Model-informed prediction of human pharmacokinetics and first-in-human dose selection of J2H-1702, a novel 11(3-hydroxysteroid dehydrogenase type 1 inhibitor for metabolic-associated steatohepatitis

Abstract

Metabolic dysfunction-associated steatohepatitis (MASH) is a progressive liver disease with limited therapeutic options. J2H-1702 is a novel and selective inhibitor of 11(3-hydroxysteroid dehydrogenase type 1 (11(3-HSD1) that has demonstrated pharmacological activity in preclinical models. This study aimed to predict the human pharmacokinetics (PK) of J2H-1702 and support rational first-in-human (FIH) dose selection using a modelinformed approach. Plasma PK data following intravenous and oral administration were obtained from Sprague-Dawley rats and Beagle dogs and analyzed using non-compartmental analysis and nonlinear mixed-effects modeling. Less-than-dose-proportional increases in systemic exposure were observed following oral administration in both species. This nonlinear exposure behavior was explicitly characterized using a nonlinear bioavailability model and incorporated into an interspecies pharmacokinetic framework. Interspecies extrapolation was performed using multiple allometric scaling approaches, including body weight-based, maximum lifespan potential-adjusted, brain weight-adjusted, and multiple-exponent models. Among these, the brain weight-adjusted model showed the best predictive performance. The final model was used to simulate human PK profiles across oral dose levels. Simulations predicted that doses >= 3 mg would maintain plasma concentrations above the in vitro half-maximal inhibitory concentration for 11(3-HSD1 inhibition for approximately 48 hours. In parallel, no-observed-adverse-effect level-based calculations supported a substantially higher maximum recommended starting dose, indicating a wide safety margin. Integration of exposure-based pharmacodynamic criteria with toxicology-based constraints supported selection of 3 mg as a conservative and pharmacologically relevant FIH starting dose. These findings demonstrate the utility of model-informed PK analysis for translating preclinical data into rational FIH dose selection in early clinical development.