Introduction

Anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) is an autoimmune disease characterized by necrotizing small-vessel vasculitis, commonly involving the kidneys as ANCA-associated glomerulonephritis (ANCA-GN). Although traditionally classified as “pauci-immune” glomerulonephritis based on histopathology, emerging evidence suggests that abnormal activation of the complement system, particularly the alternative pathway, plays a critical role in the pathogenesis and progression of AAV1,2,3,4,5.

Previous research has confirmed that complement activation products such as C5a participate in the pathologic process of AAV1,6,7,8. However, the dynamic changes and clinical implications of C3 cleavage fragments—key intermediates of complement activation, including C3a, C3b, iC3b, C3c, and C3d—have not been fully elucidated in AAV. In particular, reliable biomarkers for real-time and non-invasive monitoring of renal-localized complement activation are still lacking9,10,11.

Therefore, this study aimed to detect the urinary expression profiles of multiple C3 cleavage products in AAV patients using enzyme-linked immunosorbent assay (ELISA). By analyzing the correlations between these fragments and clinical indicators such as the Birmingham Vasculitis Activity Score (BVAS) and proteinuria, we sought to explore the potential clinical utility of these fragments as biomarkers. Our findings may provide new insights into the role of complement activation in AAV pathogenesis and offer tools for non-invasive disease monitoring in clinical practice.

Subjects and methods

Study subjects

A total of 42 participants were enrolled, including 22 AAV patients and 20 healthy controls. Data from AAV patients were collected from the outpatient and inpatient departments of Wuhan No.1 Hospital between July 2024 and June 2025. The study protocol complied with the Declaration of Helsinki and was approved by the Medical Ethics Committee of Wuhan No.1 Hospital (Approval No. [2024]51).

Inclusion criteria for AAV patients were: diagnosis of myeloperoxidase (MPO)-AAV according to the Improving Global Outcomes (KDIGO) 2024 guideline which relies on the combination of clinical findings and results of imaging studies and laboratory tests (such as C-reactive protein level, complete blood count, kidney parameters, and urine sediment analysis12; positive MPO-ANCA or perinuclear ANCA (P-ANCA) detected by ELISA or immunofluorescence; Histopathologic confirmation, especially by kidney biopsy, is recommended when feasible and safe; age 18–80 years; and informed consent provided. Exclusion criteria included secondary AAV (e.g., drug-induced or infection-associated vasculitis); coexisting autoimmune diseases such as anti-glomerular basement membrane nephritis, systemic lupus erythematosus, IgA nephropathy, or membranous nephropathy; end-stage renal disease; severe infections; liver dysfunction (transaminases > 3× upper normal limit); malignancies; use of complement-targeting therapies within 3 months; participation in other clinical researches within 30 days; pregnancy or lactation; or other conditions deemed unsuitable by investigators.

Healthy controls were age- and sex-matched individuals with normal renal function (eGFR ≥ 90 mL/min/1.73 m², or 60–89 mL/min/1.73 m² for those over 60 with normal urinalysis), no history of kidney disease, diabetes, hypertension, autoimmune disorders, or recent infections (CRP < 5 mg/L), and no recent use of glucocorticoids or complement-affecting medications.

Methods

Data collection and testing

This cross-sectional study collected 10 mL of morning midstream urine from all participants. Samples were centrifuged at 3000 rpm for 10 min, and supernatants were stored at − 80 °C until analysis. Urinary creatinine was measured to normalize complement fragment levels. The ratio of each C3 cleavage product to urinary creatinine (ng/mg) was calculated.

Urinary C3a, C3b, iC3b, C3c, and C3d levels were quantified using commercial ELISA kits (Ruixin Biotech, Wuhan, China) according to manufacturers’ instructions. Samples were thawed at 4 °C, kept on ice, and analyzed within 1 h to avoid repeated freeze-thaw cycles.

Clinical parameters collected included serum creatinine (µmol/L), 24-hour urinary protein (g/24 h), hematuria graded 0–3+, and BVAS version 3 to assess disease activity. BVAS scoring was performed independently by two physicians, with discrepancies > 2 points resolved by a third. The urine complement data were temporally matched with the other clinical activity indicators (e.g., BVAS score) at that specific study baseline time point.

