Are All SGLT2 Inhibitors Created Equal?

Kidney and cardiovascular protective effects of sodium-glucose cotransporter-2 (SGLT2) inhibitors have been demonstrated in patients with and without CKD. While kidney-related outcomes have focused on the achievement of clinical end points, such as the doubling of the serum creatinine level or requirement for KRT, several recent large clinical trials have investigated the effect of SGLT2 inhibitors on the slope of the eGFR.

In this issue of CJASN, Cherney et al. (1) published a prespecified exploratory analysis of the Evaluation of Ertugliflozin Efficacy and Safety Cardiovascular Outcomes Trial (VERTIS CV), a randomized, placebo-controlled trial evaluating the effect of the SGLT2 inhibitor ertugliflozin on eGFR slope (assessed using the Modification of Diet in Renal Disease equation) over up to 5 years of follow-up. Individuals randomized to receive either 5 mg daily or 15 mg daily of ertugliflozin were pooled for this analysis and compared with the placebo group. The primary analysis assessed the eGFR slope from week 6 through each year of follow-up of the study to take into account the expected eGFR dip that occurs in the weeks after initiation of ertugliflozin. They found that ertugliflozin led to a clinically significant improvement in eGFR slope, defined as a between-groups improvement significantly >0.75 ml/min per 1.73 m² per year over 3 years. They reported that from week 6 to week 156, the eGFR slope was −0.32 (95% confidence interval [95% CI], −0.45 to −0.19) ml/min per 1.73 m² per year in the ertugliflozin group and −1.51 (95% CI, −1.70 to −1.32) ml/min per 1.73 m² per year in the placebo group, with a between-group difference of 1.19 (95% CI, 0.95 to 1.42) ml/min per 1.73 m² per year.

Prior publications from VERTIS CV reported that the secondary composite kidney outcome (doubling of serum creatinine, KRT, or kidney disease–related death) trended toward, but did not meet, the threshold for statistical significance (2). However, the change in eGFR and the composite outcome of sustained ≥40% reduction in eGFR, KRT, or kidney disease–related death was significantly reduced (3). This analysis adds to the previously published VERTIS CV kidney outcomes data by providing a detailed assessment of eGFR slope, which is an important surrogate of meaningful end points and offers more power for studies of individuals with earlier stages of CKD that may not reach a prespecified outcome threshold for change in creatinine or eGFR (4). Sustained benefit in decreasing the eGFR slope compared with placebo has been demonstrated for canagliflozin, dapagliflozin, empagliflozin, and now ertugliflozin, and the effects on kidney function decline appears to be a class effect rather than a feature limited to individual SGLT2 inhibitors (Figure 1) (1,5–11).

Figure 1.

The difference in the eGFR slope between sodium-glucose cotransporter-2 (SGLT2) inhibitors and placebo may relate to the magnitude of the eGFR dip. The forest plot shows the difference in the eGFR slope (95% confidence interval [95% CI]) between the SGLT2 inhibitor group and the placebo group, as was reported in large randomized clinical trials that studied different agents. Studies are arranged in order of the magnitude of the absolute eGFR dip from baseline to first postbaseline measurement, which was between weeks 2 and 13 for these trials. ^aIn the EMPEROR-Reduced subgroup analysis by CKD status, CKD was defined as an eGFR <60 ml/min per 1.73 m² or a urine albumin-creatinine ratio >300 mg/g. ^bEMPA-REG OUTCOME and DAPA-HF did not report the between-groups difference (95% CI) in eGFR slope, but did report the eGFR slope in each group. For these trials, the forest plot presents the calculated difference between the means of the eGFR slope in the SGLT2 inhibitor group and the placebo group. ^cThe EMPEROR-Reduced study and the subgroup analysis by baseline eGFR reported the absolute decrease in eGFR in figures, so approximate numbers are presented. CANVAS, Canagliflozin Cardiovascular Assessment Study; CANVAS-R, Canagliflozin Cardiovascular Assessment Study-Renal; CREDENCE, Canagliflozin and Renal Events in Diabetes with Established Nephropathy Clinical Evaluation; DAPA-CKD, Dapagliflozin and Prevention of Adverse Outcomes in Chronic Kidney Disease; DAPA-HF, Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure; EMPA-REG OUTCOME, Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients—Removing Excess Glucose; EMPEROR-Reduced, Empagliflozin Outcome Trial in Patients with Chronic Heart Failure with Reduced Ejection Fraction; VERTIS CV, Evaluation of Ertugliflozin Efficacy and Safety Cardiovascular Outcomes Trial.

However, these clinical trials reported a broad range of effect of SGLT2 inhibitors on the eGFR slope, raising questions about what factors influence the degree of kidney protective benefit one might expect when prescribing these medications. The natural first question is whether there could be differences between individual SGLT2 inhibitors, but no such pattern is clearly apparent for the GFR slope. Within individual drugs, trials reported variable estimates of this effect. There is similarly no readily apparent relationship between the mean baseline eGFR and the magnitude of benefit between trials, but, among individuals with heart failure in Empagliflozin Outcome Trial in Patients with Chronic Heart Failure with Reduced Ejection Fraction, the benefit was seen more strongly in individuals without CKD than with CKD (interaction P=0.05) (9). However, a potential pattern emerges when evaluating the eGFR slope on the basis of the magnitude of the eGFR dip, with studies with a larger dip generally showing a stronger benefit of SGLT2 inhibitors compared with placebo (Figure 1).

