Abstract 297: AZD4573: Mechanistic PKPD model linking CDK9 inhibition to Mcl1 depletion and induction of apoptosis in preclinical AML model

Abstract

Abstract Cyclin-dependent kinase 9 (CDK9) regulates elongation of transcription through phosphorylation of RNA polymerase II (pSer2-RNAPII), and its short-term inhibition results in the selective downregulation of genes with short-lived transcripts and labile proteins - including the anti-apoptotic protein Mcl1. AZD4573 is a selective inhibitor of CDK9 with short pharmacokinetic (PK) half-life. Intermittent dosing of AZD4573 in mouse MV411 (AML cell line) xenograft models results in progressive reduction in tumor volume with the mechanism of action believed to be via induction of apoptosis following depletion of Mcl1. The aim of this work was to derive a quantitative understanding of the relationships between extent and duration of CDK9 inhibition, depletion of Mcl1 and rate of induction of apoptosis in MV411 tumor cells. A mechanistic model has been established that quantitatively and dynamically connects AZD4573 plasma and tumor PK to the rate and extent of modulation of pSer2-RNAPII and Mcl1 in the tumor and rate of induction of cell death (as measured by reduction in tumor volume). Tumor pSer2-RNAPII and Mcl1 pharmacodynamics were modeled using a series of linked indirect response models. Production rate of pSer2-RNAPII was modeled as being directly inhibited by AZD4573 concentration in the tumor. Production rate of Mcl1 was linked to pSer2-RNAPII via a series of transit compartments to capture the transcription/translation driven delay in onset of response. Induction of intrinsic apoptosis in the MV411 tumor cells was modelled as being inhibited by Mcl1. Tumor pSer2-RNAPII exhibited a rapid, dose-dependent decrease following IP dosing of AZD4573 in mice. The free concentration of AZD4573 that resulted in half-maximal inhibition of pSer2-RNAPII production rate was estimated to be in the range 11-21 nM. Following a brief delay, tumor Mcl1 also exhibited a relatively rapid decrease that was proportional to the pSer2-RNAPII response. Mcl1 protein half-life was estimated to be 0.3 hr. Rate of induction of apoptosis could be decribed as a saturable first-order process (Kmax ~ 0.2 hr-1) and appeared to exhibit a steep response to the depletion of Mcl1, with reduction of Mcl1 to 25% (of the baseline value) being estimated to result in half-maximal rate of induction of apoptosis in the MV411 cells. The described MV411 PKPD/efficacy model has been assumed to be representative of AML in human patients and was used to derive preliminary predictions of clinical efficacy at a range of possible IV dosing regimens. Citation Format: Douglas Ferguson, Theresa Proia, Justin Cidado, Scott Boiko, Maryann San Martin, Steven Criscione, Wenlin Shao, Lisa Drew. AZD4573: Mechanistic PKPD model linking CDK9 inhibition to Mcl1 depletion and induction of apoptosis in preclinical AML model [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 297.