Evaluation of a simple dot-blot method for the detection of anti-neutrophil cytoplasmic antibodies directed against proteinase 3 and myeloperoxidase.
Anti-neutrophil cytoplasmic antibodies (ANCAs) usually are demonstrated by indirect immunofluorescence (IF), followed by examination of patient samples for the antibodies against proteinase-3 (PR3) and myeloperoxidase (MPO). The antibody testing usually requires ELISA and may be hampered by the cost and the requirement for an ELISA reader. Dot-blot (DB) analysis can be an attractive alternative for ELISA for the identification of some autoantibodies (1,2).
In this study, we compared a capture ELISA for PR3 and MPO antibodies with a new, commercially available DB method. Sera from 74 patients were selected for the evaluation. In addition, samples with negative IF results (n = 40); samples with cytoplasmic (c)-ANCAs (n = 20), perinuclear (p)-ANCAs (n = 8), or atypical IF (n = 4) results; and samples with equivocal IF results (n = 4) were included.
Routine detection of antibodies against PR3 and MPO was performed by capture ELISA essentially as described by Goldschmeding et al. (3). Over the years, this in-house ELISA has been shown to be valid in both clinical settings and proficiency testing programs such as the United Kingdom National External Quality Assessment Scheme and the European Consensus Group. In brief, microtiter plates were successively coated with goat anti-mouse IgG, mouse antibodies against PR3 or MPO, and finally PR3 or MPO purified from human granulocytes. Subsequently, plates were incubated with diluted patient samples (1:100). Finally, bound antibodies against PR3 or MPO were detected by alkaline phosphatase-conjugated anti-human IgG. The tests were performed in duplicate.
The DB method (BMD) is a qualitative assay that uses nitrocellulose strips on which purified antigens are blotted at preset spots. The antigen source for both PR3 and MPO is human granulocytes. The test procedure was performed according to directions supplied by the manufacturer. Test strips were incubated for 10 min with a 50-fold dilution of patient serum in a phosphate-buffered saline-Tween solution supplied by the manufacturer. The test strips were then washed with phosphate-buffered saline-Tween and incubated with an alkaline phosphatase-protein A conjugate for 10 min. Finally, the test strips were stained with 5-bromo-4-choro-3-indolylphosphate/nitroblue tetrazolium for 5 min, washed with deionized water, and air-dried. Only strips on which the positive control was clearly stained were evaluated for this study. A single test requires ~30 min.
The sensitivity and specificity of the DB method for PR3 and MPO detection, compared with ELISA, are shown in Table 1. The agreement between the ELISA and the DB method was 99% for PR3 and 86% for MPO. The IF patterns of the samples that showed a discrepancy are included in the legend for Table 1. The clinical diagnoses of the corresponding patients (if present) are also included.
Studies on the analytical performance of assays for ANCA characterization are scarce. Because no gold standard for ANCA characterization exists, one must be careful with the interpretation of the presented data. In a recent study by Pollock et al. (4), the sensitivity and specificity of several commercial ELISAs for PR3 and MPO varied from 92% to 100% and 59% to 100%, respectively. In particular, for MPO those results are substantially better than in the present study. However, they possibly were inflated because only sera with clear c-ANCA or p-ANCA IF patterns were used. In our study, the observed discrepancies were predominantly caused by "difficult" sera with atypical and equivocal IF patterns (see Table 1). However, some of the samples that were ELISA-positive and DB-negative for MPO originated from patients with confirmed systemic vasculitis or a related autoimmune disease. Thus, we cannot conclude from this study that the clinical performance is equal to that of the ELISA. The analytical performance of this specific test, therefore, seems inadequate.
On the other hand, the presented data also show that the analytical performance of the DB method for detecting PR3 detection is excellent. Moreover, the test procedure is easy to master and can be performed in a batch assay. Therefore, it can be concluded that this specific test may be a simple, fast, and financially attractive method for the identification of PR3 antibodies.
(1.) Ermens AAM, Bayens AJM, van Gemert ACM, van Duiinhoven JLP. Simple dot-blot method evaluated for detection of antibodies against extractable nuclear antigens. Clin Chem 1997; 43:2420-2.
(2.) Mewis A, Marien G, Blanckaert N, Bossuyt X. Evaluation of a dot-blot method for identification of antibodies against extractable nuclear antigens and anti-cytoplasmic antibodies. Clin Chem 1999;45:311-2.
(3.) Goldschmeding R, van der Schoot CE, ten Bokkel Huinink D, Hack CE, van den Ende ME, Kallenberg CGM, von dem Borne AE. Wegener's granulomatosis autoantibodies identify a novel diisopropylfluorophosphate-binding protein in the lysosymes of normal human neutrophils. J Clin Invest 1989;84:1577-87.
(4.) Pollock W, Dunster K, Rolland JM, Koh H, Savige J. A comparison of commercial and in-house ELISAs for antineutrophil cytoplasmic antibodies directed against proteinase 3 and myeloperoxidase. Pathology 1999;31:38-43.
Anthony A.M. Ermens  *
Angelique J.M. Bayens 
Adrienne Crooymans 
Anita A.M. Broekman-van Hout 
Hans L.P. van Duijnhoven 
 Klinisch Laboratorium
5600 PD Eindhoven, The Netherlands
 Algemeen Klinisch Laboratorium
5800 AB Helmond, The Netherlands
* Author for correspondence. Fax 31-40-2335595; e-mail email@example.com.
Table 1. Test results of patient samples that were analyzed for the presence of PR3 and MPO antibodies, and sensitivity and specificity of DB method compared with ELISA. ELISA-/DB-, ELISA+/DB+, ELISA+/DB-, n n n n PR3 74 52 21 0 MPO 74 55 9 8 (c) ELISA-/DB+, n Sensitivity (a) (Specificity (a) PR3 1 (b) 1 0.98 (0.95-1.01) MPO 2 (d) 0.53 (0.41-0.64) 0.97 (0.92-1.00) (a) 95% confidence intervals in parentheses. (b) One p-ANCA (no signs of autoimmune disease). (c) Three atypical (two patients with systemic vasculitis, one patient without signs of autoimmune disease), four equivocal (three patients with paraproteinemia, one patient with rheumatoid arthritis), one negative (no signs of autoimmune disease). (d) One atypical (no signs of autoimmune disease), one c-ANCA (no signs of autoimmune disease).
|Printer friendly Cite/link Email Feedback|
|Author:||Ermens, Anthony A.M.; Bayens, Angelique J.M.; Crooymans, Adrienne; Broekman-van Hout, Anita A.M.; va|
|Article Type:||Letter to the editor|
|Date:||Oct 1, 2000|
|Previous Article:||State of the art instead of biological variation to set requirements for imprecision.|
|Next Article:||Dilution protocols for detection of hook effects/prozone phenomenon.|