Abstract
The detection rate of antineutrophil cytoplasmic antibodies (ANCA) in Chinese patients with clinically suspected small vessel vasculitis was investigated, and their antigen specificity and demographic features were analyzed. A number of sera (n = 5,604) sent to our referral laboratory for ANCA screening were tested by indirect immunofluorescence (IIF), enzyme-linked immunosorbent assays (ELISAs) for myeloperoxidase (MPO)- and proteinase 3 (PR3)-ANCA. Then the IIF-ANCA-positive sera that were negative for MPO- and PR3-ANCA were further tested by antigen-specific ELISA by using other five highly purified known ANCA antigens as solid-phase ligands. The known antigens included bactericidal/permeability-increasing protein (BPI), human leukocyte elastase (HLE), lactoferrin, cathepsin G, and azurocidins. Of the 5,604 sera, 267 (4.76%) sera were IIF-ANCA positive and 390 (7%) were antinuclear antibody (ANA) positive in the IIF assay. Of the IIF-positive samples, 213 were anti-MPO positive, 32 were anti-PR3 positive, and five cases were positive for both. Of the 48 sera positive for IIF-ANCA but negative for MPO- and PR3-ANCA, 13 sera (27%) recognized other target antigens, 7 sera recognized BPI, 5 recognized HLE, 1 recognize cathepsin G, and 1 recognized azurocidin. None of the sera recognized lactoferrin, and one serum sample recognized both BPI and HLE. The majority of ANCA-positive patients presented in summer or winter. There was no difference in gender (male/female ratio, 1:1.12) in ANCA-positive patients with a mean age of 53.1 years. The male/female ratio was 1.17:1 for patients over 60 years of age; however, it was 1:4 for patients under 20 years of age. We conclude that ANCA-related diseases are not rare in China, and the major antigens are MPO and PR3. When the IIF technique is used to detect ANCA, ANA should be carefully distinguished.
The systemic vasculitides comprise a spectrum of clinical syndromes united by a common pathology consisting of a primary inflammation and necrosis of blood vessels. Antineutrophil cytoplasmic antibodies (ANCA) are serological markers for certain primary vasculitic diseases, including Wegener's granulomatosis, microscopic polyangiitis, Churg-Strauss syndrome, and idiopathic pauci-immune necrotizing glomerulonephritis. These diseases were called ANCA-associated systemic vasculitis (AASV) (3, 7). ANCA can be identified by using indirect immunofluorescence (IIF) techniques and by overlay serum from patients with suspected vasculitis on alcohol-fixed human polymorphonuclear leukocytes (PMN). This procedure produces two staining patterns: a cytoplasmic pattern (C-ANCA) and a perinuclear pattern (P-ANCA) (2, 9). Although proteinase 3 (PR3) and myeloperoxidase (MPO) are the major C-ANCA and P-ANCA antigens (2, 9), several other neutrophil granule constituents can also be recognized by ANCA-positive sera, including bactericidal/permeability-increasing protein (BPI), cathepsin G (CG), lactoferrin, human leukocyte elastase (HLE), and azurocidin (16, 17, 18).
AASV are common autoimmune disorders in the Caucasian population but were not recognized in China until ANCA was available a decade ago (19). This study investigates the prevalence of ANCA and the antigen specificities, as well as demographic features of Chinese patients with AASV from a diagnostic referral ANCA screening center in the Institute of Nephrology, Peking University.
MATERIALS AND METHODS
Patients and sera.
A number of sera (n = 5,604), sent for screening ANCA from January 1996 to December 2001, were collected from patients with clinical suspected vasculitis in our referral laboratory in the Institute of Nephrology, the First Hospital, Peking University. These sera came from all around China, including Beijing and other provinces. Clinical and demographic data of all ANCA-positive patients were collected and analyzed further.
Detecting strategy.
All the serum samples were screened by IIF and enzyme-linked immunosorbent assays (ELISAs) for MPO-ANCA and PR3-ANCA. The IIF-ANCA-positive samples that were negative for MPO-ANCA and PR3-ANCA were further screened by ELISA for BPI, HLE, CG, lactoferrin, and azurocidin.
IIF assay for detecting ANCA.
