#1393 Urinary soluble CD163 correlates with glomerular inflammation in proliferative lupus nephritis

Abstract

Listen

Translate article

Abstract Background and Aims Clinical distinction between histologic activity and chronicity in lupus nephritis remains challenging. We sought to evaluate the utility of urinary soluble protein CD613 (usCD163) as a potential biomarker to reflect histologic activity in lupus nephritis (LN). Method Forty-three patients with LN with concurrent determination of usCD163 at the moment of kidney biopsy were included in the study. A first morning void prior to the kidney biopsy was collected and usCD163 was measured by a commercial ELISA assay (EUROIMMUN, Lubeck, DE). Urine CD163 values were normalized to urine creatinine and expressed as ng per mmol of urine creatinine. Results The study cohort had a mean age at the moment of kidney biopsy of 35.3 ± 12.3 years, 81.4% being females. The mean eGFR and 24-h proteinuria were 77.1 ± 33.7 ml/min/1.73 m2 and 3.28 ± 3.23 g/day, respectively. In terms of histological findings, the majority of patients had proliferative LN (class III—34.9%, class IV—34.9%, class IV+V—9.3%), while the median activity and chronicity index were 7 (IQR: 3-11) and 3 (IQR: 1-5), respectively. The median usCD163 normalized to urine creatinine was 616.4 ng/mmol (IQR: 117.6-2363.7). usCD163 significantly correlated with 24-h proteinuria (r = 0.46, p = 0.002), albumin/creatinine ratio (r = 0.61, p < 0.001) and C3 level (r = −0.32, p = 0.03), but not with hematuria (r = 0.26, p = 0.09) or eGFR (r = −0.27, p = 0.07). Regarding the histological parameters, usCD163 significantly correlated with the activity index (AI) and the individual active lesions (with the exception of fibrinoid necrosis), but not with chronicity index or any chronic lesion (Fig. 1). In addition, the level of usCD163 was higher in patients with proliferative LN compared to those with class II or isolated class V LN (Fig. 2). When evaluating the capacity of different variables to discriminate patients with a high AI (≥9), usCD163 outperformed classical clinical or immunological variables, with an area under the curve of 0.71 (95% CI, 0.55-0.86; p = 0.01). At a cutoff of 313 ng/mmol (Youden Index = 0.42), usCD163 had a sensitivity of 88.2% and specificity of 46% to identify patients with a high AI. Patients with an elevated level of usCD163 (≥313 ng/mmol) had higher 24-h proteinuria (4.41 ± 3.4 vs. 1.36 ± 1.74 g/day, p < 0.001), higher AI [10 (IQR: 6-12) vs. 3.5 (IQR: 0-6), p < 0.001] and more frequently class IV LN (48.1% vs. 12.5%, p = 0.02), compared to those with low levels of usCD163. In addition, 4 patients underwent repeat kidney biopsy post-induction therapy. In these patients, serial monitoring showed a normalization of usCD163 levels that paralleled a decrease in both proteinuria and AI. Conclusion Urinary soluble CD163 is a promising non-invasive biomarker for glomerular inflammation and could be used to evaluate histologic activity in patients with LN.