Antihyperglycemic agents as novel natriuretic therapies in diabetic kidney disease

Abstract

While sodium-glucose cotransporter-2 (SGLT2) inhibitors have been used for the routine management of type 2 diabetes for several years, it is perhaps their natriuretic effects that are most important clinically. This natriuresis activates tubuloglomerular feedback, resulting in reduced glomerular hypertension and proteinuria, leading to renal protective effects in the EMPA-REG OUTCOME and CANVAS Program trials. In the cardiovascular system, it is likely that plasma volume contraction due to natriuresis in response to SGLT2 inhibition is at least in part responsible for the reduction in the risk of heart failure observed in these trials. We compare this mechanism of action with other antidiabetics. Importantly, other diuretic classes, including thiazide and loop diuretics, have not resulted in such robust clinical benefits in patients with type 2 diabetes, possibly because these older agents do not influence intraglomerular pressure directly. In contrast, SGLT2 inhibitors do have important physiological similarities with carbonic anhydrase inhibitors, which also act proximally, and have been shown to activate tubuloglomerular feedback.

INTRODUCTION

Diabetic kidney disease (DKD) is a common microvascular complication of diabetes mellitus (DM) and is an important, independent risk factor for the development of cardiovascular disease (14). Current clinical markers of DKD progression include worsening albuminuria and/or progressive glomerular filtration rate (GFR) decline (7). Approximately 20–30% of patients with DM (type 1 and 2) develop DKD (43), despite current therapeutic strategies (70) including glycemic, lipid, and blood pressure control and blockade of the renin-angiotensin-aldosterone system (RAAS) (31, 42, 70). As a consequence, DKD remains a leading cause of end-stage renal disease (ESRD) and renal replacement therapy (RRT) in developed countries (7, 36). The identification of novel therapeutic strategies that reduce the risk of DKD initiation and progression is therefore of a research priority.

Beyond older approaches outlined above, more recent mechanistic studies and clinical trial data have highlighted the role for novel antihyperglycemic agents to reduce the risk of DKD and cardiovascular complications (48, 79). Importantly, even though these agents, including sodium-glucose cotransporter-2 (SGLT2) inhibitors and glucagon-like peptide-1 receptor agonists (GLP1-RAs), are approved as glucose-lowering therapies, the mechanisms associated with decreased cardiorenal risk are likely related to natriuretic rather than glucosuric effects. Accordingly, in this review, we focus on the potential role of natriuresis in the pathophysiology of DKD and the relationship between natriuresis and renal protection with newer antihyperglycemic agents.

THE PATHOGENESIS OF DKD

The pathogenesis of DKD is complex, multifactorial, and still incompletely understood (4, 14, 36). DKD is traditionally thought to predominantly impact the glomerulus (75), through a variety of mechanisms including intraglomerular hypertension, as described below, which promotes glomerulosclerosis and inflammation by increasing glomerular wall tension and shear stress, especially in the context of systemic hypertension. The diabetic milieu is, however, also associated with tubular injury and inflammation. Consequently, the concept of “diabetic tubulopathy” has emerged as a significant manifestation of DKD (86). Proteinuria that is typical of DKD can therefore either be due to damage to the glomerular barrier and its subsequent albuminuria and/or loss of the capacity of tubular protein absorption on the basis of tubular injury. Furthermore, the increase in glomerular protein filtration may cause direct tubular toxicity, resulting in tubular proteinuria and perpetuation of the injury process, although this remains controversial (86).

Activation of inflammatory pathways in the setting of diabetes is likely to be primarily related to ambient hyperglycemia, although other factors such as glomerular hemodynamic dysfunction and albuminuria may be involved. This inflammatory activation is achieved by several mechanisms: accumulation of advanced glycation end products and activation of oxidative stress, nuclear transcription factor-κB (NF-κB), protein kinase C-β, and other putative injury pathways (14). The final common result is extracellular matrix and mesangial expansion, thickening of the glomerular basement membrane and tubulointerstitial injury.

SODIUM AND DKD

From a renal mechanistic perspective, sodium (Na⁺) plays an important role in the pathogenesis of DKD for two main reasons. First, increased dietary Na⁺ intake and volume expansion are risk factors for hypertension, which promotes the initiation and progression of DKD (1). Second, the proximal renal tubular handling of Na⁺ likely plays an important role in the pathogenesis of renal hyperfiltration, which is generally defined as either a GFR ≥2 SD above the mean or GFR of ≥135 ml·min⁻¹·1.73 m⁻² (50). Renal hyperfiltration is thought to reflect underlying glomerular hypertension, which contributes to the initiation and progression of clinical DKD, including elevated albumin excretion and progressive renal function decline (71).

