False-positive results of gold immunochromatography assay for IgM and IgG antibodies to 2019-nCoV

ABSTRACT

Objective: To investigate the specificity of three gold immunochromatography assay (GICA) kits for IgM and IgG antibodies against 2019 novel coronavirus (2019-nCoV), and explore the interference factors causing false-positive results and methods for eliminating interference. Methods: The serum samples were retrospectively collected from the patients presented to the Peking University Third Hospital from March 6, 2020 to April 20, 2020. Three commercial GICA kits were used to detect SARS-CoV-2 antibodies in serum samples from the following 4 groups. ①Healthy control group 120 serum samples. ②Special pathogen group 139 samples, including 42 Hepatitis B virus (HBV) antibody positive sera, 27 Hepatitis C virus (HCV) antibody positive sera, 11 Treponema Pallidum (TP) antibody positive sera, 12 Cytomegalovirus (CMV) IgM positive sera, 12 Rubella virus (RV) IgM positive sera, 8 Epstein-Barr virus (EBV) IgM positive sera, 12 sera with Escherichia coli confirmed by blood cultures, 15 sera from influenza A virus infected patients and 15 influenza B virus infected patients. ③Endogenous interference group242 samples, including 20 sera with elevated total IgM, 20 sera with elevated total IgG, 15 sera with elevated total complement, 9 sera with elevated alpha fetoprotein (AFP), 84 sera with normal rheumatoid factor (RF) level and 94 sera with elevated RF level. ④Pregnant women and patients with malignant tumor 126 samples. Reverse transcription polymerase chain reaction (RT-PCR) was used to detect 2019-nCoV viral nucleic acid in the throat swabs. Statistical analysis was performed by SPSS, version 19.0. Results: The false-positive rates of 2019-nCoV IgM and IgG in healthy controls tested by A kit were 0 and 0.83% (1/120), 4.17% (5/120) by B kit, and 0 by C kit. One in 12 serum samples collected from patients with E. coli bactereia was positive for 2019-nCoV antibodies tested by three kits. The false-positive rate of 2019-nCoV antibodies in HCV antibody positive sera was 7.41% (2/27) by B kit. The false-positive rates of 2019-nCoV antibodies in sera with normal and elevated RF level tested by B kit were 9.52% (8/84) and 54.26%(51/94) (χ²=40.05, P<0.001), 3 serum samples in elevated RF group were positive for 2019-nCoV IgM/IgG antibodies by A and C kits. The overall false-positive rates of A, B and C antibody test kits in all the subjects were 0.55%(3/548), 4.2%(23/548) and 0.73% (4/548), respectively. Conclusions: The specificity of GICA kits varies greatly among different manufacturers. Rheumatoid factor is the main factor leading to false-positive results of 2019-nCoV antibodies tested by B kit. The use of high-purity antibody and suitable immunoassay blocking reagent will be helpful to improve the specificity of GICA kits.