Effects of Diabetes Medications Targeting the Incretin System on the Kidney

Incretins are gut-derived peptide hormones that modulate glucose homeostasis as well as other physiologic pathways. Pharmacologic inhibition of dipeptidyl peptidase-4 (DPP-4) potentiates incretin action by reducing their degradation (Figure 1). Glucagon-Like Peptide-1 Receptor agonists (GLP-1RAs) that are resistant to DPP-4 have also been developed. In both cases, an increased incretin effect results in a glucose-dependent reduction in blood glucose, which is valuable for the management of type 2 diabetes. Beyond glucose lowering, recent large-scale, randomized, placebo-controlled trials (RCTs) have also raised the possibility that targeting the incretin pathway may also have pleiotropic effects to protect the diabetic kidney. Although kidney end points were not the primary outcome of these trials, the recruitment of patients with or at high risk of cardiovascular disease (CVD) has provided an opportunity to explore the effects of glucose-lowering medications on CKD, which is highly prevalent in this setting.

Figure 1.

Glucagon-Like Peptide-1 (GLP-1) is sequentially degraded by dipeptidyl peptidase-4 (DPP-4) and neprilysin (NEP) into smaller peptides that may retain biologic activity independent to the Glucagon-Like Peptide-1 Receptor (GLP-1R). Inhibition of DPP-4 or the use of Glucagon-Like Peptide-1 Receptor agonist (GLP-1RA) results in increased activation of GLP-1Rs in the kidney, leading to the increased levels of cAMP and the activation of protein kinase A (PKA). DPP4i, dipeptidyl peptidase-4 inhibitor; NEPi, neprilysin inhibitor; NHE3, reducing Na/H exchange transporter isoform 3.

In phase 3 RCTs with DPP-4 inhibitors, a modest reduction in albuminuria has been reported (1–3). However, the only study to specifically examine their effect on albuminuria was negative (4). Moreover, no actions of DPP-4 inhibition on doubling of serum creatinine, change in GFR, kidney failure, or ESRD have been shown (1–3).

By comparison, the kidney outcomes in RCTs with GLP-1RAs seem more promising. In the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome (LEADER) Trial, the effects of daily injections of the GLP-1RA liraglutide were examined in 9340 subjects with type 2 diabetes at high risk for CVD (5). Over a median follow-up of 3.8 years, participants receiving liraglutide experienced a 22% reduction (hazard ratio [HR], 0.78; 95% confidence interval [95% CI], 0.67 to 0.92; P=0.003) in the prespecified kidney outcome, a composite of time to the development of new-onset persistent macroalbuminuria, persistent doubling of serum creatinine and eGFR<45 ml/min per 1.73 m², ESRD, and death due to CKD. This benefit was driven by a 26% reduction in new-onset macroalbuminuria (HR, 0.74; 95% CI, 0.60 to 0.91; P=0.004). The rate of decline in eGFR during the study was also modestly lower in participants receiving liraglutide compared with the placebo group (−7.4 versus −7.8 ml/min per 1.73 m² over 36 months; P=0.01), especially in participants with stage 3 or 4 CKD at baseline (5). However, the number of participants who experienced a persistent doubling of serum creatinine level and an eGFR≤45 ml/min per 1.73 m² was unaffected (HR, 0.89; 95% CI, 0.67 to 1.19; P=0.43), and no changes were observed in the incidence of ESRD (5).

In the Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes (SUSTAIN-6), 3297 patients with type 2 diabetes with CVD or having significant cardiovascular risk factors were randomly assigned to the GLP-1RA semaglutide (0.5 or 1 mg once weekly) or placebo (6). After a median follow-up of 2 years, new or worsening nephropathy, the same end point used in the LEADER Trial, also occurred less frequently in patients treated with semaglutide (HR, 0.64; 95% CI, 0.46 to 0.88; P<0.01). As was seen in the LEADER Trial, this outcome was largely driven by a reduction in new-onset macroalbuminuria (semaglutide versus placebo: 2.5% versus 4.9%, respectively), whereas doubling of serum creatinine concentration to an eGFR≤45 ml/min per 1.73 m², ESRD, and kidney death were unaffected.

