Comparative analysis of complement C3 and C4 serum levels for outcome prediction in ANCA-associated renal vasculitis

Background The activation of the complement system contributes essentially to the pathogenesis of anti-neutrophil cytoplasmic antibody (ANCA)-associated renal vasculitis. We here aimed to directly compare levels of C3 and C4 for outcome prediction in ANCA-associated renal vasculitis. Methods Serum levels of complement components C3 and C4 were directly compared in association with clinical and outcome data in a retrospective cohort of ANCA-associated renal vasculitis. Results As compared to poor outcome prediction by low levels of complement C3 (p = 0.0093), low levels of complement C4 did not associate with early requirement of kidney replacement therapy (KRT) or death (p = 0.2396). In the subgroup that experienced KRT or death, low C3 levels identified 11/14 (78.6%, p = 0.0071) and C4 levels 9/14 (64.3%, p = 0.1786) cases. Interestingly, 2/14 (14.3%) patients that experienced KRT or death had isolated C4 lowering, and combining low C3 and/or C4 levels identified 13/14 (92.3%, p < 0.0001) cases in this subgroup. Non-superiority to predict poor outcome by low C3 and/or C4 as compared to C3 alone in the total cohort was attributed to 4/24 (16.7%) patients with isolated C4 lowering in the subgroup that did not experience KRT or death. Conclusion While low levels of complement C3 were superior in predicting poor outcome in ANCA-associated renal vasculitis, a minor fraction with poor outcome had isolated C4 lowering not captured by serum C3 measurements. Therefore, detailed knowledge of distinct complement components contributing to kidney injury could be of relevance to improve current strategies targeting the complement system in ANCA-associated renal vasculitis. Graphical abstract


Introduction
Anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) is a small vessel vasculitis affecting multiple organ systems, including the kidney. Small vessels in the kidney include small-sized arteries, capillaries, and venules. Disease manifestations in the kidney are common and usually characterized by pauci-immune glomerulonephritis with only minor, if any, immunoglobulin and complement depositions in the vascular system. On a mechanistic level, pathogenic ANCA autoantibodies activate neutrophils, causing a release of inflammatory cytokines, reactive oxygen species, and lytic enzymes, resulting in inflammation and vascular injury. The activation of the complement system contributes essentially to the pathogenesis of ANCA-associated renal vasculitis by autoantibody-antigen recognition directed against host cells [1]. Activation of the complement system induces neutrophil adhesion and formation of neutrophilplatelet aggregates in vascular endothelial cells, implicating a direct mechanistic link between complement activation and vascular injury [1]. The measurement of serum complement C3 and C4 with immunoassays is routinely used in clinical practice to determine and monitor complement activation. Importantly, low serum complement C3 levels without hypocomplementemia per se are an independent predictor of poor renal prognosis in patients with ANCA-associated renal vasculitis [2][3][4][5]. By contrast, low serum complement C4 levels alone were not associated with renal outcome [2]. However, serum levels of complement C4 have not been directly compared and analyzed in combination with serum C3 levels. Therefore, we here aimed to directly compare levels of C3 and C4 with regard to its clinical implications in ANCA-associated renal vasculitis.

Study population
A total of 38 cases with ANCA-associated renal vasculitis at the University Medical Center Göttingen were retrospectively included between 2015 and 2020 (Table 1), the patient cohort has previously been described [3,[6][7][8][9][10]. While no formal approval was required for the use of routine clinical data, a favorable ethical opinion was granted by the local Ethics committee (no. 22/2/14 and 28/09/17). At admission, the Birmingham Vasculitis Activity Score (BVAS) version 3 was assessed [11]. Medical records were used to obtain data on age, sex, duration of disease onset before admission, diagnosis (MPA or GPA) and laboratory results. The estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation [12]. Kidney replacement therapy (KRT) was performed intermittently in all cases. Indications for KRT included severe electrolyte and acid-base abnormalities, volume overload and encephalopathy.

