Serum biomarkers in diabetic kidney disease: A comprehensive narrative review

Abstract

Diabetic kidney disease (DKD) is the leading cause of end-stage renal disease worldwide. Traditional biomarkers such as serum creatinine and estimated glomerular filtration rate lack sensitivity for early kidney injury detection, limiting timely intervention. This review comprehensively evaluates serum biomarkers for DKD, focusing on their diagnostic and prognostic utility, mechanistic insights, and clinical applicability. Novel serum biomarkers, including cystatin C, neutrophil gelatinase-associated lipocalin, kidney injury molecule-1, and tumor necrosis factor receptors (TNFRs) (TNFR1, TNFR2), demonstrate superior sensitivity for early renal damage and progression prediction. Emerging markers, such as adropin, soluble urokinase plasminogen activator receptor, monocyte chemoattractant protein-1, chitinase-3-like protein 1, endostatin, zinc-alpha-2-glycoprotein, PromarkerD, and circulating microRNAs offer additional risk stratification potential. Integrating validated serum biomarkers may enhance early DKD detection, enable precise risk stratification, and support personalized therapeutic strategies to improve patient outcomes.

Key Words: Diabetic nephropathy; Chronic kidney disease; Albuminuria; Renal function; Inflammation markers; Diabetes; Biomarkers

Core Tip: Diabetic kidney disease is the main cause of end-stage renal disease, yet early detection is limited by the poor sensitivity of serum creatinine and estimated glomerular filtration rate. Novel biomarkers such as cystatin C, neutrophil gelatinase-associated lipocalin, kidney injury molecule-1, and tumor necrosis factor receptors show superior accuracy for identifying early damage and predicting progression. Emerging markers, including adropin, soluble urokinase plasminogen activator receptor, monocyte chemoattractant protein-1, chitinase-3-like protein 1, endostatin, zinc-α2-glycoprotein, PromarkerD, and microRNAs, provide additional value for risk stratification and personalized therapy. Integration of validated multi-marker panels may transform monitoring and management, improving outcomes for patients with diabetes.

INTRODUCTION

Diabetic kidney disease (DKD) is a serious complication of diabetes and is now the leading cause of end-stage renal disease (ESRD) worldwide, imposing a significant burden on both patients and healthcare systems[1]. Although advances in glycemic control and the introduction of novel pharmacologic therapies have significantly improved the management of DKD, early diagnosis and accurate disease monitoring continue to represent major unmet clinical needs. Conventional biomarkers, including serum creatinine (Cr) and estimated glomerular filtration rate (eGFR), are widely employed to evaluate renal function decline. Nevertheless, these parameters exhibit limited sensitivity and specificity in the early stages of DKD, as substantial nephron loss may occur before measurable alterations become evident due to renal compensatory mechanisms. Within this framework, microalbuminuria (30-300 mg/day) remains the cornerstone biomarker of early glomerular injury in DKD, serving not only as an indicator of incipient nephropathy but also as a robust predictor of progression to overt renal disease and associated cardiovascular events[2,3]. Moreover, factors unrelated to kidney function, including age, muscle mass, diet, and ethnicity, can affect these biomarkers, highlighting the urgent need for more specific serum indicators. Recent research has expanded the search for biomarkers that reflect the complex pathophysiology of DKD, including markers of glomerular and tubular injury, inflammation, oxidative stress, hemodynamic changes, fibrosis, and immune activation[4]. However, the clinical adoption of these novel biomarkers is limited by the need for large-scale, prospective validation and standardized assay methods. Integrating panels that combine traditional and novel biomarkers may ultimately enable earlier diagnosis, more accurate prognosis, and individualized therapeutic strategies for patients with diabetes at risk of kidney disease. Given the large number of biomarkers reported in the literature and the high prevalence of DKD, a comprehensive review is warranted. Therefore, this article focuses on serum biomarkers for both practical and scientific reasons.

FRAMEWORK FOR SERUM BIOMARKER CLASSIFICATION IN DKD

Based on the complex and multifactorial pathophysiology of DKD, serum biomarkers can be broadly classified into several categories reflecting distinct but interrelated mechanisms of injury. These include markers of glycemic control, such as plasma glucose, glycosylated hemoglobin (HbA1c), fructosamine, and glycated albumin, which reflect chronic and intermediate-term hyperglycemia and its metabolic consequences[4,5]. Inflammatory and immune activation biomarkers encompass cytokines, chemokines, and their soluble receptors, including tumor necrosis factor-α (TNF-α), TNF receptors (TNFRs) (TNFR1 and TNFR2), interleukins (ILs), monocyte chemoattractant protein-1 (MCP-1), and chitinase-3-like protein 1 (YKL-40), mirroring the inflammatory milieu driving renal injury[5,6]. Biomarkers of hemodynamic and endothelial alterations, such as midregional (MR) proadrenomedullin (proADM), natriuretic peptides, and soluble urokinase plasminogen activator receptor (suPAR), reflect vascular dysfunction and intraglomerular pressure changes[4,6]. Oxidative stress and mitochondrial dysfunction are captured by markers such as 8-hydroxydeoxyguanosine (8-OHdG) and growth differentiation factor-15 (GDF-15), indicating reactive oxygen species (ROS)-mediated damage. Finally, markers of glomerular and tubular injury, including cystatin C, neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), beta 2-microglobulin (β2-MG), uromodulin, and heparanase (HPSE), provide direct insight into structural and functional renal damage in DKD[4-6].

METABOLIC DYSREGULATION BIOMARKERS

Metabolic dysregulation in diabetes involves more than just fluctuations in blood glucose; it arises from a complex interplay of biochemical changes, including altered insulin signaling, increased protein glycosylation, disrupted fatty acid oxidation, and changes in branched-chain amino acid metabolism. These processes collectively promote chronic inflammation and cellular stress[7]. As a result, there is increased ROS production, endothelial dysfunction, and profibrotic pathway activation, all contributing to glomerular and tubulointerstitial injury in diabetes[8]. Because these metabolic disturbances occur early, intermediates from these pathways may serve as biomarkers for detecting subclinical kidney damage, even before diabetes is fully established, and may also help predict the rate of disease progression and loss of renal function.

HbA1c

HbA1c reflects protein glycosylation and provides an estimate of average blood glucose levels over the past 2-3 months, serving as an indicator of medium-term exposure to hyperglycemia and its complications. Recent research has increasingly focused on the impact of long-term fluctuations in glycemic levels, referred to as HbA1c variability, on both microvascular and macrovascular damage. HbA1c variability, typically measured as the standard deviation (SD) of serial HbA1c values over time, is now being considered for routine management protocols to improve risk stratification and potentially reduce disease progression, especially in patients at higher risk for diabetes-related complications. Zhu et al[9] recently conducted a post-hoc analysis from the CREDENCE trial[10], demonstrating the independent association of intraindividual HbA1c variability with an increased risk of renal outcomes, namely serum Cr doubling to > 200 μmol/L, ESRD, or death due to kidney disease, in 3080 subjects with type 2 diabetes mellitus (T2DM) and DKD. A subsequent comprehensive meta-analysis of 18 studies dated 2012 to 2023 performed by Wang et al[11] revealed HbA1c variability to positively correlate with deteriorating kidney function in individuals with T2DM for high vs low variability groups, as measured by SD. Analysis of additional variability indicators, such as HbA1c coefficient of variation, HbA1c variability score, and HbA1c index produced comparable data. Meanwhile, retrospective research from Muthukumar et al[12] engaged an ethnically heterogeneous cohort of 3466 patients with type 1 diabetes mellitus (T1DM), for which HbA1c variability predicted kidney disease progression independently of traditional risk factors. Analogous results from earlier research had documented the predictive power of HbA1c variability in terms of new-onset microalbuminuria (e.g., prospective DMIDS project[13], already included in the above-mentioned meta-analysis by Wang et al[11]) and progression of renal involvement (FinnDiane Study[14]) in both T1DM and T2DM.

Overall, evidence from large retrospective, cross-sectional, and prospective cohort studies supports HbA1c as an independent marker associated with early renal injury and increased risk of DKD across different age groups and diabetes types[15-17].

Nevertheless, while observational and cohort studies highlight a continuous association between HbA1c and renal risk, evidence from interventional trials indicates that fixed glycemic thresholds, especially below the diagnostic cutoff for diabetes, may not reliably predict the onset of DKD.

In this regard, a post-hoc analysis of the ADVANCE trial[18], which involved a large cohort of 11970 patients with T2DM, underscored the absence of any significant association (P > 0.40) between mean HbA1c levels below the 6.5% threshold and the risk of microvascular outcomes, including new or worsening nephropathy. Additional research is needed to determine clear diagnostic cut-offs for glycemic profiles that signal potential renal involvement during early diabetic changes.

Advanced glycation end products

Advanced glycation end products (AGEs) are a heterogeneous group of compounds formed by the non-enzymatic glycation of proteins, lipids, and nucleic acids. Damage related to AGEs occurs directly through tissue accumulation and disruption of protein function, and indirectly via interactions with specific membrane and soluble receptors. These interactions promote inflammatory pathways, increase ROS generation, and induce apoptosis[19]. The role of AGEs as biomarkers of diabetic outcomes has been thoroughly investigated by a multitude of studies. In terms of DKD, a large-scale case-cohort study by Thomas et al[20] was performed in 3763 participants with T2DM from the ADVANCE trial[21]. Circulating AGE levels were independently associated with incipient or progressing DKD, with notably improved accuracy in prediction of 5-year risk of new or worsening nephropathy (net reclassification index in continuous model = 0.24). Similarly, in a prospective cohort of 1150 participants with T2DM from Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial[22] the overall AGE score, defined as the mean of standardized individual AGEs, was positively correlated with reduced eGFR, 30% renal function, 40% renal function loss, macroalbuminuria, and high-risk CKD. Comparable findings were more recently documented by Ding et al[23], who found that higher circulating AGEs positively associate with urinary albumin, urinary albumin-to-Cr ratio (UACR), and blood urea nitrogen in a study of 176 individuals with T2DM, with multivariate logistic regression showing elevated AGE levels as an independent risk factor for DKD progression. Overall, available research suggests a pivotal role of AGEs in the onset of renal damage and deterioration of kidney function in diabetic patients, making them a potential target for the prevention and management of DKD.

Serum adropin

Serum adropin is a newly identified regulatory protein involved in energy homeostasis and lipid metabolism. Chen et al[24] reported significantly higher adropin levels in T2DM subjects compared to healthy controls, with even higher levels in patients with advanced renal impairment (RI) and in DKD progressors vs non-progressors. In their study, adropin predicted DKD with 86% sensitivity and 70% specificity at a cut-off value of 6872.24 pg/mL, even after adjusting for age, gender, and body mass index. However, these findings contrast with previous research by Hu and Chen[25], who reported lower serum adropin concentrations in T2DM patients with DKD compared to those without renal involvement (P < 0.001), indicating an inverse correlation between adropin concentration and DKD development, independent of age and gender. Similarly, Es-Haghi et al[26] observed lower serum adropin levels in patients with DKD compared to both healthy subjects and diabetic patients without nephropathy; at a cut-off value of 3.20 mg/dL, adropin showed 80% sensitivity and 60% specificity for DKD detection, with an area under the curve (AUC) of 0.830. These conflicting results highlight the need for further research to clarify whether adropin has a protective or contributory role in DKD.

Zinc-α2-glycoprotein

In a study by Xu et al[27], higher serum concentrations of zinc-α2-glycoprotein (ZAG), a 43 kDa adipocytokine that acts as a lipid-mobilizing factor, significantly associated with mildly decreased eGFR (< 90 mL/minute/1.73 m²) in a cohort of 438 T2DM patients, especially in females with higher UACR (≥ 2.7 mg/mmol) and bigger waist circumference (≥ 85 cm). Similar findings were confirmed by Elsheikh et al[28], who identified serum ZAG as a reliable predictor of albuminuria in early DKD, with a sensitivity and specificity of 97% and 72.7%, respectively, providing an overall accuracy of 90.9% at a cut-off value of ≥ 22.5. More recently, Sonkar et al[29] delivered further promising data, assessing ZAG as a novel biomarker for both early DKD and progressing disease in 160 individuals with T2DM; serum concentrations positively correlated with UACR and negatively correlated with eGFR.