Standard IIF assays were performed according to the manufacturer's instructions (EUROIMMUN, Lübeck, Germany). Ethanol-fixed human PMN were used to detect ANCA, and monkey liver sections were used to exclude antinuclear antibodies (ANA). By use of primate liver as an additional antigen substrate, P-ANCA and ANA can be differentiated, because the PMN in the sinusoids lie in immediate proximity to the nuclei of the hepatocytes and can thus be identified optically together with them. If the PMN in the sinusoids were positive, whereas the nuclei of the hepatocytes were negative, the antibody was ANCA; whereas if the PMN in the sinusoids and the nuclei of the hepatocytes were all positive, the antibody perhaps was ANA. Cooccurrence of ANCA and ANA is not excluded without antigen-specific ELISAs, however.
Antigen-specific ELISAs.
Seven highly purified known ANCA antigens, purified as detailed in previous reports (17, 18, 20), were used as solid-phase ligands in ELISA. PR3, MPO, HLE, CG, lactoferrin, and azurocidin were diluted to 1 to 2 μg/ml with 0.05 M bicarbonate buffer, pH 9.6, and BPI was diluted to 1 μg/ml with 0.01 M phosphate-buffered saline (PBS); the wells of one half of a Costar microtiter plate were coated with it, while the wells in the other half were coated with coating buffer alone so that they could act as antigen-free wells. The volumes of this step and subsequent steps were 100 μl, all incubations were carried out at 37°C for 1 h, and plates were washed three times with PBS containing 0.1% Tween 20 (PBST) between stages. Test serum samples were diluted 1:50 with PBST, and both antigen-coated wells and antigen-free wells were coated in duplicate with them; every plate contained positive, negative, and blank (PBST) controls. The binding was detected with horseradish peroxidase-conjugated goat anti-human immunoglobulin G (Gibco BRL) (1:5,000 in PBST). The horseradish peroxidase substrate o-phenylenediamine was used at 0.4 mg/ml in citrate phosphate buffer, pH 5.0. The reaction was stopped by 2 M H2SO4, and the results were recorded as the net optical density at 490 nm (average value of antigen wells minus average value of antigen-free wells) and were expressed as a percentage of the known positive controls. Samples were considered positive if they exceeded the mean plus 3 standard deviations from 100 normal blood donors.
Diagnostic criteria.
American College of Rheumatology classification criteria (from 1990) were used to diagnose systemic vasculitis.
Statistical analysis.
To assess association between qualitative variables, the chi-square test was applied and a P of <0.05 was considered statistically significant.
RESULTS
Rate of detection of ANCA and ANA.
Of the 5,604 sera, 267 (4.76%) were IIF-ANCA positive, 232 (4.14%) were P-ANCA positive, and 35 (0.62%) were C-ANCA positive. However, 390 of the 5,604 (7%) sera were ANA positive.
ANCA antigen specificity.
Of the 5,604 sera, 213 recognized MPO, 32 recognized PR3, and 5 sera recognized both. Of the 232 P-ANCA-positive sera, 179 recognized MPO, while 5 recognized both MPO and PR3. Of the 35 C-ANCA-positive sera, 32 recognized PR3 and 3 recognized MPO. However, 31 of the 390 ANA-positive sera also recognized MPO and 2 sera recognized both MPO and PR3 (Table 1).
TABLE 1.
Relationship between IIF-ANCA and ELISAs for MPO- and PR3-ANCAa
| ELISA sample type | No. of samples with:
|
Total no. of samples per type | ||
|---|---|---|---|---|
| pANCA | cANCA | ANA | ||
| Anti-MPO positive | 179 | 3 | 31 | 213 |
| Anti-PR3 positive | 0 | 32 | 0 | 32 |
| Double positive | 5 | 0 | 0 | 5 |
| Double negative | 48 | 0 | 0 | 48 |
| Total | 232 | 35 | 31 | 298 |
Double positive, both MPO- and PR3-ANCA positive; double negative, both MPO- and PR3-ANCA negative.
For the 48 P-ANCA-positive sera yet not MPO-ANCA and PR3-ANCA, only seven sera recognized BPI, five sera recognized HLE, one recognized CG, and one recognized azurocidins, while none of the sera recognized lactoferrin and one serum sample recognized both BPI and HLE. Titers of anti-MPO versus anti-PR3 of the five double-positive sera were 94 versus 57%, 85 versus 32%, 98 versus 29%, 78 versus 35%, and 25 versus 25%, respectively. The five patients were diagnosed as having propylthiouracil-induced ASSV.
