#683 Utility of urinary soluble CD163 as a novel biomarker for the detection of IgA nephropathy

Abstract Background and Aims IgA nephropathy (IgAN) is an immune mediated disease and a leading cause of biopsy proven glomerulonephritis (GN) worldwide. CD163 is a membrane protein localised on the surface of monocytes and macrophages. Its soluble form, sCD163, produced by enzymatic splitting of the ectodomain of inflammatory cells when activated, is found in increased concentrations in acute inflammatory diseases such as sepsis and vasculitis. Urinary sCD163 levels have also been reported to be a useful biomarker of glomerular inflammation in lupus nephritis (LN) and anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis. Yet, few studies have evaluated the role of urinary sCD163 as a biomarker in IgAN, and comparing it across different GN subtypes. This study aims to explore the utility of urinary sCD163 as a non-invasive biomarker for IgAN, and across different immune mediated GN types. Method This is a retrospective and prospective cohort study from a single tertiary hospital. Frozen urine samples from patients with GN and healthy controls from the Asian Kidney Disease Study and Singapore Kidney Function Study respectively were utilised for the retrospective cohort. The prospective cohort consists patients with biopsy-proven IgAN recruited from May to December 2023 with informed consent taken. Urinary sCD163 was measured using enzyme-linked immunosorbent assay (ELISA). Univariate analyses were performed and urinary sCD163 values were plotted on receiver-operating characteristic (ROC) curve and the area under the curve (AUC) was calculated. This study obtained Institutional Ethics Board Approval (DSRB 2022/00698). Categorical variables were represented by n (%). Continuous variables with a parametric distribution were represented by their means ± standard deviation, non-parametric variables were expressed as their medians with interquartile ranges. Chi-square test and ANOVA were performed between categorical variables, while a student's t-test was performed for continuous variables. P-value of 0.05 was taken as statistically significant. A ROC curve was performed for sCD163 values in IgAN and controls, and the AUC calculated. Results In total, urine samples of 59 subjects were analysed (IgAN n = 10; LN n = 5; healthy controls n = 44). There were 30 male subjects (50.8%) and 29 female subjects (49.2%). Median age was 41.0 (IQR 31.0-52.0) years. Median serum creatinine was 78.0 (IQR: 60.0-91.0) µmol/L. Mean urine sCD163 value was 0.32 (±0.19) ng/ml. There were no baseline differences in demographics, comorbidities and medication between groups. Prospective cohort patients with IgAN were more likely to have received steroids (p = 0.03), and there were no patients of the Malay Ethnicity in the prospective cohort (p = 0.05). Median urine protein (g/day) in IgAN was 0.66 (IQR 0.16 to 3.29)¸ 0.159 in LN (IQR 0.09 to 0.26) and 0.12 in controls (IQR 0.07 to 0.20). Using Mann-Whitney U test to compare median proteinuria between IgAN and controls showed significant difference (p = 0.001), whereas median proteinuria between IgAN and LN was not statistically significant (p = 0.08). On univariate analysis compared to healthy controls, IgAN patients had a significantly higher serum creatinine (p = 0.04), compared to LN patients against controls (p = 0.30). Urine sCD163 levels were significantly higher in IgAN patients compared to controls (0.54 ± 0.26, p = 0.02). Patients with LN do not have a significant urinary sCD163 level difference compared to controls (0.21 ± 0.20, p = 0.40), which also correlates with creatinine and degree of proteinuria. There were no other significant baseline differences against controls. On the ROC curve for IgAN patients, urine sCD163 of 0.38 ng/ml has a 70% sensitivity, and 75% specificity in distinguishing IgAN against controls (AUC 0.77; 95% CI: 0.59-0.96, p = 0.008). For LN, urine SCD163 did not distinguish between LN and controls (AUC 0.29; 95% CI: 0.000-0.58, p = 0.12). Conclusion Urine sCD163 is elevated in patients with IgAN compared to controls and LN. Urine sCD163 is useful in the diagnosis of IgAN. A urine sCD163 level of 0.38 ng/ml has a 70% sensitivity and 75% specificity in identifying IgAN against healthy subjects.