Apolipoprotein A-IV, apolipoprotein C-III, and Insulin-like growth factor-binding protein 3

Peters et al[30] conducted a study to validate a novel biomarker panel, including apolipoprotein A-IV (apoA4), cluster of differentiation (CD) 5 antigen-like (CD5 L), complement C1q subcomponent subunit B (C1QB), and insulin-like growth factor-binding protein 3 (IGFBP3), as a predictive model for rapidly declining eGFR (defined as incident CKD, eGFR decline ≥ 30%, or annual decline in eGFR ≥ 5 mL/minute/1.73 m²) in DKD. Adding these four biomarkers to the clinical model improved discrimination, sensitivity, specificity, and risk classification. Similarly, a consensus model comprising apoA4, CD5 L, IGFBP3, age, serum high density lipoprotein-cholesterol, and eGFR demonstrated high accuracy in predicting incident DKD; at the optimal score cut-off, this model achieved 86% sensitivity, 78% specificity, 30% positive predictive value, and 98% negative predictive value for the 4-year risk of developing DKD in an independent cohort of individuals with T2DM over a 4-year follow-up period[31]. The same panel was also recently tested by Davis et al[32] in 91 T1DM patients, demonstrating strong predictive power and clinical utility in identifying people with T1DM at risk of future adverse renal outcomes.

Retinol-binding protein 4

Retinol-binding protein 4 (RBP4) is an adipokine produced by the liver and adipose tissue that plays a key role in insulin resistance. An extensive meta-analysis by Cao et al[33] evaluated the diagnostic potential of RBP4 for early detection of DKD across 29 studies (2012-2024) involving 5549 patients with T2DM. The analysis demonstrated promising diagnostic accuracy, with a pooled sensitivity of 0.76 and specificity of 0.81. The area under the summary receiver operating characteristic (ROC) curve was 0.85, indicating good overall diagnostic performance. The diagnostic odds ratio was 13.76, further supporting the utility of RBP4 as a diagnostic biomarker. Elevated RBP4 levels in DKD are attributed to two primary mechanisms: First, impaired renal catabolism, as the kidneys normally regulate retinol homeostasis through glomerular filtration and proximal tubular reabsorption of RBP4, leading to plasma accumulation when renal function declines; and second, worsening insulin resistance. Notably, RBP4 demonstrated higher specificity than traditional ACR for early DKD prediction, suggesting its potential as a complementary biomarker.

GLOMERULAR AND TUBULAR INJURY BIOMARKERS

DKD is a leading cause of ESRD globally, particularly among individuals with T2DM. The progression of DKD is often insidious, with early stages being asymptomatic. Therefore, early detection and monitoring are crucial to prevent irreversible kidney damage. While traditional markers like serum Cr and eGFR have been employed to assess renal function, their limitations in detecting early RI necessitate the exploration of additional serum and urinary biomarkers.

Serum Cr

Creatine is synthesized primarily in the kidneys and liver from the amino acids arginine, glycine, and methionine, and is subsequently transported to skeletal muscle, where it is phosphorylated to phosphocreatine, functioning as a rapid energy buffer for adenosine triphosphate resynthesis. Phosphocreatine undergoes spontaneous, non-enzymatic cyclization at a relatively constant rate, resulting in the formation of Cr, which is released into the circulation and cleared by the kidneys and, to a lesser extent, by metabolism in the gut microbiota[34]. Serum Cr level is the most used clinical marker for assessing kidney function in DKD patients because of its inverse relationship with eGFR, providing an indirect but accessible measure of renal function[35].

Notably, Lee et al[36] demonstrated that serum Cr levels were significantly higher in patients with T2DM compared to non-diabetic controls, correlating with reduced eGFR and indicating impaired renal function.

However, serum Cr has significant limitations. Barr et al[37] highlighted that cystatin C-based eGFR equations may provide a more accurate reflection of actual GFR in selected populations compared to the traditional Cr-based formula. As a biomarker for early diagnosis and prognosis of DKD, it typically tends to increase only after substantial nephron loss has occurred, thus reducing its sensitivity for detecting early-stage kidney injury[38]. Moreover, serum Cr levels are affected by extrarenal factors such as advanced age, physical activity, protein-rich diets, male sex, medications, and ethnicity, all of which reduce its specificity and reliability as a standalone marker[39].

Even though the measurement of both serum Cr and UACR has been shown to improve the early diagnosis of kidney damage, a significant proportion of DKD patients show a non-albuminuric phenotype, which further limits diagnostic accuracy. These considerations underscore the clinical need for more precise indicators of kidney dysfunction[40].

Cystatin C

Cystatin C is a low molecular weight (13 kDa) cysteine protease inhibitor produced by all nucleated cells at a constant rate and freely filtered by the glomerulus. Unlike Cr, its serum concentration is less influenced by muscle mass, age, or diet, making it a more reliable endogenous marker for eGFR[41]. Serum cystatin C levels tend to rise earlier than Cr in the course of RI, which supports its use as a sensitive biomarker for the early detection of CKD, including DKD[42]. In DKD, cystatin C has shown promise across both pediatric and adult populations.

Trutin et al[43] investigated pediatric patients with T1DM and found that cystatin C, together with renal resistance index and KIM-1, effectively predicts early DKD in subclinical RI. Another study by Salem et al[44] broadened this evidence by evaluating children with T1DM, where serum cystatin C and urinary cyclophilin A, an intracellular protein secreted in response to inflammatory stimuli, were reliable markers of early renal dysfunction and offered a non-invasive approach to monitoring DKD.

Complementing these pediatric findings, Kang et al[45] conducted a cross-sectional analysis in adults with T2DM, revealing that serum cystatin C levels and transforming growth factor-beta 1 possess substantial diagnostic value, linking its elevation to progressive renal damage, proteinuria and fibrosis, thus supporting its integration into routine clinical assessment. These conclusions are shared with meta-analyses in adults with T2DM that highlight the superior sensitivity and specificity of serum cystatin C over Cr for early DKD diagnosis[46].

Wang et al[47] conducted a longitudinal study investigating the trajectory of serum cystatin C levels in patients with DKD. Their analysis revealed that dynamic changes in cystatin C concentrations over time serve as a sensitive marker of progression and that its temporal patterns can predict the rate of kidney function decline. This finding further underscores the potential of serial cystatin C measurements as a prognostic biomarker that may capture renal damage earlier than traditional markers. Moreover, Pan et al[48] compared eGFR equations that incorporate both serum Cr and cystatin C in a cohort of patients with T2DM and DKD and found that combined Cr-cystatin C equations provide superior accuracy in estimating eGFR compared to Cr-based formulas alone, emphasizing cystatin C-based equations in long-term renal outcomes.

Genetic studies support a causal relationship between elevated cystatin C and DKD susceptibility. Mendelian randomization analyses suggest that higher genetically determined cystatin C levels are linked to increased DKD risk, indicating that cystatin C may play a role beyond a simple filtration marker[49]. Taha et al[50] evaluated the diagnostic efficacy of serum cystatin C in conjunction with angiotensin-converting enzyme (ACE) gene polymorphisms in patients with T2DM. Their results demonstrated that elevated cystatin C levels were significantly associated with early renal dysfunction and that specific ACE gene variants, notably the I/D polymorphism, modulated the predictive value of cystatin C. This genetic interaction suggests a precision-medicine approach to nephropathy risk assessment, wherein cystatin C serves not only as a functional biomarker but also as a tool for identifying genetically predisposed individuals.

In a large cohort of Ethiopian patients with T2DM, elevated cystatin C levels and dyslipidemia were independently associated with DKD, and their combined assessment improved the overall predictive performance for early kidney damage[51].

Additionally, cystatin C could be considered a potential biomarker of microvascular damage. Zhao et al[52] conducted a cross-sectional study to investigate the association between serum cystatin C levels and the presence of diabetic foot ulceration (DFU) in patients with T2DM. They found that serum cystatin C concentrations were significantly higher in patients with DFU. Notably, elevated cystatin C levels were independently associated with the presence of DFU, suggesting that cystatin C may reflect systemic microvascular damage and endothelial dysfunction, potentially serving as a broader biomarker of vascular complications in diabetes.

NGAL

NGAL is a protein expressed by renal tubular epithelial cells in response to various forms of injury to mitigate oxidative stress and promote epithelial repair and regeneration[4]. Serum NGAL has emerged as a promising tubular biomarker for early detection and monitoring of DKD, especially because unlike traditional markers such as UACR, NGAL levels can be elevated prior to the onset of albuminuria, offering a potential advantage in early diagnosis. Several studies have demonstrated that serum NGAL levels are significantly higher in patients with DKD compared with healthy controls.

He et al[4] performed a systematic review and meta-analysis to evaluate the utility of NGAL in diagnosing DKD. They analyzed 15 studies from various databases and included both serum and urinary NGAL measurements. The results showed that patients with DKD had significantly higher levels of NGAL compared to healthy individuals. In particular, the average NGAL level in DKD patients was approximately 168 ng/mL, while it was approximately 75 ng/mL in healthy controls. These findings suggest that NGAL may help detect kidney damage early in diabetic patients. However, the study also highlighted that there is no standardized cutoff for NGAL levels and different labs use different testing methods, making it difficult to compare results across clinics. In a more recent meta-analysis, Prashant et al[53] also focused on the role of NGAL in DKD. Consistent with He et al[4], the authors demonstrated that NGAL levels were significantly higher in patients with DKD than in healthy controls, with mean serum NGAL levels of 168.08 ng/mL and similarly elevated urinary NGAL levels. The study emphasized NGAL’s potential as a non-invasive biomarker for early detection and monitoring of DKD, highlighting its association with clinical parameters such as glycemic control, renal function, and albuminuria.

Moreover, Bacci et al[54] conducted a study to evaluate the diagnostic performance of NGAL and cystatin C compared with traditional markers of renal dysfunction in patients with T2DM. The results indicated that both serum NGAL and cystatin C levels were positively associated with RI; however, ACR demonstrated a slightly higher diagnostic performance compared to NGAL and cystatin C, suggesting that a combination of biomarkers may be more effective in detecting early DKD.

Another correlation study that examined serum NGAL and insulin-like growth factor binding protein 4 (IGFBP4) in the early detection of DKD showed that both biomarkers significantly increased in patients with kidney involvement. In addition, even in this case, NGAL and IGFBP4 positively correlated with serum Cr and negatively correlated with eGFR[55].

HPSE

Serum heparanase (HPSE), an endo-β-D-glucuronidase enzyme capable of degrading heparan sulfate (HS) chains within the glomerular basement membrane (GBM) and extracellular matrix, has recently emerged as a key player in DKD pathogenesis and a potential biomarker for early renal damage[56,57]. Its enzymatic activity leads to the cleavage of HS chains, critical components of the GBM, which contribute to its charge-selective filtration properties. HS degradation by HPSE results in loss of anionic charge, thereby increasing glomerular permeability and promoting proteinuria[58,59].

Furthermore, HPSE has been shown to induce differential loss of distinct HS domains, underscoring its molecular specificity of action on GBM structure[60,61].

In experimental models, genetic ablation or pharmacological inhibition of HPSE attenuates DKD development, reinforcing its pivotal role in disease progression. HPSE knockout mice are protected from diabetes-induced albuminuria and glomerular damage, establishing HPSE as essential in DKD pathogenesis[62]. Moreover, HPSE modulates the expression and activity of transforming growth factor-β, a key profibrotic cytokine, thereby promoting renal fibrosis[63].

HPSE expression is regulated through multiple signaling pathways in DKD. An et al[64] found that AGEs activate HPSE expression in podocytes via the receptor for AGEs (RAGE) and nuclear factor kappa B (NF-κB) signaling, linking hyperglycemia-induced metabolic stress to HPSE upregulation. Recent studies have expanded on this, demonstrating that endothelial nitric oxide synthase deficiency causes podocyte injury by activating nuclear factor of activated T cell (NFAT) 2, which subsequently increases HPSE expression, illustrating a nexus between endothelial dysfunction and HPSE-mediated glomerular injury[65]. Additionally, Xu et al[66] showed that AGE-stimulated macrophages secrete pro-inflammatory mediators that induce HPSE expression in glomerular endothelial cells, disrupting their function and contributing to local inflammation and barrier dysfunction. Moreover, Maxhimer et al[67] demonstrated that high glucose alone upregulates HPSE transcription in renal epithelial cells, highlighting a direct glucose-dependent mechanism in HPSE upregulation.