Renal Biopsy Status: In this cohort, 9 out of 22 patients underwent renal biopsy. All biopsies confirmed typical pathological findings consistent with AAV glomerulonephritis (e.g., necrotizing crescentic glomerulonephritis). The remaining patients (n = 13) were diagnosed based on the KDIGO principles and clinical judgment (detailed diagnostic approach see supplementary Methods), primarily due to factors such as advanced kidney injury requiring dialysis (making biopsy high-risk and low-yield) or patient/family refusal due to perceived procedural risks. A descriptive comparison was performed between patients with and without kidney biopsy to assess potential differences in baseline characteristics (see supplementary Table 1).

Statistical methods

Normally distributed data were analyzed by independent sample t-tests and presented as mean ± standard deviation. Non-normally distributed data used Mann–Whitney U tests and were expressed as median (interquartile range). Spearman’s correlation assessed associations between urinary complement fragments and clinical indicators. Multivariate linear regression with a stepwise approach identified independent predictors (p < 0.05). Model fitness was evaluated by R2 and F-tests, and multicollinearity assessed via variance inflation factor (VIF). Analyses were performed using SPSS 26.0 (IBM Corp), with two-sided p < 0.05 considered statistically significant.

Results and discussion

Results

Demographic and general data

The study included 22 AAV patients (3 males, 19 females) with a mean age of 71.9 ± 6.2 years, and 20 healthy controls (2 males, 18 females) with a mean age of 69.4 ± 9.5 years. AAV patients had a mean proteinuria of 1.74 ± 1.01 g/24 h, mean serum creatinine of 261.27 ± 120.44 µmol/L, and mean BVAS score of 7.14 ± 7.97. Healthy controls had normal renal function and no proteinuria or hematuria (see Table 1).

Table 1 Clinical and general data comparison between AAV and healthy control groups.
Full size table

Comparison of urinary complement C3 fragments between AAV patients and healthy controls

All measured urinary complement C3 fragments (C3a, C3b, iC3b, C3c, C3d), normalized to urinary creatinine, were significantly elevated in AAV patients compared to healthy controls (all p < 0.01) (see Table 2; Fig. 1)

Table 2 Comparison of urinary complement C3 fragments between AAV patients and healthy controls.
Full size table
Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Comparison of corrected urinary C3 fragment levels between AAV patients and healthy controls. (A) C3a/Cr; (B) C3b/Cr; (C) iC3b/Cr; (D) C3c/Cr; (E) C3d/Cr. Note: Horizontal lines represent mean/median, and dots represent individual values.

Correlation and multivariate analysis of urinary complement C3 fragments with clinical parameters

Spearman correlation showed positive correlations between urinary C3 fragments and BVAS scores (r = 0.793–0.900, all p < 0.001), with C3c exhibiting the strongest correlation (r = 0.900). Proteinuria (r = 0.551–0.735) and hematuria (r = 0.643–0.752) were also significantly correlated with all C3 fragments (all p < 0.01). No significant correlations were found between urinary C3 fragments and serum creatinine (p > 0.05) (See Table 3; Fig. 2).

Table 3 Spearman correlation analysis of variables in the AAV group.
Full size table

Further multivariate regression analysis indicated that BVAS score emerged as an independent predictor of all corrected urinary complement C3 fragments (β = 0.458–0.760, all p < 0.05). Proteinuria was independently predictive of all fragments except iC3b (β = 0.302–0.455, all p < 0.05). The regression models demonstrated good fit (R2 = 0.578–0.880), with no multicollinearity detected (VIF < 2) (see Table 4). The correlation analysis results between urinary C3 fragments and clinical parameters with independent predictive value are shown in Fig. 2.

Table 4 Multivariate regression analysis of urinary complement components and clinical parameters (stepwise regression).
Full size table
Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Correlation analysis of urinary C3 fragments with independent predictors in the AAV group. (A,B) C3a/Cr with BVAS score and proteinuria; (C,D) C3b/Cr with BVAS score and proteinuria; (E) iC3b/Cr with BVAS score; (F,G) C3c/Cr with BVAS score and proteinuria; (H,I) C3d/Cr with BVAS score and proteinuria.

