How Does Empagliﬂozin Reduce
Cardiovascular Mortality? Insights
From a Mediation Analysis of the
EMPA-REG OUTCOME Trial
Empagliﬂozin, a highly selective sodium–glucose cotransporter 2 (SGLT2) inhibitor, was the ﬁrst glucose-lowering agent to demonstrate a reduction in cardiovascular (CV) death in patients with type 2 diabetes and high CV risk (1). In the EMPA-REG OUTCOME trial, over a median observation time of 3.1 years, treatment with empagliﬂozin versus placebo in addition to standard of care led to a 14% reduction in the risk of three-point major adverse CV events (MACE [the composite of CV death, nonfatal myocardial infarction, and nonfatal stroke]) (hazard ratio [HR] 0.86 [95% CI 0.74, 0.99]; P = 0.04). This was driven by a 38% reduction in the risk of
The ﬁnding led the U.S. Food and Drug Administration to extend the indication for empagliﬂozin to include reducing the risk of CV death in patients with type 2 diabetes and established CV disease (2). Empagliﬂozin also reduced the risk of hospitalization for heart fail-ure (HR 0.65 [95% CI 0.50, 0.85]; P =0.002) and all-cause mortality (HR 0.68 [95% CI 0.57, 0.82]; P , 0.001) versus placebo (1).
As with other CV outcome trials, EMPA-REG OUTCOME was not designed to determine the mechanisms under-pinning its results. Several explanations for the reduction in CV death with empagliﬂozin have been proposed, in-cluding hemodynamic changes related to plasma volume reduction, a switch in use of fuel, and direct cardiac effects (3–9). The very early reduction in CV death observed in this trial and the het-erogeneity of the hazard ratios for the components of three-point MACE sug-gest that the predominant mechanism, at least in the early part of the trial, was not primarily an attenuation of athero-sclerosis, the traditional consideration in CV outcome trials in patients with diabetes. The equally rapid reduction in the risk of hospitalization for heart failure suggests that the cardioprotec-tive beneﬁt of empagliﬂozin may be re-lated to improved hemodynamic status (10).
However, effects on atherogenic processes, the myocardium, ventricular remodeling, or vessel integrity cannot be ruled out because the beneﬁts on CV death were sustained over the course of the trial, and additional mech-anisms may have contributed to the re-duction in the risk of CV death observed with prolonged treatment. In the cur-rent exploratory post hoc mediation analysis of data from the EMPA-REG OUTCOME trial, we identiﬁed the ex-tent to which treatment-induced changes in speciﬁc variables, either alone or in combination, contributed to the reduc-tion in the risk of CV death observed with empagliﬂozin versus placebo and, thereby, considered potential media-tors of this beneﬁt.
RESEARCH DESIGN AND METHODS
The design of the EMPA-REG OUTCOME trial has beendescribedpreviously (1,11). In brief, adults with type 2 diabetes who were drug-naive with an HbA1c of 7.0– 10.0% (53–86 mmol/mol) or were taking any background glucose-lowering medi-cation with an HbA1c of 7.0–9.0% (53–
75 mmol/mol) and who had established CV disease were eligible for inclusion. Pa-tients were randomized (1:1:1) to receive empagliﬂozin10mg,empagliﬂozin25mg, or placebo. Glucose-lowering medication was to remain unchanged for the ﬁrst 12 weeks, although intensiﬁcation was permitted if the patient had a conﬁrmed fasting glucose level of .240 mg/dL; in cases of medical necessity, dose reduc-tion or discontinuation of background medication could occur. After week 12, investigators were encouraged to adjust glucose-lowering medication to achieve glycemic control according to local guide-lines. Throughout the trial, investigators were encouraged to treat other CV risk factors according to local guidelines. The trial was to continue until $691 patients experienced an adjudicated primary out-come event (three-point MACE). Patients who prematurely discontinued study medication continued to be followed for ascertainment of CV outcomes and vital status.
