Random and nonrandom processes in metastasis, and metastatic inefficiency.

Basically two hypotheses can be used to account for the metastatic inefficiency of cancer cells. The first is that the inefficiency is apparent, and reflects the small proportions of cells of a preexisting metastatic phenotype in primary tumors which are available for metastasis; thus, metastasis and metastatic inefficiency result from nonrandom processes. The opposed view is that most cancer cells have metastatic potential which is not realized unless they randomly enter a transient metastatic compartment which increases their chances of random survival in the metastatic cascade; thus, metastasis and metastatic inefficiency are basically random events. Discussion of this issue has been confused by the indisputable existence of cancer cell heterogeneity and cell lines with 'high' and 'low' metastatic or metastasis-related behavior. However, the relationship of these lines to naturally occurring metastasis is not at all clear. The evidence suggests that metastasis and metastatic inefficiency are determined by a series of random events, and although the presence of metastatic variants could alter the rate or efficiency with which metastasis occurs, they are not expected to determine whether they occur or not, or to produce major modifications in either.