HPSE also triggers an inflammatory cascade critical for DKD progression. Goldberg et al[68] highlighted that HPSE enzymatic activity facilitates the release of HS-bound cytokines and chemokines, promoting macrophage infiltration and sustained renal inflammation. This inflammatory milieu exacerbates glomerular and tubular damage, further advancing renal decline. Moreover, Chang et al[69] described a role for HPSE in mediating endothelial-to-mesenchymal transition via extracellular signal-regulated kinase signaling in diabetic glomerular endothelial cells, linking HPSE to renal fibrosis through cellular trans-differentiation processes. This contributes to extracellular matrix expansion and progressive sclerosis characteristic of DKD. Supporting this fibrotic mechanism, Masola et al[70] reported that HPSE also enhances transforming growth factor-beta signaling, directly contributing to the fibrogenic process, and sulodexide has been shown to counteract this effect by inhibiting HPSE and preventing fibroblast growth factor (FGF) 2-induced epithelial-to-mesenchymal transition.

Several clinical studies support the utility of serum HPSE as a possible biomarker of DKD. Shafat et al[71] documented elevated HPSE levels in the plasma and urine of T2DM patients, with levels correlating with blood glucose concentrations, indicating that HPSE reflects metabolic control and renal injury. Similarly, Zhao et al[72] found that plasma HPSE correlates with blood glucose but not with early microalbuminuria, suggesting that HPSE might detect renal injury earlier than traditional markers. The early increase of circulating HPSE may mirror subclinical renal HS degradation and basement membrane alterations preceding overt proteinuria. Thus, HPSE levels in serum holds promise as a non-invasive biomarker to predict the onset and progression of DKD, enabling earlier therapeutic intervention. This is further supported by data showing that HPSE levels mirror alterations in glomerular endothelial glycocalyx. Garsen et al[73] linked HPSE activity to endothelial glycocalyx degradation, a key driver of albumin leakage. In experimental models, glycocalyx injury paralleled increases in HPSE expression, suggesting that enzymatic cleavage of HS proteoglycans compromises the endothelial barrier and contributes to albumin leakage. This insight supports the idea that circulating HPSE levels may reflect not only glomerular injury but also systemic endothelial injury in DKD. In follow-up studies, endothelin-1 was shown to induce HPSE expression and subsequent proteinuria by damaging the glomerular glycocalyx[74].

Given HPSE’s multifaceted role in DKD pathogenesis, targeting HPSE offers a promising therapeutic avenue. Buijsers et al[75] recently showed that HPSE-2, a naturally occurring HPSE isoform, exerts protective effects against experimental DKD by antagonizing HPSE activity and attenuating renal injury. These findings suggest that HPSE inhibitors or HPSE-2 mimetics could be developed to slow or halt DKD progression. Gamez et al[76] further demonstrated that systemic HPSE inhibition protects the endothelial glycocalyx and prevents diabetic microvascular complications, suggesting that HPSE-targeted therapies could have systemic benefits beyond the kidney.

Furthermore, modulating upstream signaling pathways regulating HPSE expression, such as the RAGE/NF-κB and NFAT2 pathways, may provide additional therapeutic benefits. Integrating serum HPSE measurement with clinical parameters could also refine patient stratification and treatment monitoring in DKD.

Serum uromodulin

Uromodulin, also known as Tamm-Horsfall protein, is the most abundant protein excreted in normal human urine. Synthesized exclusively by the epithelial cells of the thick ascending limb of Henle’s loop and the early distal convoluted tubule, uromodulin plays a pivotal role in maintaining renal homeostasis. It contributes to the regulation of sodium and potassium balance, modulates urinary calcium excretion, and exhibits anti-inflammatory properties. Several studies have investigated the association between serum uromodulin levels and DKD, yielding insights into its potential as a biomarker for early detection and prognosis. A systematic review and meta-analysis showed that serum uromodulin levels were significantly decreased in DKD patients[77].

Serum uromodulin also holds particular promise in pediatric patients, where early detection is crucial. Studies in children and adolescents with type 1 diabetes report lower serum uromodulin levels even before overt microalbuminuria manifests[78,79], suggesting its role as an early biomarker for tubular injury. Beyond correlation, studies propose that serum uromodulin reduction reflects tubulointerstitial damage, which is increasingly recognized as a key mediator of DKD progression. This damage leads to diminished synthesis and secretion of uromodulin into the bloodstream, providing a biologically plausible explanation for the observed biomarker trends[80]. In DKD, chronic hyperglycemia induces glomerular damage characterized by mesangial expansion, basement membrane thickening, and podocyte loss, which eventually compromises the tubulointerstitial compartment. Tubular injury is increasingly recognized as a critical driver of DKD progression, with tubular epithelial cell dysfunction and loss leading to decreased uromodulin synthesis[81]. Additionally, hyperglycemia and oxidative stress activate pro-inflammatory and profibrotic signaling pathways, such as transforming growth factor-beta and NF-κB, resulting in tubular cell apoptosis and decreased uromodulin production[82]. Moreover, early DKD stages can exhibit increased urinary excretion of uromodulin due to altered tubular permeability or increased uromodulin shedding from injured tubular cells. This augmented urinary loss may contribute to the depletion of circulating serum uromodulin[83]. Conversely, as tubular mass diminishes with disease progression, uromodulin synthesis declines, leading to persistently low serum concentrations.

Overall, the decline in serum uromodulin in DKD reflects a combination of reduced tubular cell mass, impaired synthesis due to metabolic and inflammatory stress, increased urinary loss, and genetic predisposition. These mechanisms underscore the utility of serum uromodulin as an indicator of tubular health and early tubular injury in DKD[77]. The growing body of evidence supporting serum uromodulin as a biomarker for DKD has important clinical implications that could transform current diagnostic and prognostic paradigms. By reflecting tubular health, serum uromodulin complements glomerular markers and may enable earlier intervention, potentially improving patient outcomes. Studies have demonstrated that higher baseline serum uromodulin levels predict lower odds of both coronary artery calcification progression and incident DKD over 12 years in adults with T1DM. Higher baseline serum uromodulin levels were associated with lower odds of coronary artery calcification progression, incident elevated albumin excretion, rapid eGFR decline, and impaired GFR. These associations remained significant after adjusting for traditional cardiovascular risk factors, underscoring the potential of serum uromodulin as a predictive biomarker for both renal and cardiovascular outcomes in the diabetic population[84].

Clinically, serum uromodulin measurement may offer a minimally invasive, cost-effective tool for identifying early tubular damage before overt albuminuria develops and for monitoring disease progression and response to therapy in diabetic patients. Its utility spans from identifying early tubular damage before overt albuminuria develops to monitoring disease progression and response to therapy. Despite these promising prospects, several challenges must be addressed before widespread clinical implementation. Standardization of assay methods and establishment of reference ranges specific to diabetic populations are essential to ensure reproducibility and interpretability. Moreover, longitudinal studies across diverse cohorts, including both type 1 and type 2 diabetes, are needed to validate serum uromodulin’s predictive power and define clinically relevant thresholds. Furthermore, elucidation of the molecular pathways linking uromodulin expression and secretion to diabetic tubular pathology may unveil novel therapeutic targets. Genetic studies implicating UMOD variants in DKD susceptibility underscore the potential for personalized medicine approaches tailored to individual genetic risk profiles, with evidence showing significant heterogeneity in UMOD variant effects across different population groups[85]. In conclusion, serum uromodulin represents a compelling biomarker candidate with the potential to improve DKD diagnosis and management. Realizing its full clinical utility will require concerted efforts in methodological standardization, large-scale validation studies, and translational research exploring its biological functions and therapeutic implications.

Serum beta-2-microglobulin

Beta 2-microglobulin (β2-MG), a low molecular weight protein that forms the light chain of the major histocompatibility complex class I (MHC-I) molecules, and expressed ubiquitously on nucleated cell membranes, is continuously shed into circulation and primarily eliminated by glomerular filtration, followed by proximal tubular reabsorption and catabolism. Consequently, increased serum and urinary β2-MG levels reflect impaired renal filtration and tubular dysfunction, respectively, both of which are pathophysiological hallmarks of DKD[86]. This phenomenon is primarily mediated by immune cells, including activated lymphocytes, monocytes, and macrophages, which upregulate MHC class I molecule synthesis and consequently release increased amounts of β2-MG into circulation[87]. In chronic inflammatory diseases, including diabetes mellitus and DKD, persistent immune activation leads to sustained β2-MG production[88]. Additionally, proinflammatory cytokines such as TNF-α, IL-1, and interferon-gamma have been shown to stimulate β2-MG gene expression at the transcriptional level in various cell types, including renal tubular epithelial cells[89,90]. This cytokine-driven upregulation suggests that elevated β2-MG levels may reflect an inflammatory milieu intrinsic to DKD pathogenesis beyond renal clearance impairment.

Several clinical studies have investigated the diagnostic and prognostic value of serum β2-MG in patients with DKD, providing compelling evidence for its role as a biomarker in different populations and disease stages. Chen and Li[91] conducted a clinical study focusing on early detection of DKD by measuring serum levels of β2-MG and cystatin C in patients with T2DM. They found that β2-MG was significantly elevated in diabetic patients with microalbuminuria compared to those without nephropathy and healthy controls, highlighting its sensitivity for early RI. This study also emphasized that β2-MG had a higher diagnostic accuracy than serum Cr, suggesting it could serve as a non-invasive marker for identifying incipient nephropathy. In a cross-sectional analysis, Yang et al[92] evaluated serum β2-MG alongside HbA1c and vascular endothelial growth factor (VEGF) in patients with varying stages of DKD. They demonstrated a strong positive correlation between β2-MG levels and nephropathy severity, as assessed by albuminuria and decreased eGFR. Moreover, β2-MG levels were associated with elevated VEGF, a known mediator of diabetic microvascular complications, proposing a link between β2-MG and endothelial dysfunction in DKD pathogenesis.

A significant contribution came from Uemura et al[93], who performed a longitudinal study on patients with biopsy-confirmed DKD. Their findings revealed that baseline serum β2-MG was a robust independent predictor of adverse renal outcomes, including progression to ESRD and doubling of serum Cr. Notably, β2-MG outperformed traditional markers such as serum Cr and proteinuria in prognostic stratification. This study highlighted the potential of β2-MG for risk assessment in advanced DKD, supporting its use in clinical practice for individualized patient management.

In a comprehensive meta-analysis, Gholaminejad et al[94] synthesized data from multiple prospective cohorts assessing circulating β2-MG and α1-microglobulin in diabetic patients. Their analysis confirmed that elevated β2-MG was consistently associated with faster progression of nephropathy and increased risk of renal function decline. The meta-analysis also highlighted the biomarker’s value in both type 1 and type 2 diabetes populations and across different ethnicities, reinforcing its generalizability.

The earlier study by Aksun et al[95] compared serum β2-MG and cystatin C in T2DM patients with and without renal involvement. They reported that β2-MG was significantly higher in diabetic patients with DKD and inversely correlated with Cr clearance, suggesting early tubular involvement before overt glomerular damage. Their work advocated the combined use of β2-MG and cystatin C to improve detection of subtle renal dysfunction in diabetes.

Finally, more recently Alekseienko et al[96] explored the relationship between β2-MG levels, proinflammatory cytokines, and renal function decline in DKD. Their study revealed that increased serum β2-MG correlated with elevated inflammatory markers such as TNF-α and IL-6, as well as worsening renal functional parameters. These findings support the dual role of β2-MG as a marker reflecting both impaired renal clearance and systemic inflammation, highlighting its potential for monitoring disease activity and progression.

Despite promising data, the clinical adoption of serum β2-MG as a routine biomarker for DKD is hampered by several factors, including analytical variability, influence of non-renal factors (e.g., inflammation, malignancy), and lack of standardized reference ranges, limiting interpretability. Furthermore, longitudinal studies with larger, ethnically diverse populations are warranted to validate its predictive capacity and establish cut-offs[97]. Future research should focus on integrating β2-MG measurements within multi-marker panels and exploring its role in guiding therapeutic decisions. Advances in assay standardization and understanding of β2-MG’s pathophysiological roles may enhance its utility in personalized medicine for DKD.

Plasma beta-trace protein

Beta-trace protein (BTP), also known as lipocalin-type prostaglandin D synthase, is a low-molecular-weight (23-29 kDa) glycoprotein predominantly expressed in the central nervous system and renal tubular cells. Functionally, BTP enzymatically catalyzes the conversion of prostaglandin H2 to prostaglandin D2, a lipid mediator involved in several physiological processes, including sleep regulation, nociception, and inflammation modulation[98]. Notably, BTP is freely filtered at the glomerulus and is minimally influenced by muscle mass or diet, making it a promising endogenous marker of GFR[99]. Because of its dual nature of small enzymatic and freely filtered molecule, in combination with its renal stability, plasma BTP can be considered a potential early serum and urinary biomarker of DKD, able to detect tubular injury before changes in serum Cr or albuminuria.

