(T-053) Modeling and inter-species scaling of plasma DHO as a biomarker to inform first-in-human dose selection for a novel DHODH inhibitor

Joo Young Na, Ph.D. – postdoc, The Ohio State University; Damien Cronier, Ph.D. – NA, Bexon Clinical Consulting LLC; Sebastian Biglione, PharmD, PhD, MLA, CCRP – Director of Clinical and Regulatory Strategy, Drug Development Institute |OSUCCC; Chad Bennett, Ph.D. – Senior Director of Chemistry, Drug Development Institute |OSUCCC; Chris Coss, Ph.D. – Associate professor, The Ohio State University; Mitch Phelps, Ph.D. – Professor, The Ohio State University

Objectives: In rapidly proliferating cancer cells, the de novo pyrimidine synthesis pathway is highly activated and enhances the tumor’s supply of pyrimidine nucleotides. HOSU-53 is under development as an orally bioavailable, small-molecule inhibitor of dihydroorotate dehydrogenase (DHODH), a mitochondrial enzyme that catalyzes the rate-limiting step of de novo pyrimidine nucleotide biosynthesis, conversion of dihydroorotate (DHO) to orotate1. Biological testing verified the efficacy of HOSU-53 in acute myeloid leukemia, multiple myeloma, small-cell lung cancer, melanoma, and other cancer cell lines. Our goals were to develop PK/PD models to scale across species and utilize plasma DHO exposure as a biomarker for both efficacy and on-target toxicity to target a safe, yet potentially effective starting dose in a first-in-human (FIH) trial.

Methods: Plasma HOSU-53 and DHO concentration vs. time data were available from Good Laboratory Practice (GLP) and non-GLP pharmacokinetics (PK), toxicity, and efficacy studies in mice, rats, and beagle dogs. One-, two-, and three-compartment models for HOSU-53 PK were investigated and then linked to various PD models for DHO accumulation via DHODH inhibition, and a comprehensive dataset combining pharmacokinetics/pharmacodynamics (PK/PD) data across the three species was used for fitting and estimation of random and fixed effects, including allometric scale factors. Simulations of various dose regimens in humans have been made based on the final PK/PD model selected.

Results: A 2-compartment model with first-order absorption and linear elimination with allometric factors adequately characterized the observed HOSU-53 PK profiles across species. DHO response to HOSU-53 was best characterized using a turnover model to capture the delay between plasma drug concentration and biomarker response. Compared with a previously developed physiologically based PK (PBPK) model, the PK/PD model is in good agreement in predicting human PK profiles. However, the HOSU-53 exposure-DHO response relationships did not scale well across species, suggesting yet unidentified mechanistic components may be required within the model.

Conclusions: A translational modeling and simulation approach was used to identify a target dose range for the FIH study of a novel DHODH inhibitor using data from preclinical studies. Human simulations suggest an oral FIH starting dose of 5 mg QD will be safe and near the low end of the predicted efficacy range based on preclinical DHO exposure data. The discrepancy in PD predictions between species indicates better understanding of the pharmacological mechanisms of HOSU-53, DHODH, de novo pyrimidine synthesis, and DHO pharmacodynamics across species is needed to aid the development of HOSU-53 in humans.

Citations: [1] Elgamal, O. A.; Fobare, S.; Vibhute, S.; Mehmood, A.; Vroom, D. C.; Johnson, M. L.; Stearns, B.; Lerma, J. R.; Truxall, J.; Stahl, E.; Carmichael, B.; Orwick, S. J.; Mims, A. S.; Curran, E.; Santhanam, R.; Tridandapani, S.; Phelps, M. A.; Xie, Z.; Coss, C. C.; Baker, S. D.; Patrick, J.; Ezzell, J. K.; Rai, J.; Pan, J.; Rai, S. N.; Stillwell, C.; Wunderlich, M.; Abdulrahim, M.; Goodwin, T. E.; Hilinski, G.; Bennett, C. E.; Hertlein, E.; Byrd, J. C. Pyrimidine Depletion Enhances Targeted and Immune Therapy Combinations in Acute Myeloid Leukemia. JCI Insight 2024, 9 (8). https://doi.org/10.1172/jci.insight.173646.