Etiology of ANCA-positive patients.
According to the request forms and/or questionnaires, 254 (85%) ANCA-positive patients had microscopic polyangiitis or Wegener's granulomatosis, four patients had ulcerative colitis, and two patients had hemolytic anemia, while the others had unknown etiology.
Epidemiologic and demographic features.
ANCA prevalence increased chronologically. The rate of detection of ANCA in 2000 and 2001 was significantly (P < 0.05) higher than that in 1998 and before (Table 2). Most patients presented from June to December with a peak in July (Fig. 1).
TABLE 2.
ANCA-positive samples screened chronologicallya
| Yr | No. of samples | No. of ANCA-positive samples (%) | No. of P-ANCA-positive samples (%) | No. of C-ANCA-positive samples (%) |
|---|---|---|---|---|
| ∼1997d | 1,034 | 34 (3.3) | 28 (2.7) | 6 (0.6) |
| 1998 | 815 | 34 (4.2) | 27 (3.3) | 7 (0.9) |
| 1999 | 873 | 44 (5.0) | 40 (4.6) | 4 (0.4) |
| 2000 | 1,361 | 89 (6.5)b,c | 76 (5.6) | 13 (0.9) |
| 2001 | 1,521 | 93 (6.1)b,c | 88 (5.8) | 5 (0.3) |
Patients with diseases of nonvasculitic origin such as inflammatory bowel disease were excluded.
P < 0.05 versus ANCA-positive percentage before 1998 (chi-square test).
P < 0.05 versus ANCA-positive percentage in 1998 (chi-square test).
∼1997, before and in 1997.
FIG. 1.
Distribution of ANCA-positive samples in different months. (A) P-ANCA-positive sera. Patients with diseases of nonvasculitic origin such as inflammatory bowel disease were excluded. (B) C-ANCA-positive sera.
Out of the 254 patients with ANCA-associated vasculitides, 209 had complete clinical data. They had an average age of 53.1 (7 to 79) years, and the male/female ratio was 1:1.12. However, the male/female ratio was 1.17:1 in patients over 60 years of age and was 1:4 in patients below 20 years of age (Fig. 2). The gender ratio for patients under 20 years of age is significantly different from that for patients over 60. The average duration from onset of the disease to final diagnosis was 76 days.
FIG. 2.
Distribution by gender of patients of different ages having AASV. For the chi-square test, * signifies a P of <0.05, versus the group of patients under 20 years of age.
DISCUSSION
Although AASV are common in Caucasians, the clinical diagnosis was still difficult until the discovery and clinical application of ANCA (8, 10). In the 1980s, the incidence of AASV was 10 per 106 in the United Kingdom, but after ANCA were clinically available in the 1990s, the incidence of AASV increased to 20 per 106 in the United Kingdom (4, 11, 12, 13, 15). The incidence of AASV in the Chinese population is unknown. In the 1980s, not a single patient with primary systemic vasculitis was diagnosed in our institute; however, after IIF-ANCA and ELISA with crude neutrophil acid extracts as solid-phase ligands were established in the early 1990, a retrospective study of 50 hospitalized Chinese patients with crescentic glomerulonephritis or end-stage renal disease revealed that 5 of the 50 patients were ANCA positive (19). After commercial IIF-ANCA kits became available and purified MPO and PR3 were used in routine ANCA-screening assays, the prevalence of AASV increased substantially. The present study suggested that AASV were common in the Chinese population as well; however, in our practice, many Chinese physicians still do not have awareness of AASV and ANCA assays. Furthermore, ELISAs for MPO-ANCA and PR3-ANCA are still not commonly available in many provinces; therefore, the diagnoses of many patients with AASV are delayed or misdiagnosed in China.