Several studies have elucidated the biological role of BTP in DKD. Ragolia et al[98] employed prostaglandin D synthase knockout mice, a BTP-deficient model, and revealed markedly accelerated glucose intolerance, development of nephropathy, and atherosclerosis compared to wild-type controls. These findings imply that BTP has protective functions in metabolic regulation and renal homeostasis, possibly through modulation of inflammatory pathways and oxidative stress responses. The exacerbation of renal damage in BTP-deficient animals suggests that reduced BTP expression or activity may not only be a marker but also a contributor to disease progression.

Notably, Motawi et al[6] provided one of the most comprehensive analyses, demonstrating that serum BTP levels were markedly elevated in diabetic patients with incipient nephropathy compared to those without renal involvement. Their study reported a diagnostic sensitivity of 82% and specificity of 78% for plasma BTP at an optimized cut-off value, with an AUC of 0.86. This indicates excellent discriminative capacity, suggesting that plasma BTP can effectively differentiate patients with early kidney involvement from those without renal complications. Similarly, Kobata et al[99] confirmed these findings in a cohort of T2DM patients. They observed that plasma BTP levels inversely correlated with eGFR and progressively increased with advancing stages of DKD. The study reported AUC values ranging from 0.80 to 0.89 for plasma BTP, outperforming serum Cr in predicting early RI. Importantly, the elevation of plasma BTP often preceded the onset of microalbuminuria, highlighting its potential role as an early prognostic marker rather than merely a reflection of established damage. More recently, Bacci et al[100] extended these findings by investigating the combined use of plasma BTP with other endogenous markers such as cystatin C. They found that the integration of these biomarkers improved diagnostic accuracy, suggesting that plasma BTP might be optimally employed as part of a multi-marker panel to enhance the sensitivity and specificity of DKD detection.

BTP has also emerged as a urinary biomarker of DKD, which has been greatly facilitated by the development of sensitive and specific immunoassays. Notably, Oda et al[101] pioneered a practical enzyme-linked immunosorbent assay (ELISA) specifically designed for quantifying human urinary BTP. This assay demonstrated excellent analytical performance, including high specificity for BTP isoforms and low intra- and inter-assay variability, enabling reliable detection of even subtle changes in urinary BTP levels. Such methodological advances have paved the way for clinical applications, especially in the early detection of renal injury. Building on this foundation, Uehara et al[102] conducted a rigorous multicenter, prospective study involving patients with T2DM. Their data revealed that elevated urinary BTP excretion preceded the development of overt proteinuria and RI, underscoring its potential as a prognostic biomarker. These findings suggest that urinary BTP reflects tubular dysfunction or injury early in the course of DKD, potentially before glomerular damage becomes clinically apparent. Complementing these findings, Hirawa et al[103] provided early clinical evidence that urinary BTP excretion is increased in diabetic patients even before the onset of classic nephropathy markers such as microalbuminuria. This early elevation supports the hypothesis that BTP may serve as a sensitive indicator of subclinical renal injury, enabling earlier intervention and potentially better outcomes.

Collectively, the reviewed literature suggests that plasma and urinary BTP may be promising biomarkers for early detection and prognosis of DKD. Limitations include variability in assay methods, small sample sizes in some studies, and heterogeneity of diabetic populations. Future studies should focus on standardizing cut-off values, longitudinal validation in larger cohorts, and integrating BTP measurement in multi-marker panels.

KIM-1

KIM-1, also known as HAVcr-1, or TIM-1, is a type I transmembrane glycoprotein belonging to the immunoglobulin and T-cell mucin domain family, predominantly expressed in the proximal tubular cells of the kidney. Under normal conditions, KIM-1 is minimally detectable, whereas, its expression is markedly upregulated in response to ischemic, toxic and inflammatory renal injury in the context of tubular damage. This upregulation is thought to play a crucial role in the repair and regeneration of renal tissues by mediating the phagocytosis of apoptotic and necrotic cells, thereby facilitating tissue remodeling and recovery. KIM-1 mediates tubular epithelial cell phagocytosis of apoptotic cells and participates in tubular repair mechanisms. In DKD characterized by both glomerular and tubular damage, KIM-1 reflects early tubular injury preceding overt albuminuria, thus representing a candidate biomarker for early disease detection and progression monitoring[104].

Several studies have explored urinary KIM-1 as an early indicator of DKD, but few investigations focused about serum KIM-1. Notably, Balu et al[105] conducted a comparative cross-sectional study that included a cohort of T2DM patients stratified according to the presence or absence of microalbuminuria, alongside healthy controls, aiming to evaluate serum KIM-1 levels in early RI stages. The authors reported a statistically significant elevation of serum KIM-1 in diabetic patients compared to controls (P < 0.001), including those without microalbuminuria, supporting that tubular injury may precede glomerular dysfunction detectable by albuminuria. The study also found positive correlations between serum KIM-1 and markers of glycemic control (such as HbA1c), suggesting a link between hyperglycemia-induced tubular injury and biomarker expression. Additionally, Colombo et al[106] investigated whether a small panel of serum biomarkers, specifically KIM-1 and β2-MG, could predict rapid decline in kidney function in patients with T2DM. The authors analyzed data from several well-characterized cohorts of individuals with T2DM and baseline eGFR between 30 and 75 mL/minute/1.73 m². Rapid decline was defined as a loss of more than 20% of eGFR over time and was assessed over a median follow-up period of 7 years. They found that KIM-1 and β2-MG consistently showed the strongest associations with rapid kidney function decline. In statistical terms, the addition of KIM-1 and β2-MG to a clinical model (including age, sex, baseline eGFR, and albuminuria) significantly improved the model’s predictive ability. This finding suggests that a minimal panel including KIM-1 and β2-MG could offer a simpler, more cost-effective approach to risk prediction in DKD.

Furthermore, Sabbisetti et al[107] investigated serum KIM-1 as a prognostic biomarker in patients with T1DM and acute kidney injury (AKI). In particular, this study extended into the CKD setting by examining a well-characterized cohort of patients with T1DM and proteinuria. The authors found that baseline serum KIM-1 levels independently predicted both rapid eGFR decline and progression to ESRD, even after adjusting for confounders such as albuminuria, glycemic control, and baseline renal function. This observation is particularly notable because it emphasizes the value of tubular biomarkers in a field historically dominated by glomerular injury markers. This study strongly supports the integration of KIM-1 into future risk stratification frameworks for DKD, potentially enabling earlier interventions before irreversible decline occurs.

Despite promising data, serum KIM-1 as a DKD biomarker faces limitations. There is notable heterogeneity in assay methodologies and a lack of standardized cut-off values, complicating inter-study comparability. Elevated KIM-1 can also be detected in AKI and other chronic renal conditions, which may reduce specificity in clinical contexts with mixed etiologies. Moreover, KIM-1’s independent prognostic value beyond conventional markers like albuminuria and eGFR requires further validation through prospective, longitudinal studies.

Future directions should focus on standardizing KIM-1 assay techniques, establishing universally accepted thresholds, and integrating KIM-1 with other tubular biomarkers (e.g., NGAL, cystatin C) and clinical variables to enhance early diagnosis and prognosis of DKD. Large-scale, multicenter studies are imperative to confirm the clinical applicability and cost-effectiveness of serum KIM-1 levels. In conclusion, KIM-1 is a promising tubular injury biomarker with significant potential to complement traditional markers in the early detection and management of DKD.

INFLAMMATORY AND CYTOKINE ACTIVITY BIOMARKERS

Inflammatory biomarkers

Hemodynamic, metabolic and inflammatory axes are involved in DKD pathogenesis and progression[108]. In particular, hyperglycemia causes cellular injury, which triggers the release of proinflammatory mediators such as TNF-α and IL-1, adhesion molecules and damage-associated molecular pattern. Circulating proinflammatory molecules promote recruitment of innate immune cells to the kidney, where their accumulation in the glomeruli further increases the production of proinflammatory cytokines, ROS and proteases, ultimately leading to kidney damage and fibrosis[109].

TNF-α, TNFR1, and TNFR2

Several studies have reported that high serum TNF-α levels are related with DKD pathophysiology and, together with elevated levels of TNFRs, may predict the decline in kidney function, contributing to renal hypertrophy and hyperfiltration[110].

TNF-α mediates its activity through two specific receptors expressed at the cell surface, TNFR1, which primarily activates pro-inflammatory pathways, and TNFR2, which promotes anti-apoptotic reactions. In DKD patients, elevated levels of circulating TNFRs have been shown to be early predictors of end-stage kidney disease (ESKD)[110].

In a cohort study of patients with T1DM and albuminuria, TNFR2 levels were the strongest predictor of DKD. In addition, an elevated TNFR concentration was a strong predictor of eGFR decline in patients with T2DM[111]. Forsblom et al[112] examined serum concentrations of proinflammatory cytokines in patients with T2DM patients with and without renal functional damage and found elevated circulating TNFR1 levels, highlighting its important role in inflammatory responses associated with DKD and suggesting TNFR1 as a strong predictor of renal impairment. Araújo et al[113] reported that TNFR1 levels were increased in patients with T2DM and RI, indicating its potential to predict declining renal function. In contrast, TNFR2 levels were decreased in patients with T2DM without RI and increased in those with RI. Another recent study[114] showed that patients with T2DM had significantly higher plasma levels of not only TNFR1 but also TNF-α and TNFR2 compared with control subjects. A review by Murakoshi et al[115] showed that TNFR1 and TNFR2 levels are independent of albuminuria, HbA1c and baseline renal function. Progression to DKD may occur with or without albuminuria but appears to be consistently associated with elevated TNFR1 and TNFR2 levels. In animal models, suppression of TNF-TNFR signaling (e.g., with etanercept or infliximab) leads to improved renal function. Krolewski et al[116] showed that elevated TNFR2 levels were linked to DKD in 55% of cases and TNFR1 was linked to a higher chance of developing ESKD. Pavkov et al[117] evaluated TNFR1 and TNFR2 as predictors of ESKD, independent of baseline renal function. A meta-analysis demonstrated that TNFR1 and TNFR2 are strong predictors of DKD progression, independent of albuminuria, HbA1c and baseline renal function, with hazard ratios (HR) of 9.4 for TNFR1 and 7.6 for TNFR2 for progression to ESKD[118].

IL-6

Among the cytokines involved in the inflammatory processes underlying DKD, IL-6 plays a pivotal role by acting as a key mediator of endothelial activation, vascular smooth muscle cell proliferation and adhesion molecule production.

In patients with diabetes, chronic hyperglycemia causes IL-6 overproduction, which activates innate immune responses, leading to inflammation associated with overexpression of C-reaction protein and fibrinogen, and subsequent exacerbated vascular injury through oxidative stress and endothelial function damage[119]. In the kidney, IL-6 promotes both podocyte proliferation and damage, stimulates mesangial cell proliferation. Chemokine release causes fibrosis but reduces type IV collagen and fibronectin production by on tubular endothelial cells, promoting epithelial to mesenchymal transition, fibrosis and immune cell accumulation[120]. In T1DM, pathogenesis is initially characterized by increased IL-6 production. T2DM is characterized by chronic low-grade inflammation and elevated IL-6 levels. This cytokine has a central role in extracellular matrix production, contributing to fibrotic tissue deposition in the glomeruli and tubulointerstitial space. A recent study[121] demonstrated that IL-6 was significantly increased in T2DM patients compared to T1DM due to the amplified chronic inflammation in T2DM. IL-6 is weakly correlated with renal function indicators, such as eGFR and UACR. IL-6 is higher in DM patients compared with control group (P < 0.001)[122]. Serum IL-6 levels can also potentially predict atherosclerosis and kidney disease progression in DKD[123]. In their study, Sanchez-Alamo et al[124] reported that baseline IL-6 levels were independently associated with DKD progression, with values > 4.84 pg/mL predicting a faster occurrence of the composite endpoint (≥ 50% increase in Cr, ESKD, or death). This relationship remained significant after adjusting for eGFR and proteinuria, suggesting that IL-6 may offer prognostic value beyond conventional markers. These results support a possible contribution of systemic inflammation to DKD progression, though additional research is required to validate IL-6 as a dependable risk-stratification marker.

suPAR

suPAR originates from urokinase plasminogen activator receptor, which is a membrane receptor expressed on a variety of cells, including immunologically active cells, endothelial cells, and podocytes, from which it is released in a soluble form during inflammatory processes. Both soluble and membrane-bound forms are involved in cell adhesion and migration through integrin binding. suPAR has also been studied in the pathogenetic process of kidney disease, in particular focal segmental glomerulosclerosis. Wang et al[125] demonstrated that serum suPAR levels were elevated in patients with DKD. The optimal cutoff value of suPAR for predicting early DKD was 49933. In this study, ROC analysis demonstrated that suPAR is a sensitive indicator of DKD stage, suggesting that serum suPAR may serve as a biomarker for early diagnosis of DKD in patients with T2DM. Moreover, suPAR levels are associated with diabetes severity and represent an independent risk factor for onset of cardiovascular events, eGFR decline and mortality[126].

