An exposure-response model characterizing the time-course of ACR20, ACR50, and ACR70 in rheumatoid arthritis (RA) patients on baricitinib

BACKGROUND: Baricitinib (LY3009104) is a Janus kinase (JAK) inhibitor in development for the treatment of rheumatoid arthritis (RA) The objective of this study was to characterize the relationship between dose, exposure and the time‐course of 20, 50, and 70% improvement in the American College of Rheumatology criteria (ACR20, ACR50, and ACR70, respectively) related to baricitinib treatment in a Phase 2b dose‐ranging study METHODS: An exposure‐response model was developed via nonlinear mixed‐effects modeling in NONMEM (Version VII) to describe dose‐concentration‐clinical ACR response rate relationships following oral administration of baricitinib and placebo. Data from 299 RA patients receiving placebo, 1, 2, 4, 8 mg QD up to 24 weeks with rerandomization of the placebo and 1 mg groups to 2 mg BID or 4mg QD from Week 12 to 24, and corresponding ACR 20/50/70 response rates, were analyzed RESULTS: A previously published latent variable model was utilized to characterize the time‐course of exposure‐response relationships It has the ability to simultaneously describe baricitinib concentrations, a continuous latent‐dependent variable (ACR latent [ACRL]) and the binary outcome measures of efficacy ACR20, ACR50, and ACR70. ACRL is an assumed variable representing RA disease progression in signs and symptoms. Based on its known mechanism of action, baricitinib and placebo were assumed to exhibit an indirect inhibitory effect on disease activity or the ACRL variable CONCLUSION: The model adequately described the exposureresponse relationships for ACR20, 50 and 70 and fitted well the timecourse of ACR response rates within the range of exposures studied After 12 weeks of treatment, a 4 mg QD dose would be adequate in achieving close to maximal response.