Multicellular organisms need checks on how often and by how much a group of their cells can multiply. A liver must contain a specific amount of liver cells, and cancer occurs when these cells are able to continually divide and eventually activate non liver specific genes.
The cells of a multicellular organism become highly altruistic because each cell of the organism gives up its ability to reproduce in order to provide a chance for the organism's gametes to produce a new living organism.
So wyldfire's original comment on Cancer being a 'feature' is not necessarily incorrect, however the explanation is.
Cancer occurs when the contract that states that the Liver cells, who cannot reproduce, become able to continually divide without checks and balances. You can also almost think of Cancer as Cellular natural selection and evolution within the body. Cancer is a feature and consequence of multicellularity and likely not adaptive in any way such as sickle cell anemia and malaria resistance.
An awesome book which touches on this in addition to other history or cancer is The Emperor of all Maladies http://www.amazon.com/Emperor-All-Maladies-Biography-Cancer/...
http://www.amazon.com/Emperor-All-Maladies-Biography-Cancer/...
Highly recommended reading, BTW, though it's a bit depressing.
For the average to go up only four months either we're doing really well against a tiny percentage of cancers or we're doing generally a tiny bit better against most cancers. Or some combination of the two. Either way you look at it that's an indication of lack of progress.
[1]: http://www.amazon.com/Emperor-All-Maladies-Biography-Cancer/...
My take away at the end of the book was that for all of the "war on cancer" hyperbole going back to the 50s, up until the mid-90s, we just didn't know enough about cancer to really be fighting it. I feel a lot more optimistic about the next 50 years of cancer research than the previous 50.
http://www.amazon.com/Emperor-All-Maladies-Biography-Cancer/... (non-affliate link)
For some incredibly interesting reading on the history of battle against cancer I highly recommend The Empreror of All Maladies [0].
0: http://www.amazon.com/The-Emperor-All-Maladies-Biography/dp/...
If I remember correctly, most of the breast cancer that develops early is fast-spreading, while most of the cancer later in life is slow-spreading. So as you screen younger and younger people, you quickly reach a point where you cause more harm than good. You're only finding more disease you can do nothing about, and the false positive cause tremendous hardships on tens of thousands of women.
I also seem to remember that there's some serious discussion over the idea of categorizing cancer based on where it's first detected or starts. Cancer is much more a cell-based disorder than a location-based one. We may have created this huge categorization and treatment system that describes the problem in an inefficient way.
Good read.
http://www.amazon.com/The-Emperor-All-Maladies-Biography/dp/...
.. there isn't one thing called "cancer", there are at least thousands of different varieties that have different genetic profiles, different behavior and risk factors, etc.
The thick, opaque layers represent actual human learning and experience and judgment, and you can't just skip over them because it seems more fun to jump right to the high-profile heroic stuff. You can't do the good stuff without learning the rest.
(*Ok, once in awhile people come up with breakthroughs in fields they know nothing about - and once in awhile a monkey will accidentally type something intelligible, given a typewriter. We still don't issue typewriters to monkeys when we need to get some writing done, even if it would work from time to time.)
We keep thinking we have a lock on curing cancer, but it continues to be very elusive.
As I understand it, using automation to "debug" individual cancers is scalable. That is, reading a cancer, adapting to it, and creating a vaccine is something that can be done with increasing levels of detail as automation increases.
So if they only show a 1% improvement, it's a strategy definitely worth continuing. This is very much like the hacker who keeps playing around with a broken method until it works -- only it's all automated. It very well may be that we "learn" how to contain/control many cancers without having a traditional understanding of them at all. Fascinating approach, and much different from previous forays. Watch this space. Given the history, my money says it's still going to be a long, hard slog -- perhaps decades, but still, a promising approach.
http://www.amazon.com/Emperor-All-Maladies-Biography-Cancer/...
https://www.amazon.com/Emperor-All-Maladies-Biography-Cancer...