Desethylhydroxychloroquine and CYP2D6 polymorphisms are associated with low disease activity in patients with systemic lupus erythematosus

Aims Hydroxychloroquine (HCQ) is standard therapy for systemic lupus erythematosus (SLE), but the clinical relevance of its active metabolite, desethylhydroxychloroquine (DHCQ), remains unclear. We examined whether DHCQ exposure is more closely related to disease control than HCQ exposure in Japanese patients with SLE. Methods Patients receiving HCQ for at least 6 months were enrolled. Whole‐blood and plasma HCQ and DHCQ, and plasma 4β‐hydroxycholesterol (a CYP3A activity marker), were measured by liquid chromatography–tandem mass spectrometry (LC–MS/MS), and CYP2D6 (*2, *5, *10) and CYP3A5 (*3) were genotyped. The primary outcome was low disease activity (LDA), defined as SLEDAI‐2K ≤ 4 with oral corticosteroids ≤5 mg/day. Exposures were analysed as continuous variables (log‐transformed, per 1 SD) in multivariable logistic regression; cutoff‐based analyses were exploratory, with optimism assessed by bootstrap resampling. Results Among 102 patients, 36 (35.3%) attained LDA. The 23 (22.5%) CYP2D6 intermediate metabolisers had lower DHCQ/HCQ ratios and weight‐adjusted DHCQ concentration‐to‐dose ratios than normal metabolisers in both matrices. HCQ concentrations and CYP genotypes were not associated with LDA. Adjusted odds ratios per 1 SD increase were 1.80 (95% CI 1.10–2.95) for plasma DHCQ, 2.10 (1.23–3.59) for the plasma DHCQ/HCQ ratio, 1.73 (1.04–2.88) for whole‐blood DHCQ and 2.36 (1.32–4.23) for the whole‐blood DHCQ/HCQ ratio. Discrimination was modest (AUC 0.62–0.67; lower bootstrap limits 0.50–0.56). Conclusions CYP2D6 genotype was the principal determinant of DHCQ exposure. Higher DHCQ concentrations and DHCQ/HCQ ratios, but not HCQ concentrations or genotypes, were associated with LDA, supporting DHCQ as a candidate exposure marker requiring prospective validation.