Assessment of hydroxychloroquine blood levels in Sjögren's disease patients: drug adherence and clinical associations.

Hydroxychloroquine (HCQ) has been used to treat Sjögren's disease (SjD) patients. However, there are no studies evaluating drug adherence through HCQ blood levels, pharmacy refill (PR) and medication adherence questionnaires. The relationship of HCQ blood levels with glandular/extraglandular disease parameters was also poorly assessed. This cross-sectional observational study included 74 adult SjD patients, who were receiving a stable HCQ dose (4-5.5 mg/kg/day, actual weight) for at least 3 months before study inclusion. HCQ blood levels were quantified by high-performance liquid chromatography coupled to mass spectrometry. Adherence was assessed by PR and Medida de Adesão aos Tratamentos (MAT) questionnaire. The following parameters were evaluated: Xerostomia Inventory, Ocular Surface Disease Index, EULAR (European League Against Rheumatism) Sjögren's Syndrome Disease Activity Index, EULAR Sjögren's Syndrome Patient Reported Index, Schirmer's I test and non-stimulated/stimulated salivary flow rates. HCQ blood levels were 775.3(25.0-2,568.6)ng/mL. Eleven patients (14.9%) had HCQ blood levels < 200ng/mL (non-adherent group); 11(14.9%), 200-499ng/mL (sub-therapeutic levels group); and 52(70.2%), ≥ 500ng/mL (adherent group). PR classified incorrectly all non-adherent/sub-therapeutic patients and 2/52(3.9%) adherent patients. Using MAT, the overall misclassification was 24/52(46.2%) in the adherent group, and were correctly identified 9/11(81.8%) patients in non-adherent and 7/11(63.6%) in sub-therapeutic groups. MAT sensitivity and specificity to identify non-adherent/sub-therapeutic patients were 72.7% and 53.9%, respectively. The three groups were comparable regarding glandular/extraglandular disease parameters (p > 0.05). The assessment of HCQ blood levels is a promising tool for evaluating drug adherence in SjD. This is particularly crucial as one-third of patients exhibited non-adherence/sub-therapeutic levels, and neither PR nor MAT reliably identified these patients.