Abstract 3930: AI-driven discovery and profiling of GTAEXS-617, a selective and highly potent inhibitor of CDK7

Abstract

Abstract Background: Uncontrolled cancer cell growth results from genetic lesions that disrupt mechanisms regulating cell cycle progression. Furthermore, these alterations invariably lead to dysregulated transcriptional programs that promote tumourigenesis and tumour growth. Cyclin-dependent kinase (CDK) 7 is an attractive therapeutic target in that it plays a dual role in regulating cell cycle progression and transcription. Knock-down studies and pharmacological inhibition of CDK7 have been shown to severely limit the proliferative capacity of cancer cells in vitro and in vivo. Methods: To overcome challenges associated with therapeutic inhibition of CDK7, we leveraged artificial intelligence-driven drug discovery to generate an orally bioavailable, highly potent and selective small-molecule antagonist of CDK7: GTAEXS-617 (‘617). Results: Using Exscientia’s AI drug discovery platform, we identified ‘617 through synthesis and testing of only 136 compounds. ‘617 demonstrates potent anti-proliferative activity with an average IC50 of 6.6 nM in in vitro models of high-grade serous ovarian cancer (HGSOC) and triple negative breast cancer (TNBC). Associated with its anti-proliferative activity, we demonstrate that ‘617 induces cell cycle arrest, apoptosis, and inhibits several key growth-promoting transcriptional programs. In vivo, treatment of HGSOC and TNBC xenograft tumour-bearing mice with ‘617 results in complete tumour regression, with no impact on body weight. Next, we examined the impact of ‘617 on the viability of tumour and non-transformed immune cells in primary patient ovarian cancer samples using high-content microscopy and single cell phenotypic screening. ‘617 reduced the viability of primary human tumour cells with less cytotoxicity of the immune cell compartment compared to two CDK4/6 inhibitors and another CDK7 inhibitor. Further, two response groups of patient samples began to form, indicating that further investigation may be required into target cell sensitivity that may affect clinical responses. Conclusions: In summary, Excientia’s AI platform enabled rapid and highly efficient discovery of ‘617, an orally bioavailable CDK7 inhibitor, demonstrating potent anti-tumour activity in vivo and preferential activity against primary human tumour cells vs non-transformed immune cells. These data support the advancement of ‘617 towards clinical development. Citation Format: Jérémy Besnard, James Joseph, Haiyun Bai, Ross A. Paveley, Tao Wang, Olivier R. Barbeau, Xiao-Hui Gu, Andrew S. Bell, Mate Somlyay, Christina Taubert, Christophe Boudesco, Gregory Vladimer, Fred Aswad. AI-driven discovery and profiling of GTAEXS-617, a selective and highly potent inhibitor of CDK7 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3930.