Novel Therapies for Diabetic Kidney Disease: Storied Past and Forward Paths

Currently, the available therapies for DKD include treatment of hypertension with RAS inhibition, glycemic control, and dietary interventions. Inhibition of the RAS has been the primary therapeutic intervention for DKD for two decades. Several clinical trials demonstrated that administration of single RAS inhibitors, angiotensin II receptor blockers (ARBs), or ACE inhibitors was mod-estly renal-protective in patients withDKD and “overt proteinuria/macro-albuminuria” (generally, urine protein-to-creatinine ratio >500 mg/g or albumin-to-creatinine ratio >300 mg/g) (5–7).

This result prompted testing of the hypothesis that further suppression of the RAS by “dual blockade” (combination therapy with two agents (e.g., ACE inhibitor, ARB, and/or a direct renin inhibitor) may be synergistic and result in greater renal protection (8–10). These clinical trials were stopped early because of safety concerns and high rates of adverse events, particularly hyperkalemia and acute kidney injury with dual therapy (8–10). There were no apparent benefits on outcomes from combination therapy. However, because the trials were stopped prematurely, the efficacy results from these trials are inconclusive. (See the related article by Patney et al. on p. 175 of this issue.) Despite evidence that single RAS inhibition with either an ARB or an ACE inhibitor can reduce the risk of DKD progression, there is increased risk of adverse events with dual RAS blockade and no obvious additional benefit of this form of combination therapy in recent clinical trials. Development of RAS-independent therapies is an important direction for future work.

Glycemic control is essential to the optimal management of diabetes and prevention of complications, including DKD in both type 1 and type 2 diabetes. Long-term follow-up of type 1 diabetes patients in the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) cohort study, showed durable benefits of intensive glycemic control to an A1C of ∼7% versus standard treatment to an A1C of ∼9% on preventing albuminuria and reduced kidney function (11). After approximately three decades of follow-up, there was also reduction of ESRD risk, although the absolute number of cases was small.

A strong positive association between A1C concentration and incident DKD was found after 11 years of follow-up in type 2 diabetes in the Atherosclerosis Risk in Communities study (12). The association was present in the absence of albuminuria and retinopathy at baseline. Similar to the DCCT/EDIC in type 1 diabetes, several studies in relatively early-onset type 2 diabetes found that controlling hyperglycemia (to an A1C of ∼7% vs. an A1C of ∼9%) prevented new-onset and progressive albuminuria (13–17). More recent studies enrolling older adults (generally >60 years of age) with longstanding type 2 diabetes, including ACCORD (Action to Control Cardiovascular Risk in Diabetes), ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation), and VADT (Veterans Affairs Diabetes Trial), which targeted even lower A1C goals (<6–6.5%), failed to demonstrate benefits of more intensified glycemic control on the primary CVD outcomes, although the risk of albuminuria development or progression was reduced (18,19). The ADVANCE trial eventually showed reduced risk for ESRD after the period of intensive glycemic control in the follow-up ADVANCE-Post Trial observational study, but like the DCCT/EDIC in type 1 diabetes, the absolute number of cases was low. Importantly, the risk of severe hypoglycemia (neurological compromise) was increased ∼2.5-fold in these trials, which greatly restricts use of intensive regimens in this population (1). Limitations of regimens for glycemic control highlight the need for alternate approaches to improve clinical outcomes in DKD.

Numerous mechanisms driving the development and progression of DKD have been investigated as possible targets for novel therapies (Table 1). In humans and animal models, DKD is characterized by glomerular and tubulointerstitial disease with inflammation and fibrosis figuring prominently. Glomerulosclerosis and tub-ulointerstitial fibrosis culminate in loss of kidney function (20–23). Underlying mechanisms of DKD include hemodynamic and metabolic disturbances leading to activation of myriad mediators with autocrine and paracrine actions in the kidney. Primary among the aberrant metabolic products that drive the DKD process are advanced glycation end products (AGEs) and reactive oxygen species (ROS). These products are key activators for upregulation of proinflammatory and pro-fibrotic mediator production. Multiple cell types produce these mediators, ultimately resulting in the pathogenesis of DKD.

AGEs are modified proteins, peptides, and amino acids that are nonenzymatically glycated and oxidized after interaction of amino groups with aldose sugars. AGEs increase in hyperglycemic conditions and after consumption of foods high in protein, especially animal meats cooked at high temperatures. AGEs are increased in the kidneys of patients with DKD, and serum levels of AGEs correlate with DKD severity (24,25). AGEs are nephrotoxic by mechanisms including inflammation, fibrosis, and apoptosis of kidney cells (26). A major pathway for cellular demise in response to AGEs is via the receptor for AGEs (RAGE), which initiates signals that activate transcription for mediators of these processes.

