Population modelling of functional disability in early rheumatoid arthritis treated according to a treat-to-target strategy

Aims: The treat-to-target approach in the management of rheumatoid arthritis (RA) involves frequent assessments of disease activity (measured by the 28-joint disease activity score [DAS28]) to guide drug dose and regimen adjustments until remission has been achieved. Previous work has shown that DAS28 progression with combination disease modifying anti-rheumatic drug (DMARD) therapy for RA follows a predictable population time-course, but between subject variability and residual error is large (1). Such that, the aim was to quantify and evaluate inter-individual differences in functional disability (as assessed by a modified version of the Health Assessment Questionnaire – Disability Index [mHAQ]), rather than disease activity, over 2-years for an early RA cohort treated with combination DMARD therapy according to a treat-to-target protocol using population (non-linear mixed effect) modelling.

Methods: Structural models comprised parameters that described the average maximum achievable change in mHAQ over time, the rate of change in mHAQ over time and the lag in onset to mHAQ change were tested using NONMEM® with respect to their ability to describe the average mHAQ trajectory since the initiation of therapy. Covariates and random effects were added to structural model parameters to quantify how, and by how much individuals in the study population varied from this average.

Results: The average mHAQ in the population (n = 241) had a baseline of 0.535 units and decreased by 0.450 units after 2-years of combination DMARD therapy, and 50% of the overall decrease in mHAQ was achieved by 14.4 weeks. The model sufficiently described both the overall population mHAQ trajectory and the mHAQ trajectory of each study participant. Higher baseline pain and baseline DAS28 were predictors of higher baseline mHAQ, and stringent compliance to the treat-to-target protocol was associated with improved 2-year mHAQ outcomes.

Conclusion: Population modelling accurately described and explained inter-individual differences in mHAQ within our cohort of early RA patients.

1. Wojciechowski J, Wiese MD, Proudman SM, Foster DJ, Upton RN. A population model of early rheumatoid arthritis disease activity during treatment with methotrexate, sulfasalazine and hydroxychloroquine. British journal of clinical pharmacology. 2015;79(5):777-88.