Discussion

In this study, we analyzed urinary levels of five complement C3 fragments—C3a, C3b, iC3b, C3c, and C3d—in patients with MPO-ANCA–associated vasculitis (AAV), and found that all were significantly elevated compared to healthy controls. Among clinical parameters, BVAS score emerged as a consistent independent predictor of all fragments, reflecting their strong association with disease activity. The particularly strong correlation between urinary C3 fragments and BVAS directly confirms that systemic disease activity is closely coupled with intense, localized complement cleavage within the kidney. Thus, these urinary fragments serve as sensitive, non-invasive reporters of the real-time immunological processes and active vasculitic inflammation at the renal site. Proteinuria was also an independent predictor for all fragments except iC3b, suggesting a potential link between structural glomerular damage and complement dysregulation. In contrast, none of the fragments correlated with serum creatinine, implying that these complement products more likely reflect localized renal immune activity rather than overall glomerular filtration capacity.

The lack of association between urinary complement fragments and renal function suggests that these fragments are not merely passively filtered from systemic circulation but are more likely generated or activated within the kidney itself. They may result from local complement synthesis or dysregulation in glomerular endothelial cells, mesangial cells, podocytes, or tubular epithelial cells13,14,15,16. Increasing evidence supports that complement components such as C3 can be synthesized locally under inflammatory conditions, leading to enhanced C3b deposition and cascade amplification5,13. Glomerular and tubular cells express complement regulatory proteins (e.g., CD55, CD59) and receptors (e.g., CR1), which play critical roles in maintaining complement balance5,10. For instance, CD55 accelerates C3/C5 convertase decay, CD59 inhibits MAC formation, and CR1 promotes C3b clearance13. Tubular epithelial cells may also synthesize C3, C4, and factor H, and clear C3b and iC3b via endocytosis17,18. Local complement activation in AAV may be further driven by intrarenal ANCA–complement–neutrophil feedback loops, leading to tissue injury and shedding of complement fragments into the urine. These findings support the potential of urinary complement profiling as a non-invasive indicator of renal immune activity. Proteinuria, a hallmark of glomerular barrier disruption, is strongly associated with most urinary complement fragment elevation and may itself stimulate tubular and interstitial cells to activate proinflammatory pathways and further complement cleavage18,19,20. The generation of iC3b involves factor I–mediated cleavage of C3b with the aid of factor H, and may be subject to systemic regulation rather than being solely proteinuria-driven, which warrants further investigation.

These observations are consistent with the established pathogenesis of AAV, which centers on the interplay between ANCAs, neutrophils, and the complement system, particularly the alternative pathway1,2,6,8,21,22. ANCA binding to neutrophil surface antigens (e.g., MPO) leads to activation and release of C5a, promoting neutrophil degranulation through the C5a–C5aR1 axis. This forms a positive feedback loop that amplifies inflammation2,6,23. Activated neutrophils release properdin to stabilize C3 convertase (C3bBb), initiating further C3 cleavage into C3a, C3b, and downstream fragments (iC3b, C3c, and C3d)24. Neutrophil extracellular traps (NETs), containing complement proteins and tissue factors, contribute to continued local activation and antigen presentation3,10,11,23. The elevated urinary levels of these fragments observed in this study likely reflect this ongoing renal complement activity.

Each fragment plays distinct roles in the inflammatory response6,10,11,25,26,27,28. C3a acts as a potent anaphylatoxin, increasing vascular permeability and recruiting immune cells10. C3b enhances phagocytosis through opsonization and serves as a key part of the C3 convertase complex. iC3b, though incapable of forming convertase, retains opsonizing function and helps modulate inflammation through interaction with CR3/CR4 and CRIg10,25. C3c and C3d are terminal breakdown products, reflecting complement cascade intensity. C3d also contributes to B-cell activation and antibody production. While these functions are well defined in immunology, their specific roles in AAV pathophysiology remain to be elucidated. The fragment-specific patterns observed in this study may reflect underlying differences in their generation, regulation, or renal handling. For instance, iC3b’s unique association with disease activity—but not with proteinuria—suggests that it may be more tightly controlled by inflammatory signaling and complement regulatory mechanisms (e.g., factor H, factor I), rather than glomerular barrier disruption alone. This suggests that urinary iC3b might be a more specific biomarker, warranting future research attentions.