The Exenatide Study of Cardiovascular Event Lowering Trial with once weekly exenatide also reported a reduction in new-onset macroalbuminuria (2.8% versus 2.2%; P=0.03) in patients with type 2 diabetes, 70% of whom had preexisting CVD (7). Similarly, the Evaluation of Lixisenatide in Acute Coronary Syndrome Trial also reported modestly lower albuminuria after treatment with lixixenatide compared with placebo in 6068 patients with type 2 diabetes and recent admission for acute coronary syndrome (8). Finally, A study Comparing Dulaglutide with Insulin Glargine on Glycaemic Control in Participants with Type 2 Diabetes and Moderate or Severe Chronic Kidney Disease (AWARD-7) compared once weekly dulaglutide (0.75 or 1.5 mg) with insulin glargine (with both groups also receiving prandial insulin lispro) in 576 subjects with type 2 diabetes and stage 3 or 4 CKD. Consistent with the above-mentioned trials, AWARD-7 also reported a significant reduction in albuminuria and a slowing of GFR decline, specifically in participants with macroalbuminuria at baseline (9).

Although these RCTs have documented a reduction in albuminuria associated with incretin-targeted therapies, it is noteworthy that a substantial improvement in glycemic control was also documented. Indeed, the trial showing the greatest reduction in albuminuria (the SUSTAIN-6) also had the greatest difference in glycemic control between control and treatment arms (6). Glucose lowering alone (by any strategy) will reduce new-onset macroalbuminuria by 20%–25%, similar in magnitude to that observed in the GLP-1RA trials. Reductions in insulin, body weight, and BP may also have benefits in the kidney. Nonetheless, sensitivity analysis from the LEADER Trial suggests that the effects of liraglutide were partly independent of improvement in glycemic control (−0.4% hemoglobin A1c) and body weight (−2.3 kg) between the two study arms (5). Equally, the clear reduction in albuminuria in the AWARD-7 Trial, in which glycemic equipoise was achieved (9), seems consistent with glucose-independent actions on CKD in type 2 diabetes.

The mechanisms that may underlie any direct actions in the kidney remain to be established (Figure 1). Certainly, the Glucagon-Like Peptide-1 Receptor (GLP-1R) seems to be expressed in glomeruli and arterioles. Glucagon-Like Peptide-1 induces natriuresis by reducing Na/H exchange transporter isoform 3–dependent sodium reabsorption in the proximal tubule. GLP-1R agonists have also been reported to reduce oxidative stress, inflammation, macrophage infiltration, and the accumulation of type 4 collagen in the kidney (reviewed in ref. 10). Classic signaling via the GLP-1R reduces adenylate cyclase/cAMP and protein kinase A activity, and inhibition of adenylate cyclase and protein kinase A are both able to antagonize the beneficial effects of liraglutide in the experimental kidney disease, suggesting that these pathways may be involved (2). However, kidney protective actions independent of the GLP-1R have also been proposed. DPP-4 inhibition results in supraphysiologic levels of endogenous Glucagon-Like Peptide-1 and glucose-dependent insulinotropic polypeptide, resulting in improved glucose control in type 2 diabetes. Despite the obvious mechanistic similarities with GLP-1RA, over 40 other substrates are also metabolized by DPP-4, including high-mobility group protein box 1 and stromal cell–derived factor 1α, which have both been implicated in CKD.

The key advantage of DPP-4 inhibition is its safety profile and sustained efficacy in all stages of CKD. In the setting of kidney impairment, a dose reduction is recommended for all DPP-4 inhibitors apart from linagliptin, which is not excreted by the kidney. This is to ensure that drug exposure is consistent, and inadvertent overexposure is not dangerous, because all DPP-4 inhibitors have a high therapeutic index. By comparison, their remains some limitations regarding the use of GLP-1RAs in patients with CKD. In particular, the dose-related gastrointestinal toxicity of GLP-1RAs (nausea, vomiting, and diarrhea) is increased in patients with impaired kidney function, largely due to increased drug exposure. This may underlie anecdotal reports of AKI associated with the use of GLP-1RAs, although reassuringly, the clinical trials with GLP-1RAs have not borne this out (5–9). Nonetheless, careful dosing is required for kidney-cleared GLP-1RAs (exenatide and lixisenatide), and although their clearance is predominantly hepatic, liraglutide, albiglutide, and dulaglutide should be used with caution in patients with impaired kidney function (10).

In summary, clinical trials of incretin-modulating medications have shown the clear potential to reduce albuminuria and possibly to slow the rate of decline in GFR in type 2 diabetes and CKD. No benefits have been observed on hard kidney outcomes, although their event rate is too low and follow-up is too brief to sufficiently explore these outcomes outside of new trials specifically in patients with advanced CKD. Nonetheless, it is hoped that early and sustained control of glucose and weight together with reduced progression to macroalbuminuria achieved with incretin-based therapy will ultimately translate into long-term preservation of kidney function and reduced ESRD.

Disclosures

R.J.M. was an investigator for the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Trial and the Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes involving the use of liraglutide and semaglutide, respectively.