CV outcome events and deaths were prospectively adjudicated by indepen-dent clinical events committees (1,11). Analyses of CV outcomes were prespeci-ﬁed to compare the pooled empagliﬂozin dose groups with the placebo group.
Traditional Mediation Analysis
A traditional mediation analysis as ori-ginally proposed by Baron and Kenny (12) was used, taking the time-dynamic evolvement of both the potential media-tors and the outcome CV death into ac-count. A variable must satisfy several conditions to be a mediator of the treat-ment effect. Treatment must have an ef-fect on the variable over time, and the change in the variable over time must have an effect on the outcome. As an ad-ditional condition, in an analysis where the variable is included as a time-dependent covariate over time, the effect of treat-ment on the outcome (represented as the HR) must be reduced compared with the treatment effect in an unadjusted analysis.
Analysis of the Effect of Treatment on CV Death
The primary analysis of CV death with empagliﬂozin versus placebo was based on a Cox proportional hazards regression model,withtreatmentgroupadjustedfor the baseline variables of age, sex, BMI, HbA1c, estimated glomerular ﬁltration rate (eGFR), and region, in patients who received one or more doses of study drug (1).
Analysis of Changes in Variables (Considered Potential Mediators)
On the basis of evidence from previous studies, potential mediators of the bene-ﬁt of empagliﬂozin on CV death were identiﬁed post hoc from the variables that were measured and tracked in the trial. Those chosen for analysis involved several mechanistic categories (Table 1): glycemia (HbA1c, fasting plasma glucose
[FPG]), vascular tone (systolic blood pres-sure[SBP],diastolicbloodpressure[DBP], heart rate), lipids (LDL cholesterol, HDL cholesterol,triglycerides,freefattyacids), adiposity (weight, BMI, waist circum-ference), renal function (urine albumin-to-creatinineratio[UACR],eGFRaccording to MDRD and Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI] equations),volumestatus(hematocrit,he-moglobin, albumin), and other (uric acid).
Potential mediators could be physio-logical parameters conceivably linked to the outcome (e.g., blood pressure) or bio-chemical markers of such risk (e.g., LDL cholesterol). To be a potential mediator, a variable should satisfy the above-stated conditions. Differences between treatment groups in the longitudinal changes from baseline in each potential mediator were analyzed individually by using a mixed-effects repeated-measures model to estab-lish that treatment had an effect on the variable. The model for each variable in-cluded as covariates the baseline value of that variable, baseline HbA1c, base-line eGFR, baseline BMI, region, the last weekapatientcouldhavehadameasure-ment of that variable, treatment, visit, visit by treatment interaction, baseline HbA1c by visit interaction, and the base-line value of the variable by visit interac-tion as ﬁxed effects. Because of skewed distributions, triglycerides and UACR were log-transformed before analysis. Each analysis used all data available for patients who received one or more doses of study
Mediation Analysis of EMPA-REG OUTCOME Results Diabetes Care
drug and had a baseline and postbaseline measurement for the variable in question.
Analysis of Potential Mediation of the Reduction in CV Death With Empagliﬂozin
We considered that empagliﬂozin could haveadirecteffectontheriskofCVdeath or could affectCV death indirectly through itseffectsononeormoreoftheaforemen-tioned potential mediators. Each variable was analyzed as a time-dependent covari-ate in Cox regression models with the out-come time to CV death. The effect of a speciﬁc covariate was analyzed by using two approaches.
