JIM-11762; No of Pages 6 Journal of Immunological Methods xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Journal of Immunological Methods journal homepage: www.elsevier.com/locate/jim
Review
Development and validation of a lateral flow assay (LFA) for the determination of IgG-antibodies to Pr3 (cANCA) and MPO (pANCA) Offermann N. a, Conrad K. b, Fritzler M.J. c, Fooke Achterrath M. a,⁎ a b c
Dr. Fooke Laboratorien GmbH, Germany Technical University Dresden, Institute of Immunology, Dresden, Germany Faculty of Medicine, University of Calgary Calgary, Alberta, Canada
a r t i c l e
i n f o
Article history: Received 9 October 2013 Received in revised form 18 November 2013 Accepted 19 November 2013 Available online xxxx Keywords: Lateral flow assay Myeloperoxidase, MPO Proteinase 3, Pr3 Autoantibodies pANCA cANCA
a b s t r a c t The timely diagnosis of vasculopathies, such as granulomatosis with polyangiitis, has important implications for the favorable clinical outcome of these diseases. In the clinical setting, autoantibodies to proteinase 3 (Pr3) and myeloperoxidase (MPO) have been shown to be valuable adjuncts to an early and accurate diagnosis. The sensitive and specific detection of anti-Pr3 and anti-MPO was shown using a point of care device that employed rapid Lateral Flow Technologies. The validation of the lateral flow assay (LFA) was performed with serum samples collected in two Reference Laboratories and showed excellent results that were comparable to widely accepted and used ELISA. The advantage of the LFA is the flexibility to be used as an economical, point of care diagnostic device, features that are especially important for an early and accurate diagnosis and the prompt initiation of appropriate treatment so as to avoid inevitable development of undue complications of these diseases such as disseminated organ involvement, e.g. renal failure. © 2013 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Material and methods . . . . . . . . . . . . . . . . . . . . . . . 2.1. Serum samples . . . . . . . . . . . . . . . . . . . . . . . 2.2. Assay system . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Technical performance analysis . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Technical performance characteristics . . . . . . . . . . . . 3.1.1. Adjustment of assay cut-off to optimize sensitivity and 3.1.2. Inter-assay variation (lot to lot variation) . . . . . . 3.1.3. Intra-assay variation . . . . . . . . . . . . . . . . 3.2. Diagnostic performance analysis . . . . . . . . . . . . . . . 3.2.1. Cross-reactivity between AI-LFA Pr3 and AI-LFA MPO 3.2.2. Comparison to IFA (Euroimmun) . . . . . . . . . . 3.2.3. Comparison to “state of the art” ELISA (Orgentec) . . 3.2.4. Comparison to CLIA . . . . . . . . . . . . . . . . 3.2.5. Diagnostic performance summary . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
⁎ Corresponding author at: Dr. Fooke Laboratorien GmbH, Habichtweg 16, 41468 Neuss, Germany. Tel.: +49 2131 2984 135; fax: +49 2131 2984 184. 0022-1759/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jim.2013.11.017
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
1. Introduction Vasculitis is caused by inflammation of blood vessels and it manifests itself through a broad spectrum of clinical phenotypes. Anti-neutrophil cytoplasmic antibodies (ANCAs) are one of the most common serological features of small vessel vasculitis (SVV). ANCA associated vasculitis includes microscopic polyangiitis (MPA), eosinophilic granulomatosis with polyangiitis (EGPA) and granulomatosis with polyangiitis (GPA), formerly known as Wegener's granulomatosis. (Wegener, 1987; Falk et al., 2011). The sensitive and specific detection of proteinase 3 (Pr3, cANCA) and myeloperoxidase (MPO, pANCA) autoantibodies, occurring in about 80% of GPA cases, are time-proven serological biomarkers that can aid in the diagnosis of these conditions. Only timely and appropriate treatment can avoid the development of renal failure and/or other systemic complications of these diseases (Conrad et al., 2007a,b). Our aim was to develop a sensitive and specific, semiquantitative LFA together with a newly designed portable lateral flow assay reader as an approach to a point of care technology that would be an important adjunct to diagnosing and monitoring these diseases. Minimal laboratory equipment and an assay time of only 20 min result in an economic immunoassay. We compared the technical and diagnostic performance to conventional diagnostic assays such as indirect immunofluorescence (IIF) measuring cANCA and pANCA as well as the latest generation ELISA and a new chemiluminescence immunoassay (CLIA, BioFlash, INOVA Diagnostics, San Diego, CA, USA) (Lucassen et al., 2011; Schulte-Pelkum et al., 2012).
