Sodium-glucose co-transporter 2 inhibitors: strength of evidence for a cardio-renal-metabolic therapy.

Abstract

This article refers to ‘Prevention of heart failure events with sodium–glucose co-transporter 2 inhibitors across a spectrum of cardio-renal-metabolic risk’ by K. Bhatia et al., published in this issue on pages 1002–1008. Sodium–glucose co-transporter 2 (SGLT2) inhibitors are potent antidiabetic agents for patients with type 2 diabetes mellitus (T2DM) that regulate glycaemic status through a non-insulin-dependent mechanism of stimulating glucose excretion in urine. SGLT2 inhibitor therapy is now a guideline-recommended approach for glycaemic control and cardiovascular risk reduction among high-risk T2DM patients. The SGLT2 inhibitors have had a serendipitous expedition into the arena of cardiovascular risk reduction and heart failure management in specific. The data from initial investigations of SGLT2 inhibitors demonstrated that apart from reducing hyperglycaemia, this medication class reduced arterial stiffness, vascular resistance, blood pressure, body weight, visceral adiposity, and albuminuria.1 These cardiovascular benefits prompted the evaluation of the impact of empagliflozin on cardiovascular outcomes among individuals with high-risk [having atherosclerotic cardiovascular diseases (ASCVD)] T2DM in the EMPA-REG OUTCOME trial.1 The results indicated a lower risk of adverse cardiovascular events and all-cause mortality among patients receiving empagliflozin compared with placebo. While glycaemic control is anticipated to contribute to cardiovascular risk reduction, antidiabetic agents more potent than SGLT2 inhibitors did not previously extend such a cardiovascular benefit. Spurred by the promising results, similar trials using dapagliflozin (DECLARE–TIMI 58), canagliflozin (CANVAS Program), and ertugliflozin (VERTIS CV) were completed in T2DM patients with ASCVD or having a high risk of developing ASCVD.2-4 These studies demonstrated a similar signal for reducing adverse cardiovascular outcomes.2-4 Interestingly, these initial SGLT2 inhibitor clinical trials observed a lower risk of adverse renal outcomes (defined by varying definitions of composite renal outcomes) among those randomized to the SGLT2 inhibitor medication arm.1-4 Encouraged by these compelling data, the CREDENCE study demonstrated that canagliflozin effectively reduced the risk of renal failure and cardiovascular events among people with diabetes and chronic kidney disease (CKD).5 Subsequently, the DAPA-CKD trial demonstrated that among CKD patients both with and without T2DM, dapagliflozin was efficacious in reducing the risk of renal failure, i.e. reduction in renal function, or cardiorenal cause of death.6 The recently completed SCORED trial among T2DM patients with CKD also demonstrated the efficacy of sotagliflozin, a combined SGLT1/2 inhibitor, for reducing adverse cardiovascular outcomes.7 Heart failure hospitalizations account for ∼$11 billion per annum in annual spending in the United States, with a concerning temporal trend for an increase in the heart failure hospitalization burden.8 The landmark DAPA-HF study demonstrated the efficacy of dapagliflozin in reducing heart failure hospitalizations and cardiovascular mortality among heart failure with reduced ejection fraction (HFrEF) patients.9 This was followed closely by the EMPEROR-Reduced study that demonstrated the efficacy of empagliflozin for reducing adverse cardiovascular outcomes in HFrEF.10 The recently concluded SOLOIST-WHF study that enrolled heart failure patients with T2DM across the spectrum of ejection fraction close to discharge following heart failure hospitalization demonstrated the cardiovascular benefits of sotagliflozin in worsening heart failure.11 In this issue of the Journal, Bhatia et al.12 summarize the effect of SGLT2 inhibitor therapy in preventing heart failure hospitalization and composite cardiovascular outcome (heart failure hospitalization or cardiovascular mortality) across diverse population groups. The article presents a high-quality data synthesis exercise to present cumulative evidence supporting the role of SGLT2 inhibitors as a pillar of heart failure medication armamentarium. In this meta-analysis, the authors report using aggregate data from the prior clinical trials that SGLT2 inhibitors reduce incident heart failure hospitalization by 30% [hazard ratio ((HR) 0.70, 95% confidence interval (CI) 0.62–0.78)] among those with high-risk T2DM, by 39% (HR 0.61, 95% CI 0.47–0.80) in CKD patients with T2DM, and by 31% (HR 0.69, 95% CI 0.62–0.78) in HFrEF patients. For the composite of heart failure hospitalization or cardiovascular mortality, the authors note that SGLT2 inhibitors reduce the risk by 21% (HR 0.79, 95% CI 0.71–0.88) among those with high-risk T2DM, by 27% (HR 0.73, 95% CI 0.65–0.82) in CKD patients, and by 26% (HR 0.74, 95% CI 0.68–0.81) in HFrEF patients. Meta-analyses are valuable tools in generating, summarizing, and interpreting the cumulative evidence in a particular area of disease management. However, even well-conducted meta-analyses such as the one by Bhatia et al.12 may have limitations, many of which have been previously described.13 A well-designed meta-analysis starts with a clear, specific research question and a thorough literature review. There must be conceptual homogeneity in the study populations included, with the assessment of similar treatments (medication class or dose) against a similar control (placebo or active control), and a clear a priori definition for the outcomes of interest. In the study by Bhatia et al., the authors perform a thorough literature