Very small embryonic-like stem cells: A predictor synergizing with complement and inflammatory factors for immunoglobulin A nephropathy prognosis

Abstract

BACKGROUND

Immunoglobulin A nephropathy (IgAN), a common primary glomerular disease, may progress to end-stage renal disease (ESRD). Stem cells regulate renal repair by countering mesangial injury, but the prognostic value very small embryonic-like stem cells (VSELs) and their synergy with complement/inflammatory biomarkers in IgAN remain unclear. Single-pathway biomarkers are inaccurate for ESRD prediction, and combined stem cell-complement/inflammatory risk stratification awaits validation.

AIM

To clarify the independent prognostic value of VSELs, complement factor I (CFI), C3a, and tumor necrosis factor-α (TNF-α) in IgAN patients and explore whether their combined detection can improve the accuracy of ESRD prediction.

METHODS

A retrospective analysis was conducted on 255 IgAN patients from our hospital, who were stratified into ESRD (31 cases) and non-ESRD (224 cases) groups based on follow-up outcomes. Complement alternative pathway indices included complement factor B, CFI, C3a, complement factor H, and sC5b-9; inflammatory biomarkers included monocyte chemoattractant protein-1, interleukin-6, TNF-α, B-cell activating factor, and C-X-C motif chemokine ligand 10; stem cell indicators included VSELs, hematopoietic stem cells, and endothelial progenitor cells. Statistical methods included t-tests, χ² tests, Spearman correlation, Logistic regression, receiver operating characteristic curves, and Kaplan-Meier survival analysis.

RESULTS

IgAN patients in ESRD had lower complement factor B, CFI, C3a, and higher monocyte chemoattractant protein-1, interleukin-6, TNF-α, and VSELs compared to non-ESRD (all P < 0.001), with CFI, C3a, TNF-α, and VSELs as independent ESRD predictors (all P < 0.05). The combined detection of CFI, TNF-α, and VSELs showed a higher predictive value for ESRD [area under the curve (AUC) = 0.916] than single indicators (CFI: AUC = 0.660; TNF-α: AUC = 0.818; VSELs: AUC = 0.749). Spearman correlation analysis revealed that VSELs were weakly positively correlated with CFI and TNF-α, while estimated glomerular filtration rate was positively correlated with CFI and negatively correlated with TNF-α (all P < 0.05). High VSELs indicated poorer renal prognosis (P < 0.001), with no significant differences in renal survival for CFI or TNF-α (all P > 0.05).

CONCLUSION

VSELs, along with CFI and TNF-α, are independent predictors of ESRD in IgAN patients, with VSELs providing a more reliable indicator of poor renal prognosis. Their combined detection achieves high predictive accuracy (AUC = 0.916), offering a non-invasive prognostic tool independent of Oxford classification features, thus enhancing risk stratification and individualized intervention strategies.

Key Words: Complement alternative pathway; Stem cell indicators; Inflammatory markers; Immunoglobulin A nephropathy; Estimated glomerular filtration rate; End-stage renal disease

Core Tip: As a key stem cell indicator, very small embryonic-like stem cells, combined with complement alternative pathway indices and inflammatory biomarker tumor necrosis factor-α, are independent predictors of end-stage renal disease in immunoglobulin A nephropathy patients. Their combined detection yields high predictive accuracy, outperforming single indicators and enabling reliable risk stratification. High very small embryonic-like stem cells correlate with poor renal prognosis, and these serological markers predict outcomes independent of Oxford classification, providing a non-invasive alternative to renal biopsy and enriching individualized intervention biomarkers.

INTRODUCTION

Immunoglobulin A nephropathy (IgAN) is the most prevalent primary glomerulonephritis worldwide. Its hallmark is dominant or co-dominant immunoglobulin A (IgA) deposition in the mesangium[1]. The disease is driven by an aberrantly glycosylated form of IgA1 whose hinge-region lacks galactose. This glyco-deficient IgA1 elicits auto-antibodies that form circulating immune complexes. These complexes become trapped in the mesangium, triggering inflammation and progressive glomerular injury[2,3]. Diagnosis still requires a renal biopsy that shows IgA-dominant deposits by immunofluorescence, yet only the galactose-deficient IgA1 fraction appears pathogenic. The “multi-hit” hypothesis - aberrant IgA1 glycosylation, auto-antibody generation, immune-complex formation and mesangial activation - best explains the clinical heterogeneity of IgAN, although the precise molecular choreography remains incompletely understood[4,5]. The estimated glomerular filtration rate (eGFR) is the most widely used metric of kidney filtration. Both the absolute value and the slope of eGFR decline predict the risk of progression to end-stage renal disease (ESRD)[6]. IgAN, the commonest primary glomerulonephritis worldwide, carries one of the highest lifetime risks of ESRD among glomerular diseases[7]. In IgAN, eGFR is strongly and independently associated with ESRD. Incorporating the degree of proteinuria refines this risk stratification further[8]. Hypertension, heavy proteinuria and hypoalbuminemia - core clinical phenotypes of IgAN - track closely with eGFR decline and subsequent renal failure[9,10]. Thus, serial eGFR measurements serve as a dynamic biomarker that captures the interplay between structural damage and clinical phenotype, guiding both prognosis and the timing of nephro-protective interventions in patients with IgA nephropathy.

