These juxtaposed sentences come across as a bit contradictory. If telomeres are one of the cell's defenses against cancer, then telomerase activation should defend against cancer.
Most cancers do activate telomerase, but telomerase is not an oncogene.[1]
1 = http://www.nature.com/onc/journal/v21/n4/abs/1205076a.html
I think we can all agree that telomeres are critical for preventing non-homologous end joining where it shouldn't be occurring.
I thought the idea was that telomeres are like chaff that the DNA drops over the course of many replication cycles instead of losing "more important" bases of DNA. In that sense telomeres might protect against loss of function (cancers). I did not know they prevented non-homologous end joining, but see how they could. I take it non-homologous end joining could lead to gain of function (cancers).
So what is the divide? Do some people think that telomeres protect cancer?
A comment above (http://news.ycombinator.com/item?id=4846137) suggests something like this. First a cell is converted into a cancerous state, perhaps even due to a lack of telomeres and subsequent loss or gain of function. Then telomerase somehow is upregulated. If there was some error correction process (to correct the loss or gain of function cancer), could the sudden increase in telomerase and telomere length prevent that error correction from occurring?
It's quite simple and doesn't require any interaction with other cellular machinery.
That is interesting to me too though. I don't know why cells should stop dividing without telomeres, or with telomeres below a critical length. That suggests to me there is interaction with other cellular machinery at this onset of senescence.