Kritsana JanyapoonPantakit JivakanontRungkarn SurbrsingWijitporn SiriprapapanThanarat TachawuttiwatSunee KorbsrisateRangsit UniversityMahidol University2018-06-212018-06-212005-02-01Pathology. Vol.37, No.1 (2005), 63-68003130252-s2.0-14644416467https://repository.li.mahidol.ac.th/handle/20.500.14594/17067Aim: To compare different sources of DNA for use in ELISA-based assays for anti-dsDNA antibody detection in systemic lupus erythematosus (SLE) diagnosis. Method: Bacterial genomic DNA from Flavobacterium menignosepticum, Proteus vulgalis, Seratia marcescens, Streptococcus pyogenes and Salmonella typhimurium and genomic DNA from human blood were used as antigens for IgG anti-dsDNA detection by enzyme-linked immunosorbent assay (ELISA). Eighty-six sera were tested, 28 derived from patients with SLE, 28 from patients with other rheumatic diseases and 30 from normal human subjects. Results: Genomic DNA from Flavobacterium menignosepticum and human blood had high sensitivity (75%, 82%) and specificity (91%, 91%) for anti-dsDNA detection in diagnosis of SLE. However, human genomic DNA was the most effective antigen of all antigens studied. The assay had a higher sensitivity but lower specificity than commercial ELISA (61% sensitivity and 95% specificity). There was a high level of correlation between commercial ELISA and ELISA using human genomic DNA as antigen (r=0.776, p<0.001) and they exhibited a high level of diagnostic agreement with each other (κ=0.890, p<0.001). Conclusion: The genomic DNA from human blood is a potentially useful source of antigen for the detection of anti-dsDNA by ELISA. However, further studies are required to compare the performance of ELISA using this source of antigen against commercial radioimmunoassays for anti-dsDNA detection. © 2005 Royal College of Pathologists of Australasia.Mahidol UniversityMedicineArticleSCOPUS10.1080/09638280400025036