Are animal models predictive for humans?

Cancer arises from a stepwise accumulation of genetic changes that liberates neoplastic cells from the homeostatic mechanisms that govern normal cell proliferation. In humans, at least four to six mutations are required to reach this state, but fewer seem to be required in mice. By rationalizing the shared and unique elements of human and mouse models of cancer, we should be able to identify the molecular circuits that function differently in humans and mice, and use this knowledge to improve existing models of cancer.