AB0466 ADVERSE EVENTS AND THE RELAPSE RISK OF SYSTEMIC LUPUS ERYTHEMATOSUS DURING HYDROXYCHLOROQUINE TREATMENT

Abstract

Background The antimalarial drug hydroxychloroquine (HCQ) is used worldwide to control the disease activity of systemic lupus erythematosus (SLE). In Japan, HCQ was not approved until September 2015 due to the problem of retinopathy induced by chloroquine. There is insufficient evidence of the effects of HCQ or adverse events linked to HCQ treatment. Objectives Here, we evaluated adverse events and the relapse risk of SLE during HCQ treatment. Methods We retrospectively analyzed the data of 109 patients diagnosed with SLE and treated with HCQ for ≥12 months at Nagasaki University Hospital and community hospitals. The demographic data included the patient‘s age at the onset of SLE, gender, the disease duration of SLE (the time from the diagnosis of SLE until the renal biopsy), comorbidities of Sjogren syndrome (SS)/anti-phospholipid syndrome (APS), and the treatment for induction. We identified the risk of adverse events and relapse at 12 months after the introduction of HCQ. Decision tree models predicting relapse were built with the Classification and Regression Trees (CART) algorithm. The data of the length of time between a patient‘s HCQ induction to the first observation of relapse was analyzed by the Kaplan-Meier method with a log-rank test. Results Most of the patients were female (88.1%). The median age at the introduction of HCQ was 40.0 years (interquartile range [IQR] 30.5–50.0 years), and the SLE disease duration was 95 months (IQR 38.0–184.5 months). The mean observation period after HCQ introduction was 12 months. The comorbidity rates of SS and APS were 25.7% and 20.2%, respectively. The SELENA-SLEDAI scores were significantly decreased at 3 months post-HCQ introduction. The dose of oral prednisolone was significantly decreased at 6 months post-HCQ introduction. Eighty-six patients (78.9%) were continuing HCQ at 12 months post-introduction. Adverse events occurred in 27 patients (24.8%), including skin rashes (n=11, 10.1%) and gastrointestinal symptoms (n=6, 5.5%). The sole predictive factor for adverse events was the white blood cell (WBC) count at baseline (odds ratio: 0.9997, 95%CI: 0.9994–0.9999, p=0.0285). Twelve of 86 patients (14.0%) experienced relapse that required the start of prednisolone or an immunosuppressant or an increased prednisolone dose. The multivariate analysis revealed that the C4 value at baseline was a predictive factor of relapse (odds ratio: 0.841, 95%CI: 0.718–0.984, p=0.0097). The cut-off point determined by the CART algorithm showed that C4 ≥9.3 mg/dl at baseline provides the best performance for predicting relapse. The Kaplan-Meier analysis showed that compared to C4 Conclusion A lower C4 value at HCQ introduction was a predictive factor for the relapse of SLE, and a lower WBC count was a predictive factor for adverse events.