2. Material and methods 2.1. Serum samples The assay was developed by utilizing 87 commercially available serum samples representing 40 healthy donors, 5 unrelated disease controls, 20 cANCA positive and 22 pANCA positive individuals. ELISA and IIF ANCA results were available for these samples as supplied by several commercial institutions. International reference sera for anti-Pr3 (Pr3-ANCA human reference serum #16, Product Package Insert Catalog #IS2721 Lot #07-0002) and anti-MPO antibodies (MPO–ANCA human reference serum #15, Product Package Insert Catalog #IS2720 Lot #07-0001) were obtained from the Center for Disease Control and Prevention (CDC, Atlanta, GA, USA). The quantitative content of anti-Pr3 and anti-MPO antibodies in the reference sera had been determined as 100 IU/ml by different laboratories. For validation of the LFA, a different set of serum samples was collected in two different reference laboratories: Technical University Dresden, Institute of Immunology (n = 40, cANCA positive n = 20 and pANCA positive n = 20) and the University of Calgary, Calgary, Alberta, Canada (n = 64, healthy donors
n = 34 and ANCA positive n = 30). This study was approved by respective institutional human ethics review boards or was part of quality assurance and quality control processes. All patient identities remained anonymous in keeping with the latest version of the Helsinki Declaration of Human Research Ethics. The serum samples were tested by commercially available Pr3 and MPO ELISA tests and a new CLIA technology. IFA results were available for 40 sera collected in Dresden (pANCA 37/40; cANCA 39/40). Results for IFA are given in titers. The serum titer is defined as the highest serial dilution of a serum sample (expressed as the dilution factor) that still yields a positive result. Titers above 1/40 were considered as positive. All serum samples were tested for anti-Pr3 and anti-MPO antibodies in the new AI-LFA assays, measured with the newly developed lateral flow assay reader and compared to the results obtained with the other methods. 2.2. Assay system The autoimmune lateral flow assay system (AI-LFA) is based on a universal test cassette which can be used for all available IgG-antibody AI-parameters (Fig. 1). Purified native Pr3 and MPO antigens, both isolated from human neutrophils were biotinylated and used as liquid sources. The assay was performed as follows: 10 μl of undiluted serum was applied onto the sample application point (S) of the Basis Set module (Fig. 1b). Immediately after, the antigen solution was applied. The serum antibody to antigen binding occurs during the 20 min incubation while the mixture of antigen solution, conjugate and serum is driven through the device by capillary flow. The antigen specific IgG of the sample binds specifically to its corresponding antigen in the solution and these soluble antibody-(sIgG)-bound antigens are retained at the test line (T) by a capture molecule (Fig. 1). At the same time, the sIgG bound to the antigen is bound by an antibody coupled to gold particles (conjugate). The intensity of the color reaction at the test line is directly proportional to the amount of immune complexes consisting of ligand tagged antigen, sIgG, and anti-IgG specific conjugate. The signal intensity ranges from faintly pink color (low titer of sIgG) to dark ruby red (high titer of sIgG). Excess conjugate, which is not bound at the test line, forms a dark ruby control line (C) after 20 min of incubation. Interpretation of the test results is done by densitometry using a portable, proprietary lateral flow assay reader (Fig. 1c). Semi-quantitative results were expressed as relative units (RU), whereas the RU value corresponds to the color intensity of the test line (measured as grayscale level by a CCD camera) and, hence, to the concentration of autoantibody in the serum sample. 2.3. Technical performance analysis For Pr3 and MPO AI-LFA, the inter-assay variation was analyzed with 3 different lots of antigen and 3 different lots
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
Fig. 1. (a) Principle of Pr3/MPO AI-LFA, (b) AI-LFA cassette and (c) lateral flow assay reader.
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
3
4
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
Table 1 Lot-to-lot variation of 3 independent lots of AI-LFA basis sets measured with 2 positive and 2 negative serum samples per antigen. (HD = Healthy Donors).
3.2. Diagnostic performance analysis
Lot to lot variation (inter-assay) basis sets Antigen
Serum sample
Mean [RU]
Std dev [RU]
CV [%]
Pr3
HD HD Pos Pos HD HD Pos Pos
89.69 107.53 1715.69 1753.24 171.12 140.00 2043.07 1249.06
4.15 14.25 127.90 118.19 25.04 7.88 69.48 80.11
4.63 13.25 7.45 6.74 14.63 5.63 3.40 6.41
MPO
for MPO-ANCA. The results of the negative sera were below the established RU cut-off (see Table 3).
of LFA basis sets (LFA membranes and conjugate filters). The intra-assay variation was evaluated by fivefold replicate analysis of four serum samples (2 positive, 2 negative per antigen). 3. Results 3.1. Technical performance characteristics 3.1.1. Adjustment of assay cut-off to optimize sensitivity and specificity A ROC analysis was performed to determine the optimum cut-off for anti-Pr3 and anti-MPO detection. The best diagnostic performance (sensitivity and specificity) was obtained at a cut-off of 200 RU for anti-Pr3 and 450 RU for anti-MPO antibodies (see Supplementary Fig. 1). All of the anti-Pr3 and anti-MPO commercially procured positive serum samples were also positive in the AI-LFA: The ROC analysis demonstrated a sensitivity of 100% for Pr3-ANCA (20/20 samples positive) and MPO-ANCA (22/22 samples positive). The specificity was calculated to be 98% (95% CI 0.88–1.0) for both Pr3-ANCA (43/44 samples negative) and MPO-ANCA (44/45 samples negative). ROC analysis revealed area under the curve (AUC) values of 1.0 for Pr3 AI-LFA and for MPO AI-LFA, respectively (see Supplementary Fig. 1). In addition, the CDC international reference serum 16 (positive for Pr3) and reference serum 15 (positive for MPO) were also found to be positive by AI-LFA with 2249.5 and 758.7 RU, respectively. 3.1.2. Inter-assay variation (lot to lot variation) Lot to lot variation was calculated from the results of 3 different lots of LFA basis sets using the same lots of PR3 and MPO (see Table 1) and 3 different lots of LFA Pr3 and MPO antigen solutions using the same lot of LFA basis set (see Table 2). The coefficients of variation (CV) of the positive sera were between 1.25% and 7.45% for Pr3-ANCA and between 1.24% and 6.41% for MPO-ANCA. The results of the negative sera were below the required cut-off.
