulted in attenuation of the eGFR change. This is possibly the result of blunted physiologic effects seen with this low dose of empagliflozin, which still had glucosuric effects, together with chronic adaptive mechanisms involving increased reabsorption by the loop of Henle (4).
Albuminuria is a recognized surrogate marker of kidney and cardiovascular disease progression; pharmacologic reductions in UACR are linked with improved cardiorenal outcomes (5). Similar to observations in type 2 diabetes, in the pooled analysis in participants with baseline UACR ≥30 mg/g, UACR decreased by 55% with empagliflozin 25 mg versus 36% with placebo at 26 weeks (P=0.02). Accordingly, the magnitude of this effect is generally consistent with what has been reported in type 2 diabetes. Nevertheless, the 2.5-mg dose seems to be insufficient to lower UACR relative to effects observed in clinical trial settings for high doses.
Hematocrit and serum albumin increased in EASE-2 and EASE-3 with empagliflozin. The rise in hematocrit with SGLT2 inhibition is important because it is the routine laboratory parameter most closely linked with improved cardiovascular outcomes in mediation analyses. Yet, it is not known if the rise in hematocrit is on the basis of hemoconcentration and natriuresis or a rise in erythropoietin. In the context of elevations in serum albumin levels, which are not sensitive to erythropoietin, and on the basis of mechanistic studies failing to detect a long-term rise in erythropoietin, the most likely explanation involves natriuresis leading to hemoconcentration. This is especially likely as changes in hematocrit and albumin were rapidly reversible, making volume, not erythropoiesis, a more likely culprit. Natriuresis and volume-related effects may in turn protect against hospitalization for heart failure and may contribute to BP lowering. Uric acid, a similarly routine blood test, is also lowered by SGLT2 inhibition, even at very low empagliflozin doses, possibly via tubular secretion of uric acid in exchange for luminal glucose. Consequently, glucosuria is associated with increased uricosuria and, hence, serum uric acid lowering—an effect also associated with reduced cardiovascular risk in mediation analyses. Similar to observations in people with type 2 diabetes, serum uric acid decreased in people with type 1 diabetes, and changes in these parameters returned to near-baseline values at follow-up after 26 and 52 weeks of treatment.
In conclusion, empagliflozin 10 and 25 mg/d as adjunct to insulin in type 1 diabetes, but not 2.5 mg/d, resulted in short-term changes in kidney markers comparable with changes observed with SGLT2 inhibitor use in type 2 diabetes.
Disclosures
P. Bjornstad reports consultancy agreements with AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Horizon Pharma, Lilly USA, Novo Nordisk, Sanofi, and XORTX; receiving research funding from AstraZeneca, Horizon Pharma, and Merck; receiving honoraria from Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Horizon Pharma, and Novo Nordisk; and serving as a scientific advisor or member of AstraZeneca, Bayer, Boehringer Ingelheim, Horizon Pharma, Novo Nordisk, and XORTX. D. Cherney reports consultancy agreements with Abbvie, AstraZeneca, Bayer, Boehringer Ingelheim-Lilly, Bristol Myers Squibb, CSL Pharma, Janssen, Maze, Merck, Mitsubishi-Tanabe, Novo Nordisk, Prometic, and Sanofi; receiving research funding from AstraZeneca, Boehringer Ingelheim-Lilly, Janssen, Merck, Novo Nordisk, and Sanofi; receiving honoraria from Abbvie, Astellas, AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, JNJ, Lilly, Merck, Novo Nordisk, Otsuka, Prometic, and Sanofi; and serving as a scientific advisor or member of AstraZeneca, Boehringer Ingelheim, Janssen, Merck, Novo Nordisk, and Sanofi. J. Marquard, D. Neubacher, and N. Soleymanlou are employees of Boehringer Ingelheim. B. Perkins reports consultancy agreements with Abbott, Boehringer Ingelheim, and Insulet; receiving research funding from the Bank of Montreal and Novo Nordisk; receiving honoraria from Abbott, Insulet, Medtronic, Novo Nordisk, and Sanofi; and serving as a scientific advisor or member of Abbott, Boehringer Ingelheim, Insulet, and Sanofi. J. Rosenstock reports consultancy agreements with Applied Therapeutics, Boehringer Ingelheim, Eli Lilly, Hanmi, Intarcia, Janssen, Novo Nordisk, Oramed, and Sanofi; receiving research funding from Applied Therapeutics, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Genentech, GlaxoSmithKline, Hanmi, Intarcia, Janssen, Lexicon, Merck, Metacrine, Novo Nordisk, Oramed, Pfizer, REMD Bio, and Sanofi; receiving honoraria from Applied Therapeutics, Boehringer Ingelheim, Eli Lilly, Hanmi, Intarcia, Janssen, Novo Nordisk, Oramed, and Sanofi; serving as a scientific advisor or member of Applied Therapeutics, Boehringer Ingelheim, Eli Lilly, Hanmi, Intarcia, Janssen, Novo Nordisk, Oramed, and Sanofi; serving as an associate editor of Diabetes Care; and serving on speakers bureaus for Applied Therapeutics, Boehringer Ingelheim, Eli Lilly, Intarcia, Janssen, Novo Nordisk, Oramed, and Sanofi.