YKL-40

YKL-40 is a chitin binding glycoprotein that is considered an inflammatory mediator and indicator of endothelial dysfunction. It increases in T2DM and T1DM and is positively correlated with albuminuria. Elevated YKL-40 serum levels correlated with DKD progression and decreased eGFR over time, even after adjustment for potential confounders and known risk factors[127]. A meta-analysis by Kapoula et al[128] revealed that the AUC for YKL-40 was 0.91, with a sensitivity of 0.83 and specificity of 0.85, suggesting that the accuracy of this new biomarker for DKD is high. In addition, YKL-40 levels increase in parallel with DKD progression and patients with T2DM compared with the control groups, even before albuminuria onset. Another study by Shaaban et al[129] revealed that the optimal cut-off value for YKL-40 was 57.1 ng/mL, with a diagnostic sensitivity of 94.2% and specificity of 90.7%.

IL-18

IL-18 is a proinflammatory cytokine that is structurally similar to IL-1β. It facilitates Interferon gamma production and release and promotes the activation of T-helper 1 lymphocytes. IL-18 is produced by a variety of immune cells, such as macrophages and dendritic cells, as well as epithelial cells, keratinocytes, chondrocytes, and osteoblasts. In the kidney, IL-18 is released by tubular epithelial cells[130]. In DKD, its levels increase in serum and urine, making it a candidate predictive biomarker for DKD, including the development of albuminuria and decline in kidney function. It can also serve to differentiate acute from chronic damage, as it is associated with the promotion and progression of fibrosis in chronic kidney disease[131]. Furthermore, IL-18 correlates with biomarkers of podocyte and proximal tubular damage in early stages of DKD in T2DM[132]. In addition, the release of IL-18 increases free radical production and renal tissue damage, which has a fundamental impact on DKD onset[133]. In a recent study, Johnson et al[134] investigated the accuracy of this biomarker for predicting DKD onset; in the ROC analysis, the AUC was 0.6849, suggesting that IL-18 can be used as a part of a biomarker panel to detect the early stages of DKD.

MCP-1

MCP-1 is a potent chemotactic factor for monocytes. It is released by mesangial cells, podocytes and monocytes as a consequence of different inflammatory stimuli. These inflammatory cells mediate tissue and renal damage. MCP-1 is stored in the glycocalyx, forming a chemokine gradient and recruiting monocytes to the inflamed tissues and attracting several immune cells to the kidney. Moreover, it induces cytokine production. When mesangial cells are exposed to MCP-1, it causes accumulation of inflammatory factors such as IL-6 or intercellular adhesion molecule-1. Increased MCP-1 levels could be involved in DKD onset due to high glycemic levels, which lead to MCP-1 release in mesangial cells and stimulate NF-κB signaling via ROS[135]. In a recent case-control study[136] aimed at evaluating MCP-1 levels as predictors for DKD in cases of normoalbuminuria, microalbuminuria, and proteinuria, MCP-1 was identified as a promising potential predictor and prognostic biomarker. A case-cohort study by Schrauben et al[127] that including 894 subjects with diabetes and an eGFR of < 60 mL/minute per 1.73 m² at baseline revealed an increased risk of DKD progression with higher plasma MCP-1 concentrations, but only among those with moderate to severe CKD (eGFR < 45 mL/minute per 1.73 m² at baseline).

Neutrophil extracellular traps

Increased activation of neutrophilic granulocytes is associated with DKD and vascular injury. Neutrophil activation results in DNA decondensation and histone citrullination by enzymes like peptidyl arginine deiminase 4. This induces the secretion of neutrophil extracellular traps (NETs), which contain DNA, histones and neutrophil proteases (neutrophil elastase and myeloperoxidase). NETs damage vascular integrity and renal tissue, interacting with NRLP3 inflammasome, whose activation has been correlated with DKD pathogenesis[137]. NETs have been positively correlated with albuminuria; moreover, an increased presence of NETs in the glomerular structures of diabetic animals and human models has been documented. NETs are also implicated in endothelial dysfunction, which is mechanistically linked with NLRP3 inflammasome activation and IL-1 signaling in glomerular endothelial cells, resulting in pore formation in cell membranes. This is one of the basal factors of sterile inflammation and renal injury in DKD[138]. NETs have been proposed as a mechanism that mediates DKD through sterile inflammation. Recent research has suggested that inhibiting NETs could be a novel therapeutic target and biomarker for follow-up and prognosis in DKD patients[108].

TNF-related apoptosis-inducing ligand

TNF-related apoptosis-inducing ligand (TRAIL) is a pleiotropic cytokine expressed in many tissues. It can selectively induce apoptosis in cancer/transformed cells depending on local biological conditions, such as inflammatory and pro-oxidative states. In fact, oxidative stress through ROS accumulation causes the redistribution of TRAIL receptors in the membrane. Moreover, TRAIL deficiency contributes to the onset of several pathologies, such as cardiovascular disease or DKD. Circulating TRAIL levels are lower in patients with T1DM, T2DM and DKD[139]. In contrast, kidney biopsies in animal and human models of DKD have shown increased TRAIL expression, detectable in the glomeruli where it can induce panoptosis (pyroptosis, apoptosis, and necroptosis)[140]. In vitro studies revealed that TRAIL induced apoptosis of tubular cells exposed to high glucose and pro-inflammatory conditions[141]. However, in light of these conflicting data, further studies are required to define the exact role of TRAIL in DKD.

CD163

The development and progression of DKD are linked to proinflammatory and profibrotic processes. In this context, a possible inflammation biomarker is CD163, a scavenger receptor that is highly expressed on monocytes and macrophages. The ectodomain of membrane-bound CD163 is an endocytic receptor shed by matrix metalloproteinases and neutrophil elastase (ELA2), known as soluble CD (sCD) 163 (sCD163) and triggered by inflammatory stimuli and oxidative stress. Increased CD163 expression has been correlated to insulin resistance, obesity and T2DM, and increased sCD163 levels have been found in a variety of diseases, including rheumatoid arthritis, liver disease and atherosclerosis[142]. Urine CD163 has been studied as a biomarker for DKD, where it showed good predictive value[143]. Moreover, Samuelsson et al[144] investigated the influence of sCD163 plasma levels on the onset of diabetic complications. Although there were higher sCD163 plasma levels in the group that later developed DKD, the difference was not statistically significant. The authors concluded that sCD163 should be included in a panel of biomarkers designed to identify patients at high risk for DKD. In another study, Klessens et al[145] evaluated the role for glomerular and interstitial macrophages in DKD pathogenesis. They showed that diabetic nephropathy (DN) class, interstitial fibrosis, tubular atrophy, and glomerulosclerosis were all strongly correlated with glomerular CD163 + macrophages.

Calprotectin

Serum calprotectin, known as myeloid-related protein 8/14, is a member of the S100 protein family and is composed of a heterodimer of S100A8/A9 subunits[146]. It is secreted from myeloid cells in response to inflammatory stimuli and is expressed in neutrophils, monocytes, and early differentiated macrophages. The calcium-dependent interaction of S100A8/A9 complex with cytoskeletal components has a fundamental role in phagocyte trans-endothelial migration. Extracellularly, it possesses antimicrobial activity, stimulates endothelial activation, endothelial cell apoptosis and necrosis, as well as neutrophil chemotaxis, exerting proinflammatory effects. S100A8/A9 also promotes cytokine secretion, but has also been reported to have anti-inflammatory activity[147]. S100A9 showed strong correlations with macrophages, T cells, and natural killer cells, and it is involved in the dynamic regulation of immune homeostasis. Regarding the value of serum calprotectin in the DKD pathogenesis, a recent study[148] demonstrated that S100A8 and S100A9 expression significantly increased in tubular epithelial cells of diabetic kidneys. S100A8/A9 over-expression promoted renal interstitial fibrosis of diabetic mice. High S100A8/A9 expression in tubular epithelial cells under diabetic conditions activated the Toll like receptor 4/NF-κB pathway, which promoted epithelial-to-mesenchymal transition and led to renal interstitial fibrosis progression. The significant association between plasma calprotectin levels and an increased risk of new-onset kidney disease, suggests its potential as a diagnostic biomarker for DKD[149,150].

Osteopontin

Osteopontin (OPN) is a pleiotropic, multi-phosphorylated glycoprotein that plays an important role in diseases, as it activates a variety of immune cells, including T-cells, B-cells, macrophages, natural killer cells and Kupffer cells. OPN is categorized as a small integrin-binding ligand, N-linked glycoprotein family member, which regulates cell and matrix signaling, cooperating with integrins and CD44 receptors. Tubular epithelial cell expression of OPN is an important element in inflammatory responses that involve natural killer cell activity in kidney disease. In the kidney, it is most highly expressed in the loop of Henle and in nephrons. However, after kidney damage, it is upregulated in glomeruli and tubular segments[151]. A recent study[152] suggested that plasma OPN levels may increase with DKD progression, indicating that plasma OPN levels could serve as a potential diagnostic predictor for DKD. OPN overexpression and macrophage recruitment may also be involved in the tubulo-interstitial damage in DKD. Further studies are needed to evaluate OPN as a definitive biomarker for DKD.

CD14

CD14 plays a pivotal role in innate immunity as a pattern-recognition receptor of endotoxin, whose action is mediated by interactions with the Toll-like receptor 4/myeloid differentiation protein-2 (MD-2) complex. CD14 is either membrane-bound with a glycosylphosphatidylinositol anchor or present as a soluble molecule. sCD14 indirectly participates in immune cell activation by stimulating transfer of lipopolysaccharides to membrane-bound CD14 or participates directly by transferring lipopolysaccharides to Toll-like receptor 4/MD-2 complex on cells that do not express membrane-bound CD14. Poesen et al[153] showed that plasma sCD14 was negatively associated with eGFR, while Zaragoza-Garcìa et al[154] found that sCD14 was higher in DKD patients and progressively increased with DKD stage. This demonstrates that sCD14 may be a good candidate for assessing the risk of kidney disease progression.

OXIDATIVE STRESS AND MITOCHONDRIAL DYSFUNCTION BIOMARKERS

8-OHdG

The role of oxidative stress in the pathogenesis of the diabetic state has been extensively studied. ROS are continuously produced in cells due to metabolic and biochemical reactions and due to exposure to physical, chemical and biological agents. Excess ROS causes oxidative stress and leads to oxidative DNA damage. Several methods and markers are available to measure oxidative stress, including direct measurement of free radicals, antioxidants, redox balance and oxidative modifications of cellular macromolecules. In ROS-mediated DNA lesions, 8-oxoguanine and its nucleotide 8-oxo-2’-deoxyguanosine, the oxidation products of guanine and deoxyguanosine, respectively, are considered the most significant biomarkers of oxidative DNA damage[155]. 8-OHdG is a specific form of oxidized guanine that results from the damaging effects of ROS on DNA, is a widely recognized biomarker for oxidative stress, and its presence in body fluids (including urine, blood, and even cerebrospinal fluid) is often used to assess the extent of DNA damage caused by ROS (free radicals). Elevated 8-OHdG levels are often found in individuals with conditions associated with increased oxidative stress, such as certain cancers, diabetes, and cardiovascular disease. Elevated 8-OHdG levels have been associated with an increased risk of certain cancers. In diabetic patients, urinary 8-OHdG levels have been correlated with DKD severity and have also been identified as a prognostic factor for cardiovascular mortality in patients with T2DM[156]. 8-OHdG can be measured using several techniques, including high-performance liquid chromatography and ELISA. Electrochemical biosensors for rapid and portable detection of 8-OHdG are also being developed.

