DailyDoctor Archive Specialties Get app
Back to the 26 October 2026 edition

Research · 02 of 06

In lupus, the hydroxychloroquine metabolite — not the parent drug — tracked with disease control

Hydroxychloroquine's metabolite, governed by CYP2D6, tracked with lupus control; treat it as a research marker, not a test to order.

Design
Cross-sectional pharmacokinetic and pharmacogenetic study
Population
102 Japanese patients with lupus on hydroxychloroquine for at least six months
Primary outcome
Low disease activity (SLEDAI-2K 4 or less, prednisolone 5 mg/day or less)
Effect
Whole-blood metabolite/parent ratio aOR 2.36 (1.32–4.23) per SD; parent drug not associated

Japanese investigators measured hydroxychloroquine and its active metabolite desethylhydroxychloroquine in 102 patients with lupus who had taken the drug for at least six months, and genotyped CYP2D6 and CYP3A5.

CYP2D6 genotype was the main determinant of metabolite exposure: the 23 intermediate metabolisers had lower metabolite-to-parent ratios. Parent drug levels and genotype were not associated with low disease activity. Higher metabolite levels and metabolite-to-parent ratios were, with odds ratios of 1.73 to 2.36 per standard deviation. Discrimination was modest (area under the curve 0.62–0.67).

This is a cross-sectional association in one population, and the metabolite is not routinely measured. It is a candidate marker for future dosing work, not a reason to change practice now. It does raise the possibility that some apparent non-response reflects metabolism rather than adherence.

  • Hydroxychloroquine blood levels remain useful mainly for checking adherence.
  • The active metabolite, shaped by CYP2D6 genotype, may matter more for response; this is not yet actionable.
  • Do not genotype CYP2D6 routinely for hydroxychloroquine on this evidence.
  • Whether the finding holds outside this Japanese cohort, including in Indian patients, is unknown.

Why it matters

Some poor responders to hydroxychloroquine may be metabolisers, not non-adherers.

Don't overread it

A cross-sectional association with modest discrimination does not support metabolite-guided dosing yet.

The statistics, in plain English

An odds ratio of 2.36 per standard deviation means each one-SD higher metabolite ratio was associated with roughly twice the odds of low disease activity. An area under the curve of 0.62 to 0.67 is only a little better than chance (0.5) at sorting individual patients.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

QR code to install Daily Doctor
Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.

interactionspkpdtdmpharmacogenomicsdrugsafetysignal

Tomorrow morning, before your first patient

One edition a day for clinical pharmacology, written by the desk, every claim tied to its paper. Six minutes.

Get the app — free
Daily Doctor All 27 specialties, every morning. Free.
Get the app