Several therapies that inhibit the formation of AGEs, or AGE crosslink breakers, have been under investigation for DKD. The crosslink breaker alagebrium reduces inflammation, fibrosis, and overall severity of kidney damage in diabetic mice (27). Although enrollment began for a phase II randomized, placebo-controlled trial of alagebrium in patients with type 1 diabetes and DKD, it was unfortunately terminated early because of business and regulatory considerations (28).

Pyridoxamine (PDX), a derivative of vitamin B₆, inhibits the formation of AGEs. In two different rodent models of DKD, administration of PDX improved albuminuria and attenuated increases in serum creatinine (29,30). In an analysis of pooled data from two phase II clinical trials of PDX for 24 weeks’ duration in people with DKD and overt proteinuria, increases in serum creatinine levels were attenuated (31). The pro-fibrotic biomarker, urinary transforming growth factor-beta (TGF-β), was also reduced. The beneficial effect of PDX was most readily observed in patients with type 2 diabetes and a serum creatinine level >1.3 mg/dL. Another phase II, randomized trial of PDX in patients with type 2 diabetic nephropathy did not show benefit on changes in serum creatinine except in patients who had reduced kidney function at the onset of the trial (32). Currently, a phase III clinical trial is being conducted to evaluate the efficacy and safety of PDX for reducing the risk of clinical events (ESRD and eGFR loss of 50%) in patients with DKD and severely increased albuminuria receiving the standard-of-care, background RAS inhibition. Reducing AGEs remains a promising area for development of DKD therapies.

ROS are increased and contribute to induction of cell-signaling pathways that produce inflammatory and fibrotic molecules in the diabetic kidney (33). ROS promote formation of AGEs, which in turn further increase cellular ROS, eliciting an autocrine loop of AGEs and ROS production in the diabetic milieu (Figure 1). ROS also induce signaling cellular cascades, including protein kinase C (PKC) and nuclear factor κB (NF-κB), culminating in the expression of numerous proinflammatory and pro-fibrotic genes (23).

Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) is a transcription factor that functions as a regulator of an endogenous antioxidant system. Compounds that activate Nrf2 have attenuated diabetes-induced kidney hypertrophy, fibrosis, and albuminuria (34).

Bardoxolone methyl (BM), an activator of Nrf2, was studied in the phase III clinical trial Bardoxolone Methyl Evaluation in Patients with Chronic Kidney Disease and Type 2 Diabetes Mellitus: the Occurrence of Renal Events (BEACON) (35). This trial was terminated prematurely because of numerous severe adverse events, including increased rates of CVD events, particularly heart failure, as well as higher levels of albuminuria and blood pressure. The consequences of this clinical trial have raised serious concern over the use of this strategy in DKD (36). For example, caution should be exercised when a drug increases, rather than decreases, albuminuria and blood pressure (37). The abrupt increase in eGFR observed in the phase II clinical trial of BM was likely related to glomerular hyperfiltration and raised intraglomerular pressure. Preclinical experiments with two analogs of BM (dh404 and RTA 405) may provide insight into the fate of the BEACON trial (38–40). For example, in a study of RTA 405 in diabetic rats, measures of kidney damage (proteinuria, tubular damage, and glomerulosclerosis) actually worsened. To the contrary, RTA 405 attenuated increases in blood urea nitrogen and creatinine in diabetic rats in another study (40). Overall, there is limited and conflicting evidence from experimental models supporting the use of BM for DKD.

Activation of Nrf2 in the kidneys with sulforaphane (SFN) and tert-butylhydroquinone (tBHQ) has attenuated kidney damage in mouse models of diabetes (34,41). For example, 3 months of SFN reduced albuminuria along with decreases in fibrosis, inflammation, and oxidative stress in the kidneys of diabetic mice (33). Administration of tBHQ significantly reduced kidney weight and proteinuria, as well as decreased kidney levels of fibronectin while concomitantly increasing expression of Nrf2 expression and antioxidant genes (41). Antioxidant therapies, perhaps including Nrf2 activation, may be worthy of further exploration for DKD.

PKC is activated by a number of diabetes-related stimuli, such as AGEs, hyperglycemia, angiotensin II, and ROS (Figure 1). PKC conveys signals to several downstream targets, including NF-κB, the SMAD/TGF-β axis, and apoptosis systems (42–44). Thus, PKC can be conceived as a nodal point in major signaling pathways of DKD, which makes it an attractive therapeutic target.

