GLP-1 agonists in the treatment of chronic kidney disease in type 2 diabetes and obesity (original) (raw)

Gut-kidney axis: kidney sodium handling. Analogous to the gut-pancreas connection, which identified the incretin effect relating to insulin secretion, early studies proposed the existence of a so-called gut-kidney axis. The proposals were based on studies in humans showing that in a sodium-depleted state, an oral sodium load is more rapidly excreted by the kidneys than an intravenous sodium load (24). This rapid excretion occurred independent of changes in the levels of circulating regulators of plasma volume, including atrial natriuretic peptide (ANP) and aldosterone, and a similar phenomenon has also been suggested for the excretion of electrolytes including potassium (18). While several gut hormones secreted by intestinal enteroendocrine cells may contribute to the gut-kidney axis, studies have shown that an oral sodium load can increase endogenous GLP-1 concentrations (18). In addition, clinical studies with GLP-1 peptide or its receptor agonists have also pointed toward a potential role for this hormone in influencing sodium homeostasis. Indeed, increased urinary sodium excretion and increased urine volume induced by GLP-1 peptide and its receptor agonists have been confirmed in rodents (25), healthy volunteers (26, 27), insulin-resistant obese males (28, 29), and patients with T2D during acute infusion studies (30). These effects on urinary volume and sodium excretion could be blocked in mice using a GLP-1 receptor antagonist (31). As seen with inhibitors of SGLT2, this increase in urinary sodium excretion is only present acutely (32). The mechanisms by which GLP-1 receptor agonists induce natriuresis are unclear, but in both rodents and humans, GLP-1 receptor agonist infusion enhances kidney lithium clearance, a marker of proximal tubular sodium reabsorption, as well as urinary pH. This suggests that the effect involves inhibition of sodium-hydrogen antiporter (NHE3) (33). Other proposed mechanisms underlying GLP-1 receptor agonism’s role in facilitating natriuresis include reduction of RAAS activity through central nervous system signaling and alteration of ANP concentrations. To what extent altered kidney sodium handling is involved in the kidney-protective effects of GLP-1 receptor agonist treatment is, however, uncertain, and GLP-1 induced natriuresis is unlikely to be the major pathway underlying renoprotection.

Glomerular hyperfiltration. Glomerular hyperfiltration has been identified as a key factor predisposing to kidney disease. Although a generally accepted definition is lacking, glomerular hyperfiltration has been defined as a GFR above 135 mL/min/1.73 m2, corresponding to two standard deviations above mean GFR in healthy individuals (34). However, in people with reduced kidney mass, hyperfiltration can occur despite the whole-kidney GFR being in the normal range, a condition referred to as single-nephron hyperfiltration. Multiple factors contribute to hyperfiltration, including impaired regulation of pre- and postglomerular resistances as well as compromised tubuloglomerular feedback due to increased proximal reabsorption of sodium and chloride (34). Studies of the effects of different GLP-1 receptor agonists on glomerular hemodynamic function and hyperfiltration have yielded inconsistent results. In obese individuals, GLP-1 infusion acutely decreased creatinine clearance (29); however, in other infusion studies with the GLP-1 receptor agonist exenatide, measured GFR (mGFR) was increased in healthy volunteers (26) and unchanged in people with T2D (30). In an uncontrolled study in people with T2D, liraglutide initially lowered eGFR (35), but this result was not confirmed in placebo-controlled studies that assessed mGFR and kidney hemodynamic function by gold-standard tracer methods (36). In people with DKD, the once-weekly GLP-1 receptor agonist dulaglutide increased mGFR (37), although semaglutide appeared to initially reduce eGFR (15). Thus, in contrast to SGLT2 inhibitors that have a clear and consistent hemodynamic mode of action, it is unlikely that kidney hemodynamic actions contribute to the renoprotective effects of GLP-1 receptor agonists.

Kidney oxygen availability. In recent years, the interest in kidney hypoxia as a driver of DKD has been revived owing to the development of techniques that allow the study of kidney oxygen availability. The chronic kidney hypoxia theory states that in DKD, the kidneys have insufficient oxygen availability, causing tissue damage. Because of constant reabsorption, most notably of sodium, kidney tissue is second only to the heart in oxygen consumption per gram tissue. Kidney hypoxia may be driven by two mechanisms. First, it may be due to impaired oxygen delivery resulting from microvascular damage. Second, increased oxygen consumption, caused by both glomerular hyperfiltration and inefficient energy generation due to impaired insulin sensitivity, may also contribute (38).

