Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/ou_press/defective-gene-expression-of-the-membrane-complement-inhibitor-cd46-in-8CrCje0EDA
- Publisher
- Oxford University Press
- Copyright
- © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
- ISSN
- 0931-0509
- eISSN
- 1460-2385
- DOI
- 10.1093/ndt/gfy064
- Publisher site
- See Article on Publisher Site
Abstract
Abstract Background Complement is thought to play a role in immunoglobulin A nephropathy (IgAN), though the activating mechanisms are unknown. This study focused on the gene expression of CD46 and CD55, two key molecules for regulating C3 convertase activity of lectin and alternative complement pathways at a cellular level. Methods The transcriptional expression in peripheral white blood cells (WBCs) of CD46 and CD55 was investigated in 157 patients enrolled by the Validation of the Oxford Classification of IgAN group, looking for correlations with clinical and pathology features and estimated glomerular filtration rate (eGFR) modifications from renal biopsy to sampling. Patients had a previous median follow-up of 6.4 (interquartile range 2.8–10.7) years and were divided into progressors and non-progressors according to the median value of their velocity of loss of renal function per year (−0.41 mL/min/1.73 m2/year). Results CD46 and CD55 messenger RNA (mRNA) expression in WBCs was not correlated with eGFR values or proteinuria at sampling. CD46 mRNA was significantly correlated with eGFR decline rate as a continuous outcome variable (P = 0.014). A significant difference was found in CD46 gene expression between progressors and non-progressors (P = 0.013). CD46 and CD55 mRNA levels were significantly correlated (P < 0.01), although no difference between progressors and non-progressors was found for CD55 mRNA values. The prediction of progression was increased when CD46 and CD55 mRNA expressions were added to clinical data at renal biopsy (eGFR, proteinuria and mean arterial blood pressure) and Oxford MEST-C (mesangial hypercellularity, endocapillary hypercellularity, segmental glomerulosclerosis, tubular atrophy/interstitial fibrosis, presence of any crescents) score. Conclusions Patients with progressive IgAN showed lower expression of mRNA encoding for the complement inhibitory protein CD46, which may implicate a defective regulation of C3 convertase with uncontrolled complement activation. biomarkers, CD46, CD55, complement, IgA nephropathy INTRODUCTION Data suggest a role for complement activation in the development and progression of immunoglobulin A nephropathy (IgAN) [1, 2]. C3 co-deposited with IgA is detected in up to 90% of patients and complement activation breakdown products have been found in glomeruli and in circulation in significant association with IgAN activity or progression [3, 4]. C3 binding is thought to confer a particular nephrotoxicity to the immune deposits of galactose-deficient IgA1 (Gd-IgA1), which represents the first event in the pathogenesis of IgAN leading to the formation of circulating immune complexes (ICs) [5, 6]. Moreover, mesangial cells have C3a receptors and, when exposed to Gd-IgA1 or aggregated IgA, secrete C3 and matrix components [7]. The efficacy of complement activation in mesangial deposits is supposed to play a role in modulating the glomerular inflammatory reaction that follows the accumulation of Gd-IgA1 containing immune material and tissue damage. This final event might be critical in tuning the clinical outcomes of patients with IgAN, which vary from a rather benign course to progression to end-stage renal disease (ESRD). Hence the investigation of factors involved in the activation of complement in mesangial deposits of IgAN deserves attention since it might be relevant for future therapeutic choices. A key point in complement activation is the generation of C3 convertases, which proceeds via three pathways [1, 2]. The classical pathway (CP) is activated by binding of C1q to IgG or IgM containing ICs and leads to formation of the C3 convertase C4b2a. This convertase can also be assembled by the lectin pathway (LP) triggered by pattern recognition molecules such as mannan-binding lectin (MBL), ficolins and collectins recognizing carbohydrates expressed on microorganisms or altered tissues. Finally, C3 can be activated by the alternative pathway (AP) initiated by spontaneous C3 hydrolysis (tick-over process). C3b is formed and the interaction with factor B (FB) and factor D on activating surfaces (microbes or cells) assembles a labile C3bBb convertase that is then stabilized by properdin (P), which enhances the AP. Hence the AP is an amplification loop of C3 activation. Polymeric IgA and IgAIC (IgA-containing immune complexes) activate in vitro the AP and LP [8, 9], but MBL, which initiates the LP, does not interact with Gd-IgA1 due to a lack of mannose in the O-linked chains typical of IgA1 [2, 6, 10]. The C3 convertases initiated by C1q, MBL, ficolins and P generate C3a and C3b, which forms the C5 convertase leading to activation of the terminal complement pathway, with production of C5a and C5b, followed by formation of the membrane attack complex (MAC) C5b-9. C3a and C5a are highly chemotactic fragments upregulating adhesion molecules and phagocytic cells, and C5b-9 is very active in enhancing cytokines and chemokines. The efficacy of the complement system strictly depends on regulatory factors acting in fluid and in solid phase [11, 12]. Soluble complement regulators include C1 inhibitor (C1-inh), C4b-binding protein (C4BP), factor H (FH) and factor H-like protein 1 (FHL-1). FH and FHL-1 are the main fluid phase regulators of the AP, preventing the formation of C3 convertases acting as cofactors for the protease factor I (FI), which cleaves C3b and C4b and accelerates the convertase decay. This proteolysis can only occur in the presence of fluid phase cofactors and additional cofactors acting at the cellular level. Four membrane-bound regulatory proteins are known. Complement receptor 1 (C1R or CD35), mostly expressed in erythrocytes, and membrane cofactor protein (MCP; CD46) are membrane-bound complement regulator cofactors for inactivation of C3b and C4b by serum FI. Decay-accelerating factor (DAF; CD55) accelerates the decay of cell surface–assembled C3 and C5 convertases. CD59 prevents the formation of MAC in the lipid bilayer. In IgAN, activation of the complement system is thought to be initiated by the AP [8] and LP [9], as suggested by the finding of P and MBL deposits in correlation with C4b immunostaining, which is a marker of progressive cases [4]. A positive correlation between C4d deposits and various fluid phase regulatory proteins (FH, C4BP and C1-inh) has been reported [13, 14]. FH, which binds to host cells and prevents assembly of the AP C3 convertase, has competitive inhibitors, complement factor H–related (FHR) proteins, which interfere with the regulatory function of FH in a process called FH deregulation. Genome-wide association studies have detected a single polymorphism (SPN), tagging the deletion of FHR-1 and FHR-3 associated with protection from the development of IgAN [15]. Very recently, increased serum levels of FHR-1 and FHR-5 have been reported in patients with IgAN, which may favour increased AP activation [16, 17]. No study has so far investigated in patients with IgAN the gene expression of membrane-bound complement regulatory cofactors. The aim of this study was to investigate in patients with IgAN the