Mirikizumab Exposure–Response Relationships in Patients with Moderately‐to‐Severely Active Ulcerative Colitis in Randomized Phase II and III Studies

Abstract

Mirikizumab is a humanized anti‐interleukin‐23p19 monoclonal antibody being developed for ulcerative colitis (UC) and Crohn's disease. We characterized the relationship of mirikizumab systemic exposure with efficacy and safety end points in patients with UC using phase II (NCT02589665) and III (NCT03518086, NCT03524092) trial data. Exposure–response models were developed for clinical remission, clinical response, endoscopic remission, and change in modified Mayo score following induction (50–1,000 mg i.v. every 4 weeks) and maintenance (200 mg s.c. every 4 or 12 weeks) treatment. These models evaluated observed and pharmacokinetic model‐predicted mirikizumab exposures as the exposure measure. Key safety event rates were compared across mirikizumab exposure quartiles in the phase III trial. Mirikizumab efficacy in patients with UC showed an apparent positive association with systemic exposure following both induction and maintenance. However, further analysis found this relationship to be overstated by the presence of confounding factors that were not among the tested patient covariates. While prior biologic experience and baseline disease severity showed statistically significant influences on estimated placebo effect, no patient factors affected the mirikizumab effect parameters in any of the phase III exposure–response models. There was no apparent mirikizumab concentration relationship with any adverse event of special interest. When the phase II and III data and confounding are considered together, efficacy was unlikely to be strongly affected by variation in exposures across individual patients at the phase III dose. Together with the previously demonstrated mirikizumab exposure insensitivity to patient factors, these findings indicate that mirikizumab dose adjustment to patient characteristics is not required.