There are several theories that explain the onset of renal hyperfiltration in DM, which have been reviewed in detail elsewhere (61, 71). One of these theories, the tubular theory, suggests that increased Na⁺ reabsorption in DM at the proximal tubule via the SGLT1 and SGLT2 in the setting of hyperglycemia (21) leads to reduced Na⁺ delivery at the macula densa with consequent tubuloglomerular feedback (TGF) inhibition (8, 61). TGF suppression leads to afferent arteriolar vasodilation, increased glomerular pressure and renal hyperfiltration (71). The other major renal mechanism that contributes to hyperfiltration is the vascular theory, whereby RAAS activation in DM results in efferent arteriolar vasoconstriction via angiotensin II activity at the angiotensin II type 1 receptor, leading to glomerular hypertension and hyperfiltration (26, 35).

EFFECT OF THE TRADITIONAL TREATMENTS IN DKD

To date, the treatments to reduce the risk of DKD progression are inadequate and include glycemic and lipid control, management of hypertension, and reducing dietary salt intake (47). RAAS inhibitors are also used to reduce the risk of DKD progression (36, 68, 84, 85). For the impact of intensive glycemic control in the setting of type 2 DM, in the Action in Diabetes and Vascular Disease: Preterax and Diamicron MR controlled evaluation (ADVANCE) study, this strategy decreased the onset and worsening nephropathy by 33% (P = 0.005) and the risk of new onset of microalbuminuria by 25% (P < 0.001) (89). Moreover, patients who received intensive glucose lowering in ADVANCE were less likely to develop ESRD than patients who received standard care [hazard ratio (HR): 0.35; P = 0.02] (89), and this benefit persisted in the follow up trial, ADVANCE-ON (HR,: 0.54; P < 0.01) (83). Likewise, in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial (24), the risk of albuminuria was clearly reduced at the end of the study in the intensive glycemic control group (HR: 0.72; P < 0.0001) (24).

In smaller, short-term trials with antihyperglycemic agents, the thiazolidinedione rosiglitazone reduced albuminuria in patients with type 2 DM (36), and similar effects have been shown with pioglitazone (36). In the case of dipeptidyl peptidase-4 (DPP4) inhibitors, although most of the available studies do not show significant renoprotective effects for this class of drugs (77), definitive conclusions cannot be drawn until the results of large cardiovascular trials (CVOT) in patients with DKD are complete. For example, the Multicenter, International, Randomized, Parallel Group Double-blind, Placebo-controlled, Cardiovascular Safety and Renal Microvascular Outcome Study with Linagliptin (CARMELINA study; NCT01897532) (60) should be able to further clarify the renoprotective effect of linagliptin, and this issue will be further explored in the Cardiovascular Outcome Study of Linagliptin vs. Glimepiride in Patients cith Type 2 Diabetes (CAROLINA trial; NCT01243424), which is using an active sulfonylurea comparator (41).

Until recently, trials with antihyperglycemic agents have shown neutral cardiovascular effects and positive effects in the kidney but mainly on surrogate markers such as albuminuria. This paradigm has changed in the last 3 yr, due to results from GLP1-RA trials such as Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results–A Long-Term Evaluation (LEADER) (39) and Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes (SUSTAIN-6) (38), as well as positive SGLT2 inhibitor cardiovascular outcomes such as Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG OUTCOME) (88) and the CANVAS Program (48). For example, in the EMPA-REG OUTCOME and CANVAS Program trials, empagliflozin and canagliflozin, respectively, reduced the risk of the primary three-point major adverse cardiac event end point and also decreased the risk of reaching secondary, composite renal end points (48, 79). Furthermore, both agents reduced the risk of hospitalization for heart failure by >30% (13). These important clinical effects, especially related to DKD and heart failure risk have led to suggestions (22) that SGLT2 inhibitors act primarily as natriuretic agents. See Tables 1, 2, and 3 to see the summary of the results of renal outcomes in trials with antihyperglycemic agents.

Table 1.

