Natalizumab, a human immunoglobulin monoclonal antibody that targets α4β1/α4β7 integrin, is an effective therapy approved for the treatment of multiple sclerosis (MS). The objective of this analysis was to develop a population exposure-response model utilizing gadolinium-enhancing (Gd) lesion count data from four clinical studies and annualized relapse rate (ARR) data from three clinical studies. The natalizumab exposures were derived for the individuals using a population pharmacokinetic model. A log-linear exposure effect on Gd lesion count and ARR adequately characterized the relationship between exposure and disease response. In the case of the Gd lesion count model, a bimodal model that distributed subjects into two subpopulations based on low or high baseline Gd lesion count provided a superior goodness of fit. The mean (95% confidence interval) slopes from the exposure-Gd lesion count model and exposure-ARR model are -0.0903 (-0.100, -0.081) and -0.0222 (-0.026, -0.015) (mg/L)-1, respectively. From these slopes, it can be inferred that both Gd lesion count and ARR decrease with increasing exposure to natalizumab in MS subjects. Model-based simulations demonstrated that although reductions in Gd lesion count and ARR were observed with lower doses (75, 150, or 200 mg), only the dose of 300 mg every 4 weeks (q4w) was associated with an ARR ≤0.25 and was considered clinically effective. The results from the exposure-Gd lesion count and exposure-ARR models thus support the appropriateness of the approved natalizumab dose (300 mg q4w) in MS subjects.
Keywords: Annualized relapse rate; Gadolinium-enhancing; Multiple sclerosis; Once-every-4-weeks dosing regimen; Pharmacokinetics.