Azd4635 A(2a) Receptor Occupancy In Cynomolgus Monkey Using Pet And Its Application To An Oncology Clinical Development Program

Introduction AZD4635 is an A 2A receptor antagonist currently being tested as monotherapy and in combination with durvalumab in patients with advanced solid cancers. High adenosine levels found in tumors are immune suppressive and therefore AZD4635 could potentiate immune checkpoint inhibitors such as durvalumab (anti-PDL1). Predictions of A 2A receptor engagement in patients at different doses and at different time points may enable better interpretation of clinical biomarker data measuring effects on immune modulation. A quantitative assessment of the receptor occupancy in the brain of non-human primates was conducted for AZD4635 with PET imaging and the resulting PK/PD model was applied to predict occupancy in humans in tumors. Methods PET measurements of A 2A R occupancy in brain was performed using the radioligand [ 18 F]MNI-444 in three anesthetized cynomologus monkeys. PET data acquisition was performed for 120 min following IV-administration of [ 18 F]MNI-444 at baseline and following pretreatment of AZD4635. Sampling for AZD4635 plasma exposure determination was performed. As part of PK/PD analysis of the occupancy data, a novel modification of the non Invasive-LOGAN data analysis of the PET data was performed to obtain a time course of occupancy for each dose. A bio-phase PK/PD mathematical model was then used to describe the relationship of occupancy with circulating concentrations of AZD4635. In parallel, a PK model for AZD4635 in humans was developed using data from cohort 1 (Clinical trial NCT02740985) after 125 mg and used for PK predictions for alternative doses of AZD4635 in the clinic. Results A clear Exposure-Effect relationship was observed for AZD4635-driven A 2A R occupancy in cyno brain when dosed 30 min prior to PET measurement. The PK/PD analysis of cyno PET-determined receptor occupancy provided an Occ 50 that is in line with the in vitro potency for the compound under physiological concentrations of adenosine in the brain. The resulting PK/PD model has then been applied to predict the level of occupancy in human tumours at other clinically relevant doses. Different simulations were done varying the amount of endogenous adenosine levels. Conclusions AZD4635 was shown to occupy A 2A R in cyno brain in an exposure dependent manner. The resulting PK/PD model built using this dataset was used to run simulations of expected tumor receptor occupancy in man and aid clinical dose selection for AZD4635. Sensitivity analysis has shown that prediction of human occupancy in the tumour is highly dependent on adenosine concentrations in the tumour. Simulations with tumor adenosine concentrations of 1 μM indicate that AZD4635 is predicted to provide ~90% receptor occupancy over the whole dosing interval at a clinically relevant dose. Citation Format: Peter Johnstrom, Pablo Morentin Gutierrez, Katarina Varnas, Magnus Schou, Akihiro Takano, Lorraine Jones, Ganesh Mugundu, Patricia McCoon, Paul Lyne, Jeffrey Infante, Gerald Falchook, Manish Patel, Janet Karlix, Melinda Merchant, James Clarke, Alan Cross, Nicholas Seneca, Lars Farde, Miles Congreve, Jon S. Mason, Fiona H. Marshall. AZD4635 A 2A receptor occupancy in cynomolgus monkey using PET and its application to an oncology clinical development program [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2641. doi:10.1158/1538-7445.AM2017-2641