Renal outcomes with SGLT2 inhibitors

Drug	Trial	Number of Patients	Median Follow-Up Time, yr	Renal Outcome	Comment
Empagliflozin	EMPA-REG OUTCOME	7,020	3.1	44% RR reduction of doubling of SCr (1.5 vs. 2.6%); P < 0.001
				38% RR reduction of progression to MA (11.2 vs. 16.2%); P < 0.001
				55% RR reduction of initiation of RRT (0.3 vs. 0.6%); P = 0.04
				Slowing GFR decline (annual decrease 0.19 ± 0.11 vs. 1.67 ± 0.13 ml·min−1·1.73 m−2; P < 0.001)
Canagliflozin	CANVAS PROGRAM	10,142	4	Mean eGFR (ml·min−1·1.73 m−2) difference from placebo (2.0 ml/min; 95% CI: 1.5–2.6)
				Mean ratio of UACR compared with placebo, eGFR categories (ml·min−1·1.73 m−2) (P heterogeneity = 0.01)
				eGFR >90: −17%
				eGFR 60–90: −17%
				eGFR 45–60: −26%
				eGFR <45: −13%
				Doubling of Scr, the need for RRT, or death from renal causes. (HR: 0.53; 95% CI: 0.33–0.84 for all subgroups eGFR; P heterogeneity = 0.21 and >0.50, respectively)
	CREDENCE (NCT02065791)	4,461	5.5	NA	Expected completion date: June of 2019
Dapagliflozin	DECLARE (NCT01730534)DAPA-CKD (NCT03036150)	17,1504,000	4–54	NANA	Expected completion date: April 2019
					Expected completion date: November of 2020
Ertugliflozin	VERTIS (NCT01986881)	8,000	5–7	NA	Expected completion date: June 2020

Studies are as follows: CANVAS PROGRAM (48): CANagliflozin cardio Vascular Assessment Study; Cardiovascular and Renal Outcomes with Canagliflozin According to Baseline Kidney Function: Data from the CANVAS Program (49); Canagliflozin and Renal Outcomes In Type 2 Diabetes: Results From The CANVAS Program Randomized Clinical Trials (53); CREDENCE (25): Evaluation of the Effects of Canagliflozin on Renal and Cardiovascular Outcomes in Participants with Diabetic Nephropathy (NCT02065791); DAPA-CKD: a Study to Evaluate the Effect of Dapagliflozin on Renal Outcomes and Cardiovascular Mortality in Patients with CKD (NCT03036150); DECLARE-TIMI 58: Multicenter Trial to Evaluate the Effect of Dapagliflozin on the Incidence of Cardiovascular Events (NCT01730534); EMPA-REG OUTCOME (88): Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients; Empagliflozin and Progression of Kidney Disease in Type 2 Diabetes (79); VERTIS: Randomized, Double-Blind, Placebo-Controlled, Parallel-Group Study To Assess Cardiovascular Outcomes Following Treatment with Ertugliflozin (MK-8835/PF-04971729) in Subjects with Type 2 Diabetes Mellitus and Established Vascular Disease (NCT01986881). SGLT1, sodium glucose cotransporter 2; UACR, urinary albumin-to-creatinine ratio; RRT, renal replacement therapy; SCr, serum creatinine; MA, macroalbuminuria; HR, hazard ratio; RR, relative risk; eGFR, estimated glomerular filtration rate; CI, confidence interval; NA, not available.

Table 2.

Renal Outcomes with DPP4 inhibitors

Drug	Trial	Number of Patients	Median Follow-Up Time, yr	Renal Outcome Change from Baseline Placebo/Intervention	Comment
Saxagliptin	SAVOR-TIMI 53	16,492	2.1	Change in UACR (mg/g) Sstudy end: −34.3. P < 0.004
				Change in UACR category Sstudy end: lowered by saxagliptin, eGFR (ml·min−1·1.73 m−2)
				eGFR >50 (n = 10,621).
				IMPROVED (10.4% SAXA vs. 8.5% PBO). WORSENED (12.7% SAXA vs. 15.1% PBO). P < 0.0001
				eGFR ≤50 (n = 1739).
				IMPROVED (12.5% SAXA vs. 9.8% PBO). WORSENED (50.9% SAXA vs. 49.1% PBO). P = 0.041
				Doubling of Scr HR 1.1 (95% CI: 0.89–1.36)
				Initiation of RRT HR 0.90 (95% CI: 0.61–1.32)
Alogliptin	EXAMINE	5,380	1.6	Change in eGFR (ml·min−1·1.73 m−2), Δ from baseline:
				eGFR >90: −4.5 PBO/−6.7 ALOG
				eGFR 60–90: 1.0 PBO/0.6 ALOG
				eGFR 30–60: 2.1 PBO /1.1 ALOG
				eGFR <30: 1.6 PBO/0.2 ALOG
				Initiation of RRT n (%): 22(0.8) PBO vs. 24(0.9) ALOG; P = 0.88
Sitagliptin	TECOS	14,671	4	Mean difference between-group treatment difference in eGFR (ml·min−1·1.73 m−2) (95% CI) year 4: −1.34 (−1.76 to −0.91), P < 0.001
				Mean difference in UACR (mg/g) (95% CI) year 4: −0.18 (−0.35 to −0.02), P = 0.031
Linagliptin	CARMELINA (NCT01897532)	8,300	4	NA	Expected completion date: January 2018
Linagliptin (vs. glimepiride)	CAROLINA (NCT01243424)	6,000	7.6	NA	Expected completion date: September 2018

