Systematic Review of the Link Between Oxford MEST-C Classification and Complement Activation in IgA Nephropathy

Abstract

Introduction

IgA nephropathy's (IgAN’s) MEST-C classification relationship with complement activation is still not fully understood because of limited and conflicting evidence. Our study aimed to delineate this relationship through a systematic review.

Methods

We adhered to the Preferred Reporting Items for Systematic Review and Meta-analysis guidelines and conducted a systematic review, utilizing databases like MEDLINE (PubMed), Embase, Scopus, and Cochrane from January 2016 (year of updated MEST-C classification) to January 2023. We specifically selected studies that employed established methods to evaluate complement activation and the MEST-C classification.

Results

A total of 34 studies with 10,082 patients were included. Among these, 7 studies focused on the pediatric population (500 patients), and 22 studies involved 8128 patients from Asian populations. C4d, C3, C5b9, MBL, C4, and factor H-related protein 5 (FHR5) were the most frequently studied complement proteins in relation to the MEST-C classification. Complement activation assessment was primarily conducted using immunofluorescence and immunohistochemistry on kidney biopsy specimens. All complement proteins investigated showed associations with the C1-2 class. Notably, FB, FH, MASP1/3, MASP2, C5a, and C5b9 from the alternative, lectin, and terminal pathways were uniquely present in the C1-2 class. Whereas C3, FHR5, C4, and C4d were associated with all the MEST-C classes.

Conclusion

We found evidence supporting the involvement of alternative and lectin complement pathways across all MEST-C classes. All examined complement factors were associated with the C1-2 class, emphasizing the critical role of complement activation, possibly at the endothelial surface. These findings may guide the development of personalized treatment strategies targeting complement pathways in relation to the MEST-C lesions.

Graphical abstract

IgAN, the most prevalent form of primary glomerulonephritis, is an autoimmune disease defined by the deposition of dominant or codominant IgA1 in the mesangium.^1,2 IgAN commonly affects young adults, with 15% to 40% of patients developing end-stage kidney disease within 20 years of diagnosis.3, 4, 5 However, the clinical course of IgAN is highly variable, with some patients experiencing stable disease for decades, whereas others rapidly progress to end-stage kidney disease within few years of diagnosis.³ The development of the Oxford classification by the International IgA Nephropathy Network and the Renal Pathology Society aimed to provide a more thorough evaluation of disease severity and prognosis.⁶ Since 2016, the classification system encompasses crescents (C1-2), in addition to the previously established mesangial hypercellularity (M1), endocapillary hypercellularity (E1), segmental glomerulosclerosis (S1), and tubular atrophy/interstitial fibrosis (T1-2).⁷ Studies have shown that the updated MEST-C classification system provides a more accurate prediction of long-term renal outcomes in IgAN and can potentially help with the management of the disease.^8,9 Importantly, MEST lesions are incorporated to a recent, validated, consensual prediction tool which encompasses both bioclinical variables and Oxford classification, becoming the cornerstone of risk prediction in IgAN .

Mounting evidence suggests that the dysregulation of the complement system, particularly the involvement of the alternative pathway and the lectin pathway, plays a pivotal role in the onset and progression of IgAN.^4,10 However, the relationship between the MEST-C classification and complement activation in IgAN is not fully understood, and the available evidence is limited and conflicting. Some studies have suggested that complement activation is associated with more severe MEST-C scores,11, 12, 13, 14, 15 whereas others have reported no significant correlation between the 2.16, 17, 18, 19

Understanding the relationship between the MEST-C lesions and the complement system may help to identify potential therapeutic targets in IgAN. Therefore, if complement activation is found to be associated with more severe MEST-C classes, targeting complement activation pathways may be a viable treatment strategy to slow disease progression and improve outcomes in patients with advanced disease. Moreover, if complement activation is found to be associated with the disease severity assessed by Oxford system, it could improve our ability to predict disease progression. Therefore, we aimed to perform a systematic review to evaluate the relationship between the updated MEST-C classification and complement activation in IgAN.

Methods

To ensure reliable results, our systematic review was conducted in accordance with the updated Preferred Reporting Items for Systematic Review and Meta-analysis guidelines.²⁰

Data Source and Search Strategy

To conduct a comprehensive search for relevant studies, the following electronic databases were searched: MEDLINE (PubMed), Embase, Scopus, and Cochrane databases. The search strategy was developed using a combination of controlled vocabulary (e.g., MeSH terms) and keywords related to the MEST-C classification and complement activation (Supplementary Table S1). The search strategy was tailored for each database, with the following general search terms being used: “IgA nephropathy,” “immunoglobulin A nephropathy,” “Berger's disease,” “IgAN,” “Oxford classification,” “Oxford score,” “MEST classification,” “MEST score,” “complement,” “complement activation,” “C3,” “C4,” “C4d,” “MBL,” “mannose-binding lectin,” “MASP,” “mannan-binding lectin serine protease,” “C5b9,” “MAC,” “membrane attack complex,” “FB,” “factor B,” “FH,” “factor H,” “FHR,” “CFHR,” “factor H related protein,” “C5,” and “properdin.” The search was limited to studies published in English from January 2016 to January 2023. In addition, the reference lists of all relevant articles were manually searched to identify any additional studies that were missed in the electronic search. Two reviewers independently screened the titles and abstracts of all identified articles to determine eligibility for inclusion in the systematic review.

