Whole transcriptome profiling has long been proposed as a method of identifying cancer-specific gene expression profiles. Indeed, a multitude of these studies have generated vast amounts of expression data for many types of cancer, and most have identified specific gene signatures associated with a given cancer. These studies however, often contradict with each other, and gene lists only rarely overlap, challenging clinical application of cancer gene signatures. To understand this issue, the biological basis of transcriptome dynamics needs to be addressed. Chromosome instability (CIN) is the main contributor to genome heterogeneity and system dynamics, therefore the relationship between CIN, genome heterogeneity, and transcriptome dynamics has important implications for cancer research. In this review, we discuss CIN and its effects on the transcriptome during cancer progression, specifically how stochastic chromosome change results in transcriptome dynamics. This discussion is further applied to metastasis and drug resistance both of which have been linked to multiple diverse molecular mechanisms but are in fact driven by CIN. The diverse molecular mechanisms that drive each process are linked to karyotypic heterogeneity through the evolutionary mechanism of cancer. Karyotypic change and the resultant transcriptome change alter network function within cells increasing the evolutionary potential of the tumor. Future studies must embrace this instability-induced heterogeneity in order to devise new research and treatment modalities that focus on the evolutionary process of cancer rather than the individual genes that are uniquely changed in each tumor. Care is also needed in evaluating results from experimental systems which measure average values of a population.