3.2.1. Cross-reactivity between AI-LFA Pr3 and AI-LFA MPO Of 38 serum samples which were positive in the Pr3 AI-LFA, no samples were false positive in the MPO LFA. This corresponds to a specificity of 100% for the MPO AI-LFA for serum samples with anti-Pr3 IgG antibodies (see Supplementary Table 1). Of 25 serum samples which were positive in the MPO AI-LFA, three samples were false positive in the Pr3 LFA. This corresponds in this group to a specificity of 88% for the Pr3 AI-LFA for serum samples with anti-MPO IgG antibodies (see Supplementary Table 2). 3.2.2. Comparison to IFA (Euroimmun) 3.2.2.1. AI-LFA Pr3. Of 39 serum samples, 20 were positive for cANCA (IIF, Euroimmun) and also positive for AI-LFA Pr3; this corresponds to a sensitivity of 100% (95% CI 0.83–1.00). Of the 19 disease control samples, 18 were negative in the AI-LFA Pr3. This corresponds to a specificity of 94.7% (95% CI 0.74–1.00; see Supplementary Fig. 2a). 3.2.2.2. AI-LFA MPO. Of 37 serum samples, 20 were positive for pANCA (IIF, Euroimmun) and also positive for AI-LFA MPO; this corresponds to a sensitivity of 100% (95% CI 0.83–1.00). Of the 17 disease control samples, all 17 were negative in the AI-LFA MPO; this corresponds to a specificity of 100% (95% CI 0.81–1.00; see Supplementary Fig. 2b). 3.2.3. Comparison to “state of the art” ELISA (Orgentec) 3.2.3.1. AI-LFA Pr3. 104 serum samples were tested with the AI-LFA Pr3 and compared to a highly sensitive anti-Pr3 ELISA (Orgentec). Of 39 serum samples which tested positive for Pr3 in the ELISA, 38 were identified as positive in the AI-LFA Pr3 system. In addition, 62 of 65 control serum samples were identified as negative in the AI-LFA Pr3. This corresponds to a sensitivity of 97.4% (95% CI 0.87–1.00) and a specificity of 93.8% (95% CI 0.85–0.98; see Supplementary Fig. 3a). 3.2.3.2. AI-LFA MPO. The 104 serum samples were also tested with the AI-LFA MPO and compared to anti-MPO ELISA (Orgentec). Of the 25 samples that were positive for MPO in the ELISA, all were also positive in the AI-LFA MPO system. 79 serum samples were found to be negative in both systems. This corresponds to a sensitivity of 100% (95% CI 0.86–1.00) and a specificity of 100% (95% CI 0.95–1.00; see Supplementary Fig. 3b). 3.2.4. Comparison to CLIA
3.1.3. Intra-assay variation Intra-assay precision of the two assays was performed by calculating the CVs for four samples per antigen each with five determinations using the same lot of LFA basis sets and antigen solutions. The CVs of the positive sera were between 5.94% and 8.58% for Pr3-ANCA and between 5.16% and 9.28%
3.2.4.1. AI-LFA Pr3. 83 serum samples were tested with the AI-LFA Pr3 and compared to the anti-Pr3 CLIA. Of the 38 serum samples which tested positive for Pr3, 37 tested positive in the AI-LFA Pr3 system. 43/45 serum samples were identified as negative in the AI-LFA Pr3. This corresponds to a sensitivity of
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
5
Table 2 Lot-to-lot variation of 3 independent antigen lots of A) Pr3 and B) MPO measured with 6 sera per antigen. A)
B)
Pr3 antigen lot to lot variation (3 lots)
MPO antigen lot to lot variation (3 lots)
Serum sample
Mean [RU]
Std dev [RU]
CV [%]
Serum sample
Mean [RU]
Std dev [RU]
CV [%]
HD
136.73 132.21 132.79 2355.40 1319.71 5031.44
9.32 16.66 14.79 47.32 72.81 62.88
6.82 12.60 11.14 2.01 5.52 1.25
HD
250.54 328.99 232.89 2358.79 1267.14 677.17
15.22 19.56 13.35 29.17 47.96 14.07
6.07 5.95 5.73 1.24 3.78 2.08
Pr3 positive
97.4% (95% CI 0.86–0.99) and a specificity of 97.4% (95% CI 0.82–0.99; see Supplementary Fig. 4a). 3.2.4.2. AI-LFA MPO. 83 serum samples were tested with the AI-LFA MPO and the anti-MPO CLIA. Of 26 serum samples which tested positive for MPO by CLIA, 24 were found to be positive in the AI-LFA MPO system. 56/57 serum samples were found to be negative in the AI-LFA MPO. This corresponds to a sensitivity of 92.3% (95% CI 0.75–0.99) and a specificity of 98.2% (95% CI 0.91–1.00; see Supplementary Fig. 4b). 3.2.5. Diagnostic performance summary ROC analysis of AI-LFA compared with the established assay methods revealed AUC values between 0.96 and 1.00 (see Table 4). An AUC of 1.0 indicates that the performance of the two compared tests is identical, which is the case for AI-LFA MPO compared with IFA pANCA as well as MPO ELISA (see Table 4). 4. Discussion The current diagnosis of ANCA-associated small vessel vasculitis typically relies on a combination of IIF cANCA, pANCA, atypical ANCA (aANCA) and ELISA, each detecting IgG-autoantibodies to Pr3 and MPO (Savige et al., 1999; Holle et al., 2005). A positive result for IIF ANCA screening is normally confirmed by ELISA, both with an assay time between 1.5 and 3 h. Most of the anti-MPO ELISAs show a diagnostic performance comparable with IIF-ANCA (Holle et al., 2005). Anti-Pr3 ELISAs often show a reduced sensitivity, whereas the specificity is comparable to IIF-ANCA (Roggenbuck et al., 2009). It has been suggested that the higher sensitivity of third generation
Table 3 Reproducibility of 5-fold AI-LFA measurements for 4 sera (2 healthy donors, 2 positive serum samples) per antigen. Antigen
Serum sample
Replicates
Mean (RU)
Std dev (RU)
CV (%)
Pr3
HD HD Pos Pos HD HD Pos Pos
5 5 5 5 5 5 5 5
78.62 87.37 1553.22 1447.24 154.81 148.78 2236.96 1257.34
8.82 12.86 133.19 85.98 12.64 14.34 115.37 116.70
11.22 14.72 8.58 5.94 8.16 9.64 5.16 9.28
MPO
MPO positive
ELISAs compared with previous ELISAs was achieved by the presentation of autoantigenic epitopes via novel protein anchoring techniques (Roggenbuck et al., 2009). The limitations of ELISA may be related to observations that human ANCA bind mostly to conformational epitopes of Pr3, and that the adsorption of the antigen to solid phases could dramatically reduce the binding of ANCA (Specks, 2000; Silva et al., 2010; Kuhl et al., 2010). Accordingly, the focus of assay development has been to represent the antigen in as close to its native conformation as technically possible in order to enable antibody binding to all clinically relevant epitopes. Thus, the most important innovations were to overcome these problems by using anchor techniques, which increase the sensitivity by presenting a higher number of epitopes for binding of specific autoantibodies (Roggenbuck et al., 2009; Mahler et al., 2012). Our new AI-LFA lateral flow assay for the detection of anti-Pr3 