PK-biomarker-safety modelling aids choice of recommended phase II dose and schedule for AZD6738 (ATR inhibitor)

Introduction: AZD6738 is a potent inhibitor of the ATR protein kinase that is being tested in patients with solid malignancies. In Phase I dose escalation studies, AZD6738 was dosed as monotherapy (continuous schedule from 40 - 480 mg per day), or in combination (intermittent schedule from 80 - 320 mg per day) with olaparib, durvalumab, carboplatin, or radiation. In those studies, thrombocytopenia was the main dose limiting toxicity and changes in peripheral monocytes and proliferating T-cells were seen with AZD6738, with up to 80% decrease on-drug and return to baseline upon cessation of dosing. This effect on monocytes and proliferating T cells was not observed with either single agent olaparib or durvalumab. Based on these observations, we developed two PKPD models linking AZD6738 dose, AZD6738 plasma exposure, and differential blood cell counts to support dose and schedule selection. Methods and Results: The first model characterized the quantitative relationship between the plasma exposure of AZD6738 monotherapy, and the difference between monocyte decrease (on-target activity) and platelet decrease (on-target toxicity). This model was a sigmoid function and predicted biological activity starting at exposures of 2.5 ug/mL (aligned with the in vitro IC90 in LoVo cells, and obtained in ≥ 50% of patients with 80 mg per day) increasing in a concentration dependent manner to 10 ug/mL (obtained in almost all patients dosed at 480 mg per day), after which negligible monocyte response over platelet decrease was observed. Those data were consistent with the preclinical observations which showed tumour regression after at least 3-day concomitant treatment of olaparib and AZD6738 at exposure maintaining IC90[1]. The second model was a PK-safety model describing the temporal relationship between circulating platelets and exposure of AZD6738 administered in combination with olaparib dosed at 300 mg twice daily continuously. This model allowed generation of various platelet/ time profiles in the most sensitive patient group (10% of the patients) for different AZD6738 doses and schedules. The simulations indicated that, for doses considered biologically relevant by the first model, a period of 21-days free of AZD6738 is needed to give full recovery of potential thrombocytopenia in a 28-day cycle. Conclusion: Using monocyte and proliferative T-cell decreases as biological activity indicators and platelet decrease as a safety indicator, we quantified the exposure/activity/safety relationship for AZD6738. The recommended Phase II dose of AZD6738 (in combination with olaparib) was then driven by maintaining maximally active exposure consistent with manageable safety. Reference: [1]. 2017 AACR Abstract 2494. Acknowledgements: T.A. Yap (The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research, London, UK) and J.C. Soria (Institut Gustave Roussy, Paris, France) Citation Format: Alienor Berges, S. Y. Amy Cheung, Andrew J. Pierce, Emma Dean, Brunella Felicetti, Nathan Standifer, Simon Smith, James Yates, Alan Lau, Christine Stephens, Matthew Krebs, Kevin Harrington, Simon J. Hollingsworth. PK-Biomarker-Safety modelling aids choice of recommended Phase II dose and schedule for AZD6738 (ATR inhibitor) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr CT118.