Treatment of Nondiabetic Kidney Diseases with Sodium-Glucose Cotransporter-2 Inhibitors

Beginning with the report of the EMPA-REG OUTCOME study, one large clinical trial after another has demonstrated that sodium-glucose cotransporter-2 (SGLT2) inhibitors reduce major adverse cardiovascular events, progressive heart failure, mortality, kidney failure, and progression of kidney disease in individuals with type 2 diabetes.^1–3 Because protective kidney and cardiovascular effects of SGLT2 inhibitors were independent of systemic glucose lowering in these trials,⁴ subsequent studies were undertaken to determine the effect of long-term SGLT2 therapy on individuals with or at risk of cardiovascular or kidney diseases but without diabetes. The results of these more recent studies confirmed that many individuals without diabetes are protected by SGLT2 inhibitors. But which kidney patients should be treated with these protective agents? Conversely, which groups of patients with CKD should avoid taking them, at least for now? A close look at the major randomized controlled trials (RCTs) that included large numbers of participants with nondiabetic kidney diseases can help provide some answers.

The Dapagliflozin and Prevention of Adverse Outcomes in Chronic Kidney Disease (DAPA-CKD)¹ and Study of Heart and Kidney Protection with Empagliflozin (EMPA-KIDNEY)⁵ trials directly demonstrated the salutary effects of SGLT2 inhibitors in participants with nondiabetic CKDs. In DAPA-CKD, around 30% of the 4000-plus participants had CKD without diabetes, while that number was even greater in EMPA-KIDNEY, in which almost 70% had nondiabetic kidney diseases. Both studies showed similar outcomes, and the relatively small differences between the results from each study can largely be explained by important differences in the duration of each study and in the severity of kidney disease and cardiovascular risk at enrollment. Specifically, the EMPA-KIDNEY population had a lower cardiovascular risk profile with a higher proportion of normalbuminuric and nondiabetic participants. Therefore, the primary end point in this study was driven much more by kidney events than cardiovascular ones. Despite these differences, the two studies together demonstrated that SGLT2 inhibitors prevented kidney disease progression, kidney failure, cardiovascular events, and cardiovascular death just as well in nondiabetic as in diabetic CKD.

Although critical for the development of clinical guidelines, RCT results are not always ideal tools to guide daily clinical practice. Conversely, careful meta-analyses can often provide summary guidance for the best use of new treatments and can be updated quickly as new evidence is reported. A recent collaborative meta-analysis that was published immediately after EMPA-KIDNEY is particularly helpful in guiding the current use of SGLT2 inhibitors in nondiabetic patients. This meta-analysis included all large, double-blind, placebo-controlled SGLT2 inhibitor trials of at least 6 months duration, including DAPA-CKD and EMPA-KIDNEY, with a total of over 90,000 adult participants, including over 10,000 without diabetes and approximately 5000 with nondiabetic CKDs.⁶ The authors of this collaborative meta-analysis worked with RCT investigators to assess relevant kidney end points that were not reported in the original RCT publications. This analysis confirmed that SGLT2 inhibitors reduced the risk of kidney disease progression in participants with nondiabetic CKDs as well as those with diabetic kidney disease and that relative protection was similar for participants in all the nondiabetic CKDs included in the RCTs. Confidence intervals were large for all nondiabetic CKDs, so the meta-analysis could not exclude outcome differences between the different nondiabetic kidney disease groups, but the important point is that reduction in risk of kidney disease progression as well as cardiovascular protection was found for all CKDs included in the analysis. Given the greatly expanded numbers of glomerulonephritis participants in EMPA-KIDNEY compared with those in DAPA-CKD, it is now clear that the protective effect of SGLT2 inhibitors extends to all patients with primary glomerular diseases as well as to patients with vascular or hypertensive CKDs.

To translate these results into real-world scenarios, the authors estimated that for every 1000 patients with CKDs treated for 1 year with an SGLT2 inhibitor, 11 kidney disease progression events (reduction in eGFR ≥50%) would be prevented in patients with diabetes, compared with 15 in patients without diabetes. In addition, SGLT2 inhibitor treatment for a year would result in approximately 4–5 fewer acute kidney injury episodes in both sets of patients. Conversely, only two hospitalizations for congestive heart failure exacerbations or cardiovascular death would be prevented in nondiabetic CKD patients compared with 11 such events in patients with diabetic CKD. Importantly, there was no lower eGFR limit to the effectiveness of SGLT2 inhibition in either diabetic or nondiabetic groups. Therefore, treatment with SGLT2 inhibitors can and should be initiated in patients with eGFRs as low as 20 ml/min per 1.73 m².