Stem cells are emerging as multi-layered regulators of outcome in IgAN[11]. Very small embryonic-like stem cells (VSELs) - a rare population that expresses pluripotency markers and can generate derivatives of all three germ layers - are selectively mobilized in IgAN, suggesting an endogenous attempt to counteract chronic mesangial injury[12,13]. Hematopoietic stem cells (HSCs) and endothelial progenitor cells (EPCs) add further complexity. EPCs home to sites of microvascular damage, amplify angiogenesis and accelerate renal repair, correlating with better preservation of eGFR after acute flares[14]. HSCs, in turn, sustain both hematopoietic and tissue-resident cell pools, and their mobilization kinetics predict histologic severity and long-term function in IgAN cohorts[15]. Collectively, these data position stem-cell trafficking as a real-time barometer of reparative capacity and imply that harnessing VSELs, EPCs or HSCs could open novel diagnostic and therapeutic avenues in IgA nephropathy.

The complement alternative pathway (AP) and its inflammatory footprints have moved to the center of IgAN research. Constitutively active at low levels, the AP can be amplified by surface-bound IgA1-containing immune complexes; once unleashed, it deposits C3 and the terminal C5b-9 membrane-attack complex in the mesangium, creating a tight linear correlation between complement load and the rate of eGFR loss[16]. Glomerular C3 fragments and soluble C3a/C5a anaphylatoxins perpetuate inflammation by engaging C3aR/C5aR on resident cells, an axis now recognized as a principal driver of tubulo-interstitial scarring and glomerulosclerosis[16,17]. Serum C4 - classically viewed as part of the lectin or classical routes - paradoxically rises in parallel with AP activity and inversely tracks eGFR, serving as an easily measurable surrogate of complement-dependent injury[18]. These mechanistic insights have already advanced into the therapeutic pipeline: Selective inhibitors of factor B, properdin, and C5aR are being repurposed from rare complement-mediated disorders to IgAN, offering the prospect of pathway-specific nephroprotection[19,20].

A rapidly expanding literature now maps the inflammatory signature of IgAN onto its clinical trajectory. In the urine, a panel of cytokines - B-cell activating factor (BAFF), monocyte chemoattractant protein-1 (MCP-1), C-X-C motif chemokine ligand 10 (CXCL10) and interleukin-6 (IL-6) - rises in tandem with proteinuria and predicts future eGFR loss with an accuracy that rivals histologic grading[21]. Systemically, soluble tumor necrosis factor (TNF) receptors and other circulating inflammatory proteins mirror the intensity of tubulo-interstitial damage, offering a non-invasive read-out of disease activity that correlates tightly with interstitial fibrosis at biopsy[22]. Collectively, these soluble mediators serve as a real-time molecular gauge: Serial measurements can identify subsets of patients whose renal function is deteriorating rapidly, long before serum creatinine levels start to rise. This refines risk stratification and facilitates timely, targeted therapeutic interventions. The AP is more than a complement cascade - it is a self-amplifying inflammatory loop. Primed by TNF-α or IL-1β, neutrophils assemble surface-bound AP C3 convertase; the ensuing C5a feeds back to up-regulate CD11b and escalate the respiratory burst, turning each granulocyte into a miniature complement factory that sustains and magnifies inflammation[23]. In membranoproliferative glomerulonephritis and C3 glomerulopathy, auto-antibodies such as C3 nephritic factor freeze the C3bBb enzyme in its active state, unleashing uncontrolled C3a and C5b-9 generation that directly injures the glomerular filter[24]. Conversely, blockade of properdin - the AP’s essential stabilizer - collapses this circuit: C3a and sC5b-9 levels plummet, neutrophil CD11b dims and platelet activation wanes, confirming the pathway’s role as a proximal hub of inflammatory amplification[25]. Thus, AP activity is not merely a biomarker; it is a druggable driver shared across diverse inflammatory diseases.

Given that IgAN is the most prevalent primary glomerulonephritis worldwide with a substantial risk of progression to ESRD, there remains a lack of precise prognostic indicators in clinical practice. Additionally, research on the combined application of alternative complement pathway indices, inflammatory biomarkers, and stem cell indicators for IgAN prognosis is still scarce. Therefore, this study retrospectively enrolled 255 IgAN patients, who were stratified into the ESRD group and non-ESRD group. Employing a range of analytical approaches including group comparisons, logistic regression, receiver operating characteristic (ROC) curve analysis, Spearman correlation analysis, and Kaplan-Meier survival analysis, we systematically investigated the associations between alternative complement pathway indices, inflammatory biomarkers, stem cell indicators and the progression of IgAN to ESRD. We identified independent predictors, validated the prognostic value of combined detection of multiple indicators, and analyzed the correlations between key indicators and eGFR as well as Oxford classification pathological features. This study aims to provide a scientific basis for risk stratification, prognostic assessment, and individualized intervention of IgAN, thereby filling the research gap regarding the combined application of multi-dimensional biomarkers in this field.

