Not disagreeing with you since as far as I know, there are no human cancers that are contagious. There are at least 2 mammalian cancers I know of, and possibly more, that are contagious. One is a disease wiping out Tasmanian devils. Another is a sexually transmitted cancer among dogs. Its thousands of years old and shows up as small tumors around the genital regions of affected animals. There was a TED talk describing this in some detail. Don't have the time to search for a link at the moment, but you can probably find it.
Heh--according to the article, there are 3 transmissible cancers, and you named 2 of them. ... Wow, I find this kind of astonishing.
The tumor cells are themselves the infectious agents, and the tumors that form are not genetically related to the host dog. Although the genome of CTVT is derived from a canid (probably a dog, wolf or coyote), it is now essentially living as a unicellular, asexually reproducing (but sexually transmitted) pathogen. Sequence analysis of the genome suggests it diverged from canids over 6,000 years ago; possibly much earlier. However, the most recent common ancestor of extant tumors is more recent: it probably originated 200 to 2,500 years ago.