gene expression of CD46 and CD55, two key molecules regulating at the cellular level and the C3 convertase activity of the AP and LP, largely transcribed in almost all peripheral blood cells and tissues [11, 12], including mesangial cells [18]. In this study, CD46 and CD55 messenger RNA (mRNA) expression was assessed in peripheral white blood cells (WBCs) of patients with IgAN enrolled by the European collaborative study group Validation of the Oxford Classification of IgAN (VALIGA) [19]. They were investigated after a follow-up of several years, aimed at detecting correlations between cell surface complement regulatory protein mRNA expression and previous clinical course and rate of renal function decline. MATERIALS AND METHODS Patients VALIGA was a European multicentre retrospective study that enrolled 1147 patients with IgAN followed over a median of 4.7 years [19]. Patients with Henoch–Schönlein nephritis, chronic hepatitis, diabetes or cancer were excluded. The present study enrolled 157 patients with IgAN belonging to 11 centres from seven European countries of the VALIGA network. CD46 and CD55 mRNA expression quantification Blood samples were collected in PAXgene tubes (QIAGEN, Hilden, Germany), incubated 4 h at room temperature and stored at −80°C. RNA was extracted from WBCs with the PAXgene RNA System Kit (QIAGEN, Hilden, Germany). Total RNA (500 ng) was retro-transcribed in cDNA using Reverse Transcription Reagents (Life Technologies, Carlsbad, CA, USA) and using a 2720 Thermal Cycler (Applied Biosystems, Waltham, MA, USA). Quantitative real-time polymerase chain reaction was performed using TaqMan Universal PCR Master Mix (Life Technologies). For the detection of mRNAs encoding for CD46 and CD55, 900 nM of forward primer, 500 nM of reverse primer and 200 nM of the fluorogenic probe were used (Sigma-Aldrich, St. Louis, MO, USA): CD46 primer sequence forward: TGGTGACAATTCAGTGTGGAGTC; reverse: CCAAATCCTGATATCTGTTTTCCATTTTCGA and probe: 6FAM-TGCTCCAGAGTGTAAAGTGGTCAAATGTCG-TAMRA and CD55 primer sequence forward: TGCTCTGCAAGTTAGACCTTTTGA; reverse: TGACCTAGAGGATGCACATCATCT and probe: 6FAM-TGTCTGGGTCATCCCACATTTCTTCA. AAA-TAMRA cDNA 50 ng was used for amplification by the ABI Prism 7500 Sequence Detection System in 96-well plates (Life Technologies). Relative quantification of mRNA expression of CD46 and CD55 genes was achieved by normalization to the reference gene GAPDH [20]. Relative quantification of target gene expression in patients was performed with the ΔΔCt method and the relative CD46 and CD55 fold changes were determined as previously detailed; hence results are expressed in corresponding arbitrary units (U) [20]. As healthy controls, 150 Caucasian subjects, matched for age and sex, were investigated as well. Clinical data set and definitions Definition of data followed the original VALIGA study [19]. Briefly, glomerular filtration rate (GFR) was estimated using the four-variable Modification of Diet in Renal Disease formula and proteinuria expressed in grams per day. Mean arterial blood pressure (MAP) was calculated as one-third of the pulse pressure. At renal biopsy, 41 subjects were children (<18 years old): in these subjects, GFR was estimated using the Schwartz formula (constant K = 0.55) with a maximum estimated GFR (eGFR) set at 120 mL/min/1.73 m2; proteinuria was expressed in g/day/1.73 m2. MAP was adjusted for gender and age, as in the original report [19]. At sampling, only three subjects were <18 years old. ESRD was defined as eGFR <15 mL/min/1.73 m2 in all patients. Time-averaged (TA) proteinuria was determined for each year of observation. Immunosuppressive treatment was considered as intent to treat regardless of the type or duration. Renal biopsies were scored according to the Oxford Classification: mesangial hypercellularity, M0/M1 (≤ and >50% of glomeruli with more than four mesangial cells/area); endocapillary hypercellularity, E0/E1 (present/absent); segmental glomerulosclerosis, S0/S1 (present/absent) and tubular atrophy/interstitial fibrosis, T0/T1/T2 (<25, 25–50 and >50%) [19]. Moreover, the presence of crescents (C1) was considered [20]. Statistical methods The rate of renal function decline (eGFR slope) was determined by fitting a straight line through available eGFR data as previously described [19]. Disease progression was defined according to the median value of eGFR slope in the study cohort. The functional form of all continuous variables was assessed, with log transformation of CD46 and CD55 expression to improve linearity. Data were tested for normal distribution using the Shapiro–Wilk test. Descriptive variables were presented as median [interquartile range (IQR)], owing to non-normal distributions, or frequency (count) and compared across relevant groups using the Mann–Whitney U test or Wilcoxon signed-rank test for continuous variables and Fisher’s exact test for categorical variables. The Kendall test was used to investigate correlations. Group tests were two-sided with P < 0.05 considered statistically significant. Kendall correlation analysis was used to test the association between gene expression levels and eGFR slope. Univariate and multivariate regression analyses were performed to calculate associations between gene expression and other clinical variables, including disease progression. Receiver operating characteristics (ROC) curves served to identify the optimal threshold point of CD46 expression based on disease progression status. Multivariate logistic regression analyses were chosen to investigate the accuracy of disease progression prediction models. Four different statistical models were created: M1: clinical data at biopsy alone (eGFR, proteinuria and MAP); M2: clinical data at biopsy with MEST-C (mesangial hypercellularity, endocapillary hypercellularity, segmental glomerulosclerosis, tubular atrophy/interstitial fibrosis, with addition of the presence of any crescents C); M3: clinical data at biopsy with MEST-C and CD46 mRNA expression; M4: clinical data at biopsy with MEST-C and CD46 and CD55 mRNA expressions. Renal function at sampling was not included in the models for consistency with the previously used methods [19, 34]. All variables were corrected for age, sex and renal function at biopsy. Overall, model fit was assessed using McFadden pseudo-R2, Nagelkerke’s R2 and the Akaike Information Criteria (AIC), with an increase in R2 and a reduction in AIC suggesting better model fit. Discrimination was assessed using the area under the curve (AUC). When comparing two models, a ΔAUC significantly greater than zero suggested improvement in discrimination. Confidence intervals (CIs) were generated using 1000 bootstrap samples. Analyses were performed and figures generated using R 3.4.1 (R Project for Statistical Computing, Vienna, Austria). RESULTS The 157 subjects with IgAN enroled in this study had, at the time of sampling, a median follow-up of 6.4 years (IQR 2.8–10.7) after renal biopsy. Baseline demographic and clinical data both at the time of renal biopsy and at sampling are reported in Table 1. The median value of the eGFR slope rate of decline was −0.41 mL/min/1.73 m2/year (IQR −1.91–0.87) and it was used to categorize patients into 79 progressive patients, denoted ‘progressors’ (eGFR decrease ≥−0.41 mL/min/1.73 m2/year) and 78 patients without progression, denoted ‘non-progressors’ (Figure 1). Progressors had, during the previous follow-up, a median loss of eGFR of −1.91 (IQR −3.81 to −1.09) ml/min/1.73 m2/year and non-progressors had a median improvement in eGFR of 0.89 (IQR 0.15–6.03) mL/min/1.73 m2/year (P < 0.01). Nine patients in the progressor group had a 50% decline in eGFR; of them, five reached eGFR <15 mL/min. Baseline clinical and pathology data at sampling are reported in Table 2. The two groups of progressors and non-progressors did not differ in clinical data at renal biopsy, including age, eGFR, proteinuria and MAP. Tubular atrophy or interstitial fibrosis (T1–T2 lesions) was more frequent in progressors (P = 0.004); all seven patients who had T2 lesions were progressors. The percentage of patients exposed to corticosteroid/immunosuppressive treatment or to renin–angiotensin system blockade (RASB) drugs during the follow-up was similar in progressors and non-progressors (Table 2). Since the group included five patients with ESRD (eGFR <15 mL/min), a subgroup of 152 patients with a minimum eGFR >15 mL/min was identified and investigated (Supplementary data, Table S1). According to their median eGFR slope of −0.40 mL/min/1.73 m2/year, the 152 patients were divided into progressors and non-progressors, with median eGFR changes of −1.79 (IQR −3.34 to −1.05) mL/min/1.73 m2/year and 1.04 (IQR 0.24–6.42) mL/min/1.73 m2/year, respectively (data reported in Supplementary data, Table S2). We focused on the whole group of 157 patients. Table 1 Demographic and clinical data at renal biopsy and at sampling of the cohort of 157 patients with IgAN Clinical data At renal biopsy At sampling Number of patients 157 Gender (female), n (%) 51 (32.5) Age (years) 36.8 (23.1–49.4) 44.4 (31.9–56.4) eGFR (mL/min/1.73 m2) 70.92 (48.48–98.72) 73.3 (45.9–89.8) Proteinuria (g/day/1.73 m2) 1.1 (0.4–2.04) 0.4 (0.17–0.8) MAP (mmHg) 100 (86.7–106.7) 93.3 (84.3–97.8) Biopsy features, % M1 54.7 E1 21.6 S1 57.3 T1–T2 29.3 C1 12.7 Follow-up data Duration of follow-up (years) 6.4 (2.8–10.7) TA proteinuria (g/day/1.73 m2) 0.74 (0.32–1.31) RASB treatment, % 83.4 Cs/Is treatment, % 34.4 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −0.41 (−1.91–0.87) 50% loss of eGFR from baseline, % 5.7 CD46 mRNA log U −0.25 (−0.68–0.25) CD55 mRNA log U −0.47 (−0.98 to −0.01) Clinical data At renal biopsy At sampling Number of patients 157 Gender (female), n (%) 51 (32.5) Age (years) 36.8 (23.1–49.4) 44.4 (31.9–56.4) eGFR (mL/min/1.73 m2) 70.92 (48.48–98.72) 73.3 (45.9–89.8) Proteinuria (g/day/1.73 m2) 1.1 (0.4–2.04) 0.4 (0.17–0.8) MAP (mmHg) 100 (86.7–106.7) 93.3 (84.3–97.8) Biopsy features, % M1 54.7 E1 21.6 S1 57.3 T1–T2 29.3 C1 12.7 Follow-up data Duration of follow-up (years) 6.4 (2.8–10.7) TA proteinuria (g/day/1.73 m2) 0.74 (0.32–1.31) RASB treatment, % 83.4 Cs/Is treatment, % 34.4 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −0.41 (−1.91–0.87) 50% loss of eGFR from baseline, % 5.7 CD46 mRNA log U −0.25 (−0.68–0.25) CD55 mRNA log U −0.47 (−0.98 to −0.01) eGFR calculated by the modified Schwartz or Modification of Diet in Renal Disease formula (see Materials and Methods section. MEST-C available from 124 patients; M1, mesangial hypercellularity (>50 of glomeruli with mesangial hypercellularity); E1, presence of endocapillary hypercellularity; S1, presence of segmental glomerular sclerosis; T1–2, tubular atrophy/interstitial fibrosis in ≥25% of renal biopsy tissues; C1, presence of any crescents; TA, time average; RASB, renin–angiotensin system blockade; Cs, corticosteroids; Is, immunosuppressive drugs. Values are expressed as median (IQR) unless stated otherwise. Table 1 Demographic and clinical data at renal biopsy and at sampling of the cohort of 157 patients with IgAN Clinical data At renal biopsy At sampling Number of patients 157 Gender (female), n (%) 51 (32.5) Age (years) 36.8 (23.1–49.4) 44.4 (31.9–56.4) eGFR (mL/min/1.73 m2) 70.92 (48.48–98.72) 73.3 (45.9–89.8) Proteinuria (g/day/1.73 m2) 1.1 (0.4–2.04) 0.4 (0.17–0.8) MAP (mmHg) 100 (86.7–106.7) 93.3 (84.3–97.8) Biopsy features, % M1 54.7 E1 21.6 S1 57.3 T1–T2 29.3 C1 12.7 Follow-up data Duration of follow-up (years) 6.4 (2.8–10.7) TA proteinuria (g/day/1.73 m2) 0.74 (0.32–1.31) RASB treatment, % 83.4 Cs/Is treatment, % 34.4 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −0.41 (−1.91–0.87) 50% loss of eGFR from baseline, % 5.7 CD46 mRNA log U −0.25 (−0.68–0.25) CD55 mRNA log U −0.47 (−0.98 to −0.01) Clinical data At renal biopsy At sampling Number of patients 157 Gender (female), n (%) 51 (32.5) Age (years) 36.8 (23.1–49.4) 44.4 (31.9–56.4) eGFR (mL/min/1.73 m2) 70.92 (48.48–98.72) 73.3 (45.9–89.8) Proteinuria (g/day/1.73 m2) 1.1 (0.4–2.04) 0.4 (0.17–0.8) MAP (mmHg) 100 (86.7–106.7) 93.3 (84.3–97.8) Biopsy features, % M1 54.7 E1 21.6 S1 57.3 T1–T2 29.3 C1 12.7 Follow-up data Duration of follow-up (years) 6.4 (2.8–10.7) TA proteinuria (g/day/1.73 m2) 0.74 (0.32–1.31) RASB treatment, % 83.4 Cs/Is treatment, % 34.4 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −0.41 (−1.91–0.87) 50% loss of eGFR from baseline, % 5.7 CD46 mRNA log U −0.25 (−0.68–0.25) CD55 mRNA log U −0.47 (−0.98 to −0.01) eGFR calculated by the modified Schwartz or Modification of Diet in Renal Disease formula (see Materials and Methods section. MEST-C available from 124 patients; M1, mesangial hypercellularity (>50 of glomeruli with mesangial hypercellularity); E1, presence of endocapillary hypercellularity; S1, presence of segmental glomerular sclerosis; T1–2, tubular atrophy/interstitial fibrosis in ≥25% of renal biopsy tissues; C1, presence of any crescents; TA, time average; RASB, renin–angiotensin system blockade; Cs, corticosteroids; Is, immunosuppressive drugs. Values are expressed as median (IQR) unless stated otherwise. Table 2 Clinical data in patients with progressive IgAN (progressors) and non-progressive IgAN (non-progressors) Clinical data Progressors Non-progressors P-value [n=79 (50.3%)] [n=78 (49.7%)] Gender (female), n (%) 20 (25.3) 31 (39.7) 0.06 Age at biopsy (years) 36.8 (24.5–47.1) 36.5 (21.2–49.6) 0.84 eGFR at renal biopsy (mL/min/1.73 m2) 73.87 (43.01–100.83) 69.31 (51.3–96.51) 0.39 Proteinuria at renal biopsy (g/day/1.73 m2) 1.22 (0.48–2.24) 1.0 (0.32–1.91) 0.20 MAP at biopsy (mmHg) 100 (87.4–106.7) 98 (86.1–106.7) 0.72 Biopsy features, % M1 52.5 57.1 0.72 E1 14.8 28.6 0.08 S1 55.7 58.7 0.85 T1–2 39.3 19.0 0.004 C1 13.1 12.7 1 Follow-up data Age at sampling 45.1 (34.1–56.8) 42.5 (28.5–56.4) 0.29 Duration of follow-up (years) 8.0 (3.4–12.3) 5.5 (2.5–9.4) 0.03 MAP at sampling (mmHg) 93.3 (83.9–100.4) 93.3 (86.7–96.7) 0.83 Proteinuria at sampling (g/day/1.73 m2) 0.4 (0.11–0.94) 0.36 (0.2–0.66) 0.75 TA proteinuria (g/day/1.73 m2) 0.82 (0.41–1.37) 0.68 (0.31–1.15) 0.14 RASB treatment, % 67.1 82.5 1 Cs/Is treatment, % 29.1 31.7 0.70 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −1.91 (−3.81 to −1.09) 0.89 (0.15–6.03) <0.01 Decline in eGFR >50% from baseline/ESRD, % 11.4 0 <0.01 Complement inhibitors mRNA expression in WBCs CD46 mRNA log U −0.46 (−0.97–0.20) −0.14 (−0.49–0.31) 0.013 CD55 mRNA log U −0.49 (−1.01 to −0.02) −0.46 (−0.95–0.03) 0.78 Clinical data Progressors Non-progressors P-value [n=79 (50.3%)] [n=78 (49.7%)] Gender (female), n (%) 20 (25.3) 31 (39.7) 0.06 Age at biopsy (years) 36.8 (24.5–47.1) 36.5 (21.2–49.6) 0.84 eGFR at renal biopsy (mL/min/1.73 m2) 73.87 (43.01–100.83) 69.31 (51.3–96.51) 0.39 Proteinuria at renal biopsy (g/day/1.73 m2) 1.22 (0.48–2.24) 1.0 (0.32–1.91) 0.20 MAP at biopsy (mmHg) 100 (87.4–106.7) 98 (86.1–106.7) 0.72 Biopsy features, % M1 52.5 57.1 0.72 E1 14.8 28.6 0.08 S1 55.7 