Original Article
A multi-center clinical study for ANA specific autoantibodies detection by chemiluminescent immunoassay
Chaojun Hu, Jing Luo, Shulan Zhang, Chuiwen Deng, Xin Zhang, Like Zhao, Qinglin Peng, Ping Zhu, Cibo Huang, Guochun Wang, Shengyun Liu, Yongfei Fang, Xiaosan Chen, Le Liu, Qingchun Li, Jiyang Li, Mengtao Li, Xiaofeng Zeng
Published 2017-08-11
Cite as Chin J Lab Med, 2017, 40(8): 602-608. DOI: 10.3760/cma.j.issn.1009-9158.2017.08.011
Abstract
ObjectiveTo evaluate the clinical performance of chemiluminescent immunoassay (CLIA) on anti-nuclear antibody(ANA) specific autoantibodies testing.
MethodsA multi-center clinical study A total of 811 Sera samples were collected from 6 collaborating hospitals during the period of April to July 2016, and tested with CLIA and line immunoassay (LIA) in parallel for autoantibodies to ribonucleoprotein(RNP), smith antigen(Sm), SSA/Ro60, SSB/La, centromere protein B(CENPB), double-stranded DNA(dsDNA), nucleosome(Nuc), and ribosome P protein(Rib-P). The positive rate, specificity and qualitative coincidence rate for each antibody between CLIA and LIA methods were analyzed. All discrepant samples for systemic lupus erythematosus (SLE) highly specific autoantibodies (including anti-Sm, dsDNA, Nuc and Rib-P) were retested by enzyme linked immunosorbent assay (ELISA) and further analyzed with SLE disease cohort using McNemar test.
ResultsThe positive rate and specificity of CLIA and LIA for antibodies to ANA specific antigens were comparable. Excellent qualitative coincidence were found between CLIA and LIA for the detection of anti-RNP, SSA/Ro60, SSB/La and CENPB (Kappa>0.75), while the coincidence rate foranti-Sm, dsDNA, Nuc and Rib-P detection were moderate (0.4<Kappa<0.75). The data from discrepant samples retested with ELISA showed there was no significant difference between CLIA and ELISA for the detection of anti-Sm (χ2=3.333, P=0.067) and Rib-P (χ2=0.888, P=0.345), but a significant difference were observed between LIA and ELISA test results for anti-Sm (χ2=5.444, P=0.019), anti-dsDNA (χ2=5.812, P=0.015), anti-Nuc (χ2=12.071, P<0.001) and anti-Rib-P (χ2=25.861, P<0.001). When analyzing discrepant sample with SLE diagnosis, data from CLIA for anti-dsDNA (χ2=1.132, P=0.249) and anti-Nuc (χ2=0.571, P=0.449) showed no differencewith SLE diagnosis, while data from LIA for anti-Sm(χ2=21.125, P<0.001), anti-dsDNA(χ2=59.507, P<0.001), anti-Nuc(χ2=38.4, P<0.001) and anti-Rib-P (χ2=6.259, P=0.012)showed significant difference with SLE diagnosis.
ConclusionsCLIA and LIA showed similar positive rate and specificity when testing antibody to ANA specific antigens. The coincidence rate between CLIA and LIA for the detection of anti-RNP, SSA/Ro60, SSB/La and CENPB wereexcellent and moderate for anti-Sm, dsDNA, Nuc and Rib-P. CLIA test results were more comparable with ELISA and had a better correlation with SLE disease diagnosis among all discrepant samples. With the additional benefits of full automation, quantitative output, random-access and high flexibility, CLIA is a preferable test platform for the detection of ANA specific autoantibodies. (Chin J Lab Med, 2017, 40: 602-608)
Key words:
Multi center study; Antibody, anti-nuclear; Lupus erythematosus, systemic; Sjögren′s syndrome; Enzyme-linked immunosorbent assay
Contributor Information
Chaojun Hu
Department of Rheumatology and Clinical Immunology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences &
Peking Union Medical College, Key Laboratory of Rheumatology and Clinical Immunology, Ministry of Education, Beijing 100032, China
Jing Luo
Shulan Zhang
Chuiwen Deng
Xin Zhang
Like Zhao
Qinglin Peng
Ping Zhu
Cibo Huang
Guochun Wang
Shengyun Liu
Yongfei Fang
Xiaosan Chen
Le Liu
Qingchun Li
Jiyang Li
Mengtao Li
Xiaofeng Zeng