Eligibility Criteria

To assess the eligibility of the retrieved references, a multistep approach was utilized. In the first step, 2 independent investigators evaluated the titles and abstracts of the articles to determine whether they met the inclusion and exclusion criteria. Subsequently, the full text of studies that passed this initial screening was appraised to confirm eligibility based on the predetermined criteria.

Several inclusion criteria were predefined and were applied for eligibility assessment as follows: (i) original research articles published in English language; (ii) studies that used well-established methods for assessing complement activation, such as measurement of complement component levels, immunohistochemical or immunofluorescence staining, and mass spectrometry; (iii) studies that used the updated MEST-C classification to classify patients with IgAN; (iv) studies that included a clearly defined population of patients with IgAN , with diagnostic criteria and disease severity clearly stated; and (v) studies published after January 2016 because this was the year when the updated Oxford classification for IgAN was introduced.⁷

Furthermore, several key exclusion criteria were established to ensure the quality and relevance of the included studies. These criteria included unpublished data, studies available only in abstract form, overlapping populations, case reports, meta-analyses, editorials, missing data, and the inability to extract data on the enrolled population and investigated outcomes.

Data Collection and Synthesis

After eligibility assessment, 2 independent investigators extracted the following data from included studies: first author, year of publication, complement protein studied, study design, country, number of patients enrolled and their age, clinical setting (estimated glomerular filtration rate [eGFR] and proteinuria), reported relationship of interest. We performed a qualitative synthesis of included studies to provide a better understanding of the reported relationship between complement factors and the MEST-C lesions. In addition, when available, data were reported as numbers, percentages, median, or mean value.

Quality and Risk of Bias Assessment

The Newcastle-Ottawa scale was used to guide the quality assessment of the included studies.²¹ The Newcastle-Ottawa scale is a tool based on designating stars for signaling questions, which were grouped into 3 domains, namely population sampling, comparability of groups, and evaluation of outcomes of interest (Supplementary Tables S2, S3, and S4). In addition, to assess the statistical methods used in the included studies, we utilized the Kmet checklist for quality assessment (Supplementary Table S5). The scoring criteria encompass questions related to the problem statement, study design, participant characteristics, outcome measures, sample size, analytical methods, estimation of variance, result reporting, and overall conclusions. A summary score was determined for each study by summing the scores across the relevant items and then dividing by the total possible score. The scores were categorized as poor (<50%), fair (50%–64%), good (65%–84%), and excellent (>85%).

Results

Study Selection

We conducted a request in the designated databases, obtaining 436 sources. After eliminating duplicate entries (n = 298), we were left with 138 sources to review for their title or abstract. Ultimately, 70 sources underwent a full-text eligibility assessment, resulting in 34 studies being incorporated into the current systematic review. A visual representation of the selection procedure can be found in Figure 1.

Figure 1.

Preferred Reporting Items for Systematic Review and Meta-analysis flow diagram of the selected studies in the present analysis.

In total, 34 studies with 10,082 patients were analyzed. Among these, 7 studies specifically targeted the pediatric population, accounting for 500 patients. The characteristics of the included studies are summarized in Table 1. All the included studies had an observational nonrandomized design, there were 27 cohort studies, 5 cross-sectional studies and 2 case-control studies.

Table 1.

