(M-070) Model Informed Drug Development (MIDD) Approach to Support Elzovantinib (TPX-022) Recommended Phase 2 Dose

Jennifer Sheng, Ph.D – VP, Clinical Pharmacology and Pharmacometrics, Incyte Corporation; Hani Maroki Ghazarian, Ph.D – Associate Director, Clinical Pharmacology and Pharmacometrics, Zymeworks Inc.; Justine Lam, Ph.D – executive director, Clinical Pharmacology and Pharmacometrics, Erasca. Inc; Michael Cerra, Ph.D – executive director, Oncology Clinical Science, Bristol-Myers Squibb; Teresa Cascella, Ph.D – Clinical development lead, Bristol-Myers Squibb; Paul Statkevich, Pharm.D – Sr. Director, Clinical Pharmacology and Pharmacometrics, Bristol-Myers Squibb

Objectives: Elzovantinib (elzo) is a novel, type I tyrosine kinase inhibitor (TKI) that targets MET, SRC, and CSF1R. Genetic alterations in MET, including exon 14 skipping (Δex14) mutations and other oncogenic mutations, amplifications, and fusions are present in many tumor types. Elzovantinib has been studied in patients with advanced or metastatic solid tumors and in healthy subjects from the food effect cohort(s). To select an optimal elzo dose for the Phase 2 study, a model-informed drug development (MIDD) approach has been used for the comprehensive analysis.

Methods: PopPK model was established with 2473 quantifiable TPX-0022 concentrations from 90 subjects from the SHIELD-1 Phase 1 study (NCT03993873) and TPX-0022-03 food effect cohort(s). Exploratory exposure-response (ER) analyses for both safety and efficacy, and PK/PD simulations were conducted. The integrated MIDD approach, including nonclinical pharmacology data, clinical safety, efficacy, and PK data and analyses were applied to recommend the RP2D titration regimen.

Results: The PK of elzo was described by a two-compartment model with sequential zero, first order absorption and a first-order elimination. A significant ER relationship was identified between single dose AUC and treatment related dizziness of grade 2+. Further modeling and simulations suggested 40 mg QD titration to 40 mg BID dose titration regimen is predicted to have a lower probability (40%) of grade 2+dizziness compared to 80 mg QD (49%) or 40 mg QD to 80 mg QD dose titration (51%) regimens. Additionally, a significant ER relationship was identified between the predicted probability of ORR and PK exposures (AUC, Cmax, Cavg, Cmin) after first elzo dose. To maximize efficacy and achieve rapid, substantial, and durable responses, the dosage of 40 mg BID was predicted that 98.9% of the subjects would reach steady state Ctrough which is over the minimal effective concentration (MEC) and during the entire dosing interval, providing maximum PK/PD coverage “around the clock”.1

Conclusions: An integrated analysis of nonclinical pharmacology data, observed clinical safety, efficacy, and PK data, as well as population PK analyses, exploratory ER analyses for both safety and efficacy, and PK/PD simulations supported the RP2D regimen of 40 mg QD for 2 weeks and increase to 40 mg BID.

Citations: [1] Derendorf, Hartmut, and Bernd Meibohm. "Modeling of pharmacokinetic/pharmacodynamic (PK/PD) relationships: concepts and perspectives." Pharmaceutical research 16 (1999): 176-185.