58.7 0.85 T1–2 39.3 19.0 0.004 C1 13.1 12.7 1 Follow-up data Age at sampling 45.1 (34.1–56.8) 42.5 (28.5–56.4) 0.29 Duration of follow-up (years) 8.0 (3.4–12.3) 5.5 (2.5–9.4) 0.03 MAP at sampling (mmHg) 93.3 (83.9–100.4) 93.3 (86.7–96.7) 0.83 Proteinuria at sampling (g/day/1.73 m2) 0.4 (0.11–0.94) 0.36 (0.2–0.66) 0.75 TA proteinuria (g/day/1.73 m2) 0.82 (0.41–1.37) 0.68 (0.31–1.15) 0.14 RASB treatment, % 67.1 82.5 1 Cs/Is treatment, % 29.1 31.7 0.70 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −1.91 (−3.81 to −1.09) 0.89 (0.15–6.03) <0.01 Decline in eGFR >50% from baseline/ESRD, % 11.4 0 <0.01 Complement inhibitors mRNA expression in WBCs CD46 mRNA log U −0.46 (−0.97–0.20) −0.14 (−0.49–0.31) 0.013 CD55 mRNA log U −0.49 (−1.01 to −0.02) −0.46 (−0.95–0.03) 0.78 The median value of the eGFR slopes detected in the whole cohort of 157 patients was used to categorize patients into progressors (eGFR decrease ≥−0.41 mL/min/1.73 m2/year) and non-progressors. For legends, see Table 1. Values are expressed as median (IQR) unless stated otherwise. Table 2 Clinical data in patients with progressive IgAN (progressors) and non-progressive IgAN (non-progressors) Clinical data Progressors Non-progressors P-value [n=79 (50.3%)] [n=78 (49.7%)] Gender (female), n (%) 20 (25.3) 31 (39.7) 0.06 Age at biopsy (years) 36.8 (24.5–47.1) 36.5 (21.2–49.6) 0.84 eGFR at renal biopsy (mL/min/1.73 m2) 73.87 (43.01–100.83) 69.31 (51.3–96.51) 0.39 Proteinuria at renal biopsy (g/day/1.73 m2) 1.22 (0.48–2.24) 1.0 (0.32–1.91) 0.20 MAP at biopsy (mmHg) 100 (87.4–106.7) 98 (86.1–106.7) 0.72 Biopsy features, % M1 52.5 57.1 0.72 E1 14.8 28.6 0.08 S1 55.7 58.7 0.85 T1–2 39.3 19.0 0.004 C1 13.1 12.7 1 Follow-up data Age at sampling 45.1 (34.1–56.8) 42.5 (28.5–56.4) 0.29 Duration of follow-up (years) 8.0 (3.4–12.3) 5.5 (2.5–9.4) 0.03 MAP at sampling (mmHg) 93.3 (83.9–100.4) 93.3 (86.7–96.7) 0.83 Proteinuria at sampling (g/day/1.73 m2) 0.4 (0.11–0.94) 0.36 (0.2–0.66) 0.75 TA proteinuria (g/day/1.73 m2) 0.82 (0.41–1.37) 0.68 (0.31–1.15) 0.14 RASB treatment, % 67.1 82.5 1 Cs/Is treatment, % 29.1 31.7 0.70 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −1.91 (−3.81 to −1.09) 0.89 (0.15–6.03) <0.01 Decline in eGFR >50% from baseline/ESRD, % 11.4 0 <0.01 Complement inhibitors mRNA expression in WBCs CD46 mRNA log U −0.46 (−0.97–0.20) −0.14 (−0.49–0.31) 0.013 CD55 mRNA log U −0.49 (−1.01 to −0.02) −0.46 (−0.95–0.03) 0.78 Clinical data Progressors Non-progressors P-value [n=79 (50.3%)] [n=78 (49.7%)] Gender (female), n (%) 20 (25.3) 31 (39.7) 0.06 Age at biopsy (years) 36.8 (24.5–47.1) 36.5 (21.2–49.6) 0.84 eGFR at renal biopsy (mL/min/1.73 m2) 73.87 (43.01–100.83) 69.31 (51.3–96.51) 0.39 Proteinuria at renal biopsy (g/day/1.73 m2) 1.22 (0.48–2.24) 1.0 (0.32–1.91) 0.20 MAP at biopsy (mmHg) 100 (87.4–106.7) 98 (86.1–106.7) 0.72 Biopsy features, % M1 52.5 57.1 0.72 E1 14.8 28.6 0.08 S1 55.7 58.7 0.85 T1–2 39.3 19.0 0.004 C1 13.1 12.7 1 Follow-up data Age at sampling 45.1 (34.1–56.8) 42.5 (28.5–56.4) 0.29 Duration of follow-up (years) 8.0 (3.4–12.3) 5.5 (2.5–9.4) 0.03 MAP at sampling (mmHg) 93.3 (83.9–100.4) 93.3 (86.7–96.7) 0.83 Proteinuria at sampling (g/day/1.73 m2) 0.4 (0.11–0.94) 0.36 (0.2–0.66) 0.75 TA proteinuria (g/day/1.73 m2) 0.82 (0.41–1.37) 0.68 (0.31–1.15) 0.14 RASB treatment, % 67.1 82.5 1 Cs/Is treatment, % 29.1 31.7 0.70 Clinical outcomes Rate of eGFR loss (mL/min/1.73 m2/year) −1.91 (−3.81 to −1.09) 0.89 (0.15–6.03) <0.01 Decline in eGFR >50% from baseline/ESRD, % 11.4 0 <0.01 Complement inhibitors mRNA expression in WBCs CD46 mRNA log U −0.46 (−0.97–0.20) −0.14 (−0.49–0.31) 0.013 CD55 mRNA log U −0.49 (−1.01 to −0.02) −0.46 (−0.95–0.03) 0.78 The median value of the eGFR slopes detected in the whole cohort of 157 patients was used to categorize patients into progressors (eGFR decrease ≥−0.41 mL/min/1.73 m2/year) and non-progressors. For legends, see Table 1. Values are expressed as median (IQR) unless stated otherwise. FIGURE 1 View largeDownload slide Identification of patients with progressive and non-progressive IgAN, denoted ‘progressors’ and ‘non-progressors’, respectively. The median value of the eGFR slope in the 157 patients investigated (−0.41 mL/min/1.73 m2/year) allowed a categorization into progressors (eGFR decrease ≥−0.41 mL/min/1.73 m2/year) and non-progressors. (A) Histogram of frequency distribution of eGFR slopes in progressors and non-progressors. (B) Box plot of eGFR slopes in progressors and non-progressors. FIGURE 1 View largeDownload slide Identification of patients with progressive and non-progressive IgAN, denoted ‘progressors’ and ‘non-progressors’, respectively. The median value of the eGFR slope in the 157 patients investigated (−0.41 mL/min/1.73 m2/year) allowed a categorization into progressors (eGFR decrease ≥−0.41 mL/min/1.73 m2/year) and non-progressors. (A) Histogram of frequency distribution of eGFR slopes in progressors and non-progressors. (B) Box plot of eGFR slopes in progressors and non-progressors. CD46 and CD55 expression in WBCs was not significantly different in patients with IgAN and healthy controls; CD46 mRNA: −0.25 (IQR −0.68–0.25) log U versus −0.17 (−0.73–0.31) log U, P = 0.82 and CD55 mRNA: −0.47 (IQR −0.98 to −0.01) log U versus −0.28 (IQR −0.81–0.18) log U, P = 0.38, respectively. CD46 and CD55 gene expressions in IgAN patients were not correlated with baseline data at renal biopsy, including the Oxford Classification MEST-C scores [20], eGFR, proteinuria and MAP (Table 3 and Supplementary data, Table S3). No correlation was found with data at sampling, including eGFR, proteinuria or TA proteinuria over the previous follow-up from renal biopsy to sampling (Table 3 and Supplementary data, Table S3). A significant correlation was found between CD46 and CD55 mRNA expression (P < 0.01) (Table 3). Table 3 Correlation between CD46 mRNA log U and data at renal biopsy and at sampling Correlation with data at renal biopsy (RB) Coefficient 95% CI P-value Gender 0.257 0.001–0.512 0.05 Age at RB 0.002 −0.006–0.009 0.67 eGFR at RB −0.002 −0.006–0.002 0.43 Proteinuria at RB 0.049 −0.01–0.108 0.11 MAP at RB 0.001 −0.009–0.009 0.95 Biopsy features M1 0.208 −0.054–0.471 0.12 E1 −0.088 −0.408–0.231 0.59 S1 0.235 −0.029–0.499 0.08 T1 0.077 −0.238–0.392 0.63 T2 0.276 −0.302–0.854 0.35 C1 −0.290 −0.681–0.1 0.14 Correlations with data at sampling eGFR at sampling 0.003 −0.001–0.007 0.15 Proteinuria at sampling −0.003 −0.116–0.11 0.96 TA proteinuria 0.001 −0.142–0.144 0.99 CD55 mRNA log U 0.511 0.404–0.617 <0.01 Correlation with data at renal biopsy (RB) Coefficient 95% CI P-value Gender 0.257 0.001–0.512 0.05 Age at RB 0.002 −0.006–0.009 0.67 eGFR at RB −0.002 −0.006–0.002 0.43 Proteinuria at RB 0.049 −0.01–0.108 0.11 MAP at RB 0.001 −0.009–0.009 0.95 Biopsy features M1 0.208 −0.054–0.471 0.12 E1 −0.088 −0.408–0.231 0.59 S1 0.235 −0.029–0.499 0.08 T1 0.077 −0.238–0.392 0.63 T2 0.276 −0.302–0.854 0.35 C1 −0.290 −0.681–0.1 0.14 Correlations with data at sampling eGFR at sampling 0.003 −0.001–0.007 0.15 Proteinuria at sampling −0.003 −0.116–0.11 0.96 TA proteinuria 0.001 −0.142–0.144 0.99 CD55 mRNA log U 0.511 0.404–0.617 <0.01 Univariate linear regression analysis. For legends, see Table 1. Table 3 Correlation between CD46 mRNA log U and data at renal biopsy and at sampling Correlation with data at renal biopsy (RB) Coefficient 95% CI P-value Gender 0.257 0.001–0.512 0.05 Age at RB 0.002 −0.006–0.009 0.67 eGFR at RB −0.002 −0.006–0.002 0.43 Proteinuria at RB 0.049 −0.01–0.108 0.11 MAP at RB 0.001 −0.009–0.009 0.95 Biopsy features M1 0.208 −0.054–0.471 0.12 E1 −0.088 −0.408–0.231 0.59 S1 0.235 −0.029–0.499 0.08 T1 0.077 −0.238–0.392 0.63 T2 0.276 −0.302–0.854 0.35 C1 −0.290 −0.681–0.1 0.14 Correlations with data at sampling eGFR at sampling 0.003 −0.001–0.007 0.15 Proteinuria at sampling −0.003 −0.116–0.11 0.96 TA proteinuria 0.001 −0.142–0.144 0.99 CD55 mRNA log U 0.511 0.404–0.617 <0.01 Correlation with data at renal biopsy (RB) Coefficient 95% CI P-value Gender 0.257 0.001–0.512 0.05 Age at RB 0.002 −0.006–0.009 0.67 eGFR at RB −0.002 −0.006–0.002 0.43 Proteinuria at RB 0.049 −0.01–0.108 0.11 MAP at RB 0.001 −0.009–0.009 0.95 Biopsy features M1 0.208 −0.054–0.471 0.12 E1 −0.088 −0.408–0.231 0.59 S1 0.235 −0.029–0.499 0.08 T1 0.077 −0.238–0.392 0.63 T2 0.276 −0.302–0.854 0.35 C1 −0.290 −0.681–0.1 0.14 Correlations with data at sampling eGFR at sampling 0.003 −0.001–0.007 0.15 Proteinuria at sampling −0.003 −0.116–0.11 0.96 TA proteinuria 0.001 −0.142–0.144 0.99 CD55 mRNA log U 0.511 0.404–0.617 <0.01 Univariate linear regression analysis. For legends, see Table 1. A significant correlation was found between CD46 mRNA and eGFR decline slope as a continuous outcome variable (Kendall rank correlation: r = 0.14, P = 0.011; Figure 2A,B shows both non-parametric fitting and linear correlation). CD46 expression in WBCs was significantly lower in progressors versus non-progressors [−0.46 (IQR −0.97–0.20) log U and −0.14 (−0.49–0.31) log U, respectively; P = 0.013] (Figure 3 and Table 2). A significant difference in CD46 mRNA median values was also found between progressors and non-progressors when the five patients with eGFR <15 mL/min/1.73 m2 at sampling were excluded (Supplementary data, Table S2). FIGURE 2 View largeDownload slide Correlation between CD46 mRNA and eGFR slope (Kendall rank correlation test). (A) Non-parametric fitting (locally weighted smoothing) representation of the correlation curve. (B) Linear correlation representation (not strictly correct since data were not normally distributed). FIGURE 2 View largeDownload slide Correlation between CD46 mRNA and eGFR slope (Kendall rank correlation test). (A) Non-parametric fitting (locally weighted smoothing) representation of the correlation curve. (B) Linear correlation representation (not strictly correct since data were not normally distributed). FIGURE 3 View largeDownload slide CD46 mRNA expression in peripheral WBCs of patients with IgAN categorized into progressors and non-progressors. (A) Box plot of CD46 mRNA log U values and (B) one-dimensional individual CD46 mRNA log U values (solid dot represents the median). FIGURE 3 View largeDownload slide CD46 mRNA expression in peripheral WBCs of patients with IgAN categorized into progressors and non-progressors. (A) Box plot of CD46 mRNA log U values and (B) one-dimensional individual CD46 mRNA log U values (solid dot represents the median). The ROC curve of CD46 mRNA discerning progressors from non-progressors [AUC 0.61 (95% CI 0.52–0.70)] showed as best threshold, by the Youden index, of −0.38 (Figure 4). When comparing 91 patients with CD46 mRNA ≤ −0.38 log U and 66 patients with values >−0.38 log U (Table 4), a significant difference was observed for the eGFR slope (P = 0.002) and the decline in GFR of >50% from baseline value (P = 0.03). No difference was found for frequency of corticosteroid treatment at sampling or over the previous follow-up. Notably, the two groups differed for CD55 mRNA values [−0.84 (IQR −1.46 to −0.5) log U versus −0.21 (−0.58–0.27) log U; P < 0.01] (Table 4). No difference in CD55 mRNA expression was found between progressors and non-progressors (Table 2). The ROC curve of CD55 mRNA values did not discriminate between progressors and non-progressors (Figure 4). Table 4 Clinical data of two subgroups of the patients investigated presenting with low or high CD46 mRNA expression Variable CD46 mRNA ≤ −0.38 log U CD46 mRNA > −0.38 log U P-value Cases, n (%) 91 (58.0) 66 (42.0) Gender (female), n (%) 12 (18.2) 39 (42.9) 0.001 Age at biopsy (years) 35.8 (19.1–50.2) 36.9 (24.8–48.5) 0.63 Age at sampling (years) 44.3 (31.4–55.2) 44.4 (32.2–56.9) 0.86 eGFR at biopsy (mL/min/1.73 m2) 73.02 (51.13–99.52) 68.29 (46.72–98.06) 0.50 eGFR at sampling (mL/min/1.73 m2) 72.76 (41.38–83.16) 76.21 (48.64–95.14) 0.09 eGFR slope −1.13 (−3.02–0.24) 0.06 (−1.16–1.61) 0.002 Proteinuria at biopsy (g/day/1.73 m2) 1.0 (0.4–1.85) 1.2 (0.41–2.2) 0.39 Proteinuria at sampling (g/day/1.73 m2) 0.39 (0.15–0.8) 0.4 (0.17–0.8) 0.97 TA proteinuria at follow-up (g/day/1.73 m2) 0.84 (0.38–1.34) 0.67 (0.3–1.3) 0.48 Previous RASB treatment, n (%) 46 (82.1) 59 (84.3) 0.81 Previous Cs/Is treatment, n (%) 18 (32.1) 25 (35.7) 0.70 CS treatment at sampling, n (%) 20 (21.9) 8 (12) 0.14 MAP at biopsy (mmHg) 100 (90–103.3) 96.7 (84.9–106.8) 0.90 MAP at sampling (mmHg) 91.7 (86.1–95.6) 93.3 (83.4–100) 0.52 M1, % 51.9 57.1 0.59 E1, % 22.2 21.4 1 S1, % 50 62.9 0.20 T1–2, n (%) 15 (27.8) (30) 0.82 C1, n (%) 9 (16.7) 7 (10) 0.29 Follow-up (years) 7.4 (2.9–12.3) 5.6 (2.7–9.9) 0.15 Decline in eGFR >50% from baseline, n (%) 7 (10.6) 2 (2.2) 0.03 CD55 mRNA log U −0.84 (−1.46 to −0.5) −0.21 (−0.58–0.27) <0.01 Variable CD46 mRNA ≤ −0.38 log U CD46 mRNA > −0.38 log U P-value Cases, n (%) 91 (58.0) 66 (42.0) Gender (female), n (%) 12 (18.2) 39 (42.9) 0.001 Age at biopsy (years) 35.8 (19.1–50.2) 36.9 (24.8–48.5) 0.63 Age at sampling (years) 44.3 (31.4–55.2) 44.4 (32.2–56.9) 0.86 eGFR at biopsy (mL/min/1.73 m2) 73.02 (51.13–99.52) 68.29 (46.72–98.06) 0.50 eGFR at sampling (mL/min/1.73 m2) 72.76 (41.38–83.16) 76.21 (48.64–95.14) 0.09 eGFR slope −1.13 (−3.02–0.24) 0.06 (−1.16–1.61) 0.002 Proteinuria at biopsy (g/day/1.73 m2) 1.0 (0.4–1.85) 1.2 (0.41–2.2) 0.39 Proteinuria at sampling (g/day/1.73 m2) 0.39 (0.15–0.8) 0.4 (0.17–0.8) 0.97 TA proteinuria at follow-up (g/day/1.73 m2) 0.84 (0.38–1.34) 0.67 (0.3–1.3) 0.48 Previous RASB treatment, n (%) 46 (82.1) 59 (84.3) 0.81 Previous Cs/Is treatment, n (%) 18 (32.1) 25 (35.7) 0.70 CS treatment at sampling, n (%) 20 (21.9) 8 (12) 0.14 MAP at biopsy (mmHg) 100 (90–103.3) 96.7 (84.9–106.8) 0.90 MAP at sampling (mmHg) 91.7 (86.1–95.6) 93.3 (83.4–100) 0.52 M1, % 51.9 57.1 0.59 E1, % 22.2 21.4 1 S1, % 50 62.9 0.20 T1–2, n (%) 15 (27.8) (30) 0.82 C1, n (%) 9 (16.7) 7 (10) 0.29 Follow-up (years) 7.4 (2.9–12.3) 5.6 (2.7–9.9) 0.15 Decline in eGFR >50% from baseline, n (%) 7 (10.6) 2 (2.2) 0.03 CD55 mRNA log U −0.84 (−1.46 to −0.5) −0.21 (−0.58–0.27) <0.01 Patients were divided in two groups according to the discrimination point of the ROC curve of CD46 gene expression (−0.38 log U, see Figure 4). Values are expressed as median (IQR) unless stated otherwise. Table 4 Clinical data of two subgroups of the patients investigated presenting with low or high CD46 mRNA expression Variable CD46 mRNA ≤ −0.38 log U CD46 mRNA > −0.38 log U P-value Cases, n (%) 91 (58.0) 66 (42.0) Gender (female), n (%) 12 (18.2) 39 (42.9) 0.001 Age at biopsy (years) 35.8 (19.1–50.2) 36.9 (24.8–48.5) 0.63 Age at sampling (years) 44.3 (31.4–55.2) 44.4 (32.2–56.9) 0.86 eGFR at biopsy (mL/min/1.73 m2) 73.02 (51.13–99.52) 68.29 (46.72–98.06) 0.50 eGFR at sampling (mL/min/1.73 m2) 72.76 (41.38–83.16) 76.21 (48.64–95.14) 0.09 eGFR slope −1.13 (−3.02–0.24) 0.06 (−1.16–1.61) 0.002 Proteinuria at biopsy (g/day/1.73 m2) 1.0 (0.4–1.85) 1.2 (0.41–2.2) 0.39 Proteinuria at sampling (g/day/1.73 m2) 0.39 (0.15–0.8) 0.4 (0.17–0.8) 0.97 TA proteinuria at follow-up (g/day/1.73 m2) 0.84 (0.38–1.34) 0.67 (0.3–1.3) 0.48 Previous RASB treatment, n (%) 46 (82.1) 59 (84.3) 0.81 Previous Cs/Is treatment, n (%) 18 (32.1) 25 (35.7) 0.70 CS treatment at sampling, n (%) 20 (21.9) 8 (12) 0.14 MAP at biopsy (mmHg) 100 (90–103.3) 96.7 (84.9–106.8) 0.90 MAP at sampling (mmHg) 91.7 (86.1–95.6) 93.3 (83.4–100) 0.52 M1, % 51.9 57.1 0.59 E1, % 22.2 21.4 1 S1, % 50 62.9 0.20 T1–2, n (%) 15 (27.8) (30) 0.82 C1, n (%) 9 (16.7) 7 (10) 0.29 Follow-up (years) 7.4 (2.9–12.3) 5.6 (2.7–9.9) 0.15 Decline in eGFR >50% from baseline, n (%) 7 (10.6) 2 (2.2) 0.03 CD55 mRNA log U −0.84 (−1.46 to −0.5) −0.21 (−0.58–0.27) <0.01 Variable CD46 mRNA ≤ −0.38 log U CD46 mRNA > −0.38 log U P-value Cases, n (%) 91 (58.0) 66 (42.0) Gender (female), n (%) 12 (18.2) 39 (42.9) 0.001 Age at biopsy (years) 35.8 (19.1–50.2) 36.9 (24.8–48.5) 0.63 Age at sampling (years) 44.3 (31.4–55.2) 44.4 (32.2–56.9) 0.86 eGFR at biopsy (mL/min/1.73 m2) 73.02 (51.13–99.52) 68.29 (46.72–98.06) 0.50 eGFR at sampling (mL/min/1.73 m2) 72.76 (41.38–83.16) 76.21 (48.64–95.14) 0.09 eGFR slope −1.13 (−3.02–0.24) 0.06 (−1.16–1.61) 0.002 Proteinuria at biopsy (g/day/1.73 m2) 1.0 (0.4–1.85) 1.2 (0.41–2.2) 0.39 Proteinuria at sampling (g/day/1.73 m2) 0.39 (0.15–0.8) 0.4 (0.17–0.8) 0.97 TA proteinuria at follow-up (g/day/1.73 m2) 0.84 (0.38–1.34) 0.67 (0.3–1.3) 0.48 Previous RASB treatment, n (%) 46 (82.1) 59 (84.3) 0.81 Previous Cs/Is treatment, n (%) 18 (32.1) 25 (35.7) 0.70 CS treatment at sampling, n (%) 20 (21.9) 8 (12) 0.14 MAP at biopsy (mmHg) 100 (90–103.3) 96.7 (84.9–106.8) 0.90 MAP at sampling (mmHg) 91.7 (86.1–95.6) 93.3 (83.4–100) 0.52 M1, % 51.9 57.1 0.59 E1, % 22.2 21.4 1 S1, % 50 62.9 0.20 T1–2, n (%) 15 (27.8) (30) 0.82 C1, n (%) 9 (16.7) 7 (10) 0.29 Follow-up (years) 7.4 (2.9–12.3) 5.6 (2.7–9.9) 0.15 Decline in eGFR >50% from baseline, n (%) 7 (10.6) 2 (2.2) 0.03 CD55 mRNA log U −0.84 (−1.46 to −0.5) −0.21 (−0.58–0.27) <0.01 Patients were divided in two groups according to the discrimination point of the ROC curve of CD46 gene expression (−0.38 log U, see Figure 4). Values are expressed as median (IQR) unless stated otherwise. FIGURE 4 View largeDownload slide ROC curve showing true-positive versus false-positive data for logistic regression based on CD46 and CD55 mRNA expressions. Values were dichotomized on the basis of the median eGFR slope of the cohort of 157 patients (eGFR decrease ≥−0.41 mL/min/1.73 m2/year). The AUC serves as a predictive value for CD46 mRNA log and CD55 mRNA log and it is reported in percentage (%) with sensitivity and specificity in brackets. The best threshold for CD46 mRNA expression was −0.38 log U. FIGURE 4 View largeDownload slide ROC curve showing true-positive versus false-positive data for logistic regression based on CD46 and CD55 mRNA expressions. Values were dichotomized on the basis of the median eGFR slope of the cohort of 157 patients (eGFR decrease ≥−0.41 mL/min/1.73 m2/year). The AUC serves as a predictive value for CD46 mRNA log and CD55 mRNA log and it is reported in percentage (%) with sensitivity and specificity in brackets. The best threshold for CD46 mRNA expression was −0.38 log U. In order to address the value of CD46 and CD55 on the progression of IgAN, we derived ROC curves using clinical data at renal biopsy (eGFR, proteinuria and MAP), clinical data plus the revised Oxford MEST-C scores and clinical data and MEST-C scores, plus CD46 and CD55 mRNA expressions (Figure 5A–D). When MEST-C was added to clinical data at biopsy, an improvement in prognostication of progression was observed, which was further improved when we included the data of the expression of CD46 and CD55 mRNAs, as indicated by the values of AUC, McFadden pseudo-R2, Nagelkerke R2 and AIC (Table 5). Table 5 Multivariate models for the risk of being progressors Model AUC (95% CI) McFadden pseudo-R2 Nagelkerke’s R2 AIC M1: Model containing clinical data at renal biopsy 0.62 (0.53–0.71) 0.03 0.06 211.28 M2: Model containing clinical data at renal biopsy + MEST-C 0.72 (0.63–0.81) 0.14 0.23 165.49 ΔM0 0.10 (0.10–0.10) 0.11 0.17 −45.79 M3: Model containing clinical data at renal biopsy + MEST-C + CD46 mRNA 0.78 (0.70–0.86) 0.21 0.33 156 M4: Model containing clinical data at renal biopsy + MEST-C + CD46 and CD55mRNA 0.81 (0.73–0.89) 0.24 0.38 152.40 ΔM0 0.19 (0.2–0.18) 0.21 −58.88 ΔM1 0.09 (0.1–0.08) 0.10 0.15 −13.09 Model AUC (95% CI) McFadden pseudo-R2 Nagelkerke’s R2 AIC M1: Model containing clinical data at renal biopsy 0.62 (0.53–0.71) 0.03 0.06 211.28 M2: Model containing clinical data at renal biopsy + MEST-C 0.72 (0.63–0.81) 0.14 0.23 165.49 ΔM0 0.10 (0.10–0.10) 0.11 0.17 −45.79 M3: Model containing clinical data at renal biopsy + MEST-C + CD46 mRNA 0.78 (0.70–0.86) 0.21 0.33 156 M4: Model containing clinical data at renal biopsy + MEST-C + CD46 and CD55mRNA 0.81 (0.73–0.89) 0.24 0.38 152.40 ΔM0 0.19 (0.2–0.18) 0.21 −58.88 ΔM1 0.09 (0.1–0.08) 0.10 0.15 −13.09 Data were adjusted for age, gender and baseline eGFR. AUC, area under the curve for C-statistics; AIC, Akaiake information criterion. Changes are expressed as Δ. Table 5 Multivariate models for the risk of being progressors Model AUC (95% CI) McFadden pseudo-R2 Nagelkerke’s R2 AIC M1: Model containing clinical data at renal biopsy 0.62 (0.53–0.71) 0.03 0.06 211.28 M2: Model containing clinical data at renal biopsy + MEST-C 0.72 (0.63–0.81) 0.14 0.23 165.49 ΔM0 0.10 (0.10–0.10) 0.11 0.17 −45.79 M3: Model containing clinical data at renal biopsy + MEST-C + CD46 mRNA 0.78 (0.70–0.86) 0.21 0.33 156 M4: Model containing clinical data at renal biopsy + MEST-C + CD46 and CD55mRNA 0.81 (0.73–0.89) 0.24 0.38 152.40 ΔM0 0.19 (0.2–0.18) 0.21 −58.88 ΔM1 0.09 (0.1–0.08) 0.10 0.15 −13.09 Model AUC (95% CI) McFadden pseudo-R2 Nagelkerke’s R2 AIC M1: Model containing clinical data at renal biopsy 0.62 (0.53–0.71) 0.03 0.06 211.28 M2: Model containing clinical data at renal biopsy + MEST-C 0.72 (0.63–0.81) 0.14 0.23 165.49 ΔM0 0.10 (0.10–0.10) 0.11 0.17 −45.79 M3: Model containing clinical data at renal biopsy + MEST-C + CD46 mRNA 0.78 (0.70–0.86) 0.21 0.33 156 M4: Model containing clinical data at renal biopsy + MEST-C + CD46 and CD55mRNA 0.81 (0.73–0.89) 0.24 0.38 152.40 ΔM0 0.19 (0.2–0.18) 0.21 −58.88 ΔM1 0.09 (0.1–0.08) 0.10 0.15 −13.09 Data were adjusted for age, gender and baseline eGFR. AUC, area under the curve for C-statistics; AIC, Akaiake information criterion. Changes are expressed as Δ. FIGURE 5 View largeDownload slide ROC curves from multivariate logistic regression analyses of patients with progressive versus non-progressive IgAN. (A) M1, model containing clinical data at renal biopsy (eGFR, proteinuria and MAP). (B) M2, model containing clinical data at renal biopsy + MEST-C. (C) M3, model containing clinical data at renal biopsy + MEST-C + CD46 mRNA log U values. (D) M4, model containing clinical data at renal biopsy + MEST-C + CD46 mRNA and CD55 mRNA log U values. When MEST-C Oxford scores were added to clinical data at biopsy (B), there was an improvement in prognostication of progression with respect to clinical data at renal biopsy (A), which was further improved by including the data of the expression of CD46 and CD55 mRNAs (C and D), as indicated by the values of the AUC. Table 5 reports further detailed analyses. FIGURE 5 View largeDownload slide ROC curves from multivariate logistic regression analyses of patients with progressive versus non-progressive IgAN. (A) M1, model containing clinical data at