Development of clinical manifestations in antiphospholipid antibodies carriers according to the 2023 acr/EULAR serological domains

Abstract

Abstract INTRODUCTION Antiphospholipid syndrome (APS) is a thrombophilic disorder defined by thrombotic/obstetric events and persistent antiphospholipid antibodies (aPL). The 2023 ACR/EULAR classification criteria introduced a novel point-based system, but the prognostic value and ability to stratify risk remain untested. This study aims to report the incidence and timing of clinical manifestations in aPL carriers and to evaluate the predictive value of the 2023 ACR/EULAR score system. METHODS A retrospective observational study was conducted at a tertiary referral center, including patients with positive lupus anticoagulant (LA), anticardiolipin antibodies (aCL), or anti-β2-glycoprotein I (anti-β2GPI), confirmed ≥12 weeks apart (except for isolated LA) between January 2018 and March 2024. Patients meeting the 2006 Sydney criteria or with clinical domains (D) 1–4 present at the time of serological confirmation were excluded. Those with valvular disease or thrombocytopenia (D5 and D6) were considered asymptomatic and included. We calculated incidence and time to event, assessed associations with serological domains, and evaluated the predictive performance of the 2023 ACR/EULAR laboratory score using logistic regression adjusted for potential confounders and ROC curve analysis. Event-free survival (EFS; time to clinical manifestations from D1 to D4) was analysed using Cox regression.RESULTS A total of 985 patients were included [median age 53 years; interquartile range (IQR) 41–65 years], of whom 686 (69.6%) were women. The median follow-up was 2.84 years [IQR 1.12–4.74]. Among aPL carriers, the cumulative incidence of clinical events was low (<3%). Venous thrombotic events (VTE) (D1) and arterial thrombotic events (ATE) (D2) occurred in both low- and high-risk profiles. Microvascular involvement (D3) appeared in 20 patients (2%), mainly as nephropathy and cardiomyopathy. Obstetric morbidity (D4) was rare, with ≥3 fetal losses being the most common event in 13 patients (1.3%). D5 and D6 were observed in 13 (1.3%) and 91 (9.2%) patients, respectively, often preceding aPL detection in 4 (0.4%) and 70 (7.1%). Median time to event was shortest for D4 (0.68 years) and D2 (0.80 years) and longest for D3 (2.29 years). Moderate/high IgM aCL/anti-β2GPI was associated with increased ATE risk in low-risk patients (OR 2.89; 95% CI 1.15–7.30; p=0.025). No statistical associations were found between serological profile and D3, D4, or D5. Thrombocytopenia (D6) was significantly associated with isolated LA (OR 1.990; 95% CI 1.105–3.582; p=0.022); persistent LA (OR 2.489; 95% CI 1.482–4.181; p=0.001); moderate/high IgM aCL or anti-β2GPI (OR 1.676; 95% CI 1.003–2.803; p=0.049); and moderate IgG aCL and/or anti-β2GPI (OR 3.249; 95% CI 1.429–7.385; p=0.005). The likelihood of fulfilling the 2023 ACR/EULAR criteria during follow-up was significantly higher in patients with simultaneously high titers of IgG aCL and anti-β2GPI (OR 4.729; 95% CI 1.110–15.518; p=0.01). EFS was reduced in those with simultaneous high IgG titers of aCL/anti-β2GPI (HR 3.268; 95% CI 1.166–9.161; p=0.024). In a logistic regression adjusted for age and sex, the cumulative score of laboratory domains was not associated with VTE (OR 0.954, 95% CI 0.853–1.068; p=0.413, additionally adjusted for VTE risk), ATE (OR 1.020, 95% CI 0.914–1.138; p=0.721, adjusted for cardiovascular risk), microvascular thrombosis (OR 1.011, 95% CI 0.862–1.187; p=0.892, adjusted for autoimmune disease), or obstetric morbidity per ACR/EULAR (OR 1.046, 95% CI 0.927–1.180; p=0.465, adjusted for in vitro fertilization). In a global model for any manifestation (D1 to D4), adjusted for all aforementioned confounders, the laboratory score remained non-significant (OR 1.027, 95% CI 0.952–1.108; p=0.488). The area under the curve (AUC) for prediction of VTE was 0.467 (95% CI 0.375–0.559), for ATE 0.528 (95% CI 0.438–0.618), for microvascular thrombosis 0.540 (95% CI 0.429–0.652), and for obstetric morbidity 0.554 (95% CI 0.446–0.663). The AUC for the prediction of any clinical manifestation (D1-D4) was 0.535 (95% CI 0.466–0.603). CONCLUSIONS Despite IgM antibodies exclusion from the classification criteria, their presence increased ATE risk in patients with low cardiovascular risk. Isolated and persistent LA, and moderate IgM/IgG aPL levels, were associated with thrombocytopenia. The 2023 ACR/EULAR laboratory score system lacked prognostic value in aPL carriers, and the composite score failed to discriminate risk.