Background/Purpose: To study the relationship between the serum hydroxychloroquine (HCQ) concentration and flares of systemic lupus erythematosus (SLE) in a longitudinal cohort of Chinese patients.

Methods:

Patients who fulfilled ≥4 of the ACR criteria for SLE and had been treated with HCQ for >6 months were recruited in November 2011.  Patients had been prescribed HCQ at dosages of 200 mg/300 mg/400 mg/day at the discretion of physicians.  Blood was assayed for the levels of HCQ by an in-house technique using the tandem mass spectrometry (SPE-MS/MS).  Patients were stratified according to the HCQ levels; group 1 (<10 ng/ml, total non-compliance); group 2 (10-500 ng/ml, sub-therapeutic); and group 3 (≥500 ng/ml, therapeutic); and were followed longitudinally for serial assessment of disease activity (SELENA-SLEDAI) and the occurrence of mild/moderate or severe SLE flares (SELENA flare instrument).  Comparison was made among these groups in the baseline and mean summated SLEDAI over time (AUC), and the annual incidence of mild/moderate and severe flares by the non-parametric Kruskal Wallis test.  Bivariate correlation of two covariates was performed by Spearman’s rank correlation.

Results:

276 SLE patients were studied (94% women; age 41.0±13.8 years; SLE duration 8.7±6.6 years).  HCQ was used for the treatment of mucocutaneous or musculoskeletal manifestations, or both, in 93% of patients.  The proportion of patients with HCQ levels of <10, 10-500, ≥500 ng/ml was 11%, 77% and 12%, respectively.  The HCQ levels correlated significantly with the prescribed daily dosage (Rho 0.44; p<0.001), which was also correlated with the baseline SLEDAI score (Rho 0.30; p<0.001), indicating higher doses were used for more active SLE manifestations. After a follow-up of 32.5±5.5 months, 153 mild flares and 91 severe flares developed in our patients.  The mean summated SLEDAI score over time was: 2.3±2.2 (group 1); 2.6±1.9 (group 2); and 3.5±2.5 (group 3), respectively (p=0.06).  The annual incidence of mild/moderate and severe flares was: 0.21±0.32 and 0.14±0.42 (group 1); 0.19±0.33 and 0.12±0.29 (group 2); and 0.32±0.57 and 0.19±0.57 (group 3), respectively (p=NS in all).  In patients with clinical and serological remission (SLEDAI=0) (N=73), therapeutic HCQ levels (>500ng/ml) were associated with lower summated mean SLEDAI over time (0.07±0.15) and no occurrence of disease flares (mild/moderate/severe) as compared to the other groups, although the differences were not statistically significant.

Conclusion:

“Sub-therapeutic” serum HCQ levels were frequent in our SLE patients, which was probably contributed by the low dosage prescribed for maintenance treatment in many patients with quiescent disease, as well as non-compliance. Patients with higher HCQ levels tended to have a higher disease activity over time and more frequent flares, indicating they had more refractory manifestations that required higher doses of HCQ for treatment.  In patients with clinical and serological remission, therapeutic levels of HCQ were associated with lower disease activity over time and incidence of disease flares.  HCQ drug level monitoring may play a role in identifying non-compliance to enhance the long-term efficacy of HCQ.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/hydroxychloroquine-serum-levels-and-flares-of-systemic-lupus-erythematosus-a-longitudinal-cohort-analysis/