The Gene Bubble: Curing Cancer, One Misleading Article at a Time

johnson

Great news for the cancerphobic, by which I mean everyone: Scientists have cracked the cancer genome! Or at least they've done it for lung and skin cancer, two of the scariest varieties. This is an enormous breakthrough for cancer research that's going to lead to prevention and treatments--indeed, it ushers in an entire new era in the fight against cancer.

cancerI learned all this from a Reuters article ("Gene maps to transform scientists' work on cancer") describing newly published research from a team of scientists. Unfortunately, the article is stunningly misleading, so please put everything I said above right out of your mind. Sure, some really fine science took place here, and some day we may get something from it. But it's fundamental research that is of little or no foreseeable relevance to your or my prospects for getting lung or skin cancer, or for having much luck with treatment if we do.

So-called "gene maps" of cancer tumors are old hat. The idea is to identify which genes differ between a cancerous cell and a normal one, in the hopes that the results will help scientists figure out how to target these villainous genes with drugs, or at least to spot them in people early on as a trigger for preventive action or fast treatment before the cancer spreads. These maps have been done for many different types of tumors, and they always end up disappointing from the point of view of practical advances, and for a simple reason: Cancer turns out not to be a matter of a few clearly identifiable genes going bonkers ("mutating"), but rather involves misbehavior of dozens or even hundreds of genes out of rogue's gallery of as many as thousands of genes. In other words, there isn't one simple way that genetic insults bring on cancer--there are a bewildering number of ways, leading scientists, when they are being ingenuous about it, to despair of being able to look at a non-tumor cell and tell if it's likely to become a cancer cell, or to figure out which genes to target in a cancerous one. In the case of the lung and skin cancer study, some 23,000 and 33,000 gene mutations turned up, respectively. Oops!

But the Reuters article, apparently based on a press briefing provided by the scientists, provides the impression the gene maps are just the ticket scientists have been looking for when it comes to beating back cancer. I'm sure a lot of other publications will get it at least as wrong, and it's a real shame--this kind of hype raises false hopes, distorts our view of our own health and of how research works, and ultimately leads to what we might think of as "science-advice burnout": We hear of so many bold breakthroughs that ultimately fail to impact our actual health prospects that we eventually stop paying attention to what scientists are doing. In this case I put much of the blame on the researchers and that briefing, in which they apparently not only encouraged the press to see their work as being of enormous practical import, but offered up the silly but very reporter-tasty suggestion that each cigarette a person smokes can now be linked to a specific number of gene mutations. Sure, it's literally true that smoking a cigarette can in theory trigger mutations that can in turn help trigger a cancer, but we've known that for decades. (For a more balanced take on the research, which will be appearing in the journal Nature, see Nature's own news report.)

The fact is, the only clear thing the research really did, in terms of news you can use, was provide irrefutable confirmation that smoking and exposure to sun lead to gene mutations that sometimes lead to lung and skin cancer, respectively. So now you know: Don't smoke, and don't fry yourself in the sun. Isn't medical progress wonderful?

David H. Freedman's next book, Wrong, about why expert opinion is often flawed, will be published by Little, Brown in 2010. He also wrote about The Gene Bubble in the November 2009 issue of the magazine.