Patients with type 2 diabetes are at high risk of developing renal and cardiovascular complications. Sodium–glucose cotransporter 2 (SGLT2) inhibitors have garnered interest due to their glucose-independent cardiorenal protective effects, as reported in trials including participants with and without diabetes, such as Dapagliflozin And Prevention of Adverse outcomes in Chronic Kidney Disease (DAPA-CKD) (1,2). These trials have demonstrated that SGLT2 inhibitors reduce cardiovascular disease (CVD) risk, especially hospitalization for heart failure (1,2). Despite these clinical benefits, the underlying physiological mechanisms of SGLT2 inhibitors are incompletely understood, particularly in patients with chronic kidney disease (CKD). Accordingly, this analysis examined the impact of treatment with an SGLT2 inhibitor, ertugliflozin, on markers of plasma volume contraction and myocardial strain in participants with type 2 diabetes and moderate CKD.
We performed a post hoc exploratory analysis in a subset of 231 participants from the eValuation of ERTugliflozin efficacy and Safety (VERTIS) RENAL trial (clinical trial reg. no. NCT01986855, ClinicalTrials.gov) with type 2 diabetes and stage 3 CKD (estimated glomerular filtration rate [eGFR] 30–59 mL/min/1.73 m2) who were randomized to SGLT2 inhibitor therapy with ertugliflozin (5 mg or 15 mg daily; pooled herein) or placebo (3). Clinical and biomarker measurements were obtained at baseline and 26 weeks and 52 weeks postrandomization. Biomarkers were quantified with Luminex xMAP (cardiac troponin, renin, and N-terminal pro B-type natriuretic peptide [NT-proBNP]) or ELISA (atrial natriuretic peptide [ANP], human erythropoietin [EPO], ACE, and ACE2). Aldosterone was quantified by DiaSorin LIAISON XL Analyzer based on competitive chemiluminescent immunoassay. Differences in longitudinal changes in biomarkers among participants receiving either placebo or ertugliflozin were analyzed with use of linear mixed-effects models. We performed a correlation analysis between changes in hematocrit (HCT) and biomarkers using the Spearman correlation coefficient.
Median age of participants (49% of whom were male) was 67 years and duration of type 2 diabetes 12.9 years, with 49% having a history of CVD. The median glycated hemoglobin (HbA1c) level was 8.0%, eGFR 47.8 mL/min/1.73 m2, systolic blood pressure 133.5 mmHg, and urine albumin-to-creatinine ratio 28.1 mg/g. The majority (97%) of study participants were on antihypertensive therapy (renin-angiotensin-aldosterone system inhibitors 87%, diuretics 59%, α-/β-blockers 62%).
Among participants randomized to ertugliflozin, compared with placebo, plasma aldosterone was significantly higher at 26 weeks (P = 0.005) (Fig. 1), an effect that was no longer significant at 52 weeks. NT-proBNP was lower among those randomized to ertugliflozin at 26 and 52 weeks (P = 0.020 and P = 0.035, respectively). Plasma renin and ANP showed similar, nonsignificant directional trends as aldosterone and NT-proBNP, respectively. Serum albumin was significantly increased in participants treated with ertugliflozin at 26 weeks (P = 0.009) but returned to baseline at 52 weeks. HCT was increased at 26 and 52 weeks among those randomized to ertugliflozin compared with placebo (P < 0.001 at both time points). Changes in HCT correlated with changes in NT-proBNP (P = 0.03), renin (P = 0.005), aldosterone (P = 0.001), and ACE2 (at 26 weeks, P = 0.047). In contrast, HCT did not correlate with changes in EPO (P = 0.363). Randomization to ertugliflozin was not associated with significant changes in cardiac troponin, EPO, ACE, ACE2, HbA1c, potassium, blood pressure, and eGFR.
Change in biomarkers and clinical parameters during ertugliflozin treatment and placebo. Black lines/circles, placebo; red lines/circles, ertugliflozin (pooled 5 and 15 mg/day). Data are presented as median % change ± IQR or mean % change ± SD for potassium and serum albumin. All data were analyzed with a mixed-effect model for repeated measures under the missing-at-random framework based on the restricted maximum likelihood method for estimation. All mixed-effect models for repeated measures included baseline biomarker/clinical parameter levels, treatment, visit, and treatment-by-visit interaction as covariates. All biomarkers and clinical parameters were log transformed before analysis to mitigate skew, and model residuals were inspected. Visits were treated as repeated-measure units from the same individual. A contrast statement with combination of treatment and treatment-by-time interaction was used to test the overall difference in changes between the placebo and treatment groups at all time points. All statistical tests comparing treatment effects were two sided and used a 5% significance level. *P < 0.05. As this was an exploratory analysis, we did not correct for multiple hypothesis testing. All analyses were performed with SAS 9.4 for Windows (SAS Institute, Cary, NC).