and anti-MPO antibodies, complemented by the newly developed lateral flow assay reader, show results of high sensitivity and specificity when compared to IIF-ANCA and newer generation ELISA. An AUC of 1.0 indicates that the sensitivity and specificity of two compared tests are identical. This is the case for AI-LFA MPO versus IFA and ELISA (see Table 4). The sensitivity and specificity for AI-LFA Pr3 versus IFA and ELISA is also excellent with an AUC of 0.99. The high sensitivity of the AI-LFAs may be explained by the antigen epitopes being available for antibody binding while in the liquid phase of the assay. The diagnostic performance of AI-LFA to CLIA is also comparable, but minimally more discrepant compared with IIF-ANCA and third generation ELISA (see Table 4). The sensitivity and specificity for AI-LFA compared to CLIA are 97.4% and 93.3% for Pr3 (AUC 0.99) and 92.3% and 98.2% for MPO (AUC 0.96). This might be explained by the characteristics of CLIA that might have a linear dynamic range of
Table 4 Sensitivities and specificities of AI-LFA when compared to different assay systems. AI-LFA
Compared with
Sensitivity (%)
Specificity (%)
n=
Pr3 (cut-off 200 RU)
IFA cANCA hs Pr3 ELISA BIO-FLASH Pr3 IFA pANCA MPO ELISA BIO-FLASH MPO
100 97.4 97.4 100 100 92.3
94.7 93.8 93.3 100 100 98.2
39 104 83 37 104 83
MPO (cut-off 450 RU)
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
6
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
antigen–antibody binding throughout virtually all concentrations of antibody–antigen reactions (Mahler et al., 2012). The AI-LFA MPO shows a better diagnostic performance compared with the IFA and ELISA than the AI-LFA Pr3 (see Table 4). In contrast, the AI-LFA Pr3 does agree slightly better with the BIO-FLASH system. Altogether the AI-LFA Pr3 as well as the AI-LFA MPO achieve sensitivities and specificities greater than 92% compared with all three different test systems (IFA, ELISA and CLIA). Interpretation of LFA with the new digital densitometry reader provides objective semi-quantitative results, in contrast to IIF-ANCA for which an experienced technologist is required. The sensitivities of the new AI-LFA tests are competitive with the sensitivity of a third generation ELISA. In addition, the AI-LFA assays feature a time-to-results of 20 min, economical single testing and require minimal laboratory equipment. The novel AI-LFA tests together with a precise reading-out technology show a highly sensitive and specific detection of autoantibodies to Pr3 and MPO, comparable with ELISA tests while keeping all the advantages of a rapid test. 5. Conclusions The new AI-LFA lateral flow assays for the detection of anti-Pr3 and anti-MPO antibodies, read with the aid of the newly developed lateral flow assay reader, show test results of high sensitivity and specificity when compared with IFA, the latest generation ELISA- and CLIA- tests. The combination of short assay time, economical single sample testing and minimal laboratory equipment requirement makes the AI-LFA tests highly convenient for the testing of anti-Pr3 and anti-MPO autoantibodies. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jim.2013.11.017.
References Conrad, K., Schößler, W., Hiepe, F., Fritzler, M., 2007a. Myeloperoxidase antibodies. In: Conrad, K., Schößler, W., Hiepe, F., Fritzler, M. (Eds.), Autoantibodies in Systemic Autoimmune Diseases- A Diagnostic Reference, pp. 111–113 (Pabst). Conrad, K., Schößler, W., Hiepe, F., Fritzler, M., 2007b. Proteinase 3 antibodies. In: Conrad, K., Schößler, W., Hiepe, F., Fritzler, M. (Eds.), Autoantibodies in Systemic Autoimmune Diseases- A Diagnostic Reference, pp. 147–149 (Pabst). Falk, R.J., Gross, W.L., Guillevin, L., Hoffman, G.S., Jayne, D.R., Jennette, J.C., Kallenberg, C.G., Luqmani, R., Mahr, A.D., Matteson, E.L., Merkel, P.A., Specks, U., Watts, R.A., 2011. Granulomatosis with polyangiitis (Wegener's): an alternative name for Wegener's granulomatosis. Arthritis Rheum. 63, 863. Holle, J.U., Hellmich, B., Backes, M., Gross, W.L., Csernok, E., 2005. Variations in performance characteristics of commercial enzyme immunoassay kits for detection of antineutrophil cytoplasmic antibodies: what is the optimal cut off? Ann. Rheum. Dis. 64, 1773. Kuhl, A., Korkmaz, B., Utecht, B., Kniepert, A., Schonermarck, U., Specks, U., Jenne, D.E., 2010. Mapping of conformational epitopes on human proteinase 3, the autoantigen of Wegener's granulomatosis. J. Immunol. 185, 387. Lucassen, R., Schulte-Pelkum, J., Petschinka, M., Fooke, M., 2011. New sensitive and reliable lateral flow assay for the detection of proteinase 3 and myeloperoxidase antibodies. Report on the 10th Dresden Symposium on Autoantibodies, Poster 694. Mahler, M., Radice, A., Yang, W., Bentow, C., Seaman, A., Bianchi, L., Sinico, R.A., 2012. Development and performance evaluation of novel chemiluminescence assays for detection of anti-PR3 and anti-MPO antibodies. Clin. Chim. Acta 413, 719. Roggenbuck, D., Buettner, T., Hoffmann, L., Schmechta, H., Reinhold, D., Conrad, K., 2009. High-sensitivity detection of autoantibodies against proteinase-3 by a novel third-generation enzyme-linked immunosorbent assay. Ann. N. Y. Acad. Sci. 1173, 41. Savige, J., Gillis, D., Benson, E., Davies, D., Esnault, V., Falk, R.J., Hagen, E.C., Jayne, D., Jennette, J.C., Paspaliaris, B., Pollock, W., Pusey, C., Savage, C.O., Silvestrini, R., van der, W.F., Wieslander, J., Wiik, A., 1999. International consensus statement on testing and reporting of antineutrophil cytoplasmic antibodies (ANCA). Am. J. Clin. Pathol. 111, 507. Schulte-Pelkum, J., Offermann, N., Fooke, M., 2012. New highly sensitive and specific lateral flow tests for the detection of proteinase 3, myeloperoxidase and glomerular basement membrane antibodies. 8th International Congress on Autoimmunity May 9–13th 2012, Granada, Spain Poster, p. 234. Silva, F., Hummel, A.M., Jenne, D.E., Specks, U., 2010. Discrimination and variable impact of ANCA binding to different surface epitopes on proteinase 3, the Wegener's autoantigen. J. Autoimmun. 35, 299. Specks, U., 2000. What you should know about PR3-ANCA. Conformational requirements of proteinase 3 (PR3) for enzymatic activity and recognition by PR3-ANCA. Arthritis Res. 2, 263. Wegener, F., 1987. On generalised septic vessel diseases. By Friedrich Wegener, 1937 (translation). Thorax 42, 918.
