Background/Purpose: Mavrilimumab is a recombinant human monoclonal antibody which neutralizes granulocyte-macrophage colony stimulating factor (GM-CSF) activity by selectively binding to the alpha subunit of its receptor (GM-CSFRα). A Phase 2b randomized, double-blind, placebo-controlled study was conducted to evaluate the efficacy and safety of mavrilimumab in subjects with at least moderately active rheumatoid arthritis (RA). Population modeling was performed to characterize the exposure-response relationship of mavrilimumab in RA patients.

Methods: In a Phase2b study, subjects with RA received subcutaneously administered placebo or mavrilimumab (30, 100 or 150 mg) once every other week for 24 weeks. The pharmacokinetic data was pooled and analyzed using a population approach. The binary ACR20 and ACR50 responses were modeled by logistic regression. A dropout hazard function was introduced to describe the voluntary patient withdrawals during the study period. Stochastic simulations based on the final PK-ACR20(50)-Dropout model were subsequently conducted for model evaluations and exposure-response relationship assessment.   

Results: A mechanistic model incorporating the subcutaneous absorption, distribution, and parallel elimination pathways of mavrilimumab by the reticuloendothelium system and GM-CSFRα mediated internalization and intracellular degradation adequately described the observed PK profiles of mavrilimumab. The placebo effect and mavrilimumab treatment effect were integrated in the logit (log odds ratio) of ACR20 or ACR50. The treatment effect was described by an Emax model for ACR20, and by a linear relationship with mavrilimumab concentration for ACR50. The dropout hazard increased with time on study, and was significantly higher in ACR20 nonresponders than in responders. The protocol-permitted rollover of patients with inadequate ACR20 responses to an open-label extension study at any time after Week 12 substantially increased the dropout risk for nonresponders in placebo and two lower dose groups. There was no change in the dropout pattern for subjects receiving the 150 mg dose.

Conclusion: The pharmacokinetics of mavrilimumab, ACR20 and ACR50 responses in patients with moderate to severe RA were well described by a model incorporating mechanistic drug disposition, logistic regression for ACR20 or ACR50, and informative patient dropout. The population model facilitated the interpretation of Phase 2b outcome and the appropriate design of late-stage clinical trials for mavrilimumab.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/exposure-response-analysis-for-mavrilimumab-phase2b-study-in-ra-patients-with-informative-dropout/