Previous studies have mainly focused on plasma complement components, showing that plasma levels of C3a, C5a, C5b-9, and Bb are correlated positively with AAV activity21,29,30,31, while factor H (CFH), a negative regulator of the alternative pathway, shows an inverse correlation32.However, their clinical utility as independent biomarkers is limited. Studies have found that although complement activation products are significantly elevated during the active phase, clinical remission following immunosuppressive therapy does not necessarily normalize the levels of various complement components33,34,35,36. In contrast, urinary complement fragments, due to their higher organ specificity, may serve as more accurate indicators of intrarenal complement activation and immune-mediated damage. In diseases like lupus and IgA nephropathy, urinary complement levels correlate well with histopathological activity and may outperform plasma markers34,37. While some studies have explored urinary complement in AAV, findings remain limited and inconsistent22,33,35. Our study provides a more comprehensive overview of urinary C3 fragments, supporting their potential as biomarkers.

Given the predominance of MPO-ANCA in China (~ 80% of AAV cases), even among patients with GPA (60–70%)3, our focus on this population helps clarify renal injury mechanisms specific to Chinese patients—particularly the involvement of the complement pathway. Although these patients generally respond to cyclophosphamide and MMF, they face higher rates of treatment resistance and infection. Therefore, studying this group may support the development of more effective, tailored therapies, including complement-targeted strategies.

Several limitations should be acknowledged. The relatively small sample size may compromise statistical power and limit the generalizability of findings to broader AAV populations, necessitating larger, multicenter cohorts to enhance robustness. The cross-sectional design restricts our ability to evaluate dynamic changes in urinary complement fragments over time or in response to therapeutic interventions, underscoring the need for longitudinal studies to assess their predictive value. A key limitation of the current study is the lack of universal renal biopsy data. While we adhered to clinical guidelines (integrating typical clinical presentation and laboratory parameters) and strongly recommended renal biopsy for all patients, 13 individuals did not undergo the procedure due to factors such as high clinical risk assessment (e.g., advanced kidney injury, coagulopathy), significant clinical improvement, and patient refusal (attributed to cultural or economic reasons). This decision reflected the necessity of balancing diagnostic benefit against procedural safety, particularly in patients whose condition had already stabilized. The absence of histology for this subset limits the comprehensive pathological staging of the entire cohort and means we cannot definitively exclude rare coexisting conditions, such as AAV combined with anti-GBM-negative anti-GBM disease. Furthermore, due to the limited number of biopsied patients (n = 9), we were unable to perform statistically meaningful correlation analyses between pathological features (such as Berden classification or the percentage of active lesions/crescents) and urinary C3 fragment levels, which restricted the depth of our clinicopathological correlation. Not all patients underwent kidney biopsy, which may introduce selection bias; however, clinical descriptive subgroup analyses did not reveal marked differences between biopsy and non-biopsy patients. Additionally, the lack of parallel serum complement measurements or renal biopsy data hinders a comprehensive understanding of complement activation in AAV-associated glomerulonephritis. Potential confounders, such as concurrent medications, infections, or urine collection variability, may influence biomarker levels, highlighting the need for standardized protocols and adjustment for urine concentration. Furthermore, the single-center setting limits applicability across diverse populations, and the absence of functional or multi-omics analyses restricts insights into the mechanistic roles of complement fragments. Future studies should integrate multi-level complement data, employ advanced omics approaches, and validate findings in diverse, longitudinal cohorts to fully elucidate the clinical utility of these biomarkers in AAV.

In conclusion, this study demonstrated that urinary complement C3 fragments are elevated in MPO-ANCA–associated vasculitis and closely associated with disease activity. These fragments may serve as sensitive, non-invasive biomarkers for renal immune activity and inflammation, providing novel tools for disease monitoring and potential therapeutic guidance in AAV.