Oral Presentation 30th Lorne Cancer Conference 2018

Intravital imaging of the tumour bone niche reveals novel cell behaviour: implications for microenvironmental regulation of tumour cell dormancy. (#22)

Michelle M McDonald 1 , Niall Byrne 1 , Pei Ying Ng 2 , Danyal Butt 1 , Karrnan Pathmanandavel 1 , Maté Biro 3 , Rachael Terry 1 , Weng Hua Khoo 1 , Sindhu T Mohanty 1 , Marija K Simic 1 , Ryan Chai 1 , Julian Quinn 1 , Jessica A Pettitt 1 , David Ani-Hanna 1 , Rohit Jain 4 , Wolfgang Weninger 4 , Paul Baldock 1 , Michael Rogers 1 , Robert Brink 1 , Nathan Pavlos 5 , Peter Croucher 1 , Tri Phan 1
  1. Garvan Institute of Medical Research, Darlinghurst, NSW, Australia
  2. Harvard School of Dental Medicine, Harvard , Boston, MA, USA
  3. St Vincents School Faculty of Medicine, UNSW, Sydney, NSW, Australia
  4. Immune Imaging Program, Centenary Institute, Sydney, NSW, Australia
  5. Cellular Orthopaedics, University of Western Australia, Perth, WA, Australia

 

Despite improved treatments, metastatic cancers are largely incurable once established in the skeleton. This is due to high rates of disease relapse which is likely driven by the reactivation of treatment resistant dormant tumour cells. Dormant cells survive in a specialised niche in bone, however, little is known about how this niche exerts its cell-extrinsic control over tumour cells. Using a novel intravital imaging approach we directly imaged the niche on the endocortical surface of the intact tibia in live mice, allowing visualisation of the dynamic behaviour of a key niche component, the bone resorbing osteoclast, in vivo.

We showed that, in the steady-state, multi-nucleated LysozymeM+Blimp-1+Osteosense+, Cathepsin K+, osteoclasts exhibit a stellate structure forming syncytial networks on the bone surface. Treatment with soluble RANKL induced rapid retraction of cellular processes, causing loss of syncytial networks. Subsequently, RANKL activated osteoclasts were visualised undergoing cell fusion and, unexpectedly, cell fission. Daughter cells were observed to re-fuse with nearby osteoclasts, in a novel process we have termed osteoclast recycling. Treatment with osteoprotegerin-Fc fusion protein (OPG-Fc), which inhibits RANKL, led to the accumulation of small round LysM+Blimp-1+ cells, suggesting it inhibited re-fusion. Following OPG-Fc treatment withdrawal recycling osteoclasts re-fused to form networks of active enlarged osteoclasts, resembling sRANKL treatment. Interestingly, sRANKL stimulation of osteoclast function re-activated dormant tumour cells in models of both myeloma and prostate cancer bone metastases.

These data demonstrate that intravital imaging of the endosteal bone surface reveals novel osteoclast dynamics, with our discovery of osteoclast recycling providing a new paradigm for understanding osteoclast fate. Our findings also explain the paradoxical acceleration of bone loss observed upon discontinuation of the anti-RANKL therapeutic Denosumab in patients with osteoporosis. Implications therefore arise for patients with metastatic bone disease treated short term with Denosmuab, whereby upon discontinuation, osteoclast driven reactivation of dormant tumour cells may drive disease relapse. This highlights the potential of targeting the tumour microenvironment to control cancer growth.