Characteristics of the included studies

Reference	Complement protein	Country	Adult/pediatric	Number	Method	Age, years	Male, %	eGFR, ml/mina	P-uria, g/daya
Yang et al.,22 2020	C4	China	Adult	642	IF	33.2	44	85.8 (60.9–110.3)	1.4 (0.7–2.9)
Bi et al.,23 2019	C4	China	Adult	1356	serum, turbidimetric	37	50	83.8 ± 34.2	2.2 ± 2.5
Pan et al.,24 2017	C3, C4	China	Adult	403	serum, turbidimetric	37.1	45	94.5 ± 30.9	1.2 ± 1.5
Nam et al.,25 2020	C3, C4d	Korea	Adult	380	IF:C3; IHC:C4d	36.1	43	85.2 ± 27.7	0.8 (0.3–1.7)
Segarra et al.,26 2018	C4d	Spain	Adult	190	IHC	29	65	100 (86–112)	1.8 (0.8–3.9)
Sato et al.,16 2019	C4d	Japan	Pediatric	25	IHC	11	36	125 (112–137)	1.4 (0.6–2.5)
Faria et al.,27 2020	C4d	Portugal	Adult	126	IHC	42	63	69.1 ± 40.5	1.8 (0.9–3.4)
Vellaisamy et al.,28 2021	C4d	India	Adult	76	IHC	35	38	64.9 ± 60.0	NA
Zhou et al.,29 2023	C4d	China	Pediatric	65	IF	NA	NA	NA	NA
Medrano et al.,30 2022	C4d	Spain	Adult	34	IHC	34	46	85 (72–101)	2.9 (0.9–3.7)
Wang et al.,11 2022	C4d	China	Adult	34	urine, ELISA	34	56	33.4 (16.2–61.7)	3.5 (1.7–5.1)
Itami et al.,31 2020	C3, C4, C4d, C1q, C5b9, MBL, MASP1/3, MASP2, properdin, factor B	Japan	Adult	132	IF	52	55	61.1 (9.7–162)	1.3 (0.04–7.8)
Yang et al.,32 2022	C4d	China	Adult	120	IHC	32	47	85.8 (60.3–110.7)	1.0 (0.8–3.5)
Baek et al.,33 2018	C4d	Korea	Pediatric	56	IHC	12.1	59	109.5 ± 32.4	0.5 (0.2–2.0)
Wen et al.,34 2019	Factor H, MBL, C5b9	Denmark	Adult	85	plasma, urine, ELISA	38	59	71.3 (43.3–111.6)	2.8 (1.4–4.9)
Guo et al.,35 2021	FHR5	China	Adult	56	IF; plasma, ELISA	40.1	38	96.4 (56.5–119.0)	1.4 (0.6–2.7)
Ouyang et al.,36 2019	MBL	China	Adult	606	SNP genotyping	36.9	51	73.3 (15.4–163.9)	1.3 (0.03–12.7)
Park et al.,37 2020	C3	Korea	Adult	453	IF	40.3	45	89.5 ± 28.5	0.7 (0.3–1.5)
Mizerska-Wasiak et al.,17 2021	C3	Poland	Pediatric	148	IF; plasma, turbidmetric	11.6	67	95.8 ± 33.5	1.4 (0.02–9.6)
Lang et al.,18 2021	C3; IgA/C3 ratio	China	Pediatric	41	IF:C3;IgA/C3: turbidmetric	9.3	63	118.8 ± 41.2	NA
Wu et al.,38 2021	C3	China	Pediatric	98	IF; plasma, turbidmetric	9.7	67	108.5 ± 37.1	2.7 (1.1–8.3)
Xie et al.,14 2023	C3	China	Adult	821	IF	34	53	91.8 (60.2–109.3)	1.5 (0.7–2.7)
Wu et al.,39 2021	C3	China	Adult	136	IF	36	49	96.9 (72.2–125.9)	1.0 (0.5–1.8)
Jullien et al.,15 2018	FHR1, FHR3, C3	France	Adult	639	PCR: FHR; IF:C3	40.1	75	74 (50–94)	0.7 (0.2–1.8)
Juan et al.,40 2022	C5a, Ba fragment	Taiwan	Adult	84	plasma, ELISA	44.7	46	54.9 ± 33.1	2.8 ± 3.1
Medjeral-Thomas et al.,41 2017	FHR1, FHR5	UK	Adult	294	plasma, ELISA	48.2	66	52.7 (28.7–82.7)	0.4 (0.1–1.1)
Guo et al.,13 2017	MBL	China	Adult	749	plasma, ELISA	34.7	48	83.8 ± 30.6	1.2 (0.6–2.5)
Zhu et al.,42 2018	FHR5	China	Adult	1126	plasma, ELISA	33	51	82.7 ± 30.6	1.3 (0.6–2.5)
Dumont et al.,19 2020	C5b9	Canada	Pediatric	67	IF	10.9	61	NA	NA
Yu et al.,43 2022	C5b9	Korea	Adult	33	urine, ELISA	40	52	79.7 ± 22.7	1.0 ± 1.0
Stefan et al.,44 2020	IgA/C3 ratio	Romania	Adult	95	serum, turbidmetric	41	70	39.4 (25.2–56.5)	1.7 (0.8–3.0)
Karahisar Sirali and Buberci,45 2022	IgA/C3 ratio	Turkey	Adult	43	serum, turbidmetric	43	40	77 ± 22.0	NA
Pan et al.,12 2018	C3/C4 ratio	China	Adult	712	serum, turbidimetric	37.3	43	90 ± 31	1.3 (0.7–2.6)
Coppo et al.,46 2019	CD46, CD55	VALIGA	Adult/Pediatric	157	PCR	36.8	67	70.9 (48.4–98.7)	1.1 (0.4–2.0)

ELISA, enzyme-linked immunosorbent assay; FHR, factor H-related protein; IF, immunofluorescence; IHC, immunohistochemistry; MASP, MBL associated serine protease; MBL, mannan binding lectin; PCR, polymerase chain reaction; SNP, single nucleotide polymorphism.

^aexpressed as mean with standard deviation (±) or median with interquartile range as reported in the original studies.

Twenty-two studies recruited a total of 8128 patients from Asian populations, which included Chinese, Japanese, Taiwanese, and Korean patients; whereas, 10 studies mainly recruited from European populations, specifically from Spain, Portugal, France, Denmark, Poland, UK, and Romania, with a total of 1887 patients (Table 1). The mean age (29.3 vs. 31.2 years) and proteinuria (1.3 vs. 1.3 g/day) were similar across studies involving predominantly Asian patients versus those mainly involving Caucasians. However, the European cohorts had a higher mean proportion of male patients (58.4% vs. 49 %) but presented with a lower mean eGFR (70.8 versus 84.1 ml/min) at the time of diagnosis.

The included studies exhibited considerable heterogeneity in terms of the complement proteins investigated. Most of the studies centered on investigating just 1 complement protein, whereas 9 of the included studies expanded their focus to assess 2 or more complement proteins (Table 1). The most frequent complement protein studied in relation to the MEST-C classification was C4d (n = 11), followed by C3 (n = 10), C5b9 (n = 4), MBL (n = 4), C4 (n = 3), and FHR5 (n = 3).