MATERIALS AND METHODS

Study design and subjects

Based on the reported 25% ESRD incidence in patients with IgAN[26], the results demonstrated that a minimum sample size of 170 participants was required to ensure adequate statistical power for the study. We retrospectively screened 313 patients who underwent native kidney biopsy at Jiaxing Hospital of Traditional Chinese Medicine and received a pathological diagnosis of IgAN. After application of predefined inclusion and exclusion criteria, 255 patients were retained for analysis. Participants were stratified into a ESRD group and a non-ESRD group according to whether they reached the composite renal endpoint during follow-up. The study protocol was approved by the Institutional Ethics Committees of Jiaxing Hospital of Traditional Chinese Medicine (Approval No. JXTCM-IRB-2025-091) and adhered to the principles of the Declaration of Helsinki. Inclusion criteria[22,27]: (1) Primary IgAN confirmed by native kidney biopsy; (2) Complete clinical and histopathologic data available; (3) ≥ 8 glomeruli present in the biopsy specimen; and (4) Written informed consent obtained. Exclusion criteria: (1) Secondary IgAN (e.g., Henoch-Schönlein purpura nephritis, hepatitis-B-associated IgAN); (2) Co-existing chronic kidney disease entities (membranous nephropathy, diabetic nephropathy, antineutrophil cytoplasmic antibody-associated glomerulonephritis, drug-induced tubulointerstitial disease, etc.); (3) Systemic autoimmune disorders (systemic lupus erythematosus, rheumatoid arthritis, ankylosing spondylitis, etc.); (4) Chronic liver disease or active malignancy; and (5) Incomplete medical records or missing biopsy data.

Data and sample collection

The included data comprise: Demographic characteristics (gender, age, body mass index), clinical manifestations (history of hypertension, systolic blood pressure, diastolic blood pressure). Peripheral venous blood samples were collected from all enrolled patients at the time of initial hospital admission using standardized EDTA-anticoagulant vacutainer tubes. All specimens were promptly centrifuged at 1500 × g for 15 minutes at room temperature for immediate processing. The isolated plasma supernatants were aliquoted into cryopreservation vials and stored at -80 °C in ultra-low temperature freezers. The routine blood indicators include: Thrombocyte, neutrophil, white blood cell, lymphocyte counts. The blood biochemical parameters include: 24-hour proteinuria, albumin, urea nitrogen, creatinine, uric acid, triglyceride, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol.

Renal pathological evaluation

Based on the PAS staining sections of the pathological tissues obtained from each patient’s percutaneous renal biopsy, according to the pathological evaluation criteria of the Oxford classification[21,28], the following parameters were evaluated: Glomerular mesangial cell proliferation (M), capillary proliferative lesions (E), segmental sclerosis/bulbar adhesion (S), tubular atrophy/interstitial fibrosis (T), and cellular/fibrous crescent formation (C).

Enzyme-linked immunosorbent assay

Serum levels of alternative complement pathway proteins, including complement factor H (Abcam, Cambridge, Cambridgeshire, United Kingdom), complement factor B (CFB, R&D Systems, Minneapolis, MN, United States), complement factor I (CFI, Thermo Fisher Scientific, Waltham, MA, United States), soluble terminal complement complex (sC5b-9, Quidel Corporation, San Diego, CA, United States), and C3a (BD Biosciences, Franklin Lakes, NJ, United States), were quantitatively determined by enzyme-linked immunosorbent assay. Additionally, serum concentrations of inflammatory mediators - BAFF (BioLegend, San Diego, CA, United States), MCP-1 (R&D Systems, Minneapolis, MN, United States), CXCL10 (eBioscience, San Diego, CA, United States), IL-6 (Thermo Fisher Scientific, Waltham, MA, United States), and TNF-α (BD Biosciences, Franklin Lakes, NJ, United States) - were measured in strict accordance with the manufacturers’ respective standard protocols.

Flow cytometry

Circulating VSELs, HSCs, and EPCs were quantified using flow cytometry with the stain-then-lyse-then-wash protocol[12]. Briefly, 600 μL of fresh EDTA-anticoagulated whole blood was incubated with a panel of specific monoclonal antibodies for 30 minutes at room temperature in the dark. Thereafter, 3 mL of FACS lysing solution was added, and the samples were incubated for an additional 15 minutes under the same conditions. Following two washes with phosphate buffered saline, the cells were fixed with CellFix (Becton Dickinson, NJ, United States) and analyzed on a FACSCanto flow cytometer (Becton Dickinson, NJ, United States). Fluorescence-minus-one controls were employed to ensure precise compensation adjustment and gating. Data were analyzed using FlowJo software (TreeStar, Ashland, OR, United States). VSELs were gated as small (2-4 μm)/Lineage (lin)-/CD235a-/CD45-/CD133+ cells, whereas HSCs were defined as small (2-4 μm)/Lin-/CD235a-/CD45+/CD133+ cells. The 2-4 μm gate was calibrated using size-predefined beads, with a low forward scatter threshold set. For EPC quantification, peripheral blood mononuclear cells were first gated based on forward scatter and side scatter signals (debris and granulocytes excluded), followed by sequential gating for CD34+, CD34+CD309+, and ultimately CD34+CD309+CD133+ phenotypes to identify EPCs.

Follow-up and survival analysis

The composite renal function endpoint indicator is the occurrence of ESRD, defined as eGFR < 15 mL/minute/1.73 m² or initiation of renal replacement therapy[26,29,30]. Follow-up began on the biopsy date and ended at the first endpoint event or last clinic visit. The median follow-up time was 42 months.

Statistical analysis

Analyses were performed with SPSS 27.0 (IBM Corp., Chicago, IL, United States) and GraphPad Prism 9.5 (GraphPad Software, San Diego, CA, United States). Continuous variables were tested for normality (Kolmogorov-Smirnov). Normally distributed data are presented as mean ± SD and compared by independent-samples t-test. Categorical variables are given as n (%) and compared by χ² test. Correlations were assessed with Spearman’s coefficient. ROC curves evaluated predictive performance. Multivariable binary logistic regression identified independent risk factors. Kaplan-Meier survival curves were compared by log-rank test. Cox proportional-hazards models (univariable and multivariable) estimated hazard ratios with 95% confidence intervals (CI) for the composite endpoint. All tests were two-tailed; P < 0.05 was considered statistically significant.