Analysis of the current change from baseline to the most recent value available before CV death, reﬂecting the acute effect of the variable on the risk of CV death. In the example below, the current change from base-line in a mediator Y (changeY) is calcu-lated as the difference in the current value to baseline: changeYti 5 Yti 2 Y0
Analysisofthemeanvalueconsidering all prior values (updated mean), re-ﬂecting the cumulative effect of all prior values of the variable on the risk of CV death (see example illustra-tion in the Supplementary Material). In the example below, the updated mean of a mediator Y (upmeanY) is calculated as the updated weighted mean, weighted for the length of time intervals:
upmeanY ðti until ti11Þ5
The mediating effect of the current change from baseline in a variable, and of the updated mean of the variable, on the treatment effect of empagliﬂozin ver-sus placebo with respect to CV death were analyzed by using separate Cox re-gression models, with treatment group adjustedforthebaselinevalueofthevari-able and either the change from baseline in the variable or the updated mean of the variable as time-dependent covari-ates. The models provided the estimated HR for CV death associated with a 1-unit increase in the variable (current change from baseline or updated mean). The re-sultswerecomparedwithamodeladjust-ing for treatment group alone (i.e., without adjusting for the baseline value ofthevariableandthelongitudinalvalues of the variable [change or updated mean]).
Effects of Empagliﬂozin on the Time
Course of Potential Mediators Table 1 shows the mean of each potential mediator at baseline and the differences with empagliﬂozin versus placebo in changes from baseline at week 12 or 28 and week 164. Over the course of the study, empagliﬂozin was associated with small reductions versus placebo in HbA1c, weight, waist circumference, uric acid, SBP and DBP, no change in heart rate, and small increases in LDL and HDL cholesterol (1). An initial decrease in eGFR (CKD-EPI) withempagliﬂozinwasobservedfollowed bystabilizationduringprolongedtreatment in contrast to a gradual decline in eGFR in the placebo group (13). Empagliﬂozin led to signiﬁcant reductions in UACR versus placebo from week 12, regardless of albu-minuria status at baseline (14). An initial increase in hematocrit with empagliﬂozin was observed followed by stabilization compared with no notable change in the placebo group (Fig. 1).
Effects of Change in Potential
Mediators on the Risk of CV Death Inanalysesadjustingforchangefrombase-line, models showed that increases in FPG, heart rate, logUACR, and uric acid were associated with an increased risk of CV death, whereas increases in SBP, HDL
cholesterol, eGFR, weight, BMI, hemato-crit, hemoglobin, and albumin were asso-ciated with a reduced risk of CV death (Supplementary Table 1). In analyses that were based on the updated mean, models showed that increases in HbA1c,
FPG,heartrate,LDLcholesterol,logUACR, and uric acid were associated with an in-creasedriskofCVdeath,whereasincreases in HDL cholesterol, eGFR, hematocrit, he-moglobin, and albumin were associ-ated with a reduced risk of CV death (Supplementary Table 2).
Univariable Analysis: Effects of
Individual Potential Mediators on the
HR for CV Death With Empagliﬂozin
Table 2 presents the HRs for CV death with empagliﬂozin versus placebo after adjusting for the current change from baseline in each covariate and the per-centage mediation compared with the HR adjusting for treatment group alone. Treatment group differences in changes from baseline in hematocrit and hemo-globin mediated 51.8% and 48.9%, re-spectively, of the effect of empagliﬂozin versus placebo on the reduction in risk of CV death. Changes in albumin and uric acid mediated 25.5% and 24.6%, respec-tively. The other potential mediators in-vestigated, including HbA1c, FPG, weight,
BMI, SBP and DBP, LDL and HDL choles-terol, triglycerides, free fatty acids, eGFR, and UACR, had no or negligible effects in these analyses.
Table 3 presents the mediating effect after adjusting for the updated mean of each covariate and the baseline value. Adjusting for the updated mean of he-matocrit and hemoglobin mediated
51.8% and 45.7%, respectively, of the ef-fect of empagliﬂozin versus placebo on the reduction in the risk of CV death. Ad-justing for the updated mean of albumin and uric acid mediated 31.6% and 18.5%, respectively. Adjusting for the updated mean of HbA1c and FPG mediated 22.8% and 29.3%, respectively, of the treatment group effect, which were notably stronger mediating effects than in the change from baseline analysis (3.0% and 16.1%, respec-tively).Theotherpotentialmediatorsinves-tigated, including weight, BMI, SBP and DBP, LDL and HDL cholesterol, triglycerides, free fatty acids, eGFR, and UACR, had no or negligible effects in these analyses.