Data are modified from Muskiet et al. (46). Studies were as follows: CARMELINA (60): Cardiovascular and Renal Microvascular Outcome Study with Linagliptin in Patients with Type 2 Diabetes Mellitus (NCT01897532); CAROLINA (41): Cardiovascular Outcome Study of Linagliptin vs. Glimepiride in Patients with Type 2 Diabetes (NCT01243424); EXAMINE (80): Examination of Cardiovascular Outcomes with Alogliptin vs. Standard of Care; SAVOR-TIMI 53 (63): the Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus (SAVOR)–Thrombolysis in Myocardial Infarction (TIMI) 53 Trial; Effect of Saxagliptin on Renal Outcomes in the SAVOR-TIMI 53 Trial (45); TECOS (18): Trial Evaluating Cardiovascular Outcomes with Sitagliptin; Effect of Sitagliptin on Kidney Function and Respective Cardiovascular Outcomes in Type 2 Diabetes: Outcomes From TECOS (10). DPP4, dipeptidyl peptidase-4; UACR, urinary albumin-to-creatinine ratio; RRT, renal replacement therapy; SCr, serum creatinine; MA, macroalbuminuria; HR, hazard ratio; NA, not available; PBO, placebo; SAXA, saxagliptin; ALOG, alogliptin.

Table 3.

Renal outcomes wit GLP1-RA

Drug	Trial	Number of Patients	Median Follow-Up Time, yr	Renal Outcome (Change from Baseline Placebo/Intervention)	Comment
Lixisenatide	ELIXA	6,068	2.1	Change in UACR (%) month 24: 34/24; P = 0.004
Liraglutide	LEADER	9,340	3.8	Time to new onset of persistent MA; HR: 0.74 (0.60–0.91); P = 0.03
				Persistent doubling of SCr; HR: 89 (0.67–1.19); P = 0.43
				Need for continuous RRT; HR: 0.87 (0.61–1.24); P = 0.44
				Death due to renal disease; HR: 1.59 (0.52–4.87); P = 0.41
Semaglutide	SUSTAIN 6	3,297	3	New or worsening nephropathy; HR: 0.64 (0.46–0.88); P = 0.005
				Persistent MA; HR: 0.54 (0.37–0.77) P = 0.001
				Persistent doubling of Scr; HR: 1.28 (0.64–2.58) P = 0.48
				Need for continuous RRT; HR: 0.91 (0.40–2.07) P = 0.83
Exenatide	EXSCEL	14,000	3	NA
Dulaglutide	REWIND (NCT01394952)	9,622	6.5	NA	Expected completion date: April 2019
Albiglutide	HARMONY Outcomes (NCT02465515)	9,400	3–5	NA	Expected completion date: May 2019

Data are modified from Muskiet et al. (46). Studies are as follows: ELIXA (56): Evaluation of Cardiovascular Outcomes in Patients with Type 2 Diabetes After Acute Coronary Syndrome during Treatment with AVE0010 (Lixisenatide); EXSCEL (23): Exenatide Study of Cardiovascular Event Lowering Trial; HARMONY Outcomes (NCT02465515): Effect of Albiglutide, When Added to Standard Blood Glucose Lowering Therapies, on Major Cardiovascular Events in Subjects with Type 2 Diabetes Mellitus; LEADER (39): Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results–A Long-Term Evaluation; Liraglutide and Renal Outcomes in Type 2 Diabetes (37); REWIND (NCT01394952): Researching Cardiovascular Events with a Weekly Incretin in Diabetes; SUSTAIN 6 (38): Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes. GLP1-RA, glucagon-like peptide-1 receptor agonist; UACR, urinary albumin-to-creatinine ratio; RRT, renal replacement therapy; SCr, serum creatinine; MA, macroalbuminuria; HR, hazard ratio; NA, not available.

