Here's a very interesting evolution-in-cancer paper. The authors main point is that cancer cells evolve, from wherever they start, towards the primitive embryonic stem cell state. And the closer a cancer, on discovery, is to that ESS, the worse the prognosis for the patient. Thus, in cancer cells, they assert, the classical view of evolution as the generation of purposeless random mutations which are selected among by the environment, does not hold. In cancer, as the writer of this Med Express article on the paper, puts it: "all roads lead to Rome" regardless from where they start or how far-flung from "Rome" they are.
Here's the abstract:
"The essence of Darwin’s theory is that evolution is driven by purposeless mutations that are subsequently selected by natural environments, so there is often no pre-defined destination in organismal evolution. Using gene expressions of 107 cell types we built a functional space of human cells to trace the evolutionary trajectory of 18 types of solid tumour cancers. We detected a dominant evolving trend toward the functional status of embryonic stem cells (ESC) for ~3,000 tumours growing in distinct tissue environments. This pattern remained the same after excluding known cancer/ESC signature genes (~3,000 genes) or excluding all oncogenic gene sets (~12,000 genes) annotated in MSigDB, suggesting a convergent evolution of the overall functional status in cancers. In support of this, the functional distance to ESC served as a common prognostic indicator for cancers of various types, with shorter distance corresponding to poor prognosis, which was true even when randomly selected gene sets were considered. Thus, regardless of the external environments, cancer evolution is a directional process toward a defined cellular destination, a finding reconciling development and evolution, the two seemingly incompatible philosophies both adopted by the cancer research community, and also raising new questions to evolutionary biology."