Current genetic and epigenetic theories of cancer-specific drug resistance do not adequately explain: (i) the karyotypic changes that coincide with resistance, (ii) the high rates at which cancer cells acquire and enhance resistance compared to the rates of conventional mutation, (iii) the wide ranges of resistance such as multidrug resistance, (iv) the frequent occurrence of intrinsic drug resistance. We have recently proposed, that specific karyotypic alterations are sufficient for drug resistance via new transcriptomes of cooperative genes, independent of gene mutation. This mechanism generates new phenotypes just like trisomy 21 generates Down syndrome. These karyotypic changes are generated by cancer-specific aneuploidy autocatalytically, because aneuploidy destabilizes the karyotype by misbalancing teams of proteins that synthesize, repair and segregate chromosomes. Evidence for this chromosomal mechanism is as follows: (i) resistance is proportional to the number of clonal chromosomal alterations compared to drug-sensitive precursors. (ii) The high rates at which cancer cells acquire drug resistance are comparable with the rates, as high as 10(-2) per cell generation, at which their karyotypes change-dimming hopes for gene-specific therapies. (iii) Multidrug resistance probably reflects un-selected transcriptomes of karyotypes selected for resistance against specific drugs. (iv) Intrinsic drug resistance probably reflects unselected transcriptomes of karyotypes selected for oncogenicity. We also adduce evidence that resistance of chronic myeloid leukemia against the drug imatinib is chromosomal, although it is widely believed to be due to mutation of a kinase.