RESULTS

Comparison of clinical baseline data between ESRD patients and non-ESRD patients

A total of 255 patients with IgAN were enrolled in this study, including 224 patients in the non-ESRD group and 31 patients in the ESRD group. The detailed results are presented in Table 1. There were no significant differences in gender composition, age, body mass index, history of hypertension, blood pressure level, and most blood routine and blood biochemical indicators (such as platelet, neutrophil, white blood cell, 24-hour proteinuria, albumin, etc.) between the two groups (all P > 0.05). Notably, the level of serum uric acid in ESRD group was significantly higher than that in non-ESRD group (344.85 ± 30.88 μmol/L vs 321.90 ± 33.48 μmol/L, t = 3.609, P < 0.001). There were no significant differences in the distribution of mesangial cell proliferation (M), endothelial cell proliferation (E), tubular atrophy/interstitial fibrosis (T) and crescent formation (C) between the two groups (all P > 0.05). However, the incidence of segmental sclerosis (S1) in the ESRD group was significantly lower than that in the non-ESRD group (35.48% vs 59.38%, χ² = 6.323, P = 0.012).

Table 1 Comparison of clinical baseline data between end-stage renal disease patients and non-end-stage renal disease patients, mean ± SD/n (%).

Indicators	Non-ESRD (n = 224)	ESRD (n = 31)	t/χ2	P value
Sex
Male	115 (51.3)	16 (51.0)	0.001	0.977
Female	109 (48.7)	15 (49.0)
Age (years)	38.46 ± 5.79	39.72 ± 9.21	0.996	0.320
BMI (kg/m2)	23.42 ± 3.13	22.98 ± 2.19	0.756	0.450
History of hypertension	63 (28.13)	9 (29.03)	0.011	0.916
Systolic pressure	129.64 ± 21.31	132.45 ± 22.32	0.684	0.495
Diastolic pressure	83.14 ± 11.14	86.47 ± 9.72	1.583	0.115
Thrombocyte (× 109/L)	259.64 ± 45.87	267.91 ± 38.85	0.957	0.339
Neutrophil (× 109/L)	3.88 ± 0.83	3.92 ± 0.77	0.381	0.801
White blood cell (× 109/L)	6.29 ± 0.75	6.37 ± 0.81	0.543	0.588
Lymphocyte (× 109/L)	1.71 ± 0.33	1.73 ± 0.37	0.324	0.746
24-hour proteinuria (g/d)	1.28 ± 0.17	1.31 ± 0.20	0.859	0.391
Albumin (g/L)	37.49 ± 5.96	38.71 ± 5.28	1.082	0.280
Urea nitrogen (mmol/L)	5.26 ± 0.53	5.31 ± 0.62	0.492	0.62.
Creatinine (μmol/L)	102.64 ± 12.53	98.89 ± 7.79	1.619	0.107
Uric acid (μmol/L)	321.90 ± 33.48	344.85 ± 30.88	3.609	< 0.001a
Triglyceride (mmol/L)	1.59 ± 0.27	1.53 ± 0.19	1.178	0.240
Total cholesterol (mmol/L)	4.08 ± 0.57	4.26 ± 0.66	1.613	0.108
High-density lipoprotein cholesterol (mmol/L)	0.89 ± 0.13	0.87 ± 0.18	0.712	0.477
Low-density lipoprotein cholesterol (mmol/L)	3.74 ± 0.42	3.89 ± 0.51	1.810	0.071
Oxford classification
M1	183 (81.70)	27 (87.10)	0.546	0.460
E1	113 (50.45)	20 (64.52)	2.160	0.142
S1	133 (59.38)	11 (35.48)	6.323	0.012a
T1-2	85 (37.95)	12 (38.71)	0.007	0.935
C1-2	121 (54.02)	16 (51.61)	0.063	0.801

^aP < 0.05.

ESRD: End-stage renal disease; BMI: Body mass index.

Comparison of the complement AP, inflammatory markers, and stem cell indicators between ESRD patients and non-ESRD patients

As shown in Table 2, in terms of alternative complement pathway indicators, the levels of CFB, CFI and C3a in ESRD group were significantly lower than those in non-ESRD group (CFB: 361.91 ± 77.08 mg/L vs 395.88 ± 85.99 mg/L, t = 4.112, P < 0.001; CFI: 50.68 ± 6.95 mg/L vs 55.44 ± 8.91 mg/L, t = 9.452, P < 0.001; C3a: 0.65 ± 0.07 mg/L vs 0.79 ± 0.11 mg/L,t = 6.366, P < 0.001), while the levels of complement factor H and sC5b-9 were not significantly different between the two groups (both P > 0.05). The levels of MCP-1, IL-6 and TNF-α in the ESRD group were significantly higher than those in the non-ESRD group (MCP-1: 289.56 ± 55.42 pg/mL vs 227.98 ± 57.01 pg/mL, t = 5.463, P < 0.001; IL-6: 23.46 ± 5.66 pg/mLvs 19.58 ± 3.58 pg/mL, t = 5.211, P < 0.001; TNF-α: 26.95 ± 4.31 pg/mL vs 21.65 ± 3.95 pg/mL, t = 6.924, P < 0.001) while BAFF and CXCL-10 levels were not significantly different between the two groups (both P > 0.05). In terms of stem cell indicators, the level of VSELs in ESRD group was significantly higher than that in non-ESRD group (62.94 ± 12.93 events/600 μL whole blood vs 52.87 ± 6.42 events/600 μL whole blood, t = 6.986, P < 0.001). However, the levels of HSCs and EPCs were not significantly different between the two groups (both P > 0.05).