The most frequent method of complement activation assessment was immunofluorescence (n = 12) and immunohistochemistry (n = 7) on kidney biopsy specimens, followed by serum or plasma (n = 10) and urine (n = 3) determination through enzyme-linked immunosorbent assay or turbidimetry (Table 1). Five studies (15%) included patients with a median eGFR under 60 ml/min, and 24 studies (71%) included patients with proteinuria higher than 1 g/day.

Relationship Between MEST-C Classification Lesions and Complement Activation

The relationship between the studied complement proteins and the MEST-C classes is presented in Table 2 and in the bubble chart (Figure 2). An important finding is that all the studied complement proteins were reported to be associated with the C1-2 class. In addition, there were complement factors from the alternative, lectin, and terminal pathways that were present only in the C1-2 class: FB, FH, MASP1/3, MASP2, C5a, and C5b9.

Table 2.

Relationship between complement factors and the MEST-C classification in the included studies

Complement Protein		Relationship Between Complement Activation and MEST-C Classification	Reference
		Associated with M1, E1 and C1-2 classesa	Yang et al.,22 2020
	C4	Higher levels in patients with M1, T1-2 and C1-2 classesb	Bi et al.,23 2019
		Higher levels in patients with S1 and T1-2 classesb	Pan et al.,24 2017
		Associated with M1 and T1-2 classesa	Nam et al.,25 2020
		No relationship with MEST-C classesa	Segarra et al.,26 2018
		No relationship with MEST-C classesa	Sato et al.,16 2019
		Associated with T1-2 classa	Faria et al.,27 2020
		Associated with S1 and T1-2 classesa	Vellaisamy et al.,28 2021
	C4d	Associated with M1 and S1 classesa	Zhou et al.,29 2023
		Associated with T1-2 and C1-2 classesa	Medrano et al.,30 2022
Lectin pathway		Higher levels in patients with M1, E1, T1-2, C1-2 classesc	Wang et al.,11 2022
		No relationship with MEST-C classesa	Itami et al.,31 2020
		Associated with T1-2 classa	Yang et al.,32 2022
		Associated with M1 classa	Baek et al.,33 2018
		Positive correlation with C1-2 classc	Wen et al.,34 2019
		Positive correlation with C1-2 classb	Guo et al.,13 2017
	MBL	MBL-rs1800450 AA genotype increased risk of ESKD in patients with E1 and T2d	Ouyang et al.,36 2019
		No relationship with MEST-C classesa	Itami et al.,31 2020
	MASP-1/3	Associated with C1-2 classa	Itami et al.,31 2020
	MASP-2	Associated with C1-2 classa	Itami et al.,31 2020
		Associated with M1, S1, T1-2, C1-2 classesa	Park et al.,37 2020
		No relationship with MEST-C classesa	Mizerska-Wasiak et al.,17 2021
		No relationship with MEST-C classesa	Lang et al.,18 2021
		Intense staining (≥2+) associated with M1 classa	Wu et al.,38 2021
Alternative pathway	C3	Associated with S1, C1-2 classes and with higher global MEST-C scorea	Xie et al.,14 2023
		No relationship with MEST-C classesa	Itami et al.,31 2020
		Intense staining (≥2+) associated with M1 and T1-2 classesa	Nam et al.,25 2020
		Intense staining (>2+) associated with M1, S1, T1-2 and C1-2 classesa	Wu et al.,39 2021
		Correlated with more severe MEST-C classesa	Jullien et al.,15 2018
		No relationship with MEST-C classesa	Pan et al.,24 2017
	Factor B	Associated with C1-2 classa	Itami et al.,31 2020
	Fragment Ba	Elevated in patients with T1-2 classb	Juan et al.,40 2022
	Factor H	Positive correlation with C1-2 classc	Wen et al.,34 2019
	FHR1	FHR3,1Δ genotype was not related to MEST-C classesd	Jullien et al.,15 2018
	FHR3	FHR3,1Δ genotype was not related to MEST-C classesd	Jullien et al.,15 2018
		Higher levels in E1 class; higher levels in patients with MEST-C score >4 than in those with score 1b	Medjeral-Thomas et al.,41 2017
	FHR5	Associated with E1 and S1 classes; levels significantly higher in patients with M1 classa,b	Guo et al.,35 2021
		Associated with severe T1-2 and C1-2 classesb	Zhu et al.,42 2018
	Fragment C5a	Associated with C1-2 classb	Juan et al.,40 2022
		Associated with C1-2 classa	Itami et al.,31 2020
		No relationship with MEST-C classesa	Dumont et al.,19 2020
Terminal pathway	C5b9 (MAC)	No relationship with MEST-C classesb, c	Wen et al.,34 2019
		No relationship with MEST-C classesc	Yu et al.,43 2022
		No relationship with MEST-C classesb	Stefan et al.,44 2020
	IgA/C3 ratio	No relationship with MEST-C classesb	Lang et al.,18 2021
Other		Positive correlation with total MEST-C scoreb	Karahisar Sirali and Buberci,45 2022
	C3/C4 ratio	Higher proportion of M1 and T1-2 classes were observed in the low group C3/C4 ratio (i.e., <3.5)b	Pan et al.,12 2018
	CD46, CD55	No relationship with MEST-C classesd	Coppo et al.,46 2019

C, extracapillary hypercellularity; E, endocapillary hypercellularity; ELISA, enzyme-linked immunosorbent assay; ESKD, end-stage kidney disease; FHR, factor H-related protein; IF/IHC, immunofluorescence/ immunohistochemistry; M, mesangial hypercellularity; MAC, membrane attack complex; MASP, MBL associated serine protease; MBL, mannan binding lectin; PCR, polymerase chain reaction; S, segmental glomerulosclerosis; SNP, single nucleotide polymorphism; T, tubular atrophy/interstitial fibrosis.