Multivariable Analysis: Effects of Combinations of Potential Mediators on the HR for CV Death With
Empagliﬂozin Versus Placebo
In the current change from baseline anal-ysis, the strongest mediator was hemato-crit, representing the volume category (proportion of the effect of empagliﬂozin on CV death mediated: 51.8%). The addi-tion of FPG as representative of the glycemiacategoryledtoaproportionme-diated of 70.9%. The addition of uric acid increased the proportion mediated to 84.4%. Finally, the addition of logUACR as representativeof therenal function cat-egory led to an estimated HR for CV death with empagliﬂozin versus placebo of 0.931 (95% CI 0.732, 1.183) (Supplementary Table 3), with the total proportion medi-ated being 85.2%. In a sensitivity analysis whereinhematocritwasreplacedbyhemo-globin as representative of the volume cat-egory, the estimated HR for CV death with empagliﬂozin versus placebo was 0.927 (95% CI 0.730, 1.176), thus showing a sim-ilar proportion mediated of 84.4%.
In the updated mean analysis, the re-sultsweresimilar.Thestrongestmediator was again hematocrit (proportion of the effect of empagliﬂozin on CV death medi-ated: 51.8%). The addition of FPG in the next step led to a proportion mediated of 84.9%. The addition of uric acid led to a proportion mediated of 92.2%. Finally, the addition of logUACR led to an esti-mated HR for CV death of 0.974 (95% CI 0.753, 1.261) (Supplementary Table 4), with a total proportion mediated being 94.6%. In the updated mean analysis, the replacement of hematocrit by hemo-globinshowedasimilarresult(proportion mediated: 91.2%). Similarly, replacement of the updated mean FPG by the updated mean HbA1c resulted in 85.1% mediation.
Resultsofstabilityanalysesonthebasisof bootstrap resampling are shown in Supplementary Figs. 1 and 2.
Weconductedthisexploratoryanalysisto identify potential mechanisms underlying the 38% reduction in the risk of CV death observed with empagliﬂozin versus pla-cebo in patients with type 2 diabetes and established CV disease in the EMPA-REGOUTCOMEtrial.Ourapproachuseda traditional mediation analysis (15,16), taking the time-dynamic evolvement of the potential mediators and the outcome of CV death into account. This method has been used by others to determine or conﬁrm the underlying mechanisms behind a treatment strategy’s effect on disease outcomes (17–19). Therein, change in the point estimate of the HR denoting a treatment effect is measured after sequential controlling for a variety ofplausiblevariables, whichareknownto be ameliorated with therapy. The vari-ables that bring the HR closest to 1.0 are saidtobethemajormediatorsofthetreat-ment effect, suggesting, but not proving, a cause-and-effect relationship. In the Diabetes Control and Complications Trial (DCCT)/Epidemiology of Diabetes Inter-ventions and Complications (EDIC) study, for example, improvement in HbA1c was found to mediate essentially 100% of the beneﬁt of intensive insulin therapy on the development and progression of retinopathy (19). Such an outcome was logical and expected because the goal of that trial was to reduce glycemia, as mea-sured by HbA1c, in an effort to reduce long-term microvascular damage. In con-trast, multiple mechanisms contribute to the CV risk in patients with type 2 diabe-tes. The precise contributions of individu-al hypothesized mechanisms to the reduction in CV death with empagliﬂozin remain to be elucidated, and a mediation analysis such as this may be used to drive future research.
