Biomarkers for early detection of sickle nephropathy

Abstract

Renal complications affect nearly 30-50% of adults with sickle cell anemia (SCA), causing significant morbidity and mortality. Standard renal function tests like serum creatinine and glomerular filtration rate become abnormal in this disease only when renal damage has become extensive and largely irreversible. Moreover, not all patients develop sickle nephropathy (SN). Therefore, noninvasive biomarkers that predict early onset of SN are necessary. We performed a cross-sectional analysis for nephropathy in 116 patients with sickle cell disease, analyzing urinary kidney injury molecule-1 (KIM-1), liver-type fatty acid binding protein (L-FABP), N-acetyl-b-D-glucosaminidase (NAG), neutrophil gelatinase-associated lipocalin (NGAL) and transforming growth factor-β1 (TGF-β), together with conventional renal biomarkers (urine albumin and osmolality, and serum creatinine and cystatin C estimated GFR) during routine clinic visits when patients were at steady-state/baseline. We observed a distinct biomarker pattern: KIM-1 and NAG emerged as biomarkers with a strong association with albuminuria. Surprisingly, and in contrast to other acute/chronic renal disorders, NGAL, L-FABP, and TGF-β levels did not show any relationship with albuminuria in patients with SCA. Our study identifies potential biomarkers for SN, and suggests longitudinal validation of these biomarkers for early detection of SN, so that therapeutic interventions can be applied before renal damage becomes irreversible.

Figure 1

Urine albumin and osmolality in patients with sickle cell anemia and other forms of sickle cell disease (a) Albuminuria in patients with SCA (Hb SS disease) and other forms of sickle cell disease (Hb SC and Hb S-β+ thalassemia; other SCD). The percentage of patients with NoA, MiA, and MaA is indicated on the right and the three albuminuria groups are separated by dotted lines. (b) Prevalence of albuminuria in SCA patients on hydroxyurea (HU), on chronic transfusions (Txn) or on neither HU nor Txn (No HU/Txn). The three albuminuria groups are separated by dotted lines. Each symbol in panels (a) and (b) represents an individual patient. Open symbols = NoA, gray symbols = MiA, black symbols = MaA. (c) The degree and prevalence of albuminuria at different ages in patients with SCA. Column bars represent a specified age group with open bar = NoA, gray bar = MiA and black bar = MaA. (d) Urine osmolality in patients with SCA across different age groups. (e) Urine osmolality in SCA patients on HU, Txn or No HU/Txn. Each symbol represents an individual patient. Numbers listed above each group or bar represent the number of patients in each group in all panels.

Figure 2

Plasma Cystatin C Estimated GFR (eGFR) in patients with SCA (a) across different age groups and (b) different albuminuria groups. Values between dashed lines in panel (a) represent normal GFR values. Numbers of patients in each age group (n) is listed above both panels and the mean ± SEM of eGFR in the NoA, MiA, and MaA groups are listed above the columns in panel (b).

Figure 3

Urine N-acetyl-b-D-glucosaminidase (NAG) Activity and Kidney Injury Molecule-1 (KIM-1) Levels in Patients with SCA. (a) NAG activity across different age groups and (b) different albuminuria groups. Values below the dashed line represent normal NAG levels reported. (c) KIM-1 levels across different age groups, and (d) different albuminuria groups. Numbers of patients in each age group (n) is listed above the graphs and the mean ± SEM of NAG activity and KIM-1 levels in the NoA, MiA, and MaA groups are listed above the columns in panel b and d, respectively. Statistically significant differences between the albuminuria groups, adjusted for age are shown with the corresponding P value.

Figure 4

Urine neutrophil gelatinase-associated lipocalin (NGAL) and liver-type fatty acid binding protein (L-FABP) levels in patients with SCA. (a) NGAL levels across different age groups and (b) different albuminuria groups. (c) L-FABP levels across different age groups and (d) different albuminuria groups. Numbers of patients in each age group (n) is listed above the graphs. The mean ± SEM of NGAL and L-FABP levels in the NoA, MiA, and MaA groups are listed above the columns in panel (b) and (d), respectively.

Figure 5

Urine TGF-β Levels in patients with SCA with different levels of albuminuria. (a)Total TGF-β was measured in a dozen urine samples in the NoA, MiA groups and all samples in the MaA groups. Mean ± SEM are depicted above the groups. (b) Active TGF-β levels in the urine in all SCA patients. Number of urine samples analyzed from each group is shown above the graph. No active TGF-β was detectable in 89 samples and one sample showed detectable levels of TGF-β. The standards used for both types of ELISAs showed the expected levels of TGF-β (The correlation coefficient (r²) for the standard curve for the ELISA assays was 0.988 for panel (a) and 0.99 for panel (b).