Background and objectives: Rituximab is approved in patients with anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) and leads to a decrease of ANCA levels. The objectives of this study were to investigate the non-linear pharmacokinetics of rituximab and the relationship between its concentrations and ANCA levels in AAV patients.
Methods: Ninety-two AAV patients from the RAVE (Rituximab in ANCA-Associated Vasculitis) trial were assessed. Both ANCA anti-myeloperoxidase (MPO-ANCA) and anti-proteinase 3 (PR3-ANCA) levels were used as biomarkers. The pharmacokinetics of rituximab were described using a semi-mechanistic two-compartment model that included a latent target antigen turnover and allowed the estimation of specific target-mediated elimination in addition to its non-specific elimination of rituximab. The effect of rituximab on the ANCA level was described using a semi-mechanistic compartment model with a negative feedback (Friberg) model with no transit compartment. A population modeling approach was used.
Results: Our pharmacokinetic and pharmacokinetic-pharmacodynamic (PK-PD) models satisfactorily described both concentration-time and concentration-effect relationship data. The mean (inter-individual standard deviation) estimated non-specific clearance was 0.15 L/day (0.30%) and the target-mediated elimination rate constant was 2.4 × 10-5 nmol/day. The elimination half-lives for MPO-ANCA and PR3-ANCA were 24 and 18 days, respectively.
Conclusions: A non-linear target-mediated elimination of rituximab was detected in AAV patients. Our PK-PD model allowed quantification of the association between rituximab concentrations and ANCA levels. This decrease was deep but delayed, and more sustained in patients with MPO-ANCA than in those with PR3-ANCA. Our results suggest that repeating courses of rituximab might improve the clinical response to rituximab.