Funding
The EASE-2 and EASE-3 clinical trials were supported by the Boehringer Ingelheim & Eli Lilly and Company Diabetes Alliance.
Acknowledgments
The authors thank the EASE-3 participants who generously volunteered their invaluable time toward this research, all of the EASE-2 and EASE-3 investigators and site research professionals and staff who took part in the conduct of these trials, and Dr. Jens Eilbracht and Ms. Ros Swallow along with the sponsor’s clinical trial teams for their diligent oversight of the operational conduct of these trials. Editorial support, supported financially by Boehringer Ingelheim, was provided by Mr. Charlie Bellinger and Dr. Katarina Kolaric of Elevate Scientific Solutions during the preparation of this article.
Data from this study were presented in abstract form at the American Society of Nephrology Kidney Week held October 20–25, 2020.
Clinical Trial registry name and registration number: Empagliflozin as Adjunctive to inSulin thErapy over 52 Weeks in Patients with Type 1 Diabetes Mellitus (EASE-2), NCT02414958; Empagliflozin as Adjunctive to Insulin Therapy over 26 Weeks in Patients with T1DM (EASE-3), NCT02580591.
The authors are fully responsible for all content and editorial decisions, were involved at all stages of manuscript development, and have approved the final version.
P. Bjornstad, D.Z.I. Cherney, J. Marquard, D. Neubacher, and N. Soleymanlou developed the research proposal and planned analyses; B.A. Perkins was the coordinating investigator of the EASE-2 study; J. Rosenstock was the coordinating investigator of the EASE-3 study; all authors took part in interpreting the results; D.Z.I. Cherney wrote the manuscript; all authors took part in reviewing and editing the manuscript; and D.Z.I. Cherney is the guarantor of this work and as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Data Sharing Statement
The sponsor of the EASE trials (Boehringer Ingelheim) is committed to responsible sharing of clinical study reports, related clinical documents, and patient-level clinical study data. Researchers are invited to submit inquiries via the following website (https://trials.boehringer-ingelheim.com/).
References
1. Rosenstock J, Marquard J, Laffel LM, Neubacher D, Kaspers S, Cherney DZ, Zinman B, Skyler JS, George J, Soleymanlou N, Perkins BA: Empagliflozin as adjunctive to insulin therapy in type 1 diabetes: The EASE trials. Diabetes Care 41: 2560–2569, 2018
2. Cherney DZ, Perkins BA, Soleymanlou N, Maione M, Lai V, Lee A, Fagan NM, Woerle HJ, Johansen OE, Broedl UC, von Eynatten M: Renal hemodynamic effect of sodium-glucose cotransporter 2 inhibition in patients with type 1 diabetes mellitus. Circulation 129: 587–597, 2014
3. Oshima M, Jardine MJ, Agarwal R, Bakris G, Cannon CP, Charytan DM, de Zeeuw D, Edwards R, Greene T, Levin A, Lim SK, Mahaffey KW, Neal B, Pollock C, Rosenthal N, Wheeler DC, Zhang H, Zinman B, Perkovic V, Heerspink HJL: Insights from CREDENCE trial indicate an acute drop in estimated glomerular filtration rate during treatment with canagliflozin with implications for clinical practice. Kidney Int 99: 999–1009, 2021
4. Thomson SC, Rieg T, Miracle C, Mansoury H, Whaley J, Vallon V, Singh P: Acute and chronic effects of SGLT2 blockade on glomerular and tubular function in the early diabetic rat. Am J Physiol Regul Integr Comp Physiol 302: R75–R83, 2012
5. Oshima M, Neuen BL, Li J, Perkovic V, Charytan DM, de Zeeuw D, Edwards R, Greene T, Levin A, Mahaffey KW, De Nicola L, Pollock C, Rosenthal N, Wheeler DC, Jardine MJ, Heerspink HJL: Early change in albuminuria with canagliflozin predicts kidney and cardiovascular outcomes: A
post hoc analysis from the CREDENCE trial. J Am Soc Nephrol 31: 2925–2936, 2020