Table 2 Comparison of the complement alternative pathway, inflammatory markers, and stem cell indicators between end-stage renal disease patients and non-end-stage renal disease patients, mean ± SD/n (%).

Indicators	Non-ESRD (n = 224)	ESRD (n = 31)	t/χ2	P value
CFH (mg/L)	612.75 ± 185.97	547.72 ± 149.54	1.864	0.063
CFB (mg/L)	395.88 ± 85.99	361.91 ± 77.08	4.112	< 0.001a
CFI (mg/L)	55.44 ± 8.91	50.68 ± 6.95	9.452	< 0.001a
SC5b-9 (mg/L)	0.52 ± 0.06	0.54 ± 0.18	1.330	0.185
C3a (mg/L)	0.79 ± 0.11	0.65 ± 0.07	6.366	< 0.001a
BAFF (ng/mL)	4.34 ± 0.25	4.41 ± 0.31	1.431	0.154
MCP-1 (pg/mL)	227.98 ± 57.01	289.56 ± 55.42	5.463	< 0.001a
CXCL-10 (pg/mL)	137.91 ± 25.94	145.05 ± 34.18	1.377	0.170
IL-6 (pg/mL)	19.58 ± 3.58	23.46 ± 5.66	5.211	< 0.001a
TNF-α (pg/mL)	21.65 ± 3.95	26.95 ± 4.31	6.924	< 0.001a
VSELs (events/600 μL whole blood)	52.87 ± 6.42	62.94 ± 12.93	6.986	< 0.001a
HSCs (events/600 μL whole blood)	81.47 ± 6.88	83.32 ± 7.47	1.419	0.157
EPCs (events/600 μL whole blood)	40.25 ± 5.42	38.57 ± 3.99	1.673	0.096

^aP < 0.05.

ESRD: End-stage renal disease; CFH: Complement factor H; CFB: Complement factor B; CFI: Complement factor I; BAFF: B-cell activating factor; MCP-1: Monocyte chemoattractant protein-1; CXCL-10: C-X-C motif chemokine ligand 10; IL-6: Interleukin-6; TNF-α: Tumor necrosis factor-α; VSELs: Very small embryonic-like stem cells; HSCs: Hematopoietic stem cells; EPCs: Endothelial progenitor cells.

Logistic regression model was used to analyze the influencing factors of ESRD occurrence in patients with IgAN

According to Table 3, univariate logistic regression analysis showed that segmental sclerosis (S1), serum uric acid, CFB, CFI, C3a, MCP-1, IL-6, TNF-α and VSELs levels were associated with the occurrence of ESRD in patients with IgA nephropathy (all P < 0.05). Among them, S1, CFB, CFI and C3a were protective factors [odds ratio (OR) values were all < 1], and serum uric acid, MCP-1, IL-6, TNF-α and VSELs were risk factors (OR values were all > 1). After adjusting for confounding factors, multivariate logistic regression analysis showed that CFI (OR = 0.656, 95%CI: 0.492-0.874, P = 0.004), C3a (OR = 0.970, 95%CI: 0.947-0.992, P = 0.009), TNF-α (OR = 2.055, 95%CI: 1.262-3.347, P = 0.004) and VSELs (OR = 1.859, 95%CI: 1.022-3.381, P = 0.042) were independent risk factors for ESRD in patients with IgA nephropathy.

Table 3 Logistic regression model was used to analyze the influencing factors of end-stage renal disease occurrence in patients with immunoglobulin A nephropathy.

Index	Univariate analysis		Multivariate analysis
	OR (95%CI)	P value	OR (95%CI)	P value
S1	0.376 (0.172-0.823)	0.014a	0.019 (0.001-0.485)	0.016a
Uric acid	1.022 (1.009-1.034)	< 0.001a	1.160 (0.902-1.492)	0.246
CFB	0.995 (0.991-1.000)	0.040a	0.987 (0.972-1.004)	0.128
CFI	0.938 (0.896-0.981)	0.006a	0.656 (0.492-0.874)	0.004a
C3a	0.989 (0.984-0.993)	< 0.001a	0.970 (0.947-0.992)	0.009a
MCP-1	1.019 (1.011-1.028)	< 0.001a	1.032 (1.000-1.066)	0.051
IL-6	1.266 (1.141-1.404)	< 0.001a	0.260 (0.023-2.895)	0.273
TNF-α	1.400 (1.238-1.583)	< 0.001a	2.055 (1.262-3.347)	0.004a
VSELs	1.159 (1.095-1.226)	< 0.001a	1.859 (1.022-3.381)	0.042a

^aP < 0.05.

OR: Odds ratio; CI: Confidence interval; CFB: Complement factor B; CFI: Complement factor I; MCP-1: Monocyte chemoattractant protein-1; IL-6: Interleukin-6; TNF-α: Tumor necrosis factor-α; VSELs: Very small embryonic-like stem cells.

