Transformative Research in Diabetic Nephropathy

Progress in the area of diabetic kidney research leading to new therapeutics development has been very limited. Indeed, no new medicines indicated for the treatment of chronic kidney disease (CKD) have been approved since ARB's have become standard of care nearly 15 years ago. Several factors explain the limited progress including but not limited to; a) animal and cell culture models do not recapitulate human DKD b) human genetic studies so far have failed to identify reproducible genetic variants associated with DKD c) the clinical manifestation of DKD is heterogeneous and might have even changed since the original description d) DKD is a clinical diagnosis and it is not clear what percentage of patients have histological disease.

Laboratory mice have served as invaluable tools to understand human disease development. As mouse genetic tools became readily available, it enabled us to perform time and cell type specific gene manipulation in animals to generate disease models and to understand the contributions of specific pathways. Unfortunately, mouse models do not recapitulate human diabetic kidney disease as animals develop only early DKD lesions; mesangial expansion and mild albuminuria11. Most models do not develop arterial hyalinosis, tubulointerstitial fibrosis and declining glomerular filtration rate (GFR); hallmarks of progressive DKD. There are several fundamental differences in gene expression patterns and physiology of human and murine kidneys. Such differences may explain the lack of translatability between mice and humans of pharmacological approaches aimed at treating DKD. This seems to be a general trend in other disease areas as well (for example Alzheimer's disease), leading to a recent movement toward translational and clinical research with increasing reliance on human samples.

Human genetic studies made paradigm-shifting observations in relatively rare monogenic forms of kidney diseases (including polycystic kidney disease and focal segmental glomerulosclerosis). Diabetic CKD on the other hand follows a complex polygenic pattern. Currently, the most powerful method to define the genetics of complex diseases such as DKD is genome wide association (GWAS), where associations between polymorphisms and the disease state are tested. Prior studies indicate that for complex traits, such as DKD, genetic polymorphisms that are associated with disease state are localized to the non-coding region of the genome12,13. Moreover, the genetic architecture of diabetic kidney disease has not been characterized and several large collaborations are currently addressing this issue14. Thus, the next challenge is to define target genes, target cell types and the mode of dysregulation caused by non-coding snips (SNPs15). Such studies require large collection of human tissue samples from disease relevant organs.

Diabetic kidney disease (DKD) remains a clinical diagnosis. Subjects with CKD in the presence of diabetes and albuminuria are considered to have diabetic nephropathy. Such definition is used in clinical practice and in research studies including clinical trials. Studies performed in 1980 provide the basis for the practice16,17. Investigators stage DKD as a progressive disease, beginning with the loss of small amounts of albumin into the urine (30-300mg/day; known as the stage of microalbuminuria, high albuminuria, occult or incipient nephropathy), then larger amounts (>300mg/day; known as macroalbuminuria, very high albuminuria or overt nephropathy), followed by progressive decline in kidney function (eGFR), renal impairment and ultimately ESRD 17-19. This paradigm has proved useful in clinical studies, especially in type 1 diabetes, for identifying cohorts at increased risk of adverse health outcomes. However, boundaries between stages of DKD are artificial and the relationship between urinary albumin excretion and adverse health outcomes is log-linear in clinical practice. Indeed, the American Diabetes Association recently abandoned staging of albuminuria (ACR) for a more-straightforward [ACR >30 mg/g, (albuminuria present); ACR <30 mg/g (albuminuria absent)] criterion. Moreover, many patients, and especially those with type 2 diabetes, do not follow this classical course in modern clinical practice. For example, many subjects with DKD do not manifest excessive urinary albumin loss20. Indeed, of the 28% of the UKPDS cohort who developed moderate to severe renal impairment, half did not have preceding albuminuria. In the Diabetes Control and Complications Trial (DCCT), of the 11% patients with type 1 diabetes who developed an eGFR<60 ml/min/1.73m2, 40% never had experienced overt nephropathy21. In addition, most patients with microalbuminuria do not progressively exhibit an increase in urinary albumin excretion as in the classical paradigm with treatment-induced and spontaneous 'remission' of albuminuria widely observed22,23. Consequently, individuals with microalbuminuria may better be regarded as being at increased risk of developing progressive renal disease (as well as cardiovascular disease and other diabetic complications), rather than as actually having DKD per se. While over the last 40 years it became evident that the original description of DKD needs revision, no alternative criteria have emerged given the lack of solid data on the correlation between histopathological (gold standard) DKD diagnosis and clinical manifestations. It is also possible that, with the introduction of better glycemic control and anti-renin (RAAS) blockade, the disease has evolved necessitating new observational cohorts to understand the clinical disease course and manifestations.

Diabetic kidney disease presents with a variable rate of kidney function decline24. Data from large observational cohorts indicate that GFR decline frequently does not follow a linear course. Several groups are working on modeling GFR decline patterns in patients. Such studies contributed to emphasizing patients termed as "rapid progressors". There is no consensus definition for rapid progression. Many studies define rapid progressors as patients with greater than 3 cc/year GFR decrease but alternative cut points such as even 10 cc/year has also been used. Identification and clinical characterization of rapid progressors became the center of several large scale efforts as these are the patients who would likely need intensive clinical management25. Furthermore recent post-hoc analyses of the Diabetic Nephropathy (IDNT and RENAAL) studies indicate that clinical trial outcomes are mostly driven by a small number of subjects with unusually rapidly progressive GFR decline i.e. subjects that display characteristics of rapid progressors. While investigators are still awaiting accurate descriptions, biomarker and clinical descriptive studies have yielded several interesting observations. Albuminuria remains one of the strongest risk factor for "FDA-approved" (hard) renal outcomes; doubling of serum creatinine, dialysis or death. Indeed some of the latest studies indicate that using a 4 or a 6 variable model, that includes albuminuria, age, sex, serum phosphate, serum calcium and serum albumin has C-statistics score of 0.84-0.91 to predict ESRD 26,27. During the last years several new biomarkers have been identified that can potentially identify patients who are at increased risk for rapid loss of kidney function. For example blood and urinary levels of kidney injury molecule (KIM1) shows promise to identify patients who are at risk for kidney function decline. Recently, investigators showed that circulating levels of tumor necrosis factor receptor 1 and 2 levels can identify patients with rapidly declining renal function 28. While these markers are generating increased interest; the critical questions remains; why do some patients follow a rapid decline in kidney function?