^aIF/IHC

^bserum/plasma, turbidimetric/ELISA

^curine, ELISA

^dPCR/SNP

Figure 2.

Bubble chart illustrating the relationship between studied complement factors and the MEST-C classes obtained from the included studies in the systematic review. The color intensity of each bubble represents the number of studies supporting a specific association between a complement factor and a histopathological lesion as detailed in Table 2. The size of the bubble denotes the total number of included patients, that is, the population size.

Itami et al. conducted the most comprehensive assessment of complement pathways, examining the alternative (C3, properdin, and factor B), lectin (C4, C4d, C1q, MBL, MASP1/3, and MASP2), and terminal pathways (C5b9) in their study.³¹ Their study revealed that MASP 1/3 and MASP 2 from the lectin pathway were associated with the C1-2 class. In addition, factor B from the alternative pathway and C5b9 from the terminal pathway were linked to C1-2.³¹

Complement proteins that were associated with all the MEST-C classes were C3, FHR5, C4, and C4d. However, the “intensity” of association between complement activation and the histopathologic lesions was not uniform. Therefore, C3 was reported to be present most frequently in the T1-2 class, followed by S1, C1-2, and M1. Similarly, C4 was most often present in the S1, T1-2 and C1-2 classes. Most studies reported C4d to be present in the T1-2 class. The most frequent complement factor associated with E1 was FHR5 (Table 2).

Out of the 4 selected studies that reported on major attack complex, only 1 reported a relationship between C5b9 and the C1-2 class, whereas the remaining 3 found no association with the Oxford classification (Table 2).

The serum IgA/C3 ratio was not related to the MEST-C classes; however, in 1 study, there was a positive correlation with the total MEST-C score. In addition, 1 study found that a serum C3/C4 ratio below 3.5 was associated with a higher prevalence of M1 and T1-2 lesions (Table 2).

We also performed a subanalysis of the studies which used the same methodology to investigate the same complement proteins in Asian and non-Asian patients (Table 3). Both populations exhibited an association between C4d glomerular staining and S1 and T1-2 classes, but non-Asian patients additionally showed a link between C4d and the C1-2 class (Table 3). In terms of glomerular C3 staining, no distinct differences were noted between the 2 populations. Serum IgA/C3 had a positive correlation with total MEST-C score among non-Asians; however, this relationship was not observed in Asians (Table 3). At diagnosis, higher plasma FHR-5 levels were observed across both groups but were associated with different MEST-C classes: E1 in non-Asians versus C1-2 and T1-2 in Asians. Finally, a link between glomerular C5b9 staining and the C1-2 class was only observed in non-Asians (Table 3).

Table 3.

Relationship between complement proteins and MEST-C classes that were studied in both Asian and non-Asian patients using similar methods

Complement biomarker	Non-Asian				Asian
	No.	eGFR, ml/min	P-uria, g/day	MEST-C relation	No	eGFR, ml/min	P-uria, g/day	MEST-C relation
	190	100 (86–112)	1.8 (0.8-3.9)	No relationship26	120	85.8 (60.3–110.7)	1.0 (0.8–3.5)	Positive staining in T1-232
C4d	34	85 (72–101)	2.9 (0.9–3.7)	Positive staining in T1-2 and C1-230	25	125 (112–137)	1.4 (0.6–2.5)	No relationship16
	76	64.9 ± 60.0	NA	Positive staining in S1 and T1-228	56	109.5 ± 32.4	0.5 (0.2–2)	Positive staining in M133
Total	300	89.4a	2.3a	Positive stainining in S1, T1-2 and C1-2. In 33% no relationship.	201	97.0a	1.7a	Positive staining in M1 and T1-2. In 33% no relationship.
	148	108.5 ± 37.1	2.7 (1.1–8.3)	No relationship17	41	118.8 ± 41.2	NA	No relationship18
	639	74 (50–94)	0.7 (0.2–1.8)	Positive staining correlated with more severe MEST-C classes15	132	61.1 (9.7–162)	1.3 (0.04–7.8)	No relationship31
C3					403	94.5 ± 30.9	1.2 ± 1.5	No relationship24
					453	89.5 ± 28.5	0.7 (0.3–1.5)	Positive staining associated with M1, S1, T1-2, C1-237
					136	96.9 (72.2–125.9)	1.0 (0.5–1.8)	Intense staining (>2+) associated with M1, S1, T1-2 and C1-239
Total	787	78.8a	1.7a	Positive staining correlated with more severe MEST-C classes. In 50% no relationship.	1165	89.5a	1.3a	Positive staining correlated with more severe M1, S1, T1-2, C1-2. In 60% no relationship.
	43	77 ± 22	NA	Positive correlation with total MEST-C score45	41	118.8 ± 41.2	NA	No relationship18
IgA/C3	95	39.4 (25.2–56.5)	1.7 (0.8–3)	No relationship44
Total	138	52.1a	NA	Positive correlation with total MEST-C score. In 50% no relationship.	41	118.8 ± 41.2	NA	No relationship
FHR5	148	95.8 ± 33.5	1.4 (0.02–9.6)	Plasma FHR-5 levels positive correlated with E117	1126	82.7 ± 30.6	1.3 (0.6–2.5)	Plasma FHR-5 levels positive correlated with T1-2, C1-242
C5b9 (MAC)	67	NA	NA	No relationship19	132	61.1 (9.7–162)	1.3 (0.04–7.8)	Positive staining associated with C1-231

