Urinary biomarkers of diabetic kidney disease

Table 1 Biomarkers can be categorized based on the renal compartment or pathological process they primarily reflect

Biomarker	Source	Diagnostic accuracy	Clinical significance and validation status	Prospective	Contras and limitations
uACR/albuminuria	Glomerular capillary wall	AUC: 0.522[18]-0.933[19]; sensitivity 77% and specificity 92% for microabuminuria; sensitivity 88% and specificity 90% for macroalbuminuria[20]	Standard early marker for DKD; persistent elevation indicates nephropathy	Widely used; inexpensive, reproducible	Affected by blood pressure, exercise, and infections (risk of false positives). DKD may be normoalbuminuric normal uACR and albuminuria do not always exclude DKD
GPC-5	Podocyte injury	ROC curve of urinary GPC-5/creatinine ratio comparing T2DM without nephropathy and those with nephropathy reveals a sensitivity of 93.3% and a specificity of 80%[21]	Emerging marker of podocyte damage; may reflect early glomerular dysfunction in DKD; more studies needed to validate it as a diagnostic tool of incipient nephropathy	Potential early marker of podocyte injury; non-invasive detection in urine	Limited validation (small sample size studies); clinical utility and standardization not yet established
Transferrin	Large protein leakage from the glomerular wall	AUC: 0.846 (95%CI: 0.810-0.882); sensitivity 69.8%, specificity 90.1% (for a cut-off value of 1.15)[22]	Suggests early glomerular barrier compromise before albuminuria. Since significant transferrin levels have been found in normoalbuminuric T2DM patients, it could be suitable as a marker of early-stage DKD; nevertheless, very limited body of research	Earlier than macroalbuminuria; non-invasive	Limited specificity (false positives in primary glomerulonephritis and hypertension in nondiabetic populations; elevated in other proteinuric states)
Immunoglobulins (IgG and IgM)	Large protein leakage from the glomerular wall	AUC: 0.894 (95%CI: 0.848-0.930); sensitivity 75.53%, specificity 92.31% (for a cut-off value of 8.56)[23]	Indicates severe dysfunction. Limited body of evidence in T2DM	Indicative of more advanced damage; could be used in combination with transferrin to enhance diagnostic values	May be negative until late stages of disease (due to high molecular weight of IgG). Testing is more invasive on the one hand, and insufficiently standardized on the other
Nephrin	Podocyte injury	Not available for DKD	Promising marker for detection of early glomerular injury in multiple studies[24]	Specific to podocyte injury; early detection possible	It does not predict early renal dysfunction in DKD[40]. Requires specialized assays, limited availability
Type IV collagen	Matrix remodeling at the glomerular basement membrane	AUC not consistently reported; significantly elevated in early DKD[25]. More sensitive than microalbuminuria but exact percentages not uniformly available	Associated with structural glomerular damage and fibrosis. Validated in large multicentric cohort across DKD stages. Some promising data, but more investigation on larger scale is needed	Reflects chronic structural damage, therefore useful to detect fibrosis	Limited validation (small-to-moderate sample size studies; lack of ROC quantification and cut-offs)[26,30]. Limited utility in early DKD, overlaps with other markers
Podocalyxin	Podocyte injury	AUC: 0.86 (95%CI: 0.81-0.91) alone; combined with α2-MG: AUC: 0.88 (95%CI: 0.83-0.93)[27], in another small study, AUC: 0.996[29]. Sensitivity 73.3%, specificity 93.3% at cut-off 43.8 ng/mL[28]	Linked to detachment and loss of podocytes. Validated in cross-sectional and prospective cohorts of 42-200 patients[27]. Might be a highly accurate biomarker of early podocyte injury, but more studies on larger scale are needed	Sensitive for podocyte detachment, useful in active disease	Not widely validated (limited sample size studies)[27,29]; inter-assay variability; requires advanced laboratory methods

uACR: Urine albumin-to-creatinine ratio; GPC-5: Glypican-5; IgG: Immunoglobulins G; IgM: Immunoglobulins M; AUC: Area under the curve; ROC: Receiver operating characteristic; T2DM: Type 2 diabetes mellitus; CI: Confidence interval; DKD: Diabetic kidney disease; α2-MG: Alpha 2-microglobulin.

