Oral Virtual Presentation (Virtual only) ESA-SRB-ANZBMS 2021

Transcriptional mapping of the endosteal dormant tumour cell niche (#77)

Alexander P Corr 1 2 , Ryan C Chai 1 2 , Weng Hua Khoo 1 2 , James T Smith 1 2 , C. Marcelo Sergio 1 , Georgia McCaughan 2 3 , John Moore 3 , Tri Phan 2 4 , Paul A Baldock 1 2 5 , Peter I Croucher 1 2
  1. Bone Biology Group, Cancer Plasticity and Dormancy Program, Garvan Institute of Medical Research, Sydney, NSW, Australia
  2. School of Clinical Medicine, Faculty of Medicine and Health, UNSW, Sydney, NSW, Australia
  3. Department of Haematology, St Vincent's Hospital, Sydney, NSW, Australia
  4. Intravital Microscopy Group, Precision Immunology, Garvan Institute of Medical Research, Sydney, NSW, Australia
  5. School of Medicine Sydney, University of Notre Dame Australia, Sydney, NSW, Australia

Despite advances in the treatment of primary cancers skeletal metastases remain a major cause of disease relapse and mortality. Within bone, metastatic relapse results from the dissemination, long-term survival and eventual reactivation of dormant tumour cells (DTCs). DTCs from many common cancers are found associated with cells neighbouring the endosteum, through a mechanism that currently remains unclear. We hypothesised that tumour cell dormancy is controlled by specific interactions between disseminated cancer cells and cells of the endosteal bone compartment.

Using murine models of bone metastasis, we isolated both dormant and reactivated myeloma, breast and prostate cancer cells and performed single-cell sequencing to identify genes upregulated during dormancy. In parallel, we sequenced 133,942 cells from the endosteal and marrow compartments, delineating 34 endogenous bone microenvironment cell types. Using this bone cell map, we performed a ligand-receptor analysis to predict microenvironment crosstalk with DTCs, for myeloma, breast and prostate cancer.

Across all cancer models, endosteal osteoblasts were most enriched for genes predicted to interact with DTCs (adjusted p values – myeloma: 2.1x10-6, breast: 1.1x10-21, prostate: 1.4x10-24). Trajectory analysis of the osteoblast lineage revealed that DTC crosstalk enrichment was highest in Cxcl12high-Leprhigh mesenchymal stromal cells (MSCs) (myeloma: 7.6x10-9, breast: 3.0x10-11, prostate: 9.5x10-24). Comparison between cancers identified potential pan-cancer dormancy regulators, including MSC-derived Gas6 binding to TAM receptors (Tyro3/Axl/Mertk) on DTCs which, when inhibited in vivo, released myeloma cells from dormancy. A Cxcl12high-Leprhigh MSC population was identified expressing DTC crosstalk genes (including Gas6) within the bone microenvironment of myeloma patients via single-cell sequencing of trephine bone biopsies.

This study utilised transcriptional data to model intercellular communication between DTCs and the bone microenvironment. Together, these data suggest osteoblastic MSCs are a key component of the dormant tumour cell niche and identify pan-cancer therapeutic targets as candidates to prevent recurrence of skeletal metastases.