C, extracapillary hypercellularity; E, endocapillary hypercellularity; FHR, factor H-related protein; M, mesangial hypercellularity; MAC, membrane attack complex; NA, not applicable; No., number of patients; S, segmental glomerulosclerosis; T, tubular atrophy/interstitial fibrosis.

^aweighted mean: calculated by multiplying each value by its respective sample size, summing these products, and then dividing by the total sample size.

Risk of Bias and Study Quality Assessment

The quality of the included studies was evaluated with the Newcastle-Ottawa Scale modified for cross-sectional, case-control, and cohort studies (Supplementary Tables S2, S3, and S4). The results ranged from ‘fair’ to ‘good,’ reflecting a reasonable degree of methodological rigor in these investigations.

Of the 34 studies evaluated with the Kmet checklist, the majority exhibited commendable methodological quality: 13 (38%) were rated as ‘good’ and 17 (50%) as ‘fair.’ Only 4 studies (11.8%) fell into the ‘poor’ category. This underscores that most studies in our review maintained a good to fair standard in their methodologies (Supplementary Table S5).

Discussion

To the best of our knowledge, we report the first systematic review which describes the relationship between complement activation and the MEST-C classification in IgAN. We found evidence supporting the involvement of both alternative and lectin complement pathways in the development and progression of IgAN across all MEST-C classes (Figure 3). Notably, all complement factors studied showed a strong association with the C1-2 class, further highlighting the significance of complement activation in IgAN pathogenesis.

Figure 3.

The studied complement system proteins reported in the systematic review and the relationship with the MEST-C classification. The colored bar beneath the MEST-C classes indicates that the corresponding complement protein has been reported to be associated with that specific class. The pathogenic chain of events in IgA nephropathy has been described as a 4-hit model. Galactose-deficient IgA1 (hit 1) and galactose-deficient IgA1-binding element (antiglycan antibody and/or soluble form of Fc alfa-receptor) along with complement elements (hit 2) form circulating immune complexes (hit 3). These complexes then deposit in the glomerulus, where the IgA can also serve as a platform for local complement activation, leading to mesangial cells proliferation and glomerular injury (hit 4). In IgA nephropathy the complement system can be activated by the alternative pathway and the lectin pathway. The alternative pathway is a constantly activated pathway, triggered by the hydrolysis of C3 thioester bonds. C3(H₂O) is bound by factor B, which renders the complex susceptible to cleavege by factor D, resulting the C3 convertase. The lectin pathway is activated by mannose moieties found on microbial surfaces or on poorly O-galactosylated IgA1; the moieties are bound by a mannose-binding lectin, which activates MASP-1 and MASP-2. Further activation of MASP-2 leads to the formation of C4bC2b. The common pathway is triggered by any of the C3 convertases, resulting in the cleavage of C3 into C3a, an inflammatory mediator, and C3b. C3b is then further processed by the C3 convertases to generate a C5 convertase, which cleaves C5 to create C5a, a powerful inflammatory mediator, and C5b. Subsequently, C5b sequentially binds with C6, C7, C8, and C9 to form the membrane attack complex (MAC), which has the capability to induce cell lysis. C, extracapillary hypercellularity; C5b9, membrane attack complex; E, endocapillary hypercellularity; FH, factor H; FHR, factor H-related protein; Gd, galactose deficient; IgA1, immunoglobulin A1; M, mesangial hypercellularity; MASP, MBL associated serine protease; MBL, mannan binding lectin; S, segmental glomerulosclerosis; T, tubular atrophy/interstitial fibrosis.

Crescents and Complement Activation in IgAN

In the original cohort studying the Oxford classification, crescents did not predict eGFR decline or end-stage kidney disease outcomes.⁶ However, later studies, including a meta-analysis with over 5000 patients with IgAN and a multicenter study of 3000 patients with IgAN, found crescents associated with lower eGFR levels, increased risks, and faster renal function decline, leading to the introduction of a crescent score (C) in the 2017 Oxford classification for IgAN.^8,47

In this systematic review, we found a strong association between all complement protein studied and crescent class, encompassing the alternative, lectin, and terminal pathways (Figure 2, Table 2). This consistent finding across multiple studies suggests a crucial role of complement activation in the pathophysiology of crescent formation in IgAN.