Blood oxygen level–dependent magnetic resonance imaging (BOLD-MRI) has been put forward as a novel tool to measure tissue oxygenation. In people with CKD, lower oxygen availability was associated with more rapid kidney function decline in comparison with people with higher oxygen availability (39). Two studies have investigated the effects of GLP-1 and GLP-1 receptor agonist treatment on several aspects of kidney physiology as measured by multiparametric kidney MRI. In a study in which GLP-1 peptide was infused, GLP-1 increased both cortical and medullary perfusion (40). In contrast, prolonged treatment with the GLP-1 receptor agonist semaglutide lowered perfusion compared with placebo (41). In both studies, oxygen availability as assessed by BOLD-MRI was not changed. Thus, it remains uncertain whether alterations in kidney oxygen availability mediate the beneficial effects of GLP-1 receptor agonist treatment.

Inflammation. It has been well established that in both type 1 diabetes (T1D) and T2D, a chronic low-grade inflammatory state contributes to DKD, in part due to hyperglycemic insults. Both kidney cells and recruited immune cells of the innate (M1 macrophages) and the adaptive (CD4+ and CD8+ T cells) immune systems produce proinflammatory factors that damage the kidney (42). Increased levels of cytokines such as IL-6, IL-8, IL-1β, TNF-α, and IFN-γ as well as chemokines including chemokine receptor 2 (CCR2), C-C motif chemokine ligand 2 (CCL2), CCL5 (RANTES), and C-X-C motif chemokine ligand 10 (CXCL10) have been observed in kidney tissue of rodent models of diabetes, as well as in plasma and serum of people with DKD (43).

While the direct effects of GLP-1 receptor agonists on kidney inflammation are difficult to investigate in humans, in a number of studies, GLP-1 receptor agonists were shown to modulate inflammation. In peripheral blood mononuclear cells isolated from people with T2D, exendin-4 reduced the expression and production of proinflammatory cytokines and chemokines, in association with a reduction in oxidative stress (44). In people with T1D, GLP-1 infusion counteracted hypo- and hyperglycemia-induced increases in the inflammation markers IL-6 and soluble intercellular adhesion molecule-1 (sICAM-1) (45). Additional mechanisms of the effects of GLP-1 on inflammation have also been detailed in rodent models, as described below. Because human phenotyping studies have not demonstrated clear mechanisms as to how GLP-1 receptor agonists improve kidney outcomes, a reduction in kidney inflammation that is difficult to measure in humans remains a leading candidate mechanism.

Reduction of kidney fat accumulation and lipotoxicity. Beyond the general DKD risk conferred by obesity, fat accumulation in depots anatomically close to the kidney as well as intrakidney fat accumulation has been linked to kidney disease. Specifically, increased perirenal fat (46) and hilar fat (47) have been linked to CKD. Proposed mechanisms include physical compression of vessels and nerves, induction of glomerular hyperfiltration (48), and activation of the RAAS as well as stimulation of inflammation by local release of cytokines (49). In addition, parenchymal lipid deposition may induce lipotoxicity (50) through the formation of toxic metabolites including ceramides and diacylglyceride that hamper the function of multiple kidney cell types, including podocytes (51).

Given the effects of GLP-1 receptor agonists on body weight and fat, the question arises of whether a reduction in kidney-specific fat depots could contribute to the kidney-protective effects of GLP-1 receptor agonist treatment. Currently, a number of studies are employing multiparametric kidney MRI to assess the effects of semaglutide (NCT04865770) (52), tirzepatide (NCT05536804) (53), and retatrutide (NCT05936151) (54) on kidney fat accumulation and link these effects to changes in GFR and albuminuria. Regarding lipotoxicity, a study in kidneys of mice fed a high-fat diet (55) revealed that the GLP-1 receptor agonist dulaglutide changed the lipid content of the organ, including reductions in diacylglycerols, phosphatidic acids, phosphatidylglycerols, and triglycerides. On the other hand, levels of other lipids, including cardiolipins, which play a pivotal role in mitochondrial metabolism, were increased after dulaglutide treatment. These data indicate that GLP-1 receptor agonist–induced changes in local adipose tissue depots or parenchymal fat accumulation may contribute to their ability to protect the kidneys.