Beyond improving glycemic control, SGLT2 inhibitors and GLP1-RA agents have the capacity to reduce blood pressure and albuminuria, which may contribute to nephroprotective effects (38, 39, 56). In the case of SGLT2 inhibitors, these agents reduce renal hyperfiltration through natriuresis, independent of glycemic control mechanisms. Empagliflozin and canagliflozin have demonstrated nephroprotective effects on the basis of a decrease in the slope of GFR decline and albuminuria progression (48, 88).

EFFECT OF THE NEW ANTIHYPERGLYCEMIC DRUGS IN DKD

SGLT2 Inhibitors

SGLT2 inhibitors are approved as antihyperglycemic agents in countries around the world and also reduce blood pressure and weight (21). From a renal perspective, beyond empirical evidence from CVOTs, SGLT2 inhibitors exert several salutary physiological effects, including induction of a natriuresis, reduction of hyperfiltration, and slowing of the progression of DKD (i.e., renal function loss and elevated albumin excretion) in animals and humans (21). In normal physiology, SGLT2 is responsible for the reabsorption of 5% of the total tubular Na⁺. During ambient hyperglycemia, the reabsorption capacity of tubular Na⁺ increases up to 14% due to the increased mRNA expression and/or protein activity of SGLT1 and SGLT2, leading to a decrease in Na⁺ delivery to the macula densa (21). This natriuresis also reduces tissue Na⁺ content, which may be important during states of volume expansion such as heart failure (81). For example, dapagliflozin treatment for 6 wk reduces the concentration of tissue Na⁺ measured by magnetic resonance imaging, which suggests that SGLT2 inhibitors decrease both the plasma volume and the amount of Na⁺ in the interstitial fluid compartment in humans (62).

Kimura (27) also suggested that SGLT2 inhibitors exert natriuretic effects that are comparable to loop diuretics, which act by inhibiting Cl⁻ and Na⁺ reabsorption at the loop of Henle by blocking the Na⁺/K⁺/2Cl⁻ cotransporter, thereby increasing distal electrolyte delivery. Similar to loop diuretics, SGLT2 inhibitors increase transit of Na⁺ and Cl⁻ through the tubular system, which augments delivery of Na⁺ and Cl⁻ to the macula densa distally. Increased distal Na⁺ and Cl⁻ delivery with loop diuretics does not, however, activate TGF, as discussed below. In contrast, SGLT2 inhibitors also increase distal delivery of Na⁺ and Cl⁻ to the macula densa, resulting in increased Na⁺ and Cl⁻ reabsorption at this location and adenosine triphosphate catabolism to adenosine, which acts as a vasoconstrictor at the afferent arteriole via the adenosine 1 receptor (21, 59). SGLT2 inhibitor-mediated TGF promotes afferent vasoconstriction, reduces glomerular pressure, and causes a 30–50% decrease in albuminuria (21, 59). The reduction in glomerular hypertension is also postulated to account for the characteristic acute and reversible “dip” in GFR in response to SGLT2 inhibitors (3, 5, 21, 79).

Loop diuretics, which also cause an increase in distal delivery of Na⁺ to the macula densa, do not impact TGF pathways that are activated with SGLT2 inhibitors (44). Reabsorption of Na⁺ and Cl⁻ at the macula densa depends on the Na⁺/K⁺/2Cl⁻ cotransporter, which is inhibited by loop diuretics. Since loop diuretics block the same cotransporter at the loop of Henle leading to natriuresis and plasma volume contraction, TGF is unaffected in humans, likely as a consequence of the inability of Na⁺/Cl⁻ to enter macula densa cells and trigger TGF (19, 51, 87). Importantly, other diuretic classes such as thiazides that act at the distal convoluted tubule, as well as agents acting at the cortical collecting duct (e.g., amiloride and mineralocorticoid antagonists), act too distally to impact TGF (22, 40, 76, 78, 87). The mechanism of action of SGLT2 inhibitors more closely resembles that of diuretics acting at the proximal tubule, such as acetazolamide (32).

