We're at a stage in the maturity of cancer treatment that can only be described as barbaric. I think it's less humor-like, and more bleeding-of-the-patient-like (which of course itself was a vestige of the theory of humors). Blood letting actually does some good in some circumstances, so they thought, "Let's do more of this." Cancer is sometimes similar. Some chemo and some radiation can work some of the time, so we just do more, often to the point where the patient can't tolerate it. It's very sad, but it's the best we can do most of the time.
Just like with blood letting, we're missing something about the theory of life that will go a long way to explaining how to better treat cancers of all kinds, and I don't think it's going to come from the sort of piecemeal approach taken by most cancer researchers (e.g., gene X is deleted here, so we need to target its target, or gene Y is multiplied, so we need to limit its expression, etc.).
Biology suffers from a lack of a theoretical framework in which to think about experimental biology. In physics, by contrast, there are lots of theorists, and they continually propose new ideas that can be tested by the experimentalists. In biology there are basically only experimentalists who are charged with devising hypotheses, testing them, and then convincing their colleagues that their hypothesis confirmation is credible. It's neither, in my opinion, a very efficient nor a very credible way to do science. It ends up being hyper descriptive, with very little predictive power (hence all the "omics," which make me want to puke).