#4790 SERUM BIOMARKERS AS PREDICTORS OF RENAL DAMAGE IN ANCA-ASSOCIATED VASCULITIDES

Abstract

Abstract Background and aims Despite therapeutic advances, anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitides (AAV) carry poor renal and survival outcomes. Prognostic elements are needed to guide treatment. Several studies explored determinants of AAV outcomes, but predictive factors are not well established. Poor prognosis factors have been lower glomerular filtration rate, lower serum C3, C3 deposition on immunofluorescence (IF), and recently platelet count below 250 × 109/L. Histopathological lesions have also been related to renal outcomes. In this study we aim to correlate serum immune and inflammatory biomarkers with renal histological lesions’ severity. Methods We retrospectively analysed histopathological findings of adults with biopsy-proven AAV renal involvement through Berden's histopathological classification (BHC) classes, Brix's renal risk score (BRS), interstitial fibrosis and tubular atrophy (IFTA) and crescentic glomeruli percentages (%), C3 positive IF and presence of interstitial hemorrhage (IH). Patients’ ANCA titer (ANCAd) measured by ELISA, serum C3 and C-reactive protein (CRP) were documented at diagnosis. ANCA titer was also registered at last follow-up date (ANCAf). Univariate statistical analysis was used to correlate every biomarker with every histologic variable mentioned. C3 and CRP were used as continuous variables and ANCA titers as continuous and categorical variables. As a continuous variable, ANCA titers above laboratorial detection were not considered (i.e. >134). Seronegative AAV were not considered for titer analysis. BRS was considered as continuous variable for its score and as categorical for its risk group. Multivariate analysis was used to identify independent predictors renal biopsy findings. Results We included 46 patients with biopsy-proven AAV kidney involvement from January 2006 to January 2023, 65.2% male (n = 30), median age 66.5 (60.75-74.5), 78.3% (36) MPO, 15.2% (7) PR3 and 6.5% (3) seronegative; and 6.5% (3) had concomitant anti-GBM+ (all MPO). Categorization by ANCAd revealed 7.7% (3) 0-19 UI/mL, 7.7% (3) 20-39, 7.7% (3) 40-59, 20.5% (8) 60-99, 12.8% (5) 100-134 and 43.6% >134. According to BHC, biopsies were: 17.4% (8) global sclerotic, 28.3% (13) mixed, 39.1% (18) crescentic, and 15.2% (7) focal. According to BRS, 2.2% (1) had low risk, 28.3% (13) medium risk and 58.7% (27) high risk of ESKD. 5 biopsies had incomplete information for a score result. Median BRS was 9 – high risk. IF C3 was positive 28.3% (16) biopsies, and IH in 10.9% (5). Median % IFTA was 40% and median %crescentic was 33%. In univariate analysis, ANCAd did not correlate to BHC class both as continuous and categorical variable (p = 0.477 and p = 0.685 respectively), nor to %IFTA (p = 0.935, p = 0.938), nor %crescents (p = 0.975, p = 0.938), nor C3 IF staining (p = 0.690, p = 0.344), nor IH presence (p = 0.773, p = 0.897), nor to BRS risk groups (p = 0.339, p = 0.584). ANCAd titer did not correlate to BRS score value (p = 0.072, r = 0.372), both as continous variables. However, ANCAf did correlate to BRS´s risk groups (p = 0.037), with ANCAf values higher in the high-risk compared to the medium-risk group (p = 0.016). In multivariate analysis including age and gender, ANCAf persists as a predictor of BRS´s risk groups (p<0.001, model p<0.001). C3 levels at diagnosis did not have statistically significant correlation with neither histological finding: BHC class (p = 0.282), BRS score (p = 0.683), % IFTA (p = 0.753), % crescents (p = 0.989), C3 on IF (p = 0.472) and IH (p = 0.773). No significant correlation was seen when considering CRP at diagnosis: BHC class (p = 0.162), BRS (p = 0.276), %IFTA (p = 0.343), %crescents (p = 0.071), C3 on IF (p = 0.657) and IH (p = 0.951). Conclusions ANCA titer at diagnosis, serum C3 and CRP did not correlate with histological severity and chronicity lesions in our population. Nonetheless, larger cohorts, systematization of biopsy findings and studies on newer biomarkers might bring helpful information to expected prognosis and initial therapeutic approaches. A positive correlation between ANCA levels at the last follow-up date and medium and high-risk BRS groups on initial biopsy might be further explored in larger studies.