Human GDF-15

Human GDF-15 is a stress-responsive cytokine belonging to the transforming growth factor-beta superfamily. GDF-15 acts through a receptor called glial-derived neurotrophic factor-like receptor alpha, which transmits the signal through the receptor tyrosine kinase rearranged during transfection. GDF-15 is widely present across many cell types (macrophages, vascular smooth muscle cells, adipocytes, cardiomyocytes, endothelial cells, fibroblasts), tissues (adipose tissue, vessels, central and peripheral nervous system tissues) and organs (heart, brain, liver, placenta) and plays an important role in the regulation of the inflammatory response, cell growth and differentiation[157]. GDF-15 is involved in regulating homeostasis, and its expression is increased in response to injury, stress and inflammation. It performs different and partially reciprocal functions depending on the state of the cells and the microenvironment in which it is located. Studies[158-160] show that this protein has potential protective functions at the renal level and is associated with both the reduction of inflammation and the upregulation of nephroprotective factors with anti-inflammatory activity. Elevated GDF-15 levels have been linked to an increased risk of incident chronic kidney disease and a more rapid decline in kidney function in several kidney diseases, including DKD, IgA nephropathy, lupus nephritis, anti-GBM nephritis, primary membranous nephropathy, renal transplantation, Fabry disease, and amyloidosis[158]. Regarding the clinical condition of DKD, a study demonstrated the role of GDF-15 as a potential prognostic and diagnostic biomarker of DKD by analyzing the association between circulating GDF-15 levels and DKD progression, examining the underlying mechanisms[161]. A cross-sectional study[162] investigating the relationship between GDF-15 levels and the presence of CKD in patients with T2DM enrolled 60 patients with T2DM and divided them into two groups based on the presence or absence of CKD. Their serum GDF-15 levels were measured using ELISA. ROC curve analysis yielded an AUC of 0.846 with a statistically significant P value of < 0.001. The optimal cutoff value for serum GDF-15 to detect CKD was 362.80 pg/mL, with corresponding sensitivity and specificity values of 77% and 79%, respectively. Consequently, in DKD, the association between circulating GDF-15 levels and the progression of kidney damage could be considered a potential prognostic, diagnostic tool, and a potential therapeutic target.

Klotho

Klotho is a β-glucuronidase capable of hydrolyzing steroid β-glucuronides. It is produced in the kidneys, brain, and pancreas. The Klotho gene produces two molecules, a membrane-bound form and a circulating form. This protein is recognized as an anti-aging gene with pleiotropic functions[163], including inflammatory, antioxidant, antifibrotic, and tumor suppressive functions. The membrane-bound form plays a role in maintaining renal homeostasis by regulating vitamin D metabolism, phosphate balance, and FGF23 signaling[164]. Therefore, Klotho may be a mobile FGF23 coreceptor. Among its other functions, Klotho (cell-bound or soluble) counteracts inflammation and can mitigate inflammaging. It inhibits NF-κB and the NLRP3 inflammasome. This inflammasome requires NF-κB activation and produces active IL-1β, membrane pores, and cell death (pyroptosis). In this way, Klotho counteracts inflammation and cellular damage induced by toxins, damage-associated molecular patterns, cytokines, and ROS[165]. In contrast, soluble klotho is produced by the cleavage of the full-length transmembrane protein by a disintegrin and metalloproteases and exerts various physiological effects by circulating throughout the body. Patients with T2DM have reportedly significantly lower klotho expression. This reduction in klotho levels could indicate the progression of DN and suggest that klotho may be involved in multiple pathological mechanisms[166]. It may be possible to evaluate whether it plays a very important role in new therapeutic strategies[167,168]. DKD is characterized by abnormal deposition of oxidized low-density lipoproteins (ox-LDL), contributing to podocyte damage. Klotho is expressed primarily in the renal tubule and then secreted into the blood, and is a suppressor of aging as it plays a key role in protecting podocytes in DKD. One study revealed how soluble Klotho effectively inhibits glucose-induced elevated ox-LDL deposition in DKD-affected podocytes by modulating insulin-like growth factor-1 receptor/RAC1/OLR1 signaling, thus reducing podocyte damage[169]. Another study[170] evaluating the serum Klotho and insulin levels aimed to determine variability in the different stages of DKD by examining specific biochemical functions associated with the target organ. Seventy patients were divided into three groups based on their ACR. The first group consisted of patients with ACR < 30 mg/g, the second group included patients with an ACR value between 30 and 300 mg/g. The third group included patients with an ACR value > 300 mg/g. In addition, the study also included 20 healthy subjects. Serum Klotho levels were significantly lower in the patient group than those of healthy subjects. Furthermore, there were strong negative correlations between serum Klotho and both ACR and homeostatic models of insulin resistance. The AUC value was excellent, at 0.93 with a P < 0.0001, thus concluding that the serum Klotho levels of diabetic patients were lower and significantly correlated with those of patients with DKD, but also that klotho levels could be influenced by ACR and play a significant role in insulin resistance. Several studies have examined the association between serum klotho levels and the incidence of DKD and mortality in T2DM[171], demonstrating that serum klotho is inversely correlated with the prevalence of DKD and reduced all-cause mortality in individuals with DKD[172,173]. As in the case of a study that highlighted the profound link between soluble klotho levels and the risk of DKD, as well as all-cause and cardiovascular mortality among subjects with T2DM, this study involved 126 Chinese patients with T2DM and 4451 individuals included in the National Health and Nutrition Examination Survey (NHANES) database. Multivariate logistic regression was used to assess the relationship between klotho levels and DKD risk. Additionally, restricted cubic regression analysis was conducted to examine the nonlinear relationship between klotho levels and DKD incidence. In the Chinese cohort, klotho levels were significantly elevated in the T2DM group compared with the DKD group. NHANES data revealed a significant inverse relationship between Klotho levels and DKD risk. Nonlinear analysis further illustrated a substantial nonlinear relationship between Klotho levels and DKD risk. Serum Klotho levels < 880.78 pg/mL were associated with an increased risk of DKD in patients with T2DM. Compared with the type 2 diabetes group, the renal diabetes group had significantly higher all-cause and cardiovascular mortality rates, while the group with low Klotho levels had poorer survival compared to the other groups. These findings highlight the potential importance of α-klotho as both a biomarker and a therapeutic target[171].

HEMODYNAMIC STRESS AND ENDOTHELIAL DYSFUNCTION BIOMARKERS

MR-proADM

Human adrenomedullin (ADM), a vasodilatory peptide widely expressed in many tissues, including bone, adrenal cortex, kidney, lung, blood vessels, and heart, has different biological properties, including vasodilatory, positive ionotropic, diuretic, natriuretic, and bronchodilatory. Increased levels of the MR-plasma proADM (MR-proADM) fragment have been reported in septic patients and in specific pathogenic conditions, such as hypertension, cirrhosis, cancer, and multiorgan failure[174]. In T2DM patients, a study measured MR-proADM using a novel immunoluminometric method in four groups of Chinese subjects: Healthy subjects [n = 100, fasting plasma glucose (FPG) < 5.6 mmol/L], impaired fasting plasma glucose (n = 60, FPG: 5.6-6.9 mmol/L), and diabetic subjects with (n = 100) and without (n = 100) nephropathy. They found that plasma MR-proADM concentrations were elevated in subjects with T2DM, which was further accentuated with nephropathy onset, demonstrating a strong correlation. Among subjects with diabetes, plasma MR-proADM concentrations were significantly correlated with resting forearm cutaneous microcirculatory perfusion (r = 0.43, P = 0.002)[175]. A very interesting study conducted in France analyzed two prospective cohorts of patients with T2DM, starting from the hypothesis that the production of the vasodilatory peptide ADM increases in response to ischemia and hypoxia in the vascular wall and kidney, and may have a nephroprotective effect. Because increased plasma MR-proADM concentration was associated with renal outcome in patients with T2DM, the data suggested that the ADM gene modulates genetic susceptibility to nephropathy progression. Indeed, the results demonstrated that ADM increases in DKD upon renal damage, thus performing a nephroprotective function[176]. This confirmed how MR-proADM improves the prediction of renal function loss risk in patients with T2DM[177] and raised the possibility of a possible therapeutic effect of ADM receptor agonists in DKD.

N-terminal pro-brain natriuretic peptide

N-terminal pro-brain natriuretic peptide (NT-proBNP) is released by cardiac cells in response to increased hemodynamic stress and serves as an established marker for cardiac dysfunction. While primarily utilized for heart failure evaluation, patients with hypertension and chronic kidney disease also have elevated NT-proBNP levels, suggesting potential utility in DKD assessment[178].

NT-proBNP is predominantly cleared by the kidneys; therefore, reduced GFR in DKD leads to elevated circulating levels. However, elevated NT-proBNP levels in DKD are not solely due to reduced renal clearance; they also reflect subclinical cardiovascular dysfunction and increased cardiovascular risk, which are common in this population[179,180]. Moreover, NT-proBNP is an independent predictor of all-cause and cardiovascular mortality in patients with DKD, even in the absence of overt heart failure[178].

A comprehensive 2023 meta-analysis[181] of 8741 participants from 14 prospective cohorts investigated the association between cardiac biomarkers and DKD progression. The analysis demonstrated that higher baseline NT-proBNP levels in diabetic patients were significantly associated with greater probability of renal disease progression, providing robust evidence for NT-proBNP as a reliable predictive biomarker for risk stratification and prognostic assessment in DKD.

Another study also aimed to analyze the relationship between serum NT-proBNP levels and the stages of DKD by identifying the probable predictive factors of serum NT-proBNP levels. Cross-sectional analysis comparing normoalbuminuric (group I), microalbuminuric (group II), and macroalbuminuric T2DM patients (group III) with healthy controls (group IV) revealed that macroalbuminuric patients exhibited significantly higher mean serum NT-proBNP levels than all other groups, establishing NT-proBNP as a useful predictive marker for DKD[182].

A prospective study enrolling patients with biopsy-confirmed DKD revealed that elevated plasma NT-proBNP levels were significantly associated with increased risk of progression to ESKD. During the 12-month follow-up period, patients who progressed to ESKD showed higher baseline NT-proBNP concentrations, and multivariate Cox regression analysis confirmed that increased NT-proBNP levels independently predicted adverse renal outcomes after adjusting for potential confounding factors, including age, sex, blood pressure, cardiovascular disease history, eGFR, proteinuria, and pathological severity scores[183].

NT-proBNP has demonstrated broader prognostic value for diabetes-related complications. In long-term survivors of T1DM from a population-based Danish cohort, higher NT-proBNP concentrations had a strong association with nephropathy development, confirming its predictive capacity across different diabetic populations[184].

This prognostic utility extends to comprehensive cardiorenal risk assessment in T2DM. Comparative analysis against established clinical risk models, including the Joint Asia Diabetes Evaluation risk equations, demonstrated that NT-proBNP provided superior discriminatory ability for multiple cardiorenal endpoints. The biomarker showed particularly strong predictive performance for significant eGFR decline and renal failure, outperforming traditional clinical risk stratification tools and offering enhanced prognostic accuracy for cardiorenal complications[185].

The long-term mortality implications of elevated NT-proBNP further underscore its clinical significance. Prospective follow-up studies spanning over 15 years have demonstrated that NT-proBNP serves as an independent predictor of all-cause and cardiovascular mortality in diabetic patients across all stages of albuminuria. Importantly, this predictive capacity remains significant even after adjustment for conventional cardiovascular risk factors and urinary albumin excretion rates, establishing NT-proBNP as a comprehensive prognostic marker that captures cardiovascular risk beyond traditional nephropathy markers.

Copeptin

Plasma copeptin is a fragment of the vasopressin precursor hormone (ADH), released equimolarly into the blood along with ADH. It is used as a surrogate marker for ADH in the diagnosis of fluid balance disorders, particularly diabetes insipidus. Copeptin is more stable than ADH in the blood, making it a more reliable marker for measurement.

Studies in patients with T1DM have established the positive association of copeptin with DKD development and progression, demonstrating that elevated plasma copeptin levels predict increased risk of renal events, all-cause mortality, and progression to ESRD, confirming its role as a marker of declining renal function[186,187].

The validity of serum copeptin as a marker for renal function deterioration in patients with T2DM was robustly demonstrated in the landmark DIABHYCAR study[188], which followed over 3000 participants with T2DM and albuminuria over 6 years. The study revealed a striking dose-response relationship between baseline copeptin levels and renal outcomes, with renal event incidence progressively increasing across copeptin tertiles from approximately 1% in the lowest tertile to nearly 5% in the highest tertile. This gradient was even more pronounced in the high-risk subset of patients with baseline macroalbuminuria, where event rates dramatically escalated across ascending copeptin tertiles, reaching nearly 12% in those with the highest copeptin levels. Patients in the highest copeptin tertile demonstrated a nearly five-fold increased risk of renal events compared to the lowest tertile. Furthermore, annual eGFR decline showed a clear copeptin-dependent pattern in macroalbuminuric patients, with progressively faster deterioration across tertiles, establishing copeptin as a powerful predictor of renal function decline in DKD.