The combined detection of CFI, TNF-α and VSELs can assist in predicting the occurrence of ESRD in patients with IgAN

As shown in Table 4 and Figure 1, ROC curve analysis demonstrated that the AUC values of CFI, TNF-α, and VSELs for predicting ESRD in patients with IgAN were 0.660 (95%CI: 0.598-0.718, P < 0.001), 0.818 (95%CI: 0.765-0.863, P < 0.001), and 0.749 (95%CI: 0.691-0.801, P < 0.001), respectively. The AUC of the combined detection of the three indicators was 0.916 (95%CI: 0.875-0.947, P < 0.001), which was significantly higher than that of each indicator alone, and the sensitivity and specificity were 83.87% and 90.62%, respectively, suggesting that the combined detection had higher predictive accuracy for ESRD in patients with IgA nephropathy.

Figure 1 The predictive value of individual and combined detection of complement factor I, tumor necrosis factor-α and very small embryonic-like stem cells for end-stage renal disease in patients with immunoglobulin A nephropathy. CFI: Complement factor I; TNF-α: Tumor necrosis factor-α; VSELs: Very small embryonic-like stem cells.

Table 4 The predictive value of individual and combined detection of complement factor I, tumor necrosis factor-α and very small embryonic-like stem cells for end-stage renal disease in patients with immunoglobulin A nephropathy.

Index	AUC	95%CI	P value	Sensitivity (%)	Specificity (%)
CFI	0.660	0.598-0.718	< 0.001a	80.65	54.46
TNF-α	0.818	0.765-0.863	< 0.001a	61.29	91.52
VSELs	0.749	0.691-0.801	< 0.001a	61.29	95.54
Combination	0.916	0.875-0.947	< 0.001a	83.87	90.62

^aP < 0.05.

AUC: Area under the curve; CI: Confidence interval; CFI: Complement factor I; TNF-α: Tumor necrosis factor-α; VSELs: Very small embryonic-like stem cells.

Correlation analysis of CFI, TNF-α, VSELs and eGFR

Spearman correlation analysis showed that there was no significant correlation between TNF-α and CFI (P = 0.562, Figure 2A). VSELs was significantly positively correlated with CFI (r = 0.405, P < 0.001, Figure 2B). VSELs was significantly positively correlated with TNF-α (r = 0.1874, P = 0.002, Figure 2C). eGFR was positively correlated with CFI (r = 0.459, P < 0.001, Figure 2D). eGFR was negatively correlated with TNF-α (r = -0.308, P < 0.001, Figure 2E). There was no significant correlation between eGFR and VSELs (P = 0.111, Figure 2F).

Figure 2 Correlation analysis of key indicators and estimated glomerular filtration rate. A: Tumor necrosis factor-α (TNF-α) vs complement factor I (CFI); B: Very small embryonic-like stem cells (VSELs) vs CFI; C: VSELs vs TNF-α; D: Estimated glomerular filtration rate (eGFR) vs CFI; E: EGFR vs TNF-α; F: EGFR vs VSELs. TNF-α: Tumor necrosis factor-α; CFI: Complement factor I; VSELs: Very small embryonic-like stem cells; eGFR: Estimated glomerular filtration rate.

Comparison of Oxford classification of IgAN patients with different levels of CFI, TNF-α and VSELs

The Oxford classification serves as the current international gold standard for the pathological assessment of IgAN, as it directly reflects the type and severity of renal structural damage and provides a crucial pathological basis for clinically evaluating the risk of disease progression[31]. To investigate the associations between CFI, TNF-α, VSELs and the degree of renal pathological injury, we compared patients stratified by different levels of these three biomarkers with the core parameters of the Oxford classification. As illustrated in Figure 3A, patients were stratified into high and low CFI groups using a cutoff value of 47.06 mg/L. The results of χ² test showed that, there were no significant differences in the distribution of pathological features of the Oxford classification between the two groups (all P > 0.05), suggesting that CFI levels were not significantly associated with pathological features in patients with IgAN.

Figure 3 χ² analysis. A: χ² analysis of the association between complement factor I and Oxford classification in immunoglobulin A nephropathy (IgAN) patients; B: χ² analysis of the association between tumor necrosis factor-α and Oxford classification in IgAN patients; C: χ² analysis of the association between very small embryonic-like stem cells and Oxford classification in IgAN patients. CFI: Complement factor I; TNF-α: Tumor necrosis factor-α; VSELs: Very small embryonic-like stem cells.

As depicted in Figure 3B, patients were stratified into high and low TNF-α groups using a cutoff value of 23.91 pg/mL. Subsequent comparative analysis revealed no statistically significant differences in the distribution of the aforementioned Oxford classification pathological features between the two groups (all P > 0.05). These findings suggest that TNF-α levels are not significantly associated with the severity of pathological lesions in patients with IgAN.

As presented in Figure 3C, patients were stratified into high and low VSELs groups using a cutoff value of 53 events/600 μL whole blood. Comparative analysis revealed no statistically significant differences in the distribution of Oxford classification pathological characteristics between the two groups (all P > 0.05). These results indicate that circulating VSELs levels are not significantly associated with the heterogeneity of renal pathological features in patients with IgAN, thus failing to reflect differences in pathological severity.