In the present study, the majority of the patients with AASV were 50 to 70 years, but the condition could occur at any age from 7 to 79 years, in agreement with previous reports in the literature (12). Chinese patients with AASV have several distinctive features compared with Caucasian patients. The first difference concerned sexual predominance. Many reports revealed that Caucasian patients were male predominant (12, 13), while for the Chinese patients, males and females are equally affected; however, for the Chinese patients under 20 years of age, the females are clearly predominant (4:1), a finding that needs to be further studied. The second concerned antigenic predominance. MPO and PR3 are the two major ANCA antigens for Chinese patients with AASV; however, the prevalence of P-ANCA/MPO-ANCA was much higher than that of C-ANCA/PR3-ANCA, with a ratio of 7:1, a finding that agreed with reports from Hong Kong (5, 6). Limited observations suggest that the ANCA-associated disease process in northern latitudes more often produces the phenotype of Wegener's granulomatosis, while the phenotypic expression in warmer climates is more often that of microscopic polyangiitis and idiopathic pauci-immune necrotizing glomerulonephritis (14). We speculate that the P-ANCA/MPO-ANCA predominance might be racially and geographically related. A third difference concerned time of presentation of patients with ANCA-associated diseases. In the literature, patients with AASV usually presented in winter or early spring (1). However, in our study, the majority of patients with P-ANCA presented from June to December and peaked in July. Due to the limited number, we did not compare the occurrence rates of C-ANCA in different months. Whether this indicates that ANCA might be induced by infections still needs further investigation. The average duration from onset of the disease to final diagnosis was about 2 to 3 months in our study, so the course of the disease was not very assured.
The IIF assay is still the standard test for screening ANCA. Besides ANCA, some serum samples contain other cytoplasmic fluorescence patterns and/or ANA, which results in homogeneous or peripheral nuclear fluorescence. Many reports indicate that positive fluorescence alone is not specific for the diagnosis of Wegener's granulomatosis or microscopic polyangiitis. In China systemic lupus erythematosus is the most common autoimmune disease, and in the present study, the prevalence of ANA was much higher than that of ANCA; therefore, IIF-ANCA, especially P-ANCA, should be clearly distinguished from ANA. To distinguish P-ANCA and ANA, formaldehyde-fixed neutrophils, mammal liver sections and Hep2 cells are good substrates for excluding ANA. On formaldehyde-fixed neutrophils, ANA were not detected, whereas P-ANCA diffusely labeled the cytoplasm. Mammal liver can also be used to distinguish ANCA and ANA. When ANA and ANCA coexist, however, these techniques could not distinguish ANCA from ANA clearly; therefore, just using the IIF assay as serum markers to diagnose AASV is not good enough.
It has been suggested that IIF-ANCA combined with antigen-specific ELISAs, for MPO-ANCA and PR3-ANCA, had good sensitivity and specificity for diagnosing AASV (2, 9), and our results supported this conclusion. Although many ANCA antigens have been identified, only MPO and PR3 are of confirmed clinical significance. In our study, 84% of the IIF-ANCA-positive serum samples recognized MPO and PR3, indicating that MPO and PR3 are indeed the two major ANCA antigens as well for Chinese patients with AASV. Although 16% of our samples did not recognize MPO and PR3, only a few of them (13 of 48) recognized the other known ANCA antigens, such as BPI, HLE, cathepsin G, and azurocidins. The specific target antigens of the other serum samples still need to be further characterized. The sera of the four patients with ulcerative colitis showed P-ANCA but were negative for MPO- and PR3-ANCA; instead, they recognized BPI. Some of the patients diagnosed as having drug-induced ASSV had more than one ANCA specificity. All these results need to be further investigated.
In conclusion, ANCA-related diseases are not rare in China and MPO is the most common ANCA antigen.