In the current analysis, changes in hematocrit (and hemoglobin) appeared to be the variables with the largest impact ontheHRforCVdeathwithempagliﬂozin versus placebo. Therefore, these variables can be considered important medi-tors of this beneﬁt, mediating ;50% of the treatment group effect. Results were consistent in analyses that were based on the current change from baseline or the updated mean. For uric acid and mea-sures of glycemia, modest mediation ef-fectswerefound,withthoseofHbA1c and
FPG appearing to be stronger in analyses that were based on the updated mean, which assessed the more chronic effects of these measures. In contrast, the medi-ation effects of other variables, including changes in classical CV risk factors, such as BMI, blood pressure, lipids, and other parameters of renal function, were ab-sent or negligible.
In the multivariable models, a much higherpercentagemediationwasdemon-strated (up to 85.2% in the change from baseline analysis and up to 94.6% in the updatedmeananalysis).Thesemodelsin-corporated widely disparate effects of empagliﬂozin, including those on hemat-ocrit, FPG, uric acid, and UACR. These data suggest that although the major in-ﬂuence governing the reduction in CV death may have been plasma volume reduction (as reﬂected in the increased hematocrit), other variables may have played more modest, yet complementary roles, and multiple mechanisms may be responsible for the reduction in CV death with empagliﬂozin in patients with type 2 diabetes and established CV disease.
Empagliﬂozin is a selective inhibitor of SGLT2 in the proximal tubule of the kidney (20). Inhibition of SGLT2 by empagliﬂozin leads to reduced renal glucose reabsorp-tion and increased urinary glucose excre-tion (21). Treatment with empagliﬂozin reducesvolume andsodiumload through its glucuretic, diuretic, and natriuretic properties (22,23). The initial increase in hematocrit with empagliﬂozin followed by stabilization during the rest of the trial likely reﬂect hemodynamic changes re-lated to plasma volume contraction. The same pattern was observed for changes ineGFRwithempagliﬂozin,whichalsoare believed to reﬂect hemodynamic alter-ations involving renal blood ﬂow (13). The resulting decrease in circulatory load, especially reduced ventricular ﬁlling pressures and cardiac workload, could be animportantmechanismbehindthemor-tality beneﬁts seen with empagliﬂozin (5,9). This ﬁnding is supported by the observation that the most frequent modes of CV death are those typically seen in patients with heart failure (sud-den death, death as a result of heart fail-ure, and presumed CV death, the latter designated when insufﬁcient data exist for the adjudication committees to attribute a cause of death) (24). Although only 10% of patients in the EMPA-REG OUTCOME trial had heart failure at base-line, the trial population comprised indi-viduals with a mean age of 63 years, of whom 76% had coronary artery disease and 52% were obese. Therefore, many participants in this trial likely had unrec-ognized left ventricular dysfunction, par-ticularly diastolic dysfunction, which may eventually lead to clinical heart failure with preserved ejection fraction (25). Thus, a tenable conclusion is that a key contributor to the reduction in CV death with empagliﬂozin is the change in renal sodium and glucose handling with resul-tant reductions in ﬂuid burden, ventricu-lar stress, and risk of sudden cardiac decompensation.
An increase in erythropoiesis could be a complementary mechanism to the hemodynamicchanges reﬂected byan in-crease in hematocrit and hemoglobin in patients treated with empagliﬂozin. In-creased erythropoietin and a median 7% increase in red blood cell mass, measured with 51Cr-labeled erythrocytes, were ob-served in a small study (n = 30) of the SGLT2 inhibitor dapagliﬂozin in patients with type 2 diabetes (26). In a larger study, a mean increase in erythropoietin of 30–40% was observed 4 weeks after initiating empagliﬂozin, which may be re-lated to changes in blood ﬂow between therenalcortexandthemedulla(27).The extent to which this mechanism contrib-utes to the CV beneﬁts observed with empagliﬂozin is unclear, however. Al-thoughanimprovement intissue oxygen-ation in a compromised cellular milieu may be hypothesized to be beneﬁcial, such a robust effect on mortality would beunexpectedonthe basisofcurrentun-derstanding of oxygen delivery dynamics at hemoglobin concentrations within the normal range (28).
