Strong efforts are being made in order to better understand the molecular mechanisms underlying cancer dissemination. We have attempted to summarise some of the findings in this area. A large number of differences in gene expression have been described in metastatic and non-metastatic cells. In the mouse B16 melanoma system, more than 50 different markers have been described. It is likely that many of these differences reflect the same genetic alteration (i.e. a mutation in a regulatory gene alters the expression of a set of co-regulated target genes). One could argue that it is more effective to study mutations in regulatory as opposed to expression of down-stream target genes. However, we feel that proto-oncogenes are less suitable as markers compared to target genes, since it is difficult to screen for mutations at multiple levels in regulatory pathways. In contrast, measuring the expression of a small number of target genes (i.e. one of the targets in Fig. 1), the expression of which are stimulated by upstream regulators, is accomplished more easily. It is anticipated that the future of optimised panels of independent markers will sharpen cancer diagnosis and lead to individualised therapy.