Renal histopathology
A renal pathologist evaluated kidney biopsies and was blinded to data analysis. Periodic acid-Schiff stainings were performed by automated slide stainer Tissue-Tek Prisma (Sakura Finetek Europe, Alphen aan den Rijn, Netherlands) according to the manufacturer's protocol. Within a kidney biopsy, each glomerulus was scored separately for the presence of necrosis, crescents and global sclerosis. Based on these scores, histopathological subgrouping according to Berden et al. into focal, crescentic, mixed or sclerotic class was performed [13]. Furthermore, the ANCA renal risk score according to Brix et al. into low, medium or high risk was calculated [14].

C3 and C4 measurements
Plasma concentrations of human complement components C3c (9D9621, Abbott, Chicago, USA) and C4 (9D9721, Abbott, Chicago, USA) were determined by turbidimetric measurements on the ARCHITECT-C module. Normal range plasma concentrations for circulating C3c is defined between 0.82-1.93 g/L and C4 between 0.15-0.57 g/L.

Statistical methods
Variables were tested for normal distribution using the Shapiro-Wilk test. Statistical comparisons were not formally powered or prespecified. Non-normally distributed continuous variables are shown as median and interquartile range (IQR), categorical variables are presented as frequency and percentage. For group comparisons, the Mann-Whitney U-test was used to determine differences in medians. Non-parametric between-group-comparisons were performed with Pearson's Chi-square test. Serum levels of complement C3 and C4 were directly compared by simple regression analysis. Survival-curve analyses were performed using the Kaplan-Meier method, comparison of survival curves was performed with log rank (Mantel-Cox) testing. Data analyses were performed with GraphPad Prism (version 8.4.3 for MacOS, GraphPad Software, San Diego, California, USA). A probability (p) value of < 0.05 was considered statistically significant.  Fig. 1A). Low complement C4 below median serum levels (< 0.26 g/L) was only associated with older age, whereas kidney function, ANCA subtype, relapsing disease, systemic vasculitis activity, or classification of ANCA-associated renal vasculitis did not differ (Table 2). By contrast, low complement C3 below median levels (< 1.3 g/L) correlated with kidney injury requiring KRT within a follow-up of 40 days after diagnosis (p = 0.0167), not associated with any histopathological lesion or subtyping of ANCA-associated renal vasculitis (Table 3). Among 25/38 (65.8%) with low serum levels of C3 and/or C4, most cases of ANCA-associated renal vasculitis showed an overlap of low C3 and C4 levels (52%), and there were subsets of patients with isolated C4 lowering (24%, Fig. 1B). Direct comparison of serum complement C3 and C4 levels confirmed a positive association with each other (Fig. 1C). As compared to previously described prediction of poor  Fig. 1D) [3]. Combined C3 and/or C4 effectively predicted poor outcome (p = 0.0134, Fig. 1E), while direct comparison revealed no superiority of low C3 and/or C4 as compared to low complement levels of C3 alone (Fig. 1F). In the subgroup that experienced KRT or death, low serum complement C3 levels identified 11/14 (78.6%, p = 0.0071) and complement C4 levels 9/14 (64.3%, p = 0.1786) cases (Fig. 1G). Interestingly, 2/14 (14.3%) patients that experienced KRT or death had isolated C4 lowering, and combining low serum levels of C3 and/or C4 identified 13/14 (92.3%, p < 0.0001) cases in this subgroup (Fig. 1G). Non-superiority to predict poor outcome by low C3 and/or C4 as compared to low complement levels of C3 alone in the total cohort was attributed to a subset of 4/24 (16.7%) patients with isolated C4 lowering in the subgroup that did not experience KRT or death (Fig. 1H). In summary, this comparative analysis revealed superiority in predicting poor outcome by low levels of complement C3 as compared to C4 in ANCA-associated renal vasculitis. However, a