A comprehensive cross-sectional analysis[189] conducted in 2016 demonstrated strong associations between copeptin and DKD in adults with type 1 diabetes. The study revealed that participants with T1DM had significantly higher ultrasensitive copeptin concentrations compared to non-diabetic controls, establishing copeptin as a distinguishing biomarker in diabetic populations. More importantly, elevated copeptin levels in patients with T1DM were associated with increased odds of impaired eGFR (over 18-fold higher risk) and albuminuria (over 10-fold higher risk), with consistent linear relationships between ultrasensitive copeptin, eGFR, and UACR. These findings suggest that early detection of elevated copeptin levels in diabetic patients could significantly improve diabetes prognosis through timely therapeutic strategy adjustments.

A comprehensive study[190] analyzing copeptin levels across different diabetes groups and healthy controls sought to establish copeptin as a more reliable biomarker for progressive stages of diabetes mellitus. Using ELISA methodology alongside standard clinical measurements including blood urea nitrogen, Cr, HbA1c, and UACR, the study revealed that copeptin levels were significantly elevated in subjects with a positive family history of diabetes and were notably higher in prediabetic patients compared to other groups. Importantly, copeptin demonstrated meaningful correlations with established diabetes and nephropathy markers, showing the strongest association with UACR, followed by significant correlations with blood urea nitrogen, Cr, and GFR. These findings highlighted copeptin’s potential role as an early and reliable biomarker of diabetes mellitus and associated nephropathy, with its ability to detect metabolic changes even before overt diabetes develops.

FGF21

The FGF family member FGF21 is involved in cell proliferation and differentiation, cytoprotection, repair, and glucose and lipid metabolism; modulates a wide range of cellular activities by interacting with a family of transmembrane FGF receptors[191]. FGF21 stimulates glucose uptake in adipocytes through an insulin-independent mechanism, with this effect validated in both mouse and human adipocyte models[192].

FGF21 levels are influenced by multiple physiological and pathological factors, including dietary composition (particularly high-carbohydrate, low-protein diets), physical exercise, and cold environmental exposure[193]. Metabolic disorders such as diabetes mellitus, obesity, and cardiovascular disease typically increase FGF21 concentrations, reflecting the body’s compensatory response to metabolic stress. Beyond its metabolic functions, FGF21 exerts important immunomodulatory effects that are directly implicated in renal damage pathophysiology[193]. Consequently, FGF21 has emerged as a valuable biomarker for CKD progression, with plasma levels significantly increasing as renal function deteriorates, making it both a marker of kidney dysfunction and a potential therapeutic target for preserving renal health in diabetic patients.

This may be explained by altered clearance or increased production. The clinical significance of this phenomenon is particularly evident in DKD, where FGF21 levels directly correlate with the severity of albuminuria and increased risks of renal composite events, suggesting that FGF21 serves as both a consequence of and contributor to progressive renal dysfunction[194,195].

Beyond its biomarker utility, FGF21 demonstrates crucial renoprotective properties by simultaneously targeting energy metabolism and innate immunity, two interconnected pathways central to DKD pathogenesis. By protecting renal endothelial cells and mitigating the persistent microinflammation that drives kidney damage, FGF21 emerges as a promising diagnostic and prognostic marker for preserving renal function in diabetic patients[193].

A comprehensive meta-analysis[196] systematically evaluated FGF21’s potential as a renal outcomes biomarker, analyzing 28 studies encompassing over 19000 participants. There was consistent evidence supporting the diagnostic and prognostic value of FGF21; patients with chronic kidney disease demonstrated significantly higher FGF21 levels compared to healthy controls, while individuals with high baseline FGF21 concentrations showed increased risk of developing CKD. In T2DM populations, elevated FGF21 levels were associated with renal outcomes, with longitudinal cohort studies confirming that higher FGF21 concentrations predicted increased incidence of adverse renal events. These findings collectively establish FGF21 as a potentially useful biomarker capable of both identifying existing kidney dysfunction and predicting future renal deterioration across diverse patient populations.

FGF23

FGF23 is a bone-derived phosphaturic hormone that serves as regulator of phosphate homeostasis and vitamin D metabolism through its actions on the kidneys. Under normal physiological conditions, FGF23 maintains mineral balance by promoting renal phosphate excretion and suppressing 1α-hydroxylase activity, thereby reducing active vitamin D (calcitriol) synthesis and stimulating 24-hydroxylase to enhance vitamin D degradation. In diabetic patients, this regulatory system becomes dysregulated early in the disease process, with serum FGF23 levels beginning to rise well before conventional markers of mineral metabolism disturbance, such as serum calcium, phosphate, or parathyroid hormone. This early increase positions FGF23 as a sensitive biomarker for incipient DKD.

As DKD progresses through its clinical stages, FGF23 concentrations increase in a stepwise manner, correlating with declining glomerular filtration rates and progressive albuminuria. Its levels progressively increase from the microalbuminuria stage (30-300 mg/g Cr), where subtle kidney damage first becomes detectable, through macroalbuminuria (> 300 mg/g Cr), where significant proteinuria indicates substantial RI. This progressive increase reflects not only the kidney's reduced capacity to excrete phosphate and maintain mineral homeostasis, but also suggests FGF23’s potential role in the pathophysiology of DKD progression, making it both a consequence of renal dysfunction and a possible contributor to further kidney damage through its effects on mineral metabolism and potentially cardiovascular complications[197,198]. Genetic variants in the FGF23 gene may modulate risk, with certain alleles conferring protection or increased susceptibility to CKD in diabetic patients[199].

FGF23 serves dual roles as both a DKD severity marker and a predictor of cardiovascular risk and mortality in diabetes, with risk effects magnified in patients with DKD[200,201]. At the molecular level, FGF23 drives DKD pathogenesis by promoting interstitial fibrosis, amplifying inflammatory responses and dysregulating calcium-phosphate homeostasis, while simultaneously showing increased local expression within kidney tissues as disease severity progresses[202-204].

Baseline FGF23 concentrations function as an independent risk stratification tool for DKD, predicting not only disease onset but also the trajectory of renal deterioration, including increased proteinuria, lower eGFR in both type 1 and type 2 diabetes and progression to dialysis-dependent ESKD and other major adverse renal events, regardless of conventional risk factors or initial renal function status[205,206].

FIBROSIS AND EXTRACELLULAR MATRIX REMODELING BIOMARKERS

Endostatin

Circulating endostatin, a C terminal fragment of collagen XVIII expressed in renal glomeruli and peritubular capillaries, was identified as a robust, independent prognostic marker for kidney function failure in patients with T2DM, being significantly associated with composite outcome of sustained 40% decline in eGFR or ESRD. Several processes implicated in impaired kidney function, including endothelial dysfunction, matrix remodeling after kidney injury, and angiogenesis, may be reflected by endostatin levels[207]. Endostatin is an endogenous anti-angiogenic agent with blood vessel-inhibiting properties, involved both in the microvascular and macrovascular complications observed in diabetes. It has also been extensively investigated as a potential therapeutic agent for cancer[208]. A nested case-control study within the ACCORD trial and a cohort from the Mount Sinai BioMe Biobank investigated plasma endostatin as a novel biomarker for predicting kidney function decline in patients with T2DM, particularly those with preserved eGFR. Baseline plasma endostatin levels were detected. The study concluded that plasma endostatin is strongly associated with adverse kidney outcomes in T2DM with preserved eGFR and significantly enhances the ability to predict risk beyond traditional predictors. The AUC for kidney outcome improved from 0.74 to 0.77 in BioMe with the addition of endostatin to the base clinical model[207]. In a cohort study of 607 individuals with T2DM, endostatin levels were higher in patients with DKD and/or microalbuminuria and were associated with an increased risk of renal function decline in longitudinal analyses adjusted for established kidney disease markers[209]. Endostatin had a weak but significant association with microalbuminuria, as shown in a case-control study about endostatin, soluble TNFR1 (sTNFR1) and soluble TNFR2 (sTNFR2). At the beginning of the study, individuals who subsequently developed microalbuminuria exhibited elevated baseline levels of both endostatin and sTNFR1, though not sTNFR2. Multivariable analysis revealed that each doubling (log₂ increase) in endostatin concentration was associated with a modest but statistically significant 6% increase in the risk of developing microalbuminuria (adjusted HR of 1.006 and 95% confidence interval: 1.001-1.011)[210].

CD5 L

CD5 L is a protein implicated in immune and inflammatory responses. It is secreted by macrophages and regulates inflammation and lipid metabolism. It contributes to obesity-associated inflammation, insulin resistance, and autoimmunity by affecting fat breakdown and immune cell activity[211]. Elevated concentrations of CD5 L are predictive of future cardiovascular events and all-cause mortality in patients with CKD[212]. In the Fremantle Diabetes Study Phase II, four specific plasma protein biomarkers were identified: Elevated apoA4 and C1QB levels, and decreased CD5 L and IBP3 levels. Over a 4-year follow-up period, these markers predicted rapid eGFR decline in individuals with T2DM independently of other clinical predictors, including baseline eGFR and ACR. In particular, CD5 L predicted the decline of renal function. The clinical prediction model for eGFR trajectory was assessed by the AUC of the ROC, which rose from 0.75 to 0.82 (P = 0.039) after the addition of the biomarkers. Both sensitivity and specificity improved. Interestingly, the duration of diabetes ceased to be a significant predictor once these biomarkers were included in the model[30]. PromarkerD, a panel of six plasma proteins (apoA4, apolipoprotein C-III, complement C1QB, complement factor H-related protein 2 (CFHR2), IGFBP3 and CD5 L) emerged as a valid diagnostic and prognostic model of DKD. The model demonstrated strong performance with an AUC of 0.88, indicating excellent discrimination, with high sensitivity and specificity for foreseeing 4-year risk of DKD in the Fremantle Diabetes Study Phase II. Importantly, it achieved a 98% negative predictive value, meaning it was very good at identifying individuals who would not develop DKD[31]. PromarkerD was also tested in 91 T1DM patients, confirming its strong predictive accuracy and clinical utility in identifying patients with T1DM at risk of future adverse renal outcomes[32]. The PromarkerD test for DKD and its individual protein biomarkers were not consistently associated with the prevalence, incidence, or progression of diabetic retinopathy in a community-based sample of individuals with T2DM, reinforcing their specificity for predicting kidney complications[213]. CD5 L was evaluated together with apoA4 and IGFBP3 in a cohort of 857 adults to predict DKD and was compared with eGFR alone, UACR alone, and the combination of eGFR and UACR. The novel biomarker panel demonstrated significantly superior predictive performance (AUC = 0.88) compared with conventional tests (AUC range, 0.63-0.82)[214]. A next-generation application of PromarkerD, the CaptSure^TM immunoassay targeting apoA4 and CD5 L, has since been shown to be analytically robust and to provide strong risk stratification accuracy in adults with T2DM, achieving an AUC values ranging from 0.78 to 0.88 for predicting a 4-year decline in kidney function[215].

OPN

OPN is sialoprotein with major roles in bone mineralization, cellular adhesion and remodeling. Beyond bone, it is also involved in various other functions across diverse cell types, including epithelial, endothelial, and renal cells. In atherosclerosis, OPN acts as a calcium-binding phosphoglycoprotein that serves as a biomarker for vascular calcification[216]. It affects podocyte motility and signaling, leading to microalbuminuria, and has a role in tubulointerstitial damage through macrophage accumulation. Its plasma levels correlate with eGFR and are higher in DKD[8]. Moreover, plasma OPN concentrations proportionally increase with DKD progression[152]. A cross-sectional study of 467 patients with T2DM diabetes for ≥ 10 years revealed the diagnostic value of different biomarkers in relation to DKD progression of in terms of normo-, micro- and macroalbuminuria. Serum OPN levels showed excellent diagnostic accuracy for macroalbuminuria (AUC = 0.938), and lower but significative association with micro-albuminuric patients (AUC = 0.692)[217]. Higher baseline serum OPN levels were also observed in patients with T1DM who subsequently developed microalbuminuria, progressed to ESRD, experienced a cardiovascular disease event, or died during the follow-up period, thus confirming its role as an independent predictor of incipient DKD, CVD events, and all-cause mortality in individuals regardless of diabetes type[218].

