The invasive and metastatic abilities of malignant cells comprise one of the key “hallmarks of cancer” [1]; metastasis is the principal cause of death of the majority of patients diagnosed with invasive cancer [2]. Pathologists have long known that metastasis is not a random process, and that certain cancers have distinct patterns of metastasis to different organs [3]. The predictability of the organ distribution patterns of breast or lung cancers, for example, indicates that the development of distant tumors is a function of interactions between the disseminating cells and the sites of the metastases. This, in essence, is the “seed and soil” hypothesis presented by Stephen Paget in 1889 [4]. More than a century later, researchers continue their efforts to identify molecular mechanisms for the patterns of metastasis that are characteristic of different types of cancer. A common goal of research into the basic mechanisms is to find new insights into ways to prevent or control metastatic disease.

Metastasis can be viewed as the most difficult cancer phenotype to simulate and thereby study using in vitro techniques. Several tissue culture traits have been identified as potential indicators of metastatic potential, notably invasion through a basement membrane [5] and growth in semisolid agarose [6]. Development of three-dimensional tissue bioreactors – for example, with osteoblasts or hepatocytes – allows study of interactions of metastatic cells in bone and liver [7, 8].