Abstract 3387: Mechanisms driving clinical bone marrow-resident cancer cell quiescence

Abstract

Abstract Despite widespread knowledge that bone marrow-resident breast cancer cells (BM-R BCs) affect disease progression and ultimately cause death, mechanisms regulating patient-derived BM-R BC quiescence remain poorly defined. This is caused by lack of biomarkers that can isolate quiescent but viable BM-R BCs devoid of cytokeratin and/or other epithelial markers. We developed and validated an experimental strategy to capture, isolate and characterize identify and isolate solitary BC cells from blood and BM of xenograft mice using a combination of human and BC-specific markers. These procedures allowed us to interrogate not only cytokeratin-positive but also unidentified cytokeratin-negative BC cell subsets. We hypothesized that subsets of patient-isolated circulating tumor cells (CTCs) will re-colonize various organs; and that underlying CTC signaling mechanisms will reflect respective transcriptomic signatures upon xeno-transplantation. Accordingly, we established an in vivo model of “latent” metastasis in CTC-derived xenografts (CDXs) by three sequential steps: a) isolation of CTC-enriched populations from metastatic BC patients, followed by their implantation in NSG mice; b) in vivo depletion of “non-tumor” human cells over an 8-month period; and c) capture and characterization of viable BC cells from blood and BM of CDX mice. These steps were implemented on patient samples from all primary BC subtypes along with validation of ex vivo CTC and BM-R BC gene signatures by whole genome transcriptomic arrays. First, three-way transcriptomic comparison between ex vivo BM-R BCs and CTCs vs de novo CTCs (obtained from GSE99394) revealed that ex vivo BMRBCs and CTCs clustered together and were largely similar (279 differentially regulated genes), implying that either the majority of ex vivo CTCs were shed from BM or that CTC transcriptomic signatures shed from other organs were similar to BM-shed CTCs. Second, comparative analyses of ex vivo BM-R BC vs de novo CTC gene signatures identified heightened mTORC2 vis-à-vis attenuated mTORC1 signaling as the most significant parameter of human BM-R BCs. Third, augmented levels of mTORC2-downstream targets were detected in quiescent (Ki67-/RBL2+) cells of paired metastatic vs primary BC tissues. Fourth, IHC analyses of CTC xenograft tissues revealed solitary BM and tissue-resident BC cells possessing high mTORC2. Lastly, shRNA knockdown of Rictor - the essential component of the mTORC2 complex - stimulated cell cycle progression and proliferation in vitro while depleting the BM-R BC population in vivo. Collectively, these findings suggest that the balance between mTORC2 vs mTORC1 interplays regulate the primordial stages of CTC quiescence. Deciphering mTORC2 signaling in CTC quiescence will proffer novel biomarkers for quiescent CTC detection and innovative therapeutic interventions in BC patients with undetectable metastasis. Citation Format: Debasish Boral, Monika Vishnoi, Haowen Liu, DARIO MARCHETTI. Mechanisms driving clinical bone marrow-resident cancer cell quiescence [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3387.