What is the mechanism by which SGLT2 inhibitors have clinical benefits? Approximately 30% of filtered Na⁺ is reabsorbed at the proximal convoluted tubule by NHE3, which is located in the apical membrane of the tubular epithelium (28). NHE3 is directly involved in Na⁺ reabsorption and indirectly in bicarbonate ions ($HC{O}_3^{-}$) reabsorption, sharing the same domain as SGLT2 in the epithelial membrane of the proximal convoluted tubule (Fig. 1A) (16, 55). It has previously been shown in animals that phlorizin and dapagliflozin reduce NHE3 activity, thereby limiting sodium bicarbonate (NaHCO₃) reabsorption with resultant natriuresis. This action is likely independent of effects on glucose reabsorption (55). Empagliflozin also reduces the activity of NHE3 by means of phosphorylation (15, 28). Thus SGLT2 inhibitor-related natriuresis would result from the sum of inhibition of Na⁺ reabsorption by SGLT2 and NHE3 (Fig. 1B). Carbonic anhydrase is an enzyme expressed in the luminal membrane of the proximal convoluted tubules that leads to the formation of carbonic acid (H₂CO₃) and subsequent conversion to H₂O and CO₂. Carbonic anhydrase activates NHE3 and thereby favors tubular excretion of H⁺ and reabsorption of Na⁺ (28).

Fig. 1.

A: glucose reabsorption in the renal proximal tubule. Normal physiology. The basolateral sodium potassium adenosine triphosphate active transporter (Na⁺-K⁺ ATPase) pumps Na⁺ out and K⁺ into the cell to establish an inward Na⁺ gradient. This gradient is used for Na⁺ and glucose cotransport across the luminal brush border of the early proximal tubule through sodium glucose cotransporter 2 (SGLT2), and the glucose is passively returned via facilitative glucose transporter (GLUT2) to the interstitium/bloodstream. Na is reabsorbed via Na/H exchanger (NHE3). B: the SGLT2 inhibitor blocks Na/glucose reabsorption via SGLT2 and potentially inhibits NHE3. C: action of acetazolamide on carbonic anhydrase (CA) with its effect similar to that of SGLT2 inhibition. CA is an enzyme expressed in the luminal membrane of the proximal tubular cells, leading to the formation of carbonic acid (H₂CO₃) and subsequent conversion into H₂O and CO₂. CA activates NHE3 favoring mobilization of its substrates, allowing tubular excretion of H⁺ and reabsorption of Na⁺. Inhibition of this enzyme involves retention of bicarbonate (NaHCO₃) in the renal tubule, which further increases diuresis.

Acetazolamide is a diuretic that acts at the proximal convoluted tubule by inhibiting carbonic anhydrase, thereby suppressing NHE3 activity. The inhibition of carbonic anhydrase causes H⁺ to accumulate in the epithelial cell, thereby decreasing tubular Na⁺ exchange, which results in natriuresis (Fig. 1C). In clinical practice, however, acetazolamide use is limited to specific scenarios such as altitude sickness (12). In animals, acetazolamide-induced increases in distal tubular Na⁺ and Cl⁻ are associated with activation of TGF and expected renal hemodynamic effects including afferent arteriolar vasoconstriction and GFR decline, intrarenal hemodynamic changes that are not observed with furosemide administration (54, 73). In fact, in humans without diabetes, acetazolamide reduces GFR by 10–20% and also reduces renal hyperfiltration in obese nondiabetic individuals, intrarenal hemodynamic changes that are again not evident with equipotent doses of furosemide (57, 66, 87). Interestingly, in the context of acute altitude sickness, which can be associated with albuminuria, acetazolamide also reduces urine albumin excretion (6, 82). Perhaps most importantly in relation to DM, in patients with type 1 DM, acetazolamide increases proximal natriuresis assessed using renal lithium clearance techniques, while at the same time reducing GFR and albumin excretion (64, 65). SGLT2 inhibitors therefore appear to most closely resemble acetazolamide, rather than more distally acting traditional diuretic agents.