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
Contents lists available at ScienceDirect
Journal of Immunological Methods journal homepage: www.elsevier.com/locate/jim
Review
Development and validation of a lateral flow assay (LFA) for the determination of IgG-antibodies to Pr3 (cANCA) and MPO (pANCA) Offermann N. a, Conrad K. b, Fritzler M.J. c, Fooke Achterrath M. a,⁎ a b c
Dr. Fooke Laboratorien GmbH, Germany Technical University Dresden, Institute of Immunology, Dresden, Germany Faculty of Medicine, University of Calgary Calgary, Alberta, Canada
a r t i c l e
i n f o
Article history: Received 9 October 2013 Received in revised form 18 November 2013 Accepted 19 November 2013 Available online xxxx Keywords: Lateral flow assay Myeloperoxidase, MPO Proteinase 3, Pr3 Autoantibodies pANCA cANCA
a b s t r a c t The timely diagnosis of vasculopathies, such as granulomatosis with polyangiitis, has important implications for the favorable clinical outcome of these diseases. In the clinical setting, autoantibodies to proteinase 3 (Pr3) and myeloperoxidase (MPO) have been shown to be valuable adjuncts to an early and accurate diagnosis. The sensitive and specific detection of anti-Pr3 and anti-MPO was shown using a point of care device that employed rapid Lateral Flow Technologies. The validation of the lateral flow assay (LFA) was performed with serum samples collected in two Reference Laboratories and showed excellent results that were comparable to widely accepted and used ELISA. The advantage of the LFA is the flexibility to be used as an economical, point of care diagnostic device, features that are especially important for an early and accurate diagnosis and the prompt initiation of appropriate treatment so as to avoid inevitable development of undue complications of these diseases such as disseminated organ involvement, e.g. renal failure. © 2013 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Material and methods . . . . . . . . . . . . . . . . . . . . . . . 2.1. Serum samples . . . . . . . . . . . . . . . . . . . . . . . 2.2. Assay system . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Technical performance analysis . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Technical performance characteristics . . . . . . . . . . . . 3.1.1. Adjustment of assay cut-off to optimize sensitivity and 3.1.2. Inter-assay variation (lot to lot variation) . . . . . . 3.1.3. Intra-assay variation . . . . . . . . . . . . . . . . 3.2. Diagnostic performance analysis . . . . . . . . . . . . . . . 3.2.1. Cross-reactivity between AI-LFA Pr3 and AI-LFA MPO 3.2.2. Comparison to IFA (Euroimmun) . . . . . . . . . . 3.2.3. Comparison to “state of the art” ELISA (Orgentec) . . 3.2.4. Comparison to CLIA . . . . . . . . . . . . . . . . 3.2.5. Diagnostic performance summary . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
⁎ Corresponding author at: Dr. Fooke Laboratorien GmbH, Habichtweg 16, 41468 Neuss, Germany. Tel.: +49 2131 2984 135; fax: +49 2131 2984 184. 0022-1759/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jim.2013.11.017
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
1. Introduction Vasculitis is caused by inflammation of blood vessels and it manifests itself through a broad spectrum of clinical phenotypes. Anti-neutrophil cytoplasmic antibodies (ANCAs) are one of the most common serological features of small vessel vasculitis (SVV). ANCA associated vasculitis includes microscopic polyangiitis (MPA), eosinophilic granulomatosis with polyangiitis (EGPA) and granulomatosis with polyangiitis (GPA), formerly known as Wegener's granulomatosis. (Wegener, 1987; Falk et al., 2011). The sensitive and specific detection of proteinase 3 (Pr3, cANCA) and myeloperoxidase (MPO, pANCA) autoantibodies, occurring in about 80% of GPA cases, are time-proven serological biomarkers that can aid in the diagnosis of these conditions. Only timely and appropriate treatment can avoid the development of renal failure and/or other systemic complications of these diseases (Conrad et al., 2007a,b). Our aim was to develop a sensitive and specific, semiquantitative LFA together with a newly designed portable lateral flow assay reader as an approach to a point of care technology that would be an important adjunct to diagnosing and monitoring these diseases. Minimal laboratory equipment and an assay time of only 20 min result in an economic immunoassay. We compared the technical and diagnostic performance to conventional diagnostic assays such as indirect immunofluorescence (IIF) measuring cANCA and pANCA as well as the latest generation ELISA and a new chemiluminescence immunoassay (CLIA, BioFlash, INOVA Diagnostics, San Diego, CA, USA) (Lucassen et al., 2011; Schulte-Pelkum et al., 2012).
2. Material and methods 2.1. Serum samples The assay was developed by utilizing 87 commercially available serum samples representing 40 healthy donors, 5 unrelated disease controls, 20 cANCA positive and 22 pANCA positive individuals. ELISA and IIF ANCA results were available for these samples as supplied by several commercial institutions. International reference sera for anti-Pr3 (Pr3-ANCA human reference serum #16, Product Package Insert Catalog #IS2721 Lot #07-0002) and anti-MPO antibodies (MPO–ANCA human reference serum #15, Product Package Insert Catalog #IS2720 Lot #07-0001) were obtained from the Center for Disease Control and Prevention (CDC, Atlanta, GA, USA). The quantitative content of anti-Pr3 and anti-MPO antibodies in the reference sera had been determined as 100 IU/ml by different laboratories. For validation of the LFA, a different set of serum samples was collected in two different reference laboratories: Technical University Dresden, Institute of Immunology (n = 40, cANCA positive n = 20 and pANCA positive n = 20) and the University of Calgary, Calgary, Alberta, Canada (n = 64, healthy donors
n = 34 and ANCA positive n = 30). This study was approved by respective institutional human ethics review boards or was part of quality assurance and quality control processes. All patient identities remained anonymous in keeping with the latest version of the Helsinki Declaration of Human Research Ethics. The serum samples were tested by commercially available Pr3 and MPO ELISA tests and a new CLIA technology. IFA results were available for 40 sera collected in Dresden (pANCA 37/40; cANCA 39/40). Results for IFA are given in titers. The serum titer is defined as the highest serial dilution of a serum sample (expressed as the dilution factor) that still yields a positive result. Titers above 1/40 were considered as positive. All serum samples were tested for anti-Pr3 and anti-MPO antibodies in the new AI-LFA assays, measured with the newly developed lateral flow assay reader and compared to the results obtained with the other methods. 2.2. Assay system The autoimmune lateral flow assay system (AI-LFA) is based on a universal test cassette which can be used for all available IgG-antibody AI-parameters (Fig. 1). Purified native Pr3 and MPO antigens, both isolated from human neutrophils were biotinylated and used as liquid sources. The assay was performed as follows: 10 μl of undiluted serum was applied onto the sample application point (S) of the Basis Set module (Fig. 1b). Immediately after, the antigen solution was applied. The serum antibody to antigen binding occurs during the 20 min incubation while the mixture of antigen solution, conjugate and serum is driven through the device by capillary flow. The antigen specific IgG of the sample binds specifically to its corresponding antigen in the solution and these soluble antibody-(sIgG)-bound antigens are retained at the test line (T) by a capture molecule (Fig. 1). At the same time, the sIgG bound to the antigen is bound by an antibody coupled to gold particles (conjugate). The intensity of the color reaction at the test line is directly proportional to the amount of immune complexes consisting of ligand tagged antigen, sIgG, and anti-IgG specific conjugate. The signal intensity ranges from faintly pink color (low titer of sIgG) to dark ruby red (high titer of sIgG). Excess conjugate, which is not bound at the test line, forms a dark ruby control line (C) after 20 min of incubation. Interpretation of the test results is done by densitometry using a portable, proprietary lateral flow assay reader (Fig. 1c). Semi-quantitative results were expressed as relative units (RU), whereas the RU value corresponds to the color intensity of the test line (measured as grayscale level by a CCD camera) and, hence, to the concentration of autoantibody in the serum sample. 2.3. Technical performance analysis For Pr3 and MPO AI-LFA, the inter-assay variation was analyzed with 3 different lots of antigen and 3 different lots
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
Fig. 1. (a) Principle of Pr3/MPO AI-LFA, (b) AI-LFA cassette and (c) lateral flow assay reader.