IMMUNE SYSTEM AND COMPLEMENT ACTIVATION BIOMARKERS

The innate immune response and complement activation contribute to the pathophysiological mechanisms underlying T2DM development, progression and its related complications.

In DKD pathophysiology, complement activation plays a crucial role. The final step of C3 activation leads to the generation of pro-inflammatory fragments (C3a, C3b) and assembly of the membrane attack complex (C5b-9), which have been shown to trigger podocyte cytoskeletal injury, mitochondrial dysfunction, proteinuria, and tubular damage in both animal and in vitro models[219]. Glomerular deposition of C3 is one of the immune phenotypes of DKD and could correlate with kidney function[220].

Furthermore, epidemiological data from a large prospective cohort (n = ~95000) demonstrate that higher circulating baseline C3 levels confer a nearly twofold increased hazard for DKD incidence and related microvascular complications[221].

Building on PromarkerD[222], a panel of plasma protein biomarkers identified using mass spectrometry, including complement C1QB and CFHR2, has been proposed as potential DKD-specific markers, showing advantages over traditional gold standard measures such as ACR and glomerular filtration rate. C1Q is a part of the C1 complex, which activates the complement classical pathway, and is formed from six copies of three polypeptide chains (A, B and C). The B-chain of C1Q plays a critical role in the formation of the C1Q hexameric structure, which is essential for binding immune complexes and initiating complement activation[223]. Dysregulation of C1QB expression has been associated with autoimmune diseases, such as systemic lupus erythematosus[224], and more recently, it has been implicated in renal inflammatory conditions including DKD. Using microarray datasets from the Gene Expression Omnibus database, researchers performed differential gene expression analysis and constructed protein-protein interaction networks, identifying C1QB among the top hub genes significantly upregulated in DKD renal tissues. Functional enrichment analysis revealed that C1QB is closely associated with immune-related pathways, particularly complement activation and leukocyte migration. The gene’s expression was further validated using external datasets and experimental models of DKD in rats, where elevated levels of C1QB messenger RNA were confirmed by quantitative polymerase chain reaction. In the evaluation of the hub-gene C1QB, the AUC was 0.911 (0.837-0.985), with sensitivity of 0.82 and specificity of 0.833. These results suggest that C1QB plays a pivotal role in DKD pathogenesis through modulation of the immune microenvironment and complement system activation, offering potential utility as a non-invasive diagnostic marker or therapeutic target[225]. C1QB was shown to be an independent predictor of rapid eGFR decline, outperforming conventional clinical variables in a cohort from the longitudinal observational Fremantle Diabetes Study Phase II[30].

Another marker associated with a decline in eGFR is CFHR2, a member of the complement factor H-related protein family, which is involved in the regulation of the alternative complement system pathway[226]. Few data are available for its involvement as a serum biomarker in DKD, but it has been already identified in urine as a higher risk factor of death in a cohort of patients with T2DM and proteinuric DKD[227].

COMPARATIVE PERFORMANCE OF TRADITIONAL CLINICAL MARKERS AND EMERGING SERUM BIOMARKERS

Traditional clinical markers for DKD, including eGFR and UACR, remain the cornerstone of diagnosis and risk stratification. However, these markers have well-documented limitations: EGFR exhibits poor sensitivity for early tubular injury and is confounded by non-renal factors such as muscle mass, dietary protein intake, and medications affecting Cr secretion, while UACR demonstrates substantial biological variability and fails to capture non-glomerular pathology or early-stage disease, particularly in non-albuminuric DKD phenotypes[228-230].

Recent advances have identified high-performing serum biomarkers that demonstrate superior diagnostic accuracy compared with traditional markers (Figure 1).

Figure 1 The schematic diagram categorizes serum biomarkers of diabetic kidney disease according to their associated pathogenic mechanisms, including metabolic dysregulation, glomerular and tubular injury, inflammatory processes, oxidative stress and mitochondrial dysfunction, fibrosis and extracellular matrix remodeling, and hemodynamic stress with endothelial dysfunction. The central kidney illustration shows the anatomical context of these pathological processes. DKD: Diabetic kidney disease; HbA1c: Glycosylated hemoglobin; AGEs: Advanced glycation end products; ZAG: Zinc-alpha-2-glycoprotein; APOA4: Apolipoprotein A-IV; APOC3: Apolipoprotein C-III; IGFBP3: Insulin-like growth factor binding protein 3; RBP4: Retinol-binding protein 4; TGF-β1: Transforming growth factor beta 1; HPSE: Heparanase; β2-MG: Beta 2-microglobulin; BTP: Beta-trace protein; KIM-1: Kidney injury molecule-1; NGAL: Neutrophil gelatinase-associated lipocalin; 8-OHdG: 8-hydroxydeoxyguanosine; GDF-15: Growth differentiation factor-15; TNF-α: Tumor necrosis factor-alpha; IL: Interleukin; suPAR: Soluble urokinase plasminogen activator receptor; YKL-40: Chitinase-3-like protein 1; MCP-1: Monocyte chemoattractant protein 1; C1QB: C1q subcomponent subunit B; CFHR2: Complement factor H-related protein 2; NETs: Neutrophil extracellular traps; CD: Cluster of differentiation; TRAIL: Tumor necrosis factor related apoptosis inducing ligand; OPN: Osteopontin; ADM: Adrenomedullin; NT-proBNP: N-terminal pro-brain natriuretic peptide; FGF: Fibroblast growth factor.

Notably, serum NGAL, insulin-like growth factor-1/IL-6 panels, and composite multi-biomarker models show marked improvements in diagnostic performance. A combined panel of serum NGAL, β2-MG, and N-acetyl-β-D-glucosaminidase achieved an AUC of 0.926, with sensitivity of 80.33% and specificity of 90.14% for DKD detection, substantially outperforming traditional markers[231]. Similarly, the Joslin kidney panel (JKP), comprising KIM-1, TNFR2, and WFDC2, demonstrated C-indices up to 0.913 for ESKD risk prediction, exceeding the discriminatory capacity of eGFR and UACR alone[232]. Furthermore, machine learning algorithms and integrated multi-omic approaches enhance predictive accuracy by combining metabolomic and proteomic datasets, uncovering novel biomarker signatures that evade detection by conventional statistical methods[230,233].

Evidence from longitudinal cohort studies and clinical trials demonstrates that elevated concentrations of tubular injury markers such as serum NGAL and KIM-1 and inflammatory mediators, including TNFR1, TNFR2, and GDF-15, are independently associated with DKD progression, ESKD risk, and therapeutic response[127,232,234]. Multi-biomarker panels consistently demonstrate superior performance over traditional markers in predicting renal outcomes, enabling more refined risk stratification and facilitating personalized therapeutic strategies[232,234]. The JKP exemplifies this clinical utility, not only stratifying ESKD risk but also predicting differential response to renoprotective therapies such as fenofibrate, thereby supporting individualized treatment decisions[232]. Real-world evidence from the PromarkerD test, a validated blood-based biomarker panel, has meaningful impacts on clinical decision-making: High-risk results prompt intensified monitoring, increased sodium-glucose cotransporter-2 inhibitor prescription, and avoidance of nephrotoxic agents, while low-risk results appropriately reduce unnecessary interventions[235]. Collectively, these findings underscore the potential of biomarker-guided management to optimize clinical outcomes through precise patient stratification and individualized therapeutic selection, emphasizing the critical importance of thoughtful biomarker integration into precision nephrology care pathways (Table 1).

Table 1 Serum biomarkers for early detection of diabetic kidney disease.

Biomarker	Diagnostic accuracy	Main limitations	Validation status
Creatinine[1,34,39]	AUC = 0.66 for early DKD in T2DM; sensitivity = 60%; specificity = 70%	Low sensitivity in early DKD; influenced by muscle mass, age, sex, diet, ethnicity, drugs; does not reflect tubular damage or albuminuria	Routinely used worldwide; clinically established but inadequate alone for early diagnosis
Cystatin C[45-50]	CKD: Sensitivity = 0.85, specificity = 0.87, AUC = 0.92; DKD meta-analysis: AUC = 0.94; T2DM cohort: AUC = 0.914	Assay variability; cut-offs vary by age and ethnicity; limited long-term outcome validation	Widely validated and recommended; incomplete global assay standardization
NGAL[5,53-57]	Sensitivity = 0.79, specificity = 0.87; sNGAL: AUC = 0.973	Lack of standardized cut-offs; assay heterogeneity; optimal in biomarker panels	Promising; requires large multicenter longitudinal studies
KIM-1[104,128]	No quantitative AUC, sensitivity or specificity reported	Poor specificity; assay heterogeneity; lack of standardized cut-offs	Insufficient validation; prospective multicenter studies required
Heparanase[56-63]	No quantitative diagnostic performance data	Multifactorial regulation; limited clinical validation	Potential early-stage biomarker; not validated for routine use
Uromodulin[77-84]	No AUC, sensitivity or specificity reported	Lack of standardized assays; genetic variability	Clinical implementation pending
β2-microglobulin[86-96]	AUC = 0.925 (high); AUC = 0.792 (moderate in T2DM cohort)	Influenced by inflammation; assay variability; no reference ranges	Promising diagnostic and prognostic marker; further validation needed
β-trace protein[98-103]	Plasma BTP: Sensitivity = 82%, specificity = 78%, AUC = 0.86	Small cohorts; assay variability; no standardized cut-offs	Potentially useful; insufficient evidence for widespread clinical use

This table summarizes the main serum biomarkers investigated for the early detection of diabetic kidney disease, highlighting their diagnostic performance, principal limitations, and current validation status in clinical research and practice. AUC: Area under the curve; KIM-1: Kidney injury molecule-1; NGAL: Neutrophil gelatinase-associated lipocalin; sNGAL: Serum neutrophil gelatinase-associated lipocalin; DKD: Diabetic kidney disease; CKD: Chronic kidney disease; T2DM: Type 2 diabetes mellitus; BTP: Beta-trace protein.

DISCUSSION

The clinical assessment of DKD remains constrained by fundamental limitations in Cr-based filtration estimates and albuminuria, which fail to capture the diverse pathophysiological mechanisms driving progressive nephron loss[2,3,38-40]. Emerging serum biomarkers now provide direct readouts of these mechanisms, metabolic dysregulation, tubular injury, inflammation, and glycocalyx degradation at stages when intervention may still modify disease trajectory.

For example, among filtration markers, cystatin C has demonstrated reproducible superiority over Cr in early detection and prognostic stratification, particularly when incorporated into combined eGFR equations[5,41-48]. Markers of tubular injury, such as NGAL, KIM-1, β2-MG, and serum uromodulin, have been shown to have strong associations with histological damage and predict eGFR decline independently of albuminuria[4,53-76,91-107], addressing a critical gap in non-albuminuric phenotypes. In parallel, TNFR1 and TNFR2 are strong, albuminuria-independent predictors of ESRD across diverse cohorts[110-118,134-150], highlighting the central contribution of inflammatory pathways in DKD progression and suggesting potential therapeutic targets. Mechanistic studies further support these clinical associations. Experimental and translational evidence linked HPSE-mediated glycocalyx degradation[56-66] and genetic modifiers of cystatin C pathways[49,50] to renal injury, providing biological plausibility for its prognostic performance. Multi-marker panels incorporating metabolic (apoA4, CD5 L, IGFBP3)[30-32] and injury markers have shown improved prognostic value over traditional clinical models alone, though external validation and assay standardization remain incomplete.

The existing literature justifies sustained investment in their rigorous evaluation. The path forward requires three priorities: Large-scale prospective validation in multiethnic cohorts with renal biopsy correlation; Harmonization of assay platforms and reference ranges; And pragmatic trials evaluating whether biomarker-guided interventions improve patient outcomes. Until these milestones are achieved, serum biomarkers will represent just an investigational tool rather than established clinical decision endpoints (Table 1).

CONCLUSION

DKD remains a major contributor to ESRD worldwide, yet traditional biomarkers often detect injury at advanced stages, highlighting an important gap in early diagnosis and intervention. This review highlights that emerging serum biomarkers, reflecting diverse pathophysiological mechanisms including metabolic dysregulation, tubular injury, inflammation, oxidative stress, and fibrosis, show promise for improved diagnostic and prognostic performance. Advancing this field requires a systematic, multidisciplinary approach: Integration of proteomic, metabolomic, and transcriptomic platforms to identify mechanistically relevant candidates; Rigorous validation in large, diverse cohorts; Development of multi-marker panels capturing disease heterogeneity; And establishment of standardized, clinically actionable algorithms.