Objectives To study the factors determining the hydroxychloroquine (HCQ) serum concentration in a cohort of Chinese patients with systemic lupus erythematosus (SLE).
Methods Consecutive patients who fulfilled ≥4 of the 1997 ACR criteria for SLE and had received HCQ for 6 months or more were recruited from our lupus clinic and hospital admissions in a 9-month period starting from November 2011. Patients were prescribed HCQ at fixed daily dosages of 400 mg, 300 mg, 200mg or less than 200mg (eg. 200mg 3 or 5 times per week) at the discretion of their attending doctors according to disease activity, organ manifestations and risk factors for HCQ toxicities. Blood was assayed for the serum levels of HCQ by an in-house technique using the tandem mass spectrometry (SPE-MS/MS). Factors affecting HCQ serum concentrations were studied by univariate and multivariate linear regression analyses. Covariates being tested in the regression models included age, sex, body mass index (BMI), prescribed dosage of HCQ, SLE disease activity score (SLEDAI), ever smoking, estimated glomerular filtration rate (eGFR) and concomitant prednisolone.
Results 276 SLE patients were studied (94% women; mean age 41.0±13.8 years; SLE duration 8.7±6.6 years). HCQ was primarily used for the treatment of mucocutaneous or musculoskeletal manifestations, or both, in 73%, 78% and 93% of the patients, respectively. Patients were stratified into 3 groups according to the HCQ levels: (1) Total non-compliance (<10 ng/ml); (2) Sub-therapeutic (10–500 ng/ml); and (3) Therapeutic (≥500 ng/ml). The proportion of patients with HCQ levels of <10, 10–500, ≥500 ng/ml was 11%, 77% and 12%, respectively. Patients with total non-compliance to HCQ were more likely to be in clinical and serological remission when compared to the remaining patients (42% vs 24%; p=0.04). However, no difference in the clinical manifestations could be observed between patients with total non-compliance and other patients. After excluding patients with total non-compliance to HCQ therapy, the mean and median HCQ serum concentration of the remaining 245 patients was 300±87ng/ml and 276ng/ml (interquartile range 167–401), respectively. Univariate linear regression revealed that the prescribed HCQ dosage (beta 0.47; p<0.001) and the SLEDAI score at the time of recruitment (beta 0.16; p=0.02) were the significant factors associated with the HCQ serum concentrations. Multivariate linear regression with all the factors being considered in the model showed that the prescribed HCQ dosage (beta 0.50; p<0.001) and eGFR (beta -0.14; p=0.02) were independent factors associated with HCQ serum levels. Age, sex, BMI, smoking, SLEDAI score and concomitant prednisolone were not significantly associated with the HCQ serum concentrations.
Conclusions Non-compliance and sub-therapeutic HCQ serum levels were frequent in our cohort of SLE patients, which was related to the low prescribed dosage for maintenance treatment in those with clinical and serological remission. Patients with lower eGFR or receiving higher doses of HCQ were more likely to achieve higher HCQ serum concentrations. Thus, HCQ level monitoring and dosage adjustment may be helpful in SLE patients with impaired renal function to reduce the risk of toxicities.
Disclosure of Interest None declared