Comparative effectiveness and safety of different hydroxychloroquine doses for patients with systemic lupus erythematosus: nationwide analysis using target trial emulation framework

Background: Hydroxychloroquine (HCQ) is a mainstay treatment for systemic lupus erythematosus (SLE), decreasing disease activity and prevent flares. Guidelines recommend maintaining HCQ at a target dose ( < 5 mg/kg/day) to minimize the risk of retinal toxicity. However, recent study found that low-dose HCQ use was associated with an increased risk of SLE flares. Clinically, healthcare providers face a dilemma in choosing optimal HCQ dose due to the risk of both SLE flares and retinal toxicity. Regarding benefit-risk relationship of different doses, it is important to evaluate both the effectiveness and safety of different HCQ doses, particularly in patients with specific risk factors.

Objectives: To evaluate the effectiveness and safety of different HCQ doses for patients with SLE

Methods: We conducted a target trial emulation study using Taiwan's NHIRD database to evaluate the effects of different HCQ doses on patients with SLE. We included patients aged over 20 years and newly received HCQ treatment between 2010 and 2021. We classified patients into high dose HCQ (≥ 400 mg/d) and low dose HCQ ( < 400 mg/d) based on their first prescription for HCQ. To emulate randomized allocation in the target trial, we implemented propensity score approaches with IPTW. The effectiveness outcomes were atherosclerosis and vein thromboembolism during the study period. The safety outcomes were retinopathy during the study period. As-started analysis was performed for causal contrasts of interest. We applied cox proportional hazards models to generate hazard ratios for each outcome with 95% confidence intervals.

Results: We included a total of 878 patients newly receiving high-dose HCQ and 22,405 patients newly receiving low-dose HCQ with a mean age of 39 years (SD, 18.9) and 43.7 years (SD, 17.2), respectively. After implementing propensity score weighting with IPTW, the weighted population comprised 22,779 patients in the high-dose group and 23,285 patients in the low-dose group. Patients receiving high-dose HCQ had a lower risk of atherosclerosis (HR, 0.94; 95%CI, 0.89 to 0.98) and vein thromboembolism (HR, 0.51; 95%CI, 0.44 to 0.60) compared to those receiving low-dose HCQ. Patients receiving high-dose HCQ had a higher risk of retinopathy (HR, 1.28; 95%CI, 1.02 to 1.60) and stroke (HR, 1.30; 95%CI, 1.18 to 1.45) compared to those receiving low-dose HCQ.

Conclusions: Our findings found an association between the use of high-dose HCQ and a decreased atherosclerosis and vein thromboembolism while also indicating an increased risk of retinopathy and stroke. When choosing the optimal doses of HCQ, healthcare providers should aware balancing its effectiveness and safety in SLE patients with specific risk factors.