BAD LUCK?

Linda and Richard, Kalamazoo, 1981

I was lurking  just over the horizon.

 

I wish I were smart enough to figure out how go paste graphs and such on these blogs.  I have run on one that is very interesting and well worth your trouble to ponder.  You can see it by clicking on the following:  http://news.sciencemag.org/biology/2015/01/simple-math-explains-why-you-may-or-may-not-get-cancer

This arises from my efforts to understand an article that has obtained significant press traction this week, which maintains that “2/3 of all cancers are due to bad luck”.

You may react at first as I instinctively did: “What!  So the other third represents good luck?  Hell, any cancer is bad luck.  Period.  What are you talking about?”

Well, they have a point, and here (more or less) is what it is.

BioBackground:  Cancers are the result of uncontrolled cellular proliferation.  They arise when the genes that tell the cell when to divide (sometimes called proto-oncogenes) get “stuck open”, so to speak.  There also are genes that tell cells when not to divide (“tumor suppressor genes”).  If a series of mutations affects both kinds of genes, in the wrong way, a tumor is the result. 

More BioBackground:  Some people are born with defective genes of one kind or another – BRCA1,2 are well-known examples.  Others aggravate the problem by experiencing environmental factors that conduce mutations: smoke, eat red meat, play with asbestos for instance.  The authors of this paper are trying to find out how many cancers are not attributable to this kind of bad luck and/or foolish behavior.  In other words, how many cancers are due to simple, dumb, bad luck?

The method used is really rather uncomplicated.  The authors estimated how many times in a lifetime tissue at a given potential cancer locus will divide.  Then, they assume, the cancer rate should be proportional to this number.  They found that, within wide error limits, this holds true.  For instance: they estimate that a colorectal stem cell divides about 10¹² times during a lifetime, whereas a stem cell of the duodenum does its thing about 10¹⁰ times.  Thus their hypothesis predicts that colorectal cancer ought to be about 100 times more common than cancer of the duodenum – which, within very elastic error limits – it is.  Thus, using this kind of simple mathematical reasoning, they conclude that 2/3 of all cancers are the result of (this special kind of) “bad luck”.

Well, yeah – big deal.  I have maintained all along that many (most?) cancers are the result of the accumulation of mutations, a process over which we essentially have no control (once we ditch the cigarettes, asbestos, etc.  of course).  I agree with the authors that this directs us to work hard on early detection (and, of course, treatment).  But the graph that you will now examine contains some mysteries.  For instance, why is it that ovarian germ cell cancers and cancers of the small intestine have about the same lifetime risk, regardless of the fact that small intestine stem cells divide about 10,000 times more often than do ovarian cells?  Maybe there is an obvious answer, but I don’t see it.  More need for well-directed and well-funded applied BioResesarch.