Change in biomarkers and clinical parameters during ertugliflozin treatment and placebo. Black lines/circles, placebo; red lines/circles, ertugliflozin (pooled 5 and 15 mg/day). Data are presented as median % change ± IQR or mean % change ± SD for potassium and serum albumin. All data were analyzed with a mixed-effect model for repeated measures under the missing-at-random framework based on the restricted maximum likelihood method for estimation. All mixed-effect models for repeated measures included baseline biomarker/clinical parameter levels, treatment, visit, and treatment-by-visit interaction as covariates. All biomarkers and clinical parameters were log transformed before analysis to mitigate skew, and model residuals were inspected. Visits were treated as repeated-measure units from the same individual. A contrast statement with combination of treatment and treatment-by-time interaction was used to test the overall difference in changes between the placebo and treatment groups at all time points. All statistical tests comparing treatment effects were two sided and used a 5% significance level. *P < 0.05. As this was an exploratory analysis, we did not correct for multiple hypothesis testing. All analyses were performed with SAS 9.4 for Windows (SAS Institute, Cary, NC).
In this analysis, treatment with ertugliflozin was associated with markers of volume contraction. Natriuretic peptides such as NT-proBNP are secreted in response to increased hydrostatic load and are routinely used as clinical biomarkers of volume overload and myocardial dysfunction. Our findings of reduced NT-proBNP, increased HCT, and elevated aldosterone suggest plasma volume contraction after ertugliflozin treatment in patients with moderate CKD, which is consistent with findings of previous studies in patients without CKD (1,4). These physiologic effects may contribute to the reduction in hospitalization for heart failure observed with ertugliflozin (4). It is uncertain why aldosterone levels returned toward baseline at 52 weeks in our study; we speculate that these findings are due to a circulating volume resetting phenomenon. Our study provides important insight into the mechanism of increased HCT with SGLT2 inhibitor therapy. The observation that the increase in HCT with SGLT2 inhibitor therapy was neither accompanied by nor correlated with an increase in EPO suggests that it occurs predominantly due to volume contraction, rather than increased erythropoiesis, at least over the time frame of this analysis. Increased HCT levels are associated with cardioprotective effects of SGLT2 inhibitors, possibly through increasing oxygen delivery to tissues (5).
In conclusion, effects of ertugliflozin on natriuretic peptides and neurohormones are preserved in patients with moderate CKD and consistent with plasma volume reduction. The rise in HCT without a change in erythropoiesis supports the volume contraction mechanism of SGLT2 inhibitor therapy. These effects may be important physiologically and contribute to a lower risk of CVD with SGLT2 inhibitor treatment in people with type 2 diabetes.
P.R.L. and H.L. contributed equally.
Article Information
Funding. D.Z.I.C. is supported by a Department of Medicine, University of Toronto, Merit Award and receives support from the Canadian Institutes for Health Research (CIHR), Diabetes Canada, and the Heart & Stroke/Richard Lewar Centres of Excellence in Cardiovascular Research. P.R.L. is supported by the Department of Medicine, University of Toronto, and receives unrelated funding from the CIHR, the Peter Munk Cardiac Centre, the LifeArc Foundation, the Province of Ontario, and the Ted Rogers Centre for Heart Research and consulting fees from Brigham and Women’s Hospital. L.E.L. receives support from a CIHR Canada Graduate Scholarship Doctoral Award.
Duality of Interest. The Impact of ErtUgliflozin on REnal and Cardiovascular BiomarKers in Patients with DiaAbetes: An Exploratory Analysis of the Phase 3 Clinical Trial Program (EUREKA) study was an investigator-initiated study (D.Z.I.C.) and an exploratory post hoc analysis of the VERTIS RENAL funded by Merck Sharp & Dohme, a subsidiary of Merck & Co. (Kenilworth, NJ) in collaboration with Pfizer (New York, NY). D.Z.I.C. has received consulting fees or speaking honorarium or both from Janssen, Bayer, Boehringer Ingelheim, Eli Lilly, AstraZeneca, Merck & Co., Prometic, Novo Nordisk, and Sanofi and has received operating funds from Janssen, Boehringer Ingelheim, Eli Lilly, Sanofi, AstraZeneca, Novo Nordisk, and Merck & Co. P.R.L. receives consulting fees from Novartis and Corrona LLC. No other potential conflicts of interest relevant to this article were reported.
Author Contributions. P.R.L., H.L., L.E.L., and D.Z.I.C. contributed to the study concept, design, analysis, first draft of the manuscript, and interpretation of data. P.R.L., H.L., C.F., L.E.L., Y.L., D.Bu., K.D.B., D.Br., and D.Z.I.C. contributed to data analysis, provided critical edits, and reviewed and approved the final manuscript. D.Z.I.C. and P.R.L. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