Table 2 Summary of urinary tubular biomarkers important in the diagnosis of diabetic kidney disease

Biomarker	Source	Diagnostic accuracy	Clinical significance	Validation status	Prospective	Contras
NGAL	Proximal and distal tubular stress injury	AUC: 0.88 (0.84-0.90); sensitivity 0.82, specificity 0.81[57]	Early marker of tubular injury; predictive of nephropathy before albuminuria	Larger-scale, standardized prospective studies are needed to implement uNGAL as a biomarker into routine clinical use	Detects injury before traditional biomarkers; non-invasive and stable in urine	Not specific for DKD; elevated in infections or AKI; limited validation (small sample size studies only, high heterogeneity among studies in terms of results and methodologies)
KIM-1	Proximal tubular injury	AUC: 0.85 (0.82-0.88); sensitivity 0.68, specificity 0.83[58]	Sensitive marker of early tubular injury in DKD; correlates with disease progression	Some promising data, but whether it can be used as a new marker for diagnosing DKD needs further investigation	Highly specific for proximal tubular damage; FDA-qualified for nephrotoxicity	Limited data in T1DM; limited validation (high heterogeneity between studies in terms of results and methodologies)
L-FABP	Lipid metabolism and proximal tubular stress	AUC: 0.97 (0.94-1.00); sensitivity 100%, specificity 86.67%[59]	Reflects oxidative and lipid-induced tubular damage	Available research seems to highlight high predictive power even compared to other urinary markers including microalbuminuria, but more studies are necessary in order to validate it as a standard biomarker	Sensitive to oxidative stress and useful for early diagnosis	Affected by diet and comorbidities (false positives); requires special assays; limited validation (small sample size studies only, high heterogeneity among studies in terms of results and methodologies)
MCP-1	Inflammatory chemokine secreted by monocytes and macrophages	AUC: 0.546 (0.453-0.638); sensitivity 67.3%, specificity 50.0%[60]	Correlates with tubulointerstitial inflammation and progression of DKD	May not be sufficient as a standalone diagnostic marker for DKD	Correlates with inflammation and disease activity; useful in prognosis	Moderate sensitivity; poor-to-fair discrimination. Levels can fluctuate widely depending on the extent of kidney damage and the presence of inflammatory stimuli
Cystatin C	Proximal tubular function	AUC: 0.807 (0.741-0.873) sensitivity, 70.9%; specificity, 86.3%[61,62]	Indicates proximal tubular dysfunction; used in conjunction with serum cystatin C	Although not specifically approved as a stand-alone biomarker for DKD, it is well recognized as a valuable marker of kidney function and for assessment of overall kidney health, including in the context of DKD	Non-invasive, standardized assays available	Affected by inflammation, corticosteroids, thyroid dysfunction and infections (false positives); not highly specific
Creatinine	Muscle metabolism and tubular function	AUC: 0.604-0.8[63]	Used to normalize other urinary biomarkers; reflects kidney filtration but not specific	Limited diagnostic value alone; rarely used; not FDA approved	Readily available, useful for comparative purposes	Not specific or sensitive for kidney damage (it reflects filtration rather than damage); limited diagnostic value alone
Fe Mg	Tubular magnesium handling	No data on AUC, sensitivity, or specificity	Early sign of tubular reabsorption impairment	May not be sufficient and specific as a diagnostic biomarker of DKD	Potential early tubular dysfunction marker; measurable in clinical laboratories	Lack of standardization, diagnostic accuracy data, and large-scale validation (limited sample sizes, no established cut-offs)
Angiotensinogen	RAAS activity	No data on AUC, sensitivity, or specificity	Reflects intrarenal RAAS activity; elevated in DKD and hypertension	uAGT remains a candidate “theragnostic” biomarker predicting response to RAAS inhibition; further rigorous studies are needed to determine its potential diagnostic role in DKD	Correlates with DKD severity; reflects RAAS dysregulation	Values can be influenced by sex hormones (false positives). Overlap with systemic RAAS activation; requires careful interpretation. Lack of external validation (lack of diagnostic accuracy data, small sample size studies)