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
3
4
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
Table 1 Lot-to-lot variation of 3 independent lots of AI-LFA basis sets measured with 2 positive and 2 negative serum samples per antigen. (HD = Healthy Donors).
3.2. Diagnostic performance analysis
Lot to lot variation (inter-assay) basis sets Antigen
Serum sample
Mean [RU]
Std dev [RU]
CV [%]
Pr3
HD HD Pos Pos HD HD Pos Pos
89.69 107.53 1715.69 1753.24 171.12 140.00 2043.07 1249.06
4.15 14.25 127.90 118.19 25.04 7.88 69.48 80.11
4.63 13.25 7.45 6.74 14.63 5.63 3.40 6.41
MPO
for MPO-ANCA. The results of the negative sera were below the established RU cut-off (see Table 3).
of LFA basis sets (LFA membranes and conjugate filters). The intra-assay variation was evaluated by fivefold replicate analysis of four serum samples (2 positive, 2 negative per antigen). 3. Results 3.1. Technical performance characteristics 3.1.1. Adjustment of assay cut-off to optimize sensitivity and specificity A ROC analysis was performed to determine the optimum cut-off for anti-Pr3 and anti-MPO detection. The best diagnostic performance (sensitivity and specificity) was obtained at a cut-off of 200 RU for anti-Pr3 and 450 RU for anti-MPO antibodies (see Supplementary Fig. 1). All of the anti-Pr3 and anti-MPO commercially procured positive serum samples were also positive in the AI-LFA: The ROC analysis demonstrated a sensitivity of 100% for Pr3-ANCA (20/20 samples positive) and MPO-ANCA (22/22 samples positive). The specificity was calculated to be 98% (95% CI 0.88–1.0) for both Pr3-ANCA (43/44 samples negative) and MPO-ANCA (44/45 samples negative). ROC analysis revealed area under the curve (AUC) values of 1.0 for Pr3 AI-LFA and for MPO AI-LFA, respectively (see Supplementary Fig. 1). In addition, the CDC international reference serum 16 (positive for Pr3) and reference serum 15 (positive for MPO) were also found to be positive by AI-LFA with 2249.5 and 758.7 RU, respectively. 3.1.2. Inter-assay variation (lot to lot variation) Lot to lot variation was calculated from the results of 3 different lots of LFA basis sets using the same lots of PR3 and MPO (see Table 1) and 3 different lots of LFA Pr3 and MPO antigen solutions using the same lot of LFA basis set (see Table 2). The coefficients of variation (CV) of the positive sera were between 1.25% and 7.45% for Pr3-ANCA and between 1.24% and 6.41% for MPO-ANCA. The results of the negative sera were below the required cut-off.
3.2.1. Cross-reactivity between AI-LFA Pr3 and AI-LFA MPO Of 38 serum samples which were positive in the Pr3 AI-LFA, no samples were false positive in the MPO LFA. This corresponds to a specificity of 100% for the MPO AI-LFA for serum samples with anti-Pr3 IgG antibodies (see Supplementary Table 1). Of 25 serum samples which were positive in the MPO AI-LFA, three samples were false positive in the Pr3 LFA. This corresponds in this group to a specificity of 88% for the Pr3 AI-LFA for serum samples with anti-MPO IgG antibodies (see Supplementary Table 2). 3.2.2. Comparison to IFA (Euroimmun) 3.2.2.1. AI-LFA Pr3. Of 39 serum samples, 20 were positive for cANCA (IIF, Euroimmun) and also positive for AI-LFA Pr3; this corresponds to a sensitivity of 100% (95% CI 0.83–1.00). Of the 19 disease control samples, 18 were negative in the AI-LFA Pr3. This corresponds to a specificity of 94.7% (95% CI 0.74–1.00; see Supplementary Fig. 2a). 3.2.2.2. AI-LFA MPO. Of 37 serum samples, 20 were positive for pANCA (IIF, Euroimmun) and also positive for AI-LFA MPO; this corresponds to a sensitivity of 100% (95% CI 0.83–1.00). Of the 17 disease control samples, all 17 were negative in the AI-LFA MPO; this corresponds to a specificity of 100% (95% CI 0.81–1.00; see Supplementary Fig. 2b). 3.2.3. Comparison to “state of the art” ELISA (Orgentec) 3.2.3.1. AI-LFA Pr3. 104 serum samples were tested with the AI-LFA Pr3 and compared to a highly sensitive anti-Pr3 ELISA (Orgentec). Of 39 serum samples which tested positive for Pr3 in the ELISA, 38 were identified as positive in the AI-LFA Pr3 system. In addition, 62 of 65 control serum samples were identified as negative in the AI-LFA Pr3. This corresponds to a sensitivity of 97.4% (95% CI 0.87–1.00) and a specificity of 93.8% (95% CI 0.85–0.98; see Supplementary Fig. 3a). 3.2.3.2. AI-LFA MPO. The 104 serum samples were also tested with the AI-LFA MPO and compared to anti-MPO ELISA (Orgentec). Of the 25 samples that were positive for MPO in the ELISA, all were also positive in the AI-LFA MPO system. 79 serum samples were found to be negative in both systems. This corresponds to a sensitivity of 100% (95% CI 0.86–1.00) and a specificity of 100% (95% CI 0.95–1.00; see Supplementary Fig. 3b). 3.2.4. Comparison to CLIA
3.1.3. Intra-assay variation Intra-assay precision of the two assays was performed by calculating the CVs for four samples per antigen each with five determinations using the same lot of LFA basis sets and antigen solutions. The CVs of the positive sera were between 5.94% and 8.58% for Pr3-ANCA and between 5.16% and 9.28%