PKC has at least 11 isoforms. There has been substantial interest in determining which isoforms lead to DKD. Early studies implicated PKCβ as one of the primary isoforms (45). Administration of the selective PKCβ inhibitor ruboxistaurin (RBX) reduced glomerular hyperfiltration and albuminuria in diabetic rats and reduced kidney fibrosis, mesangial expansion, and glomerulosclerosis in diabetic mice (45–47). PKCα is another isoform implicated in DKD. Inhibition of PKCα abrogated albuminuria but not expression of fibrotic genes or total kidney and glomerular hypertrophy in diabetic mice (48). Recently, studies exploring reduction of both PKCα and PKCβα abolished diabetes-induced renal hypertrophy, podocyte loss, and reduced fibrosis and albuminuria in mice. In all, these data suggest that dual inhibition of PKCαα and PKCβα may be a candidate therapeutic approach (49).

A randomized, controlled phase II clinical trial examined the effects of RBX (32 mg/day) in people with type 2 diabetes and persistent albuminuria (urinary albumin-to-creatinine ratio [UACR] 200–2,000 mg/g), despite therapy with RAS inhibitors (50). UACR decreased significantly and substantially (mean reduction 24%) in those treated with RBX compared to placebo. The UACR-lowering effect of RBX appeared by 1 month. eGFR did not significantly decline in the RBX group, whereas the placebo group lost eGFR at a rate of ∼5 mL/min/1.73 m² over a 1-year period (50). Urinary TGF-β increased by 43% over 1 year in the placebo group, but not in study participants who received RBX (51).

In a post hoc safety analysis from 11 controlled clinical trials of RBX in diabetic retinopathy, the overall rate of serious adverse events was not increased in the RBX group. Indeed, the placebo group experienced more frequent serious adverse events than the RBX-treated group (23 vs. 20%, respectively) (52). Recently, another post hoc analysis of kidney-related outcomes in phase III clinical trials of RBX for diabetic peripheral neuropathy was conducted. After 3 years, patients with RBX had lower UACRs and higher eGFRs compared to placebo, suggesting that RBX might prevent or delay DKD development (53).

It is important to recognize that RBX has not moved from phase II to phase III clinical trials because of business and regulatory decisions rather than concerns regarding safety or efficacy. Studies of PKC inhibition may yet yield a novel therapy for DKD.

Serum amyloid A (SAA) is an acute-phase proinflammatory protein expressed in podocytes, mesangial cells, and tubular epithelium that may contribute to inflammatory and apop-totic mechanisms in DKD (54,55). SAA initiates an inflammatory signal-ing cascade that results in upregulation of SAA itself, along with multiple inflammatory cytokines and chemoattractant molecules in podocytes (54,56). This suggests that the podocyte may promote local SAA-mediated inflammatory responses such as monocyte and macrophage recruitment in the glomeruli. SAA expression at mRNA and protein levels are increased in glomerular and tubulointerstitial compartments of diabetic mouse models and patients with DKD (56). Furthermore, SAA may also be a DKD biomarker in that blood levels associate with prevalent albuminuria in people with type 2 diabetes and predict incident albuminuria in type 1 diabetes (57–60). In studies to date, associations of SAA with DKD are independent of traditional risk factors, suggesting that it could add to DKD risk prediction in diabetic patients (56). SAA is an encouraging new candidate for therapeutic intervention and biomarker development in DKD.

Combination drug therapies that work synergistically to ameliorate serial pathogenic mechanisms may ultimately prove to be more successful than single therapies. Because adverse safety signals have limited recent attempts at intensified or combinatorial drug regimens, safety and efficacy together must be carefully considered in the preclinical experimental phase and in the design and execution of clinical trials. Additionally, methods for targeting therapies to specific sites and targets of disease, such as nano-particle delivery systems and antisense oligonucleotides, may be useful tools to enhance treatment efficacy and safety (61,62).

DKD is a multifactorial diabetic complication with numerous mech-anistic pathways contributing to disease pathogenesis. Comprehensive mechanistic-based studies and improved strategies for clinical translation are required for the development of safe and effective new treatments. Among many possible targets, AGEs, ROS, PKC, and SAA are promising mechanisms for therapeutic and biomarker development.

Dr. Tuttle is a consultant in the area of novel therapies for DKD for Amgen, Eli Lilly and Company, and Noxxon Pharmaceuticals. No other potential conflicts of interest relevant to this article were reported.