These results are in line with a study conducted by Wang et al. on a Chinese IgAN cohort of 100 patients with varying crescent proportions, which was not included in the current review because it did not evaluate the Oxford classification.⁴⁸ The authors found that markers of complement activation in urine samples were significantly increased compared to healthy controls, especially in cases with crescents involving >50% of glomeruli. These cases showed elevated levels of common complement pathway markers (C3a, C5a, and C5b-9), alternative pathway marker (Bb), and lectin pathway markers (C4d and MBL). Moreover, urinary C4d levels exhibited a strong linear association with crescent numbers.⁴⁸ In a subgroup, immunohistochemistry on renal biopsy tissue revealed glomerular C4d staining, predominantly in the mesangial area, within crescents and sclerotic lesions, whereas glomerular C5b-9 deposition and C3d staining were observed in nearly all crescentic cases.⁴⁸

Given the association between complement activation and C1-2 lesions in IgAN, patients with these histopathological findings may benefit from novel complement-targeted therapies. Personalized treatment approaches, tailored to individual complement profiles, could optimize therapeutic outcomes for these patients.

This strong association between crescents and complement activation raises the possibility that lesions of glomerular capillary walls could be directly mediated by complement activation, driving secondary extracapillary proliferation (Figure 3). The interaction of IgA1-immune complexes with endothelial cells may trigger complement activation through the alternative and lectin pathways, resulting in the development of microangiopathic lesions, which sometimes can be accompanied by thrombosis (i.e., thrombotic microangiopathy).⁴⁹ Elevated mean arterial pressure, arterial intimal fibrosis, vascular lesions, and chronic microangiopathy in patients with IgAN are connected to the arteriolar or glomerular deposition of C4d.^27,50 In addition, the combination of IgAN, microangiopathy, and C4d deposition significantly reduces kidney survival.⁴⁹ Li et al.⁵⁰ demonstrated in an Asian population with IgAN that nearly half of the patients with microangiopathic lesions possessed rare variants in complement-related genes. They observed pathogenic variants in 37% of these case, with alternative pathway accounting for 27.9% and lectin pathway accounting for 14.0%.⁵⁰

Interestingly, MASP-2, besides cleaving C2 and C4, can trigger the coagulation cascade by activating factor XII, and convert prothrombin into thrombin.^51,52 These processes enhance fibrinogen turnover and clot formation, leading to increased coagulation and contributing to thrombotic microangiopathy lesions.⁵³

Alternative and Lectin Complement Pathways Relationship With the MEST-C Classification

In IgAN, mesangial codeposition of C3 and IgA is a common feature, found in approximately 90% of biopsies.⁵⁴ The alternative pathway serves as the primary complement cascade activator in IgAN and is responsible for C3 deposition. Mesangial C4, especially the activation fragment C4d, is often detected in IgAN biopsies. The near-total absence of C1q in deposits, along with the presence of C4d, suggests that this biomarker is exclusively tied to the lectin pathway in IgAN.^4,55 Our findings indicate that C3 and C4 are associated with all the classes of the MEST-C classification. These results highlight the involvement of alternative and lectin complement pathways in the development and progression of IgAN across all MEST-C classes (Figure 3).

In a meta-analysis by Jiang et al., which included 1251 patients and 199 kidney events, there was a strong correlation between glomerular C4d deposition and increased proteinuria, arterial hypertension, and eGFR.⁵⁶ Moreover, patients with C4d deposition displayed more severe kidney lesions in biopsy samples, such as M1, E1, S1, and T1/2 lesions.⁵⁶ In a combined multivariate analysis, glomerular C4d emerged as a potent independent risk factor for kidney failure development, even in patients with early-stage IgAN.⁵⁶ Our findings both validate and extend previous research, because we showed that C4d was also associated with the C1-2 class. This further emphasizes the significance of glomerular C4d deposition as a potential prognostic biomarker in IgAN.

In our systematic review, FHR5 was present in all the Oxford classes, and more frequent in the E1 class. FHR5 is known to inhibit the regulation of alternative pathway activation and amplification by factor H.¹⁰ Glomerular FHR5 has been detected in IgAN cases with a distribution similar to IgA and C3.⁵⁷ In a UK IgAN cohort study, glomerular FHR5 abundance correlated with the amount of glomerular C3b/iC3b/C3c, C3dg, and C5b9 deposition, and showed negative associations with glomerular FH.⁴¹ Moreover, glomerular deposition of FHR5, C3, C3dg, C4d, and C5b9 was linked to increasing IgAN severity.⁴¹ These associations have been replicated in a Chinese IgAN patient cohort, where patients with endocapillary hypercellularity and segmental sclerosis had more intense glomerular FHR5 deposition, and the presence of glomerular FHR5 correlated with glomerular C3.³⁵ Proteomic analysis of microdissected glomerular cross-sections from patients with IgAN confirmed the presence of complement proteins in glomeruli, with progressive IgAN associated with the glomerular abundance of terminal pathway proteins, C4, C4-binding protein, and FHR5.⁵⁸ These findings suggest that complement alternative pathway dysregulation by FHR5, along with lectin and terminal pathway activity, contribute to the pathogenesis of severe, progressive IgAN.

Potential Implications for Complement-Targeting Therapy

There are currently multiple complement-targeting drugs undergoing clinical investigation for the treatment of IgAN. For the first time, the field is shifting toward more mechanistic therapeutic approaches, underscoring the necessity to more effectively select patients who may benefit from these drugs, a selection process not merely based on eGFR and proteinuria.