Periostin	Tubular fibrosis	AUC: 0.78 (0.71-0.86); sensitivity 80.7%, specificity 43.3% (for a cut-off value of 1.01 ng/mg Cr)[64]	Associated with tubular injury and fibrosis; elevated in early DKD	Still sparse, although consistent body of research regarding its role as an early biomarker of DKD	Reflects fibrotic transformation early; may guide interventions	Emerging marker; limited clinical validation and availability (small and cross-sectional studies). Role not clear in early DKD
Uromodulin	Thick ascending limb health	Regarding the ability to discriminate between patients with DKD and patients with CKD without diabetes, AUC-ROC for the urinary glycated uromodulin (glcUMOD) resulted in 0.715 (0.597-0.834)[65]; urinary uromodulin with an AUC of 0.81 (sensitivity 80%, specificity 60%) and exosomal UMOD gene expression with an AUC of 0.95 (sensitivity 92%, specificity 84%) are elevated in early DKD[66]	Lower levels associated with increased mortality risk and DKD progression	glcUMOD could represent a novel biomarker for DKD; determination of the clinical value of glcUMOD requires further study	Inverse association with outcomes; not expressed in damage; potentially protective marker	Not a direct damage marker; interpretation depends on clinical context; the limited research has focused on its translational modification (glcUMOD)
EGF	Tubular epithelial regeneration	Regarding the association between lower uEGF to creatinine ratio with new-onset eGFR < 60 mL/minute per 1.73 m2: In T2DM normoalbuminuric patients: AUC-ROC: 0.85 (0.81-0.90)[67]	Involved in tubular repair	Low uEGF to creatinine ratio could serve as a biomarker of progressive renal function deterioration in normoalbuminuric patients	Inverse marker (lower levels predict DKD progression); decreases before clinical deterioration	Not routinely measured, lack of sufficient validation
VDBP	Vitamin D transport, liver and tubular cells	No data on AUC, sensitivity, or specificity. Significantly elevated in diabetics with normal, micro- and macroalbuminuria compared to controls, in stepwise fashion (SMDs: 1.52, 1.81, and 1.51, respectively; all P < 0.00001)[68]	Elevated in DKD; correlates with albuminuria and tubular injury	uVDBP could be a promising tubular injury biomarker but requires formal accuracy studies on larger scale	Strong correlation with DKD progression; measurable with multiplex assays	Lack of specificity; also elevated in other renal pathologies (false positives); lack of diagnostic accuracy data and established cut-offs
RBP4	Tubular stress, adipocytes	AUC: 0.746 (0.659-0.834), sensitivity 84.6%, specificity 62.5%[69]	Associated with tubular stress and insulin resistance in DKD	Predictive role of urinary RBP4 requires further investigation, with most studies involving circulating/serum RBP4	Associated with metabolic stress and insulin resistance; useful in early DKD	Limited normative data; influenced by obesity (false positives); it may be complementary in the diagnosis of early DKD
YKL-40	Inflammation and remodeling (monocytes and chondrocytes)	It seems associated with a greater risk of kidney function decline and mortality on a limited male cohort[70]	Elevated in DKD; linked to mortality and inflammation	Serum/plasma YKL-40 predictive power of early DKD has been validated by multiple studies	Rise relevant in DKD	Emerging marker, lacks widespread validation

NGAL: Neutrophil gelatinase-associated lipocalin; KIM-1: Kidney injury molecule-1; L-FABP: Liver-type fatty acid-binding protein; MCP-1: Monocyte chemoattractant protein-1; Fe Mg: Fractional excretion of magnesium; EGF: Epidermal growth factor; VDBP: Vitamin D-binding protein; RBP: Retinol-binding protein; YKL-40: Chitinase-3-like protein-1; RAAS: Renin-angiotensin-aldosterone system; AUC: Area under the curve; Cr: Creatinine; DKD: Diabetic kidney disease; CKD: Chronic kidney disease; ROC: Receiver operating characteristic; glcUMOD: Urinary glycated uromodulin; T2DM: Type 2 diabetes mellitus; uEGF: Urinary epidermal growth factor; eGFR: Estimated glomerular filtration rate; SMD: Standardized mean difference; uNGAL: Urinary neutrophil gelatinase-associated lipocalin; FDA: Food and Drug Administration; uAGT: Urinary angiotensinogen; uVDBP: Urinary vitamin D-binding protein; AKI: Acute kidney injury; T1DM: Type 1 diabetes mellitus.