Some groups of patients with CKD were excluded from the large RCTs reported to date and, hence, from the meta-analysis. Patients with polycystic kidney diseases, type 1 diabetes, organ transplants, and those on dialysis were the most prominent excluded groups (which also included sickle cell, lupus nephritis, and vasculitis patients). Currently, some type 1 diabetic individuals and transplant recipients are being treated with SGLT2 inhibitors for improved glycemic control, and some small prospective studies in these groups have suggested that they might benefit from SGLT2 inhibitor treatment. Analysis of the nearly 70 type 1 diabetic participants in EMPA-KIDNEY will add to this literature, and trials assessing several of these excluded groups are ongoing (e.g., https://renal-lifecycle.com/en/home-en/). However, no studies to date have comprehensively reported the long-term efficacy and safety of SGLT2 inhibitor treatment in any of these populations. Because of the heightened risk of diabetic ketoacidosis and severe infections, generalized use of SGLT2 inhibitors is not recommended for any of the populations excluded from the RCTs.

Given the clear protections conferred by SGLT2 inhibitors in most patients with CKD, it is critical to estimate the risks and benefits each individual patient will face when deciding to implement such protective treatments. Because any reduction in GFR imparts high cardiovascular risk to patients with type 2 diabetes,⁷ the 2022 Kidney Disease Improving Global Outcomes guidelines recommended the use of SGLT2 inhibitors for virtually all CKD patients with type 2 diabetes. Can such a broad recommendation now be extended to adult nondiabetic CKD patients, with the exception of those with the conditions noted above? I believe the answer is yes, with a few caveats. It is clear that given the same cardiovascular and kidney risks, nondiabetic CKD patients should garner benefits from SGLT2 inhibitors that are roughly equivalent to those of their diabetic counterparts. However, cardiovascular risks are not the same for all patients with CKD and, in general, are substantially higher in patients with diabetic CKD than those without diabetes, as the meta-analysis made clear. It is obvious that a 65-year-old individual with poorly controlled type 2 diabetes, a history of coronary artery disease, an eGFR of 27 ml/min per 1.73 m², and a urinary albumin-to-creatinine ratio of 1500 mg/g has much greater short-term and long-term risks of a major cardiovascular event than does a 45-year-old patient with IgA nephropathy, no cardiovascular history, normal cholesterol levels, minimal albuminuria, and an eGFR of 75 ml/min per 1.73 m². Should both be treated similarly with SGLT2 inhibitors? Maybe, but maybe not. Individual benefit from SGLT2 inhibitor therapy is much more likely for the patient with diabetes, but the risks are similar for both individuals, and some providers would hesitate prescribing lifelong therapy to the patient with IgA nephropathy for whom long-term treatment benefits are relatively low. Nonetheless, even low-risk, nonalbuminuric CKD patients with mild–moderate reductions in eGFR showed reduced rates of eGFR decline on SGLT2 inhibitor therapy.⁸ The question of how aggressively SGLT2 inhibitors should be used in patients with lower risk CKD remains unanswered. For now, individualized decision making by patients and providers is needed.

Given the evidence that glucagon-like peptide 1 receptor agonists and finerenone reduce cardiovascular risk, these provide alternatives or additions to SGLT2 inhibitor therapy for CKD patients with diabetes. It would not be surprising to find that these classes of medication will also help CKD patients without diabetes. A challenge ahead is to best select and combine the new protective agents, along with renin-angiotensin system inhibitors, for individual patients with different CKDs. This is a good challenge to have.

Health care providers now have options when treating patients with nondiabetic CKDs. In this rapidly evolving space, guidelines and recommendations for care will need to be rapidly updated to help guide appropriate treatment for individual patients. No longer does one size fit all. For now, SGLT2 inhibitors can be used safely and effectively in most nondiabetic CKD patients who have significant cardiovascular or kidney risks. Treatment can be initiated in most CKD patients with eGFRs of 20 ml/min per 1.73 m² or greater and even at lower levels in some individuals. These agents can be safely administered until patients develop end-stage kidney disease. As always, individualized care is critical, and benefits and risks need to be weighed for each patient. Nonetheless, we are emerging into an era in which all our patients with CKD can enjoy much lower risks of life-threatening cardiovascular disease and kidney failure than they have until now. This is welcome news indeed.

Disclosures

F.C. Brosius reports research funding from NIH/NIDDK, fees from Gilead Scientific Inc for serving on a scientific advisory board in 2021, serving as an Associate Editor for Diabetes, serving on the Editorial Boards for American Journal of Physiology and Journal of Clinical Investigation, and other interests or relationships with American Society of Nephrology Diabetic Kidney Disease Collaborative (DKD-C) Task Force.

Funding

None.

Author Contributions

Conceptualization: Frank C. Brosius.

Visualization: Frank C. Brosius.

Writing – original draft: Frank C. Brosius.

Writing – review & editing: Frank C. Brosius.