Improvements in the updated mean HbA1c and FPG explained 23% and 29%, respectively, of the beneﬁcial effect of empagliﬂozin on CV death compared with 3% and 16% for the current change from baseline. In the DCCT/EDIC study and the UK Prospective Diabetes Study (UKPDS), CV beneﬁts of intensive glucose control emerged but only after a pro-longed follow-up period (29,30). Unlike the UKPDS and the DCCT study, which were designed to achieve glycemic differ-ences, the EMPA-REG OUTCOME trial was designed as a glycemic equipoise tri-al, and only a modest difference in HbA1c between the empagliﬂozin and placebo groups was found. Furthermore, unlike the UKPDS with a cumulative follow-up of .10 years (30), trials of shorter dura-tion did not convincingly demonstrate CV beneﬁts from this degree of glucose low-ering(31).Thesmallermediationeffectof glycemia relative to hematological vari-ables, therefore, may reﬂect these factors. Change in uric acid had a modest media-tion effect (24.6% in the analysis adjusting forthechangefrombaseline).Empagliﬂozin reduces uric acid, possibly as a result of theeffectofincreasedglucoseconcentra-tion on glucose transporter 9 in the baso-lateral membrane of the proximal tubule andresultingincreaseduricosuria(22,32). Serum uric acid has been associated with an increased risk of CV death (33), but ev-idence is limited regarding the CV bene-ﬁts of reducing uric acid (34), especially within the normal range and/or to the small degree observed in the EMPA-REG
Limitations of the current analysis in-clude that it was post hoc, and the results can only be considered hypothesis gener-ating, demonstrating possible associations but not necessarily causal relationships. It cannot be inferred that similar changes in these variables achieved with approaches other than empagliﬂozin treatment will yield the mortality beneﬁts observed. Moreover, only variables measured during the trial could be examined, and in addi-tion to the potential mediators we investi-gated, other variables such as reductions in glomerular hyperﬁltration (35), reduc-tions in arterial stiffness and vascular resistance (36), direct cardiac effects through reductions in myocardial intra-cellular sodium (37), and a switch in use of cardiac fuel (4,5) have been proposed as important. Finally, although the meth-ods we used in the current analysis have beenusedpreviously,they have not been used in the context of CV death. CV death encompassesmultiplesubcategorieswith potentially disparate etiologies that may notbeinﬂuencedbythesamemechanisms. Inconclusion,thisexploratoryinvestiga-tion into potential mediators of the reduc-tion in risk of CV death with empagliﬂozin versus placebo in the EMPA-REG OUTCOMEtrialfoundthatchangesinhemat-ocrit and hemoglobindostensibly markers of the effects of the drug on volumed appeared to be important mediators of the reduction in mortality risk in univari-able and multivariable models. Changes in variables related to glycemia and urate metabolism had smaller mediating ef-fects. These, in addition to changes in UACR, contributed in the multivariable models, suggesting that the underpinnings of empagliﬂozin’s CV mortality beneﬁt are likely multifaceted. In contrast, changes in some traditional CV risk factors, including obesity, blood pressure, lipids, and renal function, made negligible contributions. Ongoing studies, including mechanistic trials, the EMPEROR outcome trials
(EMPagliﬂozin outcomE tRial in Patients With chrOnic heaRt Failure With Reduced Ejec-tion Fraction and EMPagliﬂozin outcomE tRial in Patients With chrOnic heaRt Fail-ure With Preserved Ejection Fraction) that are evaluating empagliﬂozin in pa-tients with heart failure with and without diabetes (NCT03057977, NCT03057951), and studies in patients with chronic kidney disease (38), will provide additional phys-iologicalinsightsintothecardioprotective effects of this selective SGLT2 inhibitor.
www red DiabetologNytt