Based upon the findings from our systematic review, there appears to be an association between C1-2 lesions and the potential for complement-targeting drugs. Although kidney biopsy and MEST-C assessment are routinely employed in the diagnosis of IgAN, it is important to note that relying solely on the C score as an indicator for complement therapy may not be sufficient. Although the MEST-C assessment may be more accessible, the correlation between complement activity and C score is not robust enough to suggest they are interchangeable. Therefore, precision therapy should be based on precise target detection, and the relationship between complement activity and C score warrants more in-depth investigation.

Early case reports, which used eculizumab to inhibit C5, showed temporary deceleration of kidney deterioration in patients with IgAN with rapidly progressive due to extracapillary proliferation. Despite the publication bias and low quality of evidence, those observations introduced a particular interest in blocking complement activation in IgAN.59, 60, 61

The alternative pathway, typically active in IgAN, has been a target in advanced clinical trials. Our review found a link between factor B, MASP-3, and the C1-2 class. The factor B inhibitor, iptacopan, effectively reduced proteinuria in a 6-month phase 2 study (NCT03373461) and is now in a phase 3 trial (NCT04578834).⁶² Similarly, IONIS-FB-LRx showed promising results in its phase 2 study (NCT04014335).⁶³

Our systematic review highlights the activity of the lectin pathway in E1 and C1-2 classes, evidenced by MBL presence, linking endothelial activation to crescent formation. This suggests potential benefits of anticomplement agents, such as narsoplimab, targeting the lectin pathway for patients with E1 and C1-2 lesions. A phase 2 trial demonstrated narsoplimab's safety and efficacy, with reduced urine protein excretion and slower eGFR decline.^51,64 It is now under a global phase 3 trial (ARTEMIS-IGAN; NCT03608033). Though untested for thrombotic microangiopathy lesions in IgAN, narsoplimab showed promise in other conditions, including COVID-19.^65,66

We identified the terminal pathway components exclusively in the C1-2 class. Concurrently, research on agents that target these components is ongoing. In a phase 2 pilot study, avacopan, a selective C5a receptor inhibitor, exhibited about a 50% improvement in urinary protein-to-creatinine ratio in approximately 3 of 7 patients with IgAN, all of whom were on renin-angiotensin-aldosterone system inhibitors.⁶⁷

In a phase 2 trial, cemdisiran, an RNA targeting C5 production, was tested. Results showed 31.8% of cemdisiran-treated participants (n = 22) achieved a ≥50% reduction in 24-hour urinary protein-to-creatinine ratio by week 32, compared to 12.5% in the placebo group (n = 9).⁶⁸ In addition, ravulizumab, a monoclonal anti-C5, is under evaluation in a phase 2 IgAN trial (NCT04564339).

Limitations

In discussing the results of our systematic review on the relationship between complement activation and MEST-C classification in IgAN, it is important to acknowledge several limitations. First, heterogeneity among studies; the included studies exhibited variability in patient populations, methodologies, and measurement techniques, which may impact the interpretation of the findings (Table 1). Second, publication bias; as with any systematic review, there is a possibility of publication bias, because studies with negative or nonsignificant results are less likely to be published, potentially skewing the overall findings. Third, limited generalizability; because of differences in geographic regions, ethnicities, and healthcare systems among the included studies, the generalizability of the findings to broader populations may be limited (Table 1); particularly, the number of patients in Asian studies is overrepresented in the present analysis, limiting the conclusions which could be made about other ethnicities. Fourth, small sample sizes; some of the included studies had relatively small sample sizes, which may reduce the statistical power to detect significant associations between complement activation and MEST-C classification (Table 1). Fifth, lack of standardized measurements; the lack of standardized methods for assessing complement activation and MEST-C classification across studies may introduce variability in the results, making it difficult to draw firm conclusions; particularly the decrease of some circulating (plasma or serum) complement proteins as well as its increase in urine could be affected by impairment of glomerular filtration barrier proteinuria by itself, introducing a potential measurement bias. This point is not valid regarding the associations of glomerular deposits with Oxford classification lesions. Lastly, our study did not analyze the potential relationship between vascular lesions and complement activation. This important point has to be investigated in future studies because of the potential endothelial lesions secondary to the activation of complement.

Despite these limitations, our systematic review provides valuable insights into the involvement of complement activation in the development and progression of IgAN across MEST-C classes. Further research is needed to confirm these findings and explore their implications for personalized treatment strategies targeting the complement pathways.

Conclusions

In conclusion, our systematic review uncovered evidence supporting the involvement of both alternative and lectin complement pathways in IgAN across all MEST-C classes. Notably, we observed that all complement factors investigated displayed a relationship with the C1-2 class, which further accentuates the critical role of complement activation conceivably on the endothelial surface, in the pathogenesis and outcome of IgAN. These findings may guide the development of personalized treatment strategies targeting complement pathways in relation to the MEST-C lesions.

Disclosure

All the authors declared no competing interests.

Supplementary Material

Table S1. Databases and search strategies used in present systematic review.

Table S2. Quality assessment of cross-sectional studies using the Newcastle-Ottawa scale.

Table S3. Quality assessment of cohort studies using Newcastle-Ottawa scale.

Table S4. Quality assessment of case-control studies using Newcastle-Ottawa scale.

Table S5. Quality assessments of the studies using the Kmet checklist.

Table S6. Preferred Reporting Items for Systematic Review and Meta-analysis checklist.