Cancer metastasis is selective or random depending on the parent tumour population

Abstract

The issue of whether metastases result from the random survival of cells released from the primary tumour or from the selective growth of specialized subpopulations of cells having properties that allow them to complete the metastatic process is important to our understanding of tumour biology^1,2. Previous studies have provided indirect evidence that the process of metastasis favours the survival of cells having a metastatic phenotype(s)^3–8. Direct evidence that the process is selective would be provided by the demonstration that cells populating spontaneous metastases are, in general, more metastatic than most of the parent tumour cells; some^9–11, but not all^12,13, previous reports have failed to provide such evidence, suggesting that the process is random. These discrepancies could be due to differences in the biological characteristics of the various tumour systems studied and in the experimental conditions¹⁴. In the present study, these variables were minimized by using three metastatic variant cell lines of the B16 melanoma. We report that, even in controlled conditions, the process of metastasis can appear as either selective or random depending on the nature of the initial population. Specifically, metastasis by the unselected, poorly metastatic parent B16 tumour was indeed selective. In contrast, metastasis by previously selected B16 lines appeared to be random.