Validation of the 2023 American College of Rheumatology/European League Against Rheumatism antiphospholipid syndrome classification criteria in a Chinese systemic lupus erythematosus cohort

BACKGROUND: Given the high incidence of secondary antiphospholipid syndrome (APS) in patients with systemic lupus erythematosus (SLE) and the similarity of clinical manifestations between APS and SLE, validating the 2023 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) antiphospholipid syndrome (APS) classification criteria within this special population is of critical importance. This study aimed to clarify the clinical applicability of the 2023 criteria for classifying secondary APS in Chinese SLE populations. METHODS: SLE patients with available data of antiphospholipid antibodies were retrospectively identified from the Peking University First Hospital Treat SLE to Target (STAR) cohort starting from April 1, 2007. This study included the patients from STAR between April 1, 2007 and May 1, 2024. Patients were independently classified according to both the 2023 criteria and 2006 criteria. For the 2023 criteria, two strategies were implemented (all-in or all-out) to determine whether overlapping manifestations of APS and SLE should be scored as APS when clinical differentiation was not feasible. The "gold standard" for APS classification was independently established by two rheumatologists. Sensitivity and specificity were compared between the two criteria. Subgroup analyses were conducted in terms of gender, age, body mass index, disease duration, and comorbidities. RESULTS: A total of 1001 SLE patients were included in this study. According to the two scoring strategies, if all overlapping features were considered as APS manifestations, 101 patients (10.1%) met the 2023 criteria, which showed higher sensitivity (0.878 vs. 0.768) but lower specificity (0.968 vs. 0.988) compared to the 2006 criteria. When overlapping features were excluded from APS scoring, 68 patients (6.8%) met the 2023 criteria, which demonstrated similar specificity (0.993 vs. 0.988) but lower sensitivity (0.756 vs. 0.768) compared to the 2006 criteria. CONCLUSIONS: The sensitivity and specificity of 2023 APS classification criteria varied depending on the scoring strategy used. Considering the presence of SLE as a distinct item in future APS classification criteria may be warranted.