Phase 2b dose selection for the treatment of autoimmune disorders leveraging comparator data
Thomas Kerbusch* (1), Russ Wada (2), Anthe Zandvliet (1), Kuenhi Tsai (6), Jing Su (6), Joanna Zhuoying Peng (4), Yaming Hang (6), Christine Xu (3), Richard Shames (5), Ann Horowitz (3), Diane Neiman (4), Mani Lakshminarayanan (6), Usha Barai (3), Ferdous Gheyas (3), Paul Peloso (4), Devan Mehrotra (6), Nancy Zhang (2), Hanbin Li (2), Jaap Mandema (2), Gary Herman (4), Sandy Allerheiligen (6)
(1) Merck Research Laboratories, Oss, The Netherlands; (2) Quantitative Solutions, Menlo Park, CA, USA; (3) Merck Research Laboratories, Kenilworth, NJ, USA; (4) Merck Research Laboratories, Rahway, NJ, USA; (5) Merck Research Laboratories, Palo Alto, CA, USA; 6 Merck Research Laboratories, Upper Gwynedd, PA, USA.
Objectives: Selection of dose levels for Phase 2b development resulting in an optimal probability to: (1) establish the plateau of the dose-response curve, (2) demonstrate a dose-response relationship, (3) determine the lowest dose resulting in maximal efficacy.
Methods: SCH900XXX Phase 2b doses were selected with a model-based comparative efficacy analysis across 5 comparators (adalimumab, etanercept and infliximab (TNFα antagonists) and ustekinumab and briakinumab (p40 antagonists)) using published mean study-arm level data (> 10,000 patients) and in-house Phase 1b data of SCH 900XXX. Drug potencies were compound-specific, but onset of action and Emax were assumed similar across all comparators and comparator class, respectively. Clinical trial simulations were used to select a dose range bracketing a dose resulting in maximum response and ED50.
Results: Despite the availability of very limited Phase 1b data on SCH900XXX, the comparative efficacy analysis resulted in adequate estimation of the model parameters with an acceptable level of uncertainty. The dose-response model was used for comparative landscape evaluation to predict the efficacy of SCH 900XXX relative to its comparators. Establishing dose-response in Phase 2b requires a dose level between placebo and plateau. The probability that doses of 5 and/or 25 mg were (1) less effective than the maximal response and (2) more effective than placebo was demonstrated to be high. Doses of 100 and 200 mg were predicted to be at the plateau of the dose-response relationship. A 40-fold dose-range 5-200 mg brackets a dose resulting in maximum response (≥ 50 mg) and ED50 (8.4 mg, 80% CI 3.0-21.2 mg). The selected SC doses 5, 25, 100 and 200 mg SCH 900XXX were predicted to allow for (1) establishment of the plateau of the dose-response curve, (2) demonstration of a dose-response relationship, (3) model-based determination of the lowest dose resulting in maximal efficacy. The recommended Phase 2 dose levels of SCH 900XXX were 5, 25, 100 and 200 mg.
Conclusions: The value proposal of the modeling was that a more informed and robust decision could be made by leveraging comparator data to support the limited internal data. This greatly enhancing the probability of success of establishing dose-response in the Phase 2b study.