It is plausible that this could be a physiologic relationship, in that those with a greater short-term hemodynamic response to the changes in tubuloglomerular feedback seen after SGTL2 inhibition may derive greater long-term eGFR stabilization. The inclusion criteria in each trial may also drive this relationship. The two studies with the most dramatic improvement in slope were conducted in patients with heart failure with reduced ejection fraction, in whom favorable cardiovascular effects may more strongly affect the progression of CKD (8,9). In VERTIS CV, 81% of participants received concomitant renin-angiotensin system blockade, which diverges from other clinical trials that had a higher proportion of participants taking the maximally tolerated doses of such agents (7,10,11). A limitation of this analysis is that the individuals randomized to receive 5 mg of ertugliflozin daily or 15 mg daily were pooled for analysis, so we are unable to assess a dose-response relationship between ertugliflozin and effect on the eGFR dip and slope from the presented data. Furthermore, 78% of participants had a baseline eGFR >60 ml/min per 1.73 m², assessed using the Modification of Diet in Renal Disease equation, which is less accurate at higher eGFR values.

Several considerations related to trial design may contribute to the variation in the reported effect of SGLT2 inhibitors on the eGFR slope. Because there is an acute effect on eGFR after SGLT2 initiation, trialists must decide whether to calculate the eGFR slope beginning at randomization or beginning at the eGFR dip. In existing studies, the benefit of SGLT2 inhibitors was significant whether the slope was presented in either of these ways, but calculating the slope by including the baseline eGFR value may be the more conservative approach to measure the effect of SGLT2 inhibitors on eGFR progression. In this study, the authors reported the eGFR slope, both starting from the eGFR dip at week 6 and starting from baseline. The difference between these two methods in the placebo-adjusted slope was minimal (1.02 [95% CI, 0.84 to 1.20] ml/min per 1.73 m² from the eGFR dip versus 0.96 [95% CI, 0.80 to 1.11] ml/min per 1.73 m² from baseline), with a mean eGFR dip of 3.24 ml/min per 1.73 m². This is in stark contrast to Canagliflozin and Renal Events in Diabetes with Established Nephropathy Clinical Evaluation and Dapagliflozin and Prevention of Adverse Outcomes in Chronic Kidney Disease, which reported notably large differences in the eGFR slope when calculated by the two criteria, possibly related to the mathematic consequences of the larger absolute eGFR dips in these two studies (Figure 1) (10,11).

The timing of the first postrandomization eGFR measurement may also play a role. On the basis of available data, it is unknown when precisely an eGFR nadir occurs after SGLT2 inhibitor initiation. Clinical trials measured the eGFR dip at different times (Figure 1), so it is unknown whether these trials captured the true lowest eGFR, or whether the timing of the eGFR dip differs between individual drugs.

The duration of follow-up and number of data points used in the calculation of the eGFR slope may also affect slope calculation. This VERTIS CV analysis offers unique insight into the effect of follow-up interval on the calculation of the eGFR slope by presenting the slope through weeks 52, 104, 156, 208, and 260 of follow-up. The placebo-adjusted slope beginning at week 6 was most dramatic at week 52 and decreased with longer follow-up intervals. When beginning at baseline, the placebo-adjusted slope was least dramatic at week 52, peaked at week 104, and decreased with longer follow-up. This indicates that the eGFR dip affected the calculation of eGFR slope more dramatically at shorter follow-up durations, when fewer data points contributed to the slope calculation.

These considerations in trial design and analysis thwart attempts to accurately compare the effect of SGLT2 inhibitors on eGFR slope between individual drugs. With several trials consistently showing a benefit in eGFR slope compared with placebo, this is likely a class effect of SGLT2 inhibitors. The cumulative benefit of these medications is likely to be greatest when started at higher eGFR levels; it is imperative that they be started early in eligible patients to maximize long-term effect on clinically meaningful outcomes, such as dialysis dependence. We hope the emerging compelling data should overcome therapeutic inertia for SGLT2 inhibitors to avoid the underprescription that has plagued kidney protection with renin-angiotensin system blockade.

Disclosures

L.P. Gregg reports being employed by Michael E. DeBakey Veterans Affairs (VA) Medical Center (Houston, TX) and VA Health Services Research and Development Center for Innovations in Quality, Effectiveness and Safety (Houston, TX). S.D. Navaneethan reports serving as a scientific advisor or member of American Journal of Kidney Diseases, American Journal of Nephrology, CJASN, CardioRenal Medicine, and Current Opinion in Nephrology and Hypertension; having consultancy agreements with, and receiving honoraria from, Bayer, Boehringer Ingelheim, Reata, Tricida, and Vifor; receiving research grants from Keryx; serving as a guideline writing committee member of Kidney Disease Improving Global Outcomes; and being employed by Michael E. DeBakey VA Medical Center.

Funding

S.D. Navaneethan is supported by Department of VA Health Services Research & Development grant 1I01HX002917-01A1 and a National Institute of Diabetes and Digestive and Kidney Diseases grant R01DK101500. This work was also supported, in part, by the Center for Innovations in Quality, Effectiveness and Safety, Michael E. DeBakey VA Medical Center (CIN 13-413).