Clinical evaluation of the novel digital liquid chip method for anti-dsDNA detection in SLE
Yina Bai1, Rui Yu1,2, Chiyuan Xue1
1Department of Rheumatology and Clinical Immunology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College; National Clinical Research Center for Dermatologic and Immunologic Diseases (NCRC-DID), Ministry of Science & Technology; Key Laboratory of Rheumatology and Clinical Immunology, Ministry of Education, Beijing, People's Republic of China.
Objective:
This study aims to evaluate the diagnostic performance of the digital liquid chip method (DLCM) compared with indirect immunofluorescence (IIF) and chemiluminescent immunoassay (CLIA) for anti-double-stranded DNA (dsDNA) antibody detection in SLE.
Methods:
The retrospective study consecutively enrolled 1349 patients, including 698 with SLE and 651 with other autoimmune diseases at Peking Union Medical College Hospital. Anti-dsDNA antibodies were detected using IIF (EUROIMMUN, Luebeck, Germany), CLIA (YHLO, Shenzhen, China) and DLCM (Livzon, Zhuhai, China). The sensitivity, specificity and area under the curve (AUC) of each method and combination were compared at the recommended manufacturer cut-offs. The agreement between methods and the association between antibody levels and clinical characteristics including disease activity, complement levels and organ involvement were also evaluated.
Results:
All methods exhibited high specificity, while IIF performed best (98.5%), significantly greater than CLIA (96.3%) and DLCM (96.6%) (p < 0.05). CLIA demonstrated the highest sensitivity (48.1%), outperforming IIF (36.0%) and DLCM (41.4%) (p<0.001). Cohen's kappa indicated substantial positive agreement between DLCM and CLIA (κ=0.67), and moderate agreement between IIF and the other methods (κ=0.52-0.55). Combining IIF with DLCM or CLIA improved diagnosis performance, with IIF+CLIA achieving the highest sensitivity (54.0%), accuracy (74.1%) and AUC (0.75). Moreover, anti-dsDNA positivity was strongly associated with lower complement levels (C3: 0.71 vs 0.90 g/L in DLCM+ vs DLCM-, p<0.001) and moderate-severe disease activity (65.0% DLCM positive). DLCM uniquely predicted musculoskeletal involvement (55.3% vs 44.7%, p<0.01). However, the diagnostic performance for renal involvement was limited (sensitivity 46.9%, specificity 56.3%, AUC=0.52).
Conclusions:
DLCM demonstrated substantial agreement with CLIA and held potential for SLE diagnosis and monitoring. A multiassay strategy, using a sensitive assay like CLIA or DLCM for initial screening and a highly specific assay like IIF for confirmation, optimises diagnostic performance for anti-dsDNA antibody detection in SLE.
