Background/Purpose: Therapeutic drug monitoring of whole blood Hydroxychloroquine (HCQ) can help identify patients at risk of flares due to underexposure (e.g. severe non-adherence, < 200 ng/ml) or alternatively those with heightened risk of retinal toxicity due to overexposure (e.g. HCQ >1733 ng/ml) (Petri M et al. Arthritis Rheumatol. 2018). Our purpose was to report and evaluate factors associated with HCQ exposure in a large population of patients with rheumatic diseases.

Methods: The cohort of patients evaluated for this analysis originated from clinical rheumatology practices in the United States. Physicians (n=526) submitted specimens collected from patients under HCQ to reference clinical laboratory accredited by the College of American Pathologists. Levels were measured using venous blood collected in EDTA containing tubes or capillary blood collected on volumetric absorptive micro-samplers. Recommendation was made to collect specimens after 6 months therapy (steady state). HCQ was measured using liquid chromatography coupled with tandem mass spectrometry and reported to clinicians within 5 days of specimen receipt. For each patient, mean HCQ blood level was calculated. Statistical analysis consisted of logistic regression analysis. Longitudinal changes were analyzed using linear mixed effect models.

Results: A total of 10523 specimens were collected from 6559 patients (mean [SD] age =52±15 years, 90% females). Average [SD] whole blood HCQ blood levels were 959±653 ng/ml (n=10523) and comparable between venous blood (946±6542 ng/ml, n=9391) and capillary blood (1063±654 ng/ml, n=1132). The distribution of mean HCQ levels per patient is highlighted in the Figure. The proportion of patients presenting with low HCQ blood levels (< 200 ng/ml) was 13.9% (912/6559), and these low levels associated with age (OR=0.71 CI95%:0.66-0.76 per 10 years) (p< 0.001), HCQ dose (OR=0.66 CI95%: 0.53-0.81 per 200 mg) (p< 0.001) and gender (OR=0.67 CI95%: 0.48-0.93 for females) (p< 0.001). HCQ levels greater than 500 ng/ml (indicative of compliance) were achieved in 74% patients (4901/6559) and 11% patients (716/6559) presented with potential overexposure to HCQ ( >1733 ng/ml). Total non-adherence (HCQ< 50 ng/ml) in the presence of reported HCQ dose ≥200mg daily on test requisitions was 4.8% (223/4601 patients). In 4127 specimens collected from 2527 patients with HCQ dosing available (median 400 mg/daily), linear mixed effect models revealed that HCQ levels associated with HCQ dosage (estimate= 228±18 ng/ml per 200 mg HCQ, p< 0.001), patient age (93±7 ng/ml per 10-year, p< 0.001) and female gender (141±40 ng/ml, p< 0.001). This effect of age and gender on HCQ exposure was also significant in the subset of patients (n=2048, 3505 measurements) presenting with mean HCQ levels greater than 500 ng/ml and thus likely compliant to therapy (age: 65±7 ng/ml; gender: 123±41 ng/ml) (p< 0.001) after adjusting for HCQ dosing (206±20 ng/ml, p< 0.001).

Conclusion: Whole blood HCQ levels are significantly associated with gender, age, and HCQ dosing and remain essential in assessing adherence.  As whole blood accurately reflects HCQ exposure (rather than plasma level), whole blood HCQ levels will have a future role in identifying overexposure as well.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/distribution-and-predictors-of-whole-blood-hydroxychloroquine-levels-in-clinical-rheumatology-practices-in-the-united-states/