Tumor Metastasis: Bone Marrow Progenitors Sow the Seeds for Spread

The cellular and molecular mechanisms by which tumor cells metastasize to distant sites remain largely unknown. One line of thought is that inherent molecular differences in the tumor cells and the influence of the surrounding stromal cells determine sites of metastasis. In this paper, Kaplan and colleagues report an alternate explanation and suggest that tumor metastasis results from a well-defined sequence of events: tumor cells mobilize normal bone marrow cells, cause them to migrate to particular areas in distant organs, and change the local milieu to attract and support metastatic tumor spread.

Using established murine models for lung cancer, melanoma, and lymphoma, Kaplan et al. tracked the fate of various populations of normal bone marrow-derived cells during tumor metastasis. Mice were lethally irradiated to destroy host bone marrow cells and transplanted with syngeneic whole bone marrow cells or purified populations of bone marrow cells tagged with a marker to allow tracking. After the tagged marrow cells engrafted (four weeks), the mice were injected with tagged cancer cells, known to metastasize to the lungs. It was reported that bone marrow-derived hematopoietic progenitor cells expressing vascular endothelial growth factor receptor 1 (VEGFR1) home to tumor specific pre-metastatic sites and form cellular clusters before the arrival of tumor cells. Preventing VEGFR1 function with antibodies or by removing VEGFR1⁺ cells from the marrow inoculum of wild-type mice abrogated the formation of the pre-metastatic clusters and prevented tumor metastasis. Conditioned media obtained from the tumor cells stimulated normal fibroblast cells in the future metastatic sites to produce fibronectin, an extracellular matrix protein that binds very late activation antigen-4 (VLA-4; also know as integrin α4β1) expressed on VEGFR1⁺ marrow cells that serves to "lock" the bone marrow cells to their niche. Blocking VLA-4 inhibited the binding and establishment of hematopoietic cell clusters in the lungs.

This paper provides evidence for a multi-step pathway of tumor metastasis. Rather than spread to random sites when cancer cells break from a primary tumor and enter the bloodstream, the primary tumor first secretes factors that induce fibronectin production in the pre-metastatic niche. The fibronectin acts as a "recruitment site" for circulating bone marrow progenitors via VLA-4 dependent mechanisms, and, after this pre-metastatic niche is further modified by the hematopoietic progenitor cells, the primary tumor sends cancer cells through the bloodstream to the newly colonized site. The report by Kaplan and colleagues provides substantial material to stimulate modelists and clinical trialists in the field. For the modelist, this work needs to be confirmed, the phenotype of the marrow progenitor cells better defined, and the mediators of recruitment and adherence of normal cells to the pre-metastatic cluster further elucidated. For the trialist, clinical protocols designed to test the efficacy of metastatic blockers need to be developed and the patient population defined (i.e., as adjuvant therapy?). This paper provides a greater understanding of tumor growth and spread and, in this way, we will eventually reap what we sow.