Other Beneficial Effects of SGLT2 Inhibitors in DKD

SGLT2 inhibition is associated with other beneficial renal effects beyond influences on glycemic control or low blood pressure, including suppression of fibrotic, inflammatory, and renal hypoxia pathways. Due to blockade of SGLT2 at the proximal tubule, glucose does not enter renal tubular cells thereby decreasing the generation of free radicals and inflammatory markers (NK-κB, interleukin 6, and monocyte protective protein-1) and fibrosis (fibronectin and transforming growth factor-β). The inflammatory and fibrotic effects at the renal level have been studied in experimental studies; for an example, studies using human proximal tubular cells (HK2) showed that SGLT2 inhibition decreased the production of inflammatory and fibrotic markers induced by high glucose (52). These in vitro findings suggest that SGLT2 inhibitors may provide renoprotection in diabetes by averting glucose from entering proximal tubule cells (74). In humans with type 2 DM, dapagliflozin also reduced levels of interleukin-6 and markers of proximal tubular cell injury (11).

In addition to reducing levels of inflammatory mediators, SGLT2 inhibition may prevent renal hypoxia states characteristic of DM that occur on the basis of augmented glucose cotransport leading to increased energy consumption (29). As a consequence of greater ATP-dependent solute reabsorption, more energy is consumed to reabsorb the increased load of filtered glucose, leading to reduced renal oxygen tension. Accordingly, inhibition of SGLT2 can reduce the state of renal hypoxia that is characteristic of DKD, thereby exerting a protective “β-blocker”-like effect in the kidney (17).

GLP1-RAs and DPP4 Inhibitors

Other antihyperglycemic agents, GLP1-RA and DPP4 inhibitors, are approved based on glucose-lowering activity but also have natriuretic properties. Similar to SGLT2 inhibitors, GLP1-RAs also exert weight loss and blood pressure lowering and albuminuria-lowering effects in addition to inducing natriuresis (70). Albumin excretion-related benefits have been reported in large cardiovascular safety studies with GLP1-RA such as Evaluation of Cardiovascular Outcomes in Patients with Type 2 Diabetes After Acute Coronary Syndrome during Treatment with AVE0010 (ELIXA) (56), LEADER (39), and SUSTAIN-6 (38). For example, in the ELIXA study (56) the median urinary albumin-to-creatinine ratio increased by 34% in the patients treated with placebo but only 24% in patients treated with lixisenatide (difference between treatments; P = 0.004). In the LEADER study using liraglutide, the onset of albuminuria was reduced by 26% [HR: 0.74 (0.60–0.91)] and the urinary albumin-to-creatinine ratio was reduced by 19% (confidence interval: 0.14–0.24%) (39). Finally, in SUSTAIN-6 (38), a reduction in the occurrence of albuminuria was observed in the group receiving semaglutide compared with placebo (semaglutide vs placebo, 2.5 vs. 4.9%). In cardiovascular safety trials, the DPP4 inhibitors such as saxagliptin and sitagliptin exert either neutral or modest reductions in albuminuria in the Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus (SAVOR)–Thrombolysis in Myocardial Infarction 53 trial (SAVOR-TIMI 53) (63) and Trial Evaluating Cardiovascular Outcomes with Sitagliptin (TECOS) (18) trials, respectively. In contrast with SGLT2 inhibitors and GLP1RAs, DPP4 inhibitors have not been shown to reduce blood pressure or body weight.

In comparison with TGF-related mechanisms leading to reductions in albumin excretion with SGLT2 inhibitors, the mechanisms whereby GLP-1-RA and DPP4 provide nephroprotection remain incompletely understood. GLP-1 promotes natriuresis through inhibition of NHE3 and stimulation of atrial natriuretic peptide, although this latter mechanism does not appear to be relevant in humans (34). The natriuresis should promote TGF-mediated afferent arteriolar vasoconstriction and decreased glomerular pressure and albuminuria (26). Despite the natriuresis, which is uniformly seen with GLP1-RAs in humans (2, 20, 34, 67, 72), and in contrast with SGLT2 inhibitors, GLP1-RA does not impact GFR, possibly due to direct endothelial vasorelaxation by GLP1 leading to overall neutral effect on afferent arteriolar tone and renal function (58).

Whereas SGLT2 inhibitors and GLP1RAs induce a proximal natriuresis, DPP4 inhibitor-associated natriuresis is independent of GLP1 and NHE3 in experimental models and instead depends on the activity of a chemokine called stromal cell-derived factor-1α (SDF-1α) (69). In the Akita mouse model, DPP4 inhibition induced natriuresis with linagliptin is dependent on intact SDF-1α activity and is abolished with pharmacological inhibition of SDF-1α with AMD3100 (69). In patients with type 2 diabetes, DPP4 inhibition with sitagliptin enhances natriuresis without impacting renal or systemic hemodynamic parameters (35) and also increased SDF-1α levels. In this cohort, total fractional Na⁺ excretion increased without changes in lithium-derived measures of proximal Na⁺ handling, strongly implicating distal tubular Na⁺ handling pathways. Therefore, while the specific tubular transport mechanisms are not yet defined, existing preclinical and clinical data suggest that pathways downstream from the macula densa are responsible for natriuresis with DPP4 inhibition (35). Since GLP1-RA and DPP4 inhibitors appear not to impact intrarenal hemodynamic function, albumin excretion-lowering effects are unlikely to be a consequence of changes in glomerular pressure and instead were due to suppression of anti-inflammatory pathways (30, 33, 35).