3.2.4.1. AI-LFA Pr3. 83 serum samples were tested with the AI-LFA Pr3 and compared to the anti-Pr3 CLIA. Of the 38 serum samples which tested positive for Pr3, 37 tested positive in the AI-LFA Pr3 system. 43/45 serum samples were identified as negative in the AI-LFA Pr3. This corresponds to a sensitivity of
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
5
Table 2 Lot-to-lot variation of 3 independent antigen lots of A) Pr3 and B) MPO measured with 6 sera per antigen. A)
B)
Pr3 antigen lot to lot variation (3 lots)
MPO antigen lot to lot variation (3 lots)
Serum sample
Mean [RU]
Std dev [RU]
CV [%]
Serum sample
Mean [RU]
Std dev [RU]
CV [%]
HD
136.73 132.21 132.79 2355.40 1319.71 5031.44
9.32 16.66 14.79 47.32 72.81 62.88
6.82 12.60 11.14 2.01 5.52 1.25
HD
250.54 328.99 232.89 2358.79 1267.14 677.17
15.22 19.56 13.35 29.17 47.96 14.07
6.07 5.95 5.73 1.24 3.78 2.08
Pr3 positive
97.4% (95% CI 0.86–0.99) and a specificity of 97.4% (95% CI 0.82–0.99; see Supplementary Fig. 4a). 3.2.4.2. AI-LFA MPO. 83 serum samples were tested with the AI-LFA MPO and the anti-MPO CLIA. Of 26 serum samples which tested positive for MPO by CLIA, 24 were found to be positive in the AI-LFA MPO system. 56/57 serum samples were found to be negative in the AI-LFA MPO. This corresponds to a sensitivity of 92.3% (95% CI 0.75–0.99) and a specificity of 98.2% (95% CI 0.91–1.00; see Supplementary Fig. 4b). 3.2.5. Diagnostic performance summary ROC analysis of AI-LFA compared with the established assay methods revealed AUC values between 0.96 and 1.00 (see Table 4). An AUC of 1.0 indicates that the performance of the two compared tests is identical, which is the case for AI-LFA MPO compared with IFA pANCA as well as MPO ELISA (see Table 4). 4. Discussion The current diagnosis of ANCA-associated small vessel vasculitis typically relies on a combination of IIF cANCA, pANCA, atypical ANCA (aANCA) and ELISA, each detecting IgG-autoantibodies to Pr3 and MPO (Savige et al., 1999; Holle et al., 2005). A positive result for IIF ANCA screening is normally confirmed by ELISA, both with an assay time between 1.5 and 3 h. Most of the anti-MPO ELISAs show a diagnostic performance comparable with IIF-ANCA (Holle et al., 2005). Anti-Pr3 ELISAs often show a reduced sensitivity, whereas the specificity is comparable to IIF-ANCA (Roggenbuck et al., 2009). It has been suggested that the higher sensitivity of third generation
Table 3 Reproducibility of 5-fold AI-LFA measurements for 4 sera (2 healthy donors, 2 positive serum samples) per antigen. Antigen
Serum sample
Replicates
Mean (RU)
Std dev (RU)
CV (%)
Pr3
HD HD Pos Pos HD HD Pos Pos
5 5 5 5 5 5 5 5
78.62 87.37 1553.22 1447.24 154.81 148.78 2236.96 1257.34
8.82 12.86 133.19 85.98 12.64 14.34 115.37 116.70
11.22 14.72 8.58 5.94 8.16 9.64 5.16 9.28
MPO
MPO positive
ELISAs compared with previous ELISAs was achieved by the presentation of autoantigenic epitopes via novel protein anchoring techniques (Roggenbuck et al., 2009). The limitations of ELISA may be related to observations that human ANCA bind mostly to conformational epitopes of Pr3, and that the adsorption of the antigen to solid phases could dramatically reduce the binding of ANCA (Specks, 2000; Silva et al., 2010; Kuhl et al., 2010). Accordingly, the focus of assay development has been to represent the antigen in as close to its native conformation as technically possible in order to enable antibody binding to all clinically relevant epitopes. Thus, the most important innovations were to overcome these problems by using anchor techniques, which increase the sensitivity by presenting a higher number of epitopes for binding of specific autoantibodies (Roggenbuck et al., 2009; Mahler et al., 2012). Our new AI-LFA lateral flow assay for the detection of anti-Pr3 and anti-MPO antibodies, complemented by the newly developed lateral flow assay reader, show results of high sensitivity and specificity when compared to IIF-ANCA and newer generation ELISA. An AUC of 1.0 indicates that the sensitivity and specificity of two compared tests are identical. This is the case for AI-LFA MPO versus IFA and ELISA (see Table 4). The sensitivity and specificity for AI-LFA Pr3 versus IFA and ELISA is also excellent with an AUC of 0.99. The high sensitivity of the AI-LFAs may be explained by the antigen epitopes being available for antibody binding while in the liquid phase of the assay. The diagnostic performance of AI-LFA to CLIA is also comparable, but minimally more discrepant compared with IIF-ANCA and third generation ELISA (see Table 4). The sensitivity and specificity for AI-LFA compared to CLIA are 97.4% and 93.3% for Pr3 (AUC 0.99) and 92.3% and 98.2% for MPO (AUC 0.96). This might be explained by the characteristics of CLIA that might have a linear dynamic range of
Table 4 Sensitivities and specificities of AI-LFA when compared to different assay systems. AI-LFA
Compared with
Sensitivity (%)
Specificity (%)
n=
Pr3 (cut-off 200 RU)
IFA cANCA hs Pr3 ELISA BIO-FLASH Pr3 IFA pANCA MPO ELISA BIO-FLASH MPO
100 97.4 97.4 100 100 92.3
94.7 93.8 93.3 100 100 98.2
39 104 83 37 104 83
MPO (cut-off 450 RU)
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
6
O. N. et al. / Journal of Immunological Methods xxx (2013) xxx–xxx