Discussion
Low serum complement C3 levels without hypocomplementemia per se are an independent predictor of poor renal prognosis in patients with ANCA-associated renal vasculitis [2]. We performed a comparative analysis with regard to serum C3 and C4 levels to test its clinical implications in ANCA-associated renal vasculitis. As observed for levels of C3, we identified C4 hypocomplementemia only in a minor subset. As previously described, we defined low serum C4 according to median levels and identified no association with risk of KRT requirement or death in ANCA-associated renal vasculitis [2]. While most cases of ANCA-associated renal vasculitis showed an overlap of low complement C3 and C4 levels, there were subsets of patients with isolated C4 lowering. As compared to previously described prediction of poor outcome by low levels of complement C3, low levels of complement C4 did not associate with KRT or death. Interestingly, a minor fraction that experienced KRT or death had isolated C4 lowering not captured by serum C3 measurements. These observations implicate overlapping, but also independent roles of complement C3 and C4 that might contribute to kidney injury in ANCA-associated renal vasculitis. This is in line with our previous observation in this cohort that low levels of either serum C3 or C4 correlated with distinct inflammatory lesions in ANCA-associated renal vasculitis [10]. Regarding potential clinical implications, low serum levels of C3 were associated with renal injury as previously and independently described [2][3][4][5]. In contrast, low levels of complement C4 were associated with older age in ANCA-associated renal vasculitis independent of kidney function, ANCA subtype, relapsing disease, systemic vasculitis activity, or classification of ANCAassociated renal vasculitis. While a relationship between age and distinct complement components has already been described in studies of healthy populations, this has not yet been observed in ANCA-associated renal vasculitis [15]. Its relevance related to age and treatment response remains elusive and requires further investigation. Currently, two C5a inhibitors are in clinical development for AAV: the oral C5a receptor (C5aR) inhibitor avacopan and the monoclonal C5a antibody IFX-1 [16]. Safety and efficacy with steroid-sparing effects of avacopan in patients with GPA/MPA have already been shown [16]. Furthermore, IFX-1 has entered Phase II development [16]. Therefore, detailed knowledge of distinct complement components contributing to kidney injury could be of relevance to further improve current strategies targeting the complement system, which is especially relevant in severe ANCA-associated renal vasculitis [17,18].
The main limitations of our study are its retrospective design and the small patient number. Furthermore, we only assessed outcome during the initial course of ANCA-associated renal vasculitis requiring validation in independent cohorts also regarding long-term renal and overall outcome. Nevertheless, our observation that a minor fraction with isolated C4 lowering was not captured by serum C3 measurements implies overlapping, but also independent roles of complement C3 and C4 in ANCA-associated renal vasculitis. Therefore, a detailed analysis of intrarenal complement C4 deposition in comparison with C3 deposits among other histopathological lesions would be of relevance to gain further insights into the role of distinct complement components in ANCA-associated renal vasculitis.
Author contributions BT conceived the study, analyzed data and wrote the manuscript. DT, EB and SH collected and analyzed data. All authors reviewed and approved the manuscript's content before submission.
Funding Open Access funding enabled and organized by Projekt DEAL. BT was supported by the Research program, University Medical Center, University of Göttingen (1403720). EB was funded by the Else-Kröner research program entitled "molecular therapy and prediction of gastrointestinal malignancies" (7-67-1840876). We also acknowledge support by the Open Access Publication Funds of the Göttingen University. The funding sources were not involved in the design, collection, analysis, interpretation, writing or decision to submit the article.
Availability of data and material Deidentified data are available on reasonable request from the corresponding author.

Declarations
Conflict of interest Each author certifies that he, or a member of their immediate family, has no commercial associations (e.g., consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the contents of the submitted article. The results presented in this article have not been published previously in whole or part, except in abstract format.
Ethical statement All patients consented to the use of routinely collected data as part of their regular medical care. All studies involving human participants were reviewed and approved by the Institutional Review Board of the University Medical Center Göttingen, Germany (protocol numbers 22/2/14 and 28/09/17).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.