search in their quest to understand the impact of SGLT2 inhibitors compared with placebo in the prevention of heart failure hospitalization and composite of cardiovascular mortality or heart failure hospitalization in patients with different cardiorenal and metabolic diseases. The population substrate evaluated varied from high-risk T2DM to CKD (with or without diabetes) and heart failure. The authors have performed a rigorous meta-analysis using publicly available aggregate data and presented the absolute risk reduction (ARR) and number-needed-to-treat (NNT) across the varying spectrum of population substrates for various SGLT2 inhibitors. It is important for clinicians to understand that the ARR and NNT (computed as 1/ARR) are dependent on the baseline event rate in the study populations. For example, the event rate for incident heart failure hospitalization is highest in heart failure patients, followed by CKD patients, and lowest in high-risk T2DM patients. Thus, a 30% lower risk for an incident heart failure event will translate into a wide range of ARR or NNTs based on the baseline event rate, with more attractive values in populations with higher event rates. In the presented array of NNTs and ARRs for heart failure hospitalization in the different study populations, the authors have demonstrated that despite similar HR estimates, the ARRs and NNTs vary based on the baseline risk of incident heart failure hospitalization (i.e. lower in high-risk T2DM and higher in heart failure populations). Therefore, the authors' emphasis on pooled estimates derived from diverse populations may have limited clinical implications. Although a herculean effort by the investigators,12 it may be argued that since the results from all the large-scale SGLT2 inhibitor trials were positive and concordant (i.e. in the same direction), what do analyses such as the one presented here by Bhatia et al.12 add to the literature? The important clinical learning from the article by Bhatia et al.12 is that based on the patient profile (whether the patient has high-risk T2DM, CKD, or heart failure), clinicians may be able to communicate the strength of evidence demonstrated by SGLT2 inhibitors in the respective patient populations. The critical principle in appraising meta-analyses is that conceptual homogeneity precedes statistical homogeneity, and statistical homogeneity does not always imply clinical homogeneity. Meta-regression is commonly used to address the statistical heterogeneity of measured confounders at the study level but does not adequately account for other patient-level factors driving the estimates.13 The authors performed a rigorous meta-regression analysis demonstrating statistical homogeneity in the treatment effect of SGLT2 inhibitors across study-level factors. These analyses add rigor to the investigation but still leave several unanswered questions. One could argue that a meta-analysis using patient-level data, which is not what the authors did, would have allowed us to understand the nuances of differences in SGLT2 inhibitor therapy among traditionally underrepresented population subgroups such as women and those from racial minorities. Given the robust reduction in heart failure events by the class effect of SGLT2 inhibitor drugs demonstrated by Bhatia et al.,12 emphasis needs to be placed on understanding the mechanistic basis for the observed improvement in cardiovascular outcomes. The early potentiation of natriuresis by SGLT2 inhibitors has been suggested to drive its cardiovascular benefits. However, more potent diuretic medications have failed to translate their effect of symptomatic improvement and volume reduction into a substantial reduction of adverse cardiovascular events. Recent data demonstrate that SGLT2 inhibitors rapidly reduce pulmonary artery pressures in heart failure patients, and this effect is increased over time.14 Furthermore, accumulating evidence demonstrates that SGLT2 inhibitors are efficacious in preventing heart failure hospitalizations irrespective of the baseline volume status.11 Thus, the plausible mechanisms driving the cardiovascular benefit of SGLT2 inhibitors require extensive investigation. However, in parallel, we need to augment the utilization of SGLT2 inhibitors in clinics across the globe to make this efficacious class of medication available to heart failure patients. SGLT2 inhibitors were introduced initially as antidiabetic agents and now must be recognized as a potent heart failure medication. SGLT2 inhibitors form one of the key tenets of evidence-based heart failure therapy in 2021. Multiple efficacious HFrEF management therapies are now available, but the implementation of the guideline-directed medical therapy in HFrEF is universally lacking. A pragmatic polypill-based approach incorporating SGLT2 inhibitor therapy may facilitate the implementation of guideline-directed medical therapy in HFrEF. In summary, the cumulative contemporary evidence indicates that SGLT2 inhibitors are potent cardiovascular agents that effectively reduce incident heart failure hospitalizations and the composite of heart failure hospitalization or cardiovascular mortality among high-risk T2DM patients, CKD patients, and patients with heart failure. Further investigation is needed to (i) understand the mechanistic basis of the benefits associated with SGLT2 inhibitors, and (ii) study the effects among racially diverse subgroups that are disproportionately affected by cardiometabolic diseases.15 P.A. is supported by the National Institutes of Health Mentored Patient-Oriented Research Award 5K23HL146887-02. Conflict of interest: none declared.