For the effects on neurohormones, and perhaps as expected based on the mild contraction of plasma volume, SGLT2 inhibitors agents are associated with modest increases in circulating and urinary levels of RAAS mediators in patients with diabetes (8, 9). This does not have any known clinical implications, since patients in the EMPA-REG OUTCOME and CANVAS Program trials benefitted from SGLT2i regardless of the use of background RAAS inhibitors therapies. Further work is, however, required to determine if there are synergistic effects in specific patient subgroups, such as in those with heart failure, in larger dedicated heart failure trials such as EMPEROR-preserved, EMPEROR-reduced, and DAPA-HF.

CONCLUSION

Experimental and human data suggest that proximal tubular Na⁺ avidity leads to maladaptive glomerular hypertension and hyperfiltration, thereby promoting DKD. Drugs that restore distal Na⁺ delivery to the macula densa, such as SGLT2 inhibitors and acetazolamide, attenuate glomerular hypertension and hyperfiltration and also reduce albumin excretion. Although other antihyperglycemic agents aside from SGLT2 inhibitors also promote natriuretic effects and reduce albumin excretion, these physiological effects are unlikely to be connected to their clinical benefits. Further investigation is required to determine how to optimally combine antihyperglycemic agents, such as SGLT2 inhibitors and GLP1RAs, that have complementary physiological effects around natriuresis, as well as weight loss, glucose, and blood pressure lowering, and cardiorenal protection.

GRANTS

D. Z. Cherney is supported by the Canadian Institutes of Health Research, as well as Diabetes Action Canada, a Strategy for Patient-Oriented Research initiative supported by the Canadian Institutes for Health Research. D. Z. Cherney also receives support from the Department of Medicine and Heart and Stroke Richard Lewar Centre of Excellence in Cardiovascular Research, University of Toronto and the Banting and Best Diabetes Centre, University of Toronto. D. Z. Cherney is the recipient of a University of Toronto, Department of Medicine Merit Award. P. Bjornstad receives salary and research support from National Institute of Diabetes and Digestive and Kidney Diseases Grants T32-DK-063687 and K23-DK-116720. P. Bjornstad is also a recipient of a Research Fellowship from Juvenile Diabetes Research Foundation and International Society of Pediatric and Adolescent Diabetes, and receives research support from Juvenile Diabetes Research Foundation, Thrasher Foundation, Childhood Diabetes Foundation, and Colorado Clinical and Translational Sciences Institute and Center for Women’s Health Research at University of Colorado.

DISCLOSURES

D. L. Jiménez has received speaking honoraria from Boehringer Ingelheim, Janssen, AstraZeneca, Sanofi, Novo Nordisk, Eli Lilly, and Esteve. J. P. González has received consulting fees or speaking honoraria from Boehringer Ingelheim, Janssen, AstraZeneca, Merck, Sanofi, and Novo Nordisk, Bayer. D. Z. Cherney receives operating funding from Boehringer Ingelheim, Janssen, AstraZeneca, and Merck. D. Z. Cherney has received consulting fees or speaking honoraria from Boehringer Ingelheim, Janssen, AstraZeneca, Merck, Mitsubishi-Tanabe, Sanofi, and Abbvie. P. Bjornstad has received travel support from Boehringer Ingelheim and speaking honoraria from Horizon Pharma.

AUTHOR CONTRIBUTIONS

D.L.J., D.Z.I.C., P.B., and J.P.M.G. conceived and designed research; D.L.J. and J.P.M.G. prepared figures; D.L.J., D.Z.I.C., P.B., L.C.G., and J.P.M.G. drafted manuscript; D.L.J., D.Z.I.C., P.B., L.C.G., and J.P.M.G. edited and revised manuscript; D.L.J., D.Z.I.C., P.B., L.C.G., and J.P.M.G. approved final version of manuscript.