renal biopsy (eGFR, proteinuria and MAP). (B) M2, model containing clinical data at renal biopsy + MEST-C. (C) M3, model containing clinical data at renal biopsy + MEST-C + CD46 mRNA log U values. (D) M4, model containing clinical data at renal biopsy + MEST-C + CD46 mRNA and CD55 mRNA log U values. When MEST-C Oxford scores were added to clinical data at biopsy (B), there was an improvement in prognostication of progression with respect to clinical data at renal biopsy (A), which was further improved by including the data of the expression of CD46 and CD55 mRNAs (C and D), as indicated by the values of the AUC. Table 5 reports further detailed analyses. DISCUSSION In spite of the common belief that complement activation and its downstream effects increase the nephrotoxicity of otherwise innocent—lanthanic—mesangial IgA deposits [21], the data in favour of a simple relationship between C3 deposition and progression of IgAN have not been as convincing as expected. Only limited reports indicate a negative correlation between circulating C3 levels and mesangial C3 deposits [22]. Plasma C3 was not clearly reduced in Caucasian patients with IgAN, although in Chinese and Japanese patients slightly reduced levels have been reported [22, 23]. The ratio between plasma IgA and C3 only partially highlights the mild and often undetectable decrease in C3 levels [24]. However, more sensitive biomarkers such as the complement breakdown products iC3b and C3d have detected systemic signs of complement activation in active and progressive cases [3, 25–27] in correlation with high plasma levels of complement-fixing IgAIC [28]. Apart from fluid phase complement activation, C3 in mesangial deposits may have a cellular origin produced by resident mesangial cells [7] or by inflammatory blood cells [29] infiltrating the mesangial area engulfed by IgAIC. Locally produced as well as IgAIC-transported complement factors can further generate C3a, C5a and C3b, which can interact with their receptors on resident or infiltrating cells, thus increasing glomerular inflammation and tissue damage. The interest in understanding the mechanism of complement activation in IgAN has recently been further increased by two reports of high serum levels of FHR-1 and FHR-5, which act as competitive inhibitors of FH [16, 17]. FHR-1 was negatively correlated with eGFR at sampling. The present study reports for the first time a defective gene expression of specific membrane-bound complement regulatory factors that characterize patients with progressive IgAN. The most relevant finding was that defective CD46 gene expression was not correlated with eGFR at sampling but with a faster annual loss of GFR measured over the previous long period of observation. Low CD46 mRNA correlated with a consonant low gene expression of CD55. These findings provide new insights into the area of complement dysregulation at a cellular level resulting in defective control of local complement activation in patients with IgAN. CD46 is a surface-expressed regulator acting as cofactor for serum FI, which cleaves C3b to iC3b, thereby irreversibly preventing the reassembly of AP amplifying activity [11, 12]. CD55 is a membrane-bound regulator factor that accelerates the decay of cell surface–assembled C3 and C5 convertases favouring the disassociation of Bb from C3bBb and C2b from C4b and inhibiting their re-aggregation [11, 12], hence preventing formation of the final MAC. Both regulatory factors, whose expression was detected in the present study in circulating WBCs, are expressed on monocytes and infiltrating macrophages and in resident cells, including mesangial cells. CD55 is expressed also in endothelial cells and podocytes. The correlation we found between progressive loss of renal function and gene expression of these complement regulatory factors, which may act in circulating, infiltrating or resident glomerular cells, suggests a reduced control of complement activation triggered by circulating and deposited IgA1-containing circulating immune complexes, which may favour glomerular damage and disease progression. Moreover, the intracellular complement activation is important for regulation of the adaptive immune response. After interaction, T cells and antigen presenting cells (APCs) produce and release AP components C3, FB and FD. This is regulated by a transient cell surface expression of CD55. Locally produced C3a and C5a bind to their receptors and further stimulate T cells and APCs [30]. CD46 is expressed on T cells and promotes the switch from Th1 to regulatory T cells (Tregs) [31, 32]. C3a and C5a induce phosphorylation of the transcription factor Foxo1, which results in lowered naïve Treg Foxp3 expression. We, and others as well, detected in IgAN a reduced Treg expression of Foxp3 [33]. We may speculate that continuous subtle complement activation in IgAN, due to defective complement activation control at the cellular level, results in decreased Treg activity in IgAN. This might contribute to a more progressive course. This study detected the transcriptional expression of CD46 and CD55 in peripheral WBCs, which might be different from that in infiltrating of resident glomerular cells, hence further investigations will be needed to detect these biomarkers in renal biopsy tissues. However, the present results we obtained in WBCs are of interest for the clinical correlations we observed, even in case of a lack of detection in future studies of a similar downregulation in renal cells. This study has the limitation of being a cross-sectional examination, since patients were investigated for defective gene expression of cellular complement inhibitory factors years after renal biopsy. On the other hand, it has the advantage of a previous long-term follow-up of patients very well characterized from a clinical and histological point of view. However, due to its design, the study has the limitation of only one measurement per patient and it cannot give information about the possible change in complement regulator gene expression in IgAN during the course of the disease. It will be of interest to know if the deficient gene expression in progressive cases is genetic or if it is modulated by other factors determining the progression of IgAN. Notably, there is no correlation with the degree of renal function or the uraemic milieu, and no effect of previous immunosuppressive therapy was observed, hence the hypothesis of a true factor conditioning the progression of IgAN is plausible. Finally, we confirmed in these patients the finding of a previous collaborative study [34]: combining the pathology features with clinical data at renal biopsy, the prognostication of progressive disease is improved in comparison with clinical data alone. It may be of interest to note that the addition of cross-sectional data of CD46 and CD55 mRNA expression to this model significantly improved the prognostication of outcome in our cohort. This new insight into a possible role of defective complement activation regulation in patients with progressive IgAN suggests the need for large prospective cohort studies. ACKNOWLEDGEMENTS The study was supported by the Immunopathology Working Group of the ERA-EDTA. The VALIGA previous study was granted by the first research call of the European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) in 2009. Preliminary and partial data were presented as oral communication at the ERA-EDTA 2016 Congress. CONFLICT OF INTEREST