Is Cancer 'Random'?
Cancer is mostly down to chance, argues the most unsatisfying study of 2015.
Metastatic breast cancer. Image: Ed Uthman/Flickr
It's not really a fair question. Cancer is cancer, but that's true at least somewhat in the same sense in which a virus is a virus. Meaning, there are lots of cancers. Lung cancer is almost always tied to smoking or asbestos, while brain cancer more so just happens. It's an effect without a cause, which is what we usually just call luck.
A better word is risk, perhaps. External factors can increase risk and make cancer more likely—sometimes much more likely, sometimes hardly at all—but that's not quite it either. A statistic given on Wikipedia is that 90 to 95 percent of all cancer is environmentally caused, but "environmentally caused" here just means that the cancer is not genetically inherited. Can we say that? Not genetic, thus environmental.
But doesn't "environmental" imply preventability?
A study released last week in Science has stirred up some interesting shit. The work of a pair of researchers at Johns Hopkins University, geneticist Bert Vogelstein and bioinformatics professor Cristian Tomasetti, it concludes that some two-thirds of all cancers are essentially bad luck. That is, most cancer can't be prevented.
Cancer is, in the eyes of the study, a function of cell division quantities. Human bodies are vast and full of cells busily dividing into new cells. Every division is a chance for a mutation to (re)occur, a bad copy now free to make more bad copies. Increasing opportunities for mutations to be passed on demands increasing mutation.
The formula put forth by Vogelstein and Tomasetti takes a given organ, adds up all of its constituent cells, identifies the organ's long-lived stem cells, and then determines how many times those stem cells will divide. More divisions, more risk.
So, the more divisions in a given organ, the more mutations (and possibly cancer) we would expect to occur in that organ. Using these figures, risk as mediated by number of stem cell divisions, the Hopkins researchers came up with some likelihoods for each organ to develop cancer. They compared these theoretical odds with statistics on actual cancer incidence organ by organ and found they aligned very well. The organs with the most (cumulative) cellular divisions developed cancer the most IRL.
"Some tissue types give rise to human cancers millions of times more often than other tissue types," the Hopkins duo writes. "Although this has been recognized for more than a century, it has never been explained. Here, we show that the lifetime risk of cancers of many different types is strongly correlated with the total number of divisions of the normal self-renewing cells maintaining that tissue's homeostasis."
"These results suggest that only a third of the variation in cancer risk among tissues is attributable to environmental factors or inherited predispositions," they conclude.
However obvious this conclusion might be, it's easy to see why it might be just as unsatisfying. We'd find the same correlation for, say, the cumulative total number of cars on some road and the number of car crashes on that road, but we also wouldn't call that luck. To do so is to imagine that we have no other information available to us about roads and car crashes, such as speed limits, traffic enforcement, driver demographics, and so on. Cars causing car crashes is a trivial conclusion.
That said, the Hopkins results aren't quite so trivial. Luck and randomness might be viewed as features of "not knowing something," but not knowing something doesn't mean there's a practical way of accessing that missing knowledge, now or at any time in the future. Which brings us back to preventing cancer.
In a dissenting blog post, Aaron Meyer, a cancer researcher and founder of MIT's Meyer Lab, notes something important. "While this correlation might explain variation between tissues," he writes, "it does not suggest the source of mutagenesis."
"Any factor that had similar effects throughout all tissues would vary this plot on the y-axis but have no effect on the correlation," Meyer explains. "As the data is plotted on a log axis, even for tissue-specific toxins many fold changes in the incidences of these cancers would still have no effect on the conclusions of this study." (A log scale is a way of plotting data points that vary exponentially as a linear variation.)
We can prevent car crashes by not driving, but most of our organs aren't optional. So we're left with accepting cancer, or at least the more random cancers, as the cost of doing business as highly complex biological systems. There's a strong oppositional front to this basic idea within a certain contemporary fake-green movement (where cancer is both a consequence of being "Western" and "unnatural" and not much else), but ditching cause and effect is from any perspective an absurd conclusion. As usual, it's a bit of column A (noise) and a bit of column B (cause).