The predictive value of CFI, TNF-α and VSELs for the prognosis of patients with IgAN

The results of Kaplan-Meier survival analysis are presented in Figure 4. Using a cutoff value of 47.06 mg/L, no significant difference was observed in the renal survival curves between the high CFI group and the low CFI group, with a P-value of 0.111 by the Log-rank test, indicating that different CFI levels had no significant effect on renal survival time in patients with IgAN (Figure 4A). With a cutoff value of 23.91 pg/mL, there was also no significant difference in renal survival time between the high TNF-α group and the low TNF-α group (P = 0.318, Log-rank test) (Figure 4B). When a cutoff value of 53 events/600 μL whole blood was applied, the renal survival curve of the high VSELs group was significantly lower than that of the low VSELs group (P < 0.001, Log-rank test), suggesting that elevated VSELs levels serve as a protective factor for poor renal prognosis in patients with IgAN (Figure 4C).

Figure 4 Analysis of the relationship between indicator expression and renal survival in immunoglobulin A nephropathy patients. A: Complement factor I expression and renal survival; B: Tumor necrosis factor-α expression and renal survival; C: Very small embryonic-like stem cells expression and renal survival. CFI: Complement factor I; TNF-α: Tumor necrosis factor-α; VSELs: Very small embryonic-like stem cells; HR: Hazard ratio.

DISCUSSION

IgAN is a highly prevalent primary glomerular disease worldwide, with a significant proportion of patients at risk of progressing to ESRD. However, clinically, there is a lack of multi-dimensional prognostic indicators that can accurately assess disease progression. Activation of the complement AP, imbalance of inflammatory responses, and disorders of stem cell repair function are all key links in the pathological progression of IgAN. Although the prognostic value of indicators related to a single pathway has been initially explored, the synergistic mechanism of these three pathways and the clinical application value of their combined detection remain unclear. Through a retrospective analysis of 255 IgAN patients, this study systematically investigated the association between complement AP indices, inflammatory biomarkers, stem cell indicators, and the occurrence of ESRD. We identified independent prognostic factors, verified the excellent predictive efficacy of combined detection of multiple indicators, and revealed the correlation between each key indicator and eGFR as well as the lack of significant association with Oxford classification pathological features. These findings provide a more comprehensive scientific basis for risk stratification, prognostic assessment, and individualized intervention of IgAN.

First, this study identified the characteristic differences in indicators between ESRD and non-ESRD patients through intergroup comparisons: Patients in the ESRD group had significantly lower levels of complement AP indices CFB, CFI, and C3a, while significantly higher levels of inflammatory biomarkers MCP-1, IL-6, TNF-α, and stem cell indicator VSELs. This result is consistent with the mechanism by which imbalanced inhibition of the complement AP exacerbates renal injury - CFI, as a key regulator of complement activation, its reduced level may lead to excessive complement activation, thereby aggravating glomerular and tubulointerstitial damage[32]. The elevation of inflammatory factors such as TNF-α reflects the sustained activation of systemic inflammatory responses, which is closely associated with the process of renal fibrosis[33]. Although VSELs are considered a subset of stem cells with putative tissue repair potential, our study does not provide direct evidence for their functional role in renal repair. Instead, the elevated VSELs levels observed in the ESRD group should be interpreted as a sensitive prognostic indicator reflecting the host’s compensatory or stress response to progressive renal injury[34]. Additionally, the only differences in clinical baseline data between the two groups were serum uric acid levels and the proportion of S1 type in the Oxford classification. This suggests that traditional clinical and pathological indicators have limited ability to distinguish IgAN progression[35], while complement, inflammatory, and stem cell-related indices are more reflective of the molecular characteristics of disease progression.

Multivariate logistic regression analysis further confirmed that CFI, C3a, TNF-α, and VSELs are independent predictors of ESRD in IgAN patients. More importantly, the combined detection of CFI, TNF-α, and VSELs showed extremely high predictive value, which was significantly superior to any single indicator. This finding addresses the clinical pain point of insufficient predictive accuracy of single biomarkers[36]. Correlation analysis revealed the intrinsic associations between key indicators: VSELs were weakly positively correlated with both CFI and TNF-α, suggesting a potential synergistic regulatory relationship between stem cell mobilization, complement imbalance, and inflammatory activation. Meanwhile, eGFR was positively correlated with CF and negatively correlated with TNF-α, further verifying the close association of these indicators with renal function injury[37]. Notably, there were no significant correlations between CFI, TNF-α, VSELs levels and any pathological features of the Oxford classification, indicating that these serological indicators can assess disease prognosis independently of pathological classification[38], providing a new prognostic evaluation approach for patients who cannot tolerate renal biopsy or have atypical pathological manifestations. It should be noted that blood samples were collected at the initial admission; however, the interval between disease onset, renal biopsy and sampling is not standardized, which may lead to changes in biomarker levels.

Kaplan-Meier survival analysis further refined the prognostic value of each indicator: High VSELs levels were associated with significantly poorer renal survival rates, while no significant differences in survival were observed between high and low levels of CFI or TNF-α. This result is consistent with the conclusion that VSELs serve as an independent risk factor in multivariate regression, suggesting that VSELs may be a more direct indicator of the long-term prognosis of IgAN - its sustained high expression may indicate irreversible renal injury. In contrast, the prognostic value of TNF-α may be affected by treatment interventions or fluctuations in inflammatory responses[39]. Collectively, this study constructed a “complement-inflammation-stem cell” synergistic regulatory model for IgAN progression through multi-dimensional analysis, clarifying the clinical advantages of combined detection of multiple indicators. It not only enriches the biomarker system for IgAN prognostic evaluation but also provides potential directions for subsequent targeted therapies (such as regulating complement activation via CFI, inhibiting inflammation by targeting TNF-α, or modulating VSELs repair function).