STATEMENT None declared. The authors declare that the results presented in this paper have not been published previously in whole or in part. The study had approval from local ethics committee and was performed in accordance with the Declaration of Helsinki. REFERENCES 1 Maillard N , Wyatt RJ , Julian BA et al. Current understanding of the role of complement in IgA nephropathy . J Am Soc Nephrol 2015 ; 26 : 1503 – 1512 Google Scholar Crossref Search ADS PubMed 2 Daha MR , van Kooten C. Role of complement in IgA nephropathy . J Nephrol 2016 ; 29 : 1 – 4 Google Scholar Crossref Search ADS PubMed 3 Coppo R. Biomarkers and targeted new therapies for IgA nephropathy . Pediatr Nephrol 2017 ; 32 : 725 – 731 Google Scholar Crossref Search ADS PubMed 4 Espinosa M , Ortega R , Sánchez M et al. Association of C4d deposition with clinical outcomes in IgA nephropathy . Clin J Am Soc Nephrol 2014 ; 9 : 897 – 904 Google Scholar Crossref Search ADS PubMed 5 Suzuki H , Kiryluk K , Novak J et al. The pathophysiology of IgA nephropathy . J Am Soc Nephrol 2011 ; 22 : 1795 – 1803 Google Scholar Crossref Search ADS PubMed 6 Coppo R. C4d deposits in IgA nephropathy: where does complement activation come from? Pediatr Nephrol 2017 ; 32 : 1097 – 1101 Google Scholar Crossref Search ADS PubMed 7 Schmitt R , Ståhl AL , Olin AI et al. The combined role of galactose-deficient IgA1 and streptococcal IgA-binding M protein in inducing IL-6 and C3 secretion from human mesangial cells: implications for IgA nephropathy . J Immunol 2014 ; 193 : 317 – 326 Google Scholar Crossref Search ADS PubMed 8 Hiemstra PS , Gorter A , Stuurman ME et al. Activation of the alternative pathway of complement by human serum IgA . Eur J Immunol 1987 ; 17 : 321 – 326 Google Scholar Crossref Search ADS PubMed 9 Roos A , Bouwman LH , van Gijlswijk-Janssen DJ et al. Human IgA activates the complement system via the mannan-binding lectin pathway . J Immunol 2001 ; 167 : 2861 – 2868 Google Scholar Crossref Search ADS PubMed 10 Nikolova EB , Tomana M , Russell MW. The role of the carbohydrate chains in complement (C3) fixation by solid-phase-bound human IgA . Immunology 1994 ; 82 : 321 – 327 Google Scholar PubMed 11 Schmidt CQ , Lambris JD , Ricklin D. Protection of host cells by complement regulators . Immunol Rev 2016 ; 274 : 152 – 171 Google Scholar Crossref Search ADS PubMed 12 Hovingh ES , van den Broek B , Jongerius I. Hijacking complement regulatory proteins for bacterial immune evasion . Front Microbiol 2016 ; 7 : 2004 Google Scholar Crossref Search ADS PubMed 13 Endo M , Ohi H , Satomura A et al. Regulation of in situ complement activation via the lectin pathway in patients with IgA nephropathy . Clin Nephrol 2001 ; 55 : 185 – 191 Google Scholar PubMed 14 Ohsawa I , Ishii M , Ohi H et al. Pathological scenario with the mannose-binding lectin in patients with IgA nephropathy . J Biomed Biotechnol 2012 ; 2012 : 476739 15 Kiryluk K , Li Y , Scolari F et al. Discovery of new risk loci for IgA nephropathy implicates genes involved in immunity against intestinal pathogens . Nat Genet 2014 ; 46 : 1187 – 1196 Google Scholar Crossref Search ADS PubMed 16 Medjeral-Thomas NR , Lomax-Browne HJ , Beckwith H et al. Circulating complement factor H-related proteins 1 and 5 correlate with disease activity in IgA nephropathy . Kidney Int 2017 ; 92 : 942 – 952 Google Scholar Crossref Search ADS PubMed 17 Tortajada A , Gutiérrez E , Goicoechea de Jorge E et al. Elevated factor H-related 1 and factor H pathogenic variants decrease complement regulation in IgA nephropathy . Kidney Int 2017 ; 92 : 953 – 963 Google Scholar Crossref Search ADS PubMed 18 Cosio FG , Shibata T , Rovin BH et al. Effects of complement activation products on the synthesis of decay accelerating factor and membrane cofactor protein by human mesangial cells . Kidney Int 1994 ; 46 : 986 – 992 Google Scholar Crossref Search ADS PubMed 19 Coppo R , Troyanov S , Bellur S et al. Validation of the Oxford classification of IgA nephropathy in cohorts with different presentations and treatments . Kidney Int 2014 ; 86 : 828 – 836 Google Scholar Crossref Search ADS PubMed 20 Trimarchi H , Barratt J , Cattran DC et al. Oxford classification of IgA nephropathy 2016: an update from the IgA Nephropathy Classification Working Group . Kidney Int 2017 ; 91 : 1014 – 1021 Google Scholar Crossref Search ADS PubMed 21 Glassock RJ. The pathogenesis of IgA nephropathy . Curr Opin Nephrol Hypertens 2011 ; 20 : 153 – 160 Google Scholar Crossref Search ADS PubMed 22 Kim SJ , Koo HM , Lim BJ et al. Decreased circulating C3 levels and mesangial C3 deposition predict renal outcome in patients with IgA nephropathy . PLoS One 2012 ; 7 : e40495 Google Scholar Crossref Search ADS PubMed 23 Suzuki H , Ohsawa I , Kodama F et al. Fluctuation of serum C3 levels reflects disease activity and metabolic background in patients with IgA nephropathy . J Nephrol 2013 ; 26 : 708 – 715 Google Scholar Crossref Search ADS PubMed 24 Mizerska-Wasiak M , Małdyk J , Rybi-Szumińska A et al. Relationship between serum IgA/C3 ratio and severity of histological lesions using the Oxford classification in children with IgA nephropathy . Pediatr Nephrol 2015 ; 30 : 1113 – 1120 Google Scholar Crossref Search ADS PubMed 25 Zwirner J , Burg M , Schulze M et al. Activated complement C3: a potentially novel predictor of progressive IgA nephropathy . Kidney Int 1997 ; 51 : 1257 – 1264 Google Scholar Crossref Search ADS PubMed 26 Hastings MC , Moldoveanu Z , Suzuki H et al. Biomarkers in IgA nephropathy: relationship to pathogenetic hits . Expert Opin Med Diagn 2013 ; 7 : 615 – 627 Google Scholar Crossref Search ADS PubMed 27 Sogabe A , Uto H , Kanmura S et al. Correlation of serum levels of complement C4a desArg with pathologically estimated severity of glomerular lesions and mesangial hypercellularity scores in patients with IgA nephropathy . Int J Mol Med 2013 ; 32 : 307 – 314 Google Scholar Crossref Search ADS PubMed 28 Coppo R , Basolo B , Roccatello D et al. Immunological monitoring of plasma exchange in primary IgA nephropathy . Artif Organs 1985 ; 9 : 351 – 360 Google Scholar Crossref Search ADS PubMed 29 Lubbers R , van Essen MF , van Kooten C et al. Production of complement components by cells of the immune system . Clin Exp Immunol 2017 ; 188 : 183 – 194 Google Scholar Crossref Search ADS PubMed 30 Cardone J , Le Friec G , Kemper C. CD46 in innate and adaptive immunity: an update . Clin Exp Immunol 2011 ; 164 : 301 – 311 Google Scholar Crossref Search ADS PubMed 31 Lin FJ , Jiang GR , Shan JP et al. Imbalance of regulatory T cells to Th17 cells in IgA nephropathy . Scand J Clin Lab Invest 2012 ; 72 : 221 – 229 Google Scholar Crossref Search ADS PubMed 32 Charron L , Doctrinal A , Ni Choileain S et al. Monocyte T-cell interaction regulates human T-cell activation through a CD28/CD46 crosstalk . Immunol Cell Biol 2015 ; 93 : 796 – 803 Google Scholar Crossref Search ADS PubMed 33 Donadio ME , Loiacono E , Peruzzi L et al. Toll-like receptors, immunoproteasome and regulatory T cells in children with Henoch–Schönlein purpura and primary IgA nephropathy . Pediatr Nephrol 2014 ; 29 : 1545 – 1551 Google Scholar Crossref Search ADS PubMed 34 Barbour SJ , Espino-Hernandez G , Reich HN et al. The MEST score provides earlier risk prediction in IgA nephropathy . Kidney Int 2016 ; 89 : 167 – 175 Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Journal
Nephrology Dialysis Transplantation – Oxford University Press
Published: Apr 1, 2019