One of the reasons why cancer cells are a problem is that they can express telomerase, and hence become 'immortal', capable of growing for arbitrary generations.
There has been some research into switching telomerase off, but this runs into issues with the telomeres already having been extended to
sufficient length - so there is a lag phase before cell death occurs.
I therefore propose the opposite approach. The telomerase is already expressed in exactly the cells we want to kill. Suppose we could switch this protein on and hold it on (by wedging it with a suitable drug). The cancerous cells would fill up with telomeres like a hoarder with a garage; after a while they wouldn't be able to find anything useful amongst all the junk.
This is also satisfyingly karmic justice - cancer is uncontrolled cell growth, and this would do them in with uncontrolled genomic expansion.
(Incidentally, I'm assuming that telomere length is regulated by a system involving telomerase being inhibited somehow. I don't know this is the case but can't see an obvious alternative.)