antigen–antibody binding throughout virtually all concentrations of antibody–antigen reactions (Mahler et al., 2012). The AI-LFA MPO shows a better diagnostic performance compared with the IFA and ELISA than the AI-LFA Pr3 (see Table 4). In contrast, the AI-LFA Pr3 does agree slightly better with the BIO-FLASH system. Altogether the AI-LFA Pr3 as well as the AI-LFA MPO achieve sensitivities and specificities greater than 92% compared with all three different test systems (IFA, ELISA and CLIA). Interpretation of LFA with the new digital densitometry reader provides objective semi-quantitative results, in contrast to IIF-ANCA for which an experienced technologist is required. The sensitivities of the new AI-LFA tests are competitive with the sensitivity of a third generation ELISA. In addition, the AI-LFA assays feature a time-to-results of 20 min, economical single testing and require minimal laboratory equipment. The novel AI-LFA tests together with a precise reading-out technology show a highly sensitive and specific detection of autoantibodies to Pr3 and MPO, comparable with ELISA tests while keeping all the advantages of a rapid test. 5. Conclusions The new AI-LFA lateral flow assays for the detection of anti-Pr3 and anti-MPO antibodies, read with the aid of the newly developed lateral flow assay reader, show test results of high sensitivity and specificity when compared with IFA, the latest generation ELISA- and CLIA- tests. The combination of short assay time, economical single sample testing and minimal laboratory equipment requirement makes the AI-LFA tests highly convenient for the testing of anti-Pr3 and anti-MPO autoantibodies. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jim.2013.11.017.
References Conrad, K., Schößler, W., Hiepe, F., Fritzler, M., 2007a. Myeloperoxidase antibodies. In: Conrad, K., Schößler, W., Hiepe, F., Fritzler, M. (Eds.), Autoantibodies in Systemic Autoimmune Diseases- A Diagnostic Reference, pp. 111–113 (Pabst). Conrad, K., Schößler, W., Hiepe, F., Fritzler, M., 2007b. Proteinase 3 antibodies. In: Conrad, K., Schößler, W., Hiepe, F., Fritzler, M. (Eds.), Autoantibodies in Systemic Autoimmune Diseases- A Diagnostic Reference, pp. 147–149 (Pabst). Falk, R.J., Gross, W.L., Guillevin, L., Hoffman, G.S., Jayne, D.R., Jennette, J.C., Kallenberg, C.G., Luqmani, R., Mahr, A.D., Matteson, E.L., Merkel, P.A., Specks, U., Watts, R.A., 2011. Granulomatosis with polyangiitis (Wegener's): an alternative name for Wegener's granulomatosis. Arthritis Rheum. 63, 863. Holle, J.U., Hellmich, B., Backes, M., Gross, W.L., Csernok, E., 2005. Variations in performance characteristics of commercial enzyme immunoassay kits for detection of antineutrophil cytoplasmic antibodies: what is the optimal cut off? Ann. Rheum. Dis. 64, 1773. Kuhl, A., Korkmaz, B., Utecht, B., Kniepert, A., Schonermarck, U., Specks, U., Jenne, D.E., 2010. Mapping of conformational epitopes on human proteinase 3, the autoantigen of Wegener's granulomatosis. J. Immunol. 185, 387. Lucassen, R., Schulte-Pelkum, J., Petschinka, M., Fooke, M., 2011. New sensitive and reliable lateral flow assay for the detection of proteinase 3 and myeloperoxidase antibodies. Report on the 10th Dresden Symposium on Autoantibodies, Poster 694. Mahler, M., Radice, A., Yang, W., Bentow, C., Seaman, A., Bianchi, L., Sinico, R.A., 2012. Development and performance evaluation of novel chemiluminescence assays for detection of anti-PR3 and anti-MPO antibodies. Clin. Chim. Acta 413, 719. Roggenbuck, D., Buettner, T., Hoffmann, L., Schmechta, H., Reinhold, D., Conrad, K., 2009. High-sensitivity detection of autoantibodies against proteinase-3 by a novel third-generation enzyme-linked immunosorbent assay. Ann. N. Y. Acad. Sci. 1173, 41. Savige, J., Gillis, D., Benson, E., Davies, D., Esnault, V., Falk, R.J., Hagen, E.C., Jayne, D., Jennette, J.C., Paspaliaris, B., Pollock, W., Pusey, C., Savage, C.O., Silvestrini, R., van der, W.F., Wieslander, J., Wiik, A., 1999. International consensus statement on testing and reporting of antineutrophil cytoplasmic antibodies (ANCA). Am. J. Clin. Pathol. 111, 507. Schulte-Pelkum, J., Offermann, N., Fooke, M., 2012. New highly sensitive and specific lateral flow tests for the detection of proteinase 3, myeloperoxidase and glomerular basement membrane antibodies. 8th International Congress on Autoimmunity May 9–13th 2012, Granada, Spain Poster, p. 234. Silva, F., Hummel, A.M., Jenne, D.E., Specks, U., 2010. Discrimination and variable impact of ANCA binding to different surface epitopes on proteinase 3, the Wegener's autoantigen. J. Autoimmun. 35, 299. Specks, U., 2000. What you should know about PR3-ANCA. Conformational requirements of proteinase 3 (PR3) for enzymatic activity and recognition by PR3-ANCA. Arthritis Res. 2, 263. Wegener, F., 1987. On generalised septic vessel diseases. By Friedrich Wegener, 1937 (translation). Thorax 42, 918.
Please cite this article as: N., O., et al., Development and validation of a lateral flow assay (LFA) for the determination of IgGantibodies to Pr3 (cANCA) and MPO (pANCA), J. Immunol. Methods (2013), http://dx.doi.org/10.1016/j.jim.2013.11.017