REFERENCES
- 1.Falk, R. J., S. Hogan, T. S. Carey, J. C. Jennette, et al. 1990. Clinical course of anti-neutrophil cytoplasmic autoantibody-associated glomerulonephritis and systemic vasculitis. Ann. Intern. Med. 113:656-663. [DOI] [PubMed] [Google Scholar]
- 2.Hagen, E. C., M. R. Daha, J. Hermans, K. Andrassy, E. Csernok, G. Gaskin, P. Lesavre, J. Ludemann, N. Rasmussen, R. A. Sinico, A. Wiik, and F. van der Woude. 1998. Diagnostic value of standardized assays for anti-neutrophil cytoplasmic antibodies in idiopathic systemic vasculitis. Kidney Int. 53:743-753. [DOI] [PubMed] [Google Scholar]
- 3.Kamesh, L., L. Harper, and C. O. Savage. 2002. ANCA-positive vasculitis. J. Am. Soc. Nephrol. 13:1953-1960. [DOI] [PubMed] [Google Scholar]
- 4.Lane, S. E., D. G. Scott, A. Heaton, and R. A. Watts. 2000. Primary renal vasculitis in Norfolk—increasing incidence or increasing recognition? Nephrol. Dial. Transplant. 15:23-27. [DOI] [PubMed] [Google Scholar]
- 5.Lee, S., and J. W. Lawton. 2000. Heterogeneity of anti-PR3 associated disease in Hong Kong. Postgrad. Med. J. 76:287-288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Li, P. K., J. Leung, F. M. Lai, A. Wang, S. F. Lui, C. B. Leung, and K. N. Lai. 1994. Use of antineutrophil cytoplasmic autoantibodies in diagnosing vasculitis in a Chinese patient population. Am. J. Nephrol. 14:99-105. [DOI] [PubMed] [Google Scholar]
- 7.Savage, C. O. 2001. ANCA-associated renal vasculitis. Kidney Int. 60:1614-1627. [DOI] [PubMed] [Google Scholar]
- 8.Savage, C. O., C. G. Winearls, S. Jones, P. D. Marshall, and C. M. Lockwood. 1987. Prospective study of radioimmunoassay for antibodies against neutrophil cytoplasm in diagnosis of systemic vasculitis. Lancet i:1389-1393. [DOI] [PubMed] [Google Scholar]
- 9.Savige, J., D. Gillis, E. Benson, et al. 1999. International Consensus Statement on Testing and Reporting of Antineutrophil Cytoplasmic Antibodies (ANCA). Am. J. Clin. Pathol. 111:507-513. [DOI] [PubMed] [Google Scholar]
- 10.van der Woude, F. J., N. Rasmussen, S. Lobatto, et al. 1985. Autoantibodies against neutrophil and monocytes: tool for diagnosis and marker of disease activity in Wegener's granulomatosis. Lancet i:425-429. [DOI] [PubMed] [Google Scholar]
- 11.Watts, R. A., and D. G. Scott. 2003. Epidemiology of the vasculitides. Curr. Opin. Rheumatol. 15:11-16. [DOI] [PubMed] [Google Scholar]
- 12.Watts, R. A., D. G. I. Scott, and S. E. Lane. 2002. Epidemiology of Wegener's granulomatosis, microscopic polyangiitis, and Churg-Strauss syndrome. Clevel. Clin. J. Med. 69(Suppl. II):84-86. [DOI] [PubMed] [Google Scholar]
- 13.Watts, R. A., S. E. Lane, G. Bentham, and D. G. Scott. 2000. Epidemiology of systemic vasculitis: a ten-year study in the United Kingdom. Arthritis Rheum. 43:414-419. [DOI] [PubMed] [Google Scholar]
- 14.Watts, R. A., W. Koldinges, H. Nossent, et al. 2001. Epidemiology of vasculitis in Europe. Ann. Rheum. Dis. 60:1156-1157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Watts, R. A., and D. G. Scott. 1997. Classification and epidemiology of the vasculitides. Baillieres Clin. Rheumatol. 11:191-217. [DOI] [PubMed] [Google Scholar]
- 16.Wiik, A., L. Stummann, L. Kjeldsen, et al. 1993. The diversity of perinuclear antineutrophil cytoplasmic antibodies (ANCA) antigens. Clin. Exp. Immunol. 101(Suppl. 1):15-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhao, M. H., S. J. Jones, and C. M. Lockwood. 1995. Bactericidal/permeability-increasing protein (BPI) is an important antigen for antineutrophil cytoplasmic antibodies (ANCA) in vasculitis. Clin. Exp. Immunol. 99:49-56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zhao, M. H., and C. M. Lockwood. 1996. Azurocidins is a novel antigen for anti-neutrophil cytoplasmic autoantibodies (ANCA) in systemic vasculitis. Clin. Exp. Immunol. 103:397-402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhao, M. H., H. Y. Wang, and Y. K. Zhang. 1993. The renal involvement of primary vasculitis: clinical and pathological analysis of 5 cases. Zhonghua Nei Ke Za Zhi 32:40-43. (In Chinese.) [PubMed] [Google Scholar]
- 20.Zhao, M. H., and C. M. Lockwood. 1996. A comprehensive method to purify three major ANCA antigens: proteinase 3, myeloperoxidase and bactericidal/permeability-increasing protein from human neutrophil granule acid extract. J. Immunol. Methods 197:121-130. [DOI] [PubMed] [Google Scholar]