In addition, based on recent advances in stem cell biology and immune regulation, we propose a testable mechanism model linking VSELs to IgAN progression. Recent studies have shown that VSEL exhibits strong migration ability driven by chemokines such as stromal cell-derived factor 1, sphingosine-1-phosphate and extracellular ATP, and can dynamically respond to inflammation and injury-related signals in vivo[13,40]. In chronic inflammatory diseases, there is increasing evidence that stem cell mobilization is closely associated with complement activation and inflammatory signaling[41,42]. In the context of IgAN, the continuous activation of complement APs and continuous inflammatory signal transduction (such as TNF-α, IL-6) may synergistically promote the continuous mobilization of VSELs from the bone marrow niche to the circulation. However, new evidence suggests that although VSELs have regenerative potential and can differentiate into tissue-specific progenitor cells under acute injury conditions[43], their repair capacity may be limited in a chronic inflammatory microenvironment.

Based on these observations, we hypothesized that elevated circulating VSEL in IgAN patients with poor prognosis reflects an “ineffective or dysfunctional repair” state characterized by persistent complement inflammation-driven mobilization, but insufficient functional integration with renal tissue. The model is testable: Future studies can characterize the functional phenotypes of circulating VSELs (such as aging, differentiation potential); to study the relationship between complement activation products (such as C3a/C5a) and VSELs mobilization kinetics; and evaluate whether VSEL contributes to renal repair or fibrosis under chronic inflammatory conditions in an experimental model.

Despite these meaningful findings, this study has certain limitations. First of all, a notable limitation of this study lies in its relatively limited sample size (n = 255) and the regional recruitment of all participants. This homogeneous sampling strategy may inevitably lead to selection bias, thus impairing the external validity and generalizability of the study results. In addition, patients with incomplete data were excluded, which may have introduced selection bias by potentially omitting individuals with more severe or atypical disease. Second, the study follow-up time was not explicitly extended to a sufficient period, which may not fully capture the dynamic changes in long-term renal prognosis of patients, and the long-term association analysis between indicators and disease progression was insufficient. Thirdly, this study only focused on the alternative complement pathway, some inflammatory markers and stem cell markers, without in-depth discussion of the upstream and downstream regulatory mechanisms between these indicators, nor combined with other potential influencing factors (such as gene polymorphism, intestinal flora, etc.) for comprehensive analysis, it is difficult to fully reveal the molecular mechanism network of the progression of IgAN. Fourth, detailed data on immunosuppressive regimens were not systematically available, precluding stratified analyses by treatment. Given that different therapies may influence complement, inflammatory, and stem cell-related markers, residual confounding due to treatment heterogeneity cannot be excluded. Finally, this study only detected circulating VSELs, did not verify whether they were homing to the kidney, and did not analyze their correlation with renal stem cell markers (such as CD133, PAX2). Therefore, in the future, multiple immunohistochemistry or immunofluorescence co-localization techniques should be used in renal biopsy tissues, combined with dynamic monitoring of circulating VSELs, to clarify their homing characteristics and tissue sources, so as to determine whether the level of VSELs in peripheral blood can reflect the status of local stem cell pools in the kidney.

Based on the limitations of this study, in-depth research can be carried out in the following directions in the future. First, a multi-center, large-sample prospective cohort study should be designed to expand the geographical and population representation of the study population, reduce selection bias, and further verify the predictive value of CFI, C3a, TNF-α and VSELs for ESRD in patients with IgAN. Secondly, the follow-up period should be extended to dynamically monitor the long-term association between the changes of related indicators and renal prognosis, and to clarify the early warning significance of the fluctuations of indicators on disease progression. Thirdly, with the help of molecular biology experimental techniques, the upstream and downstream regulatory mechanisms between complement AP indicators, inflammatory markers and VSELs were deeply explored, and multi-dimensional factors such as gene polymorphism and intestinal flora were combined to construct a molecular mechanism network for the progression of IgAN. Fourth, stratified studies should be carried out to analyze the effects of different treatment regimens on relevant indicators and renal prognosis, to clarify the applicability of the predictive value of indicators in different treatment backgrounds, and to provide a basis for the formulation of individualized treatment strategies. In addition, it can be explored to combine these indicators with clinical routine indicators and imaging features to build a more accurate prognosis evaluation model and improve the clinical application value. Finally, in the future, multi-label immunohistochemistry or immunofluorescence co-localization should be performed in renal biopsy tissues, combined with dynamic monitoring of circulating VSELs to clarify their homing characteristics and tissue sources.

CONCLUSION

This study confirms that complement AP indices (CFI, C3a), inflammatory biomarker (TNF-α), and stem cell indicator (VSELs) are independent predictors of progression to ESRD in patients with IgAN. The combined detection of CFI, TNF-α, and VSELs can significantly improve the accuracy of ESRD prediction, among which high VSELs level indicates poor renal prognosis, while CFI and TNF-α levels have no significant correlation with renal survival time. In addition, there is no obvious correlation between CFI, TNF-α, VSELs levels and Oxford classification pathological features, suggesting that these serological indicators can assess disease prognosis independently of pathological classification. The above findings enrich the research on the progression mechanism of IgAN and provide important references for early clinical identification of high-risk ESRD patients and optimization of prognostic assessment strategies. In the future, multi-center, large-sample prospective studies are needed to further verify the conclusions and explore the regulatory mechanisms of related indicators to better guide clinical practice.