"emphasis on very rare cancers (e.g. osteosarcoma, medulloblastoma) that together make only a small contribution to the total cancer burden."
and that it
"excludes [...] common cancers for which incidence differs substantially between populations and over time."
So it sounds like the generalization hinted at in the abstract shows a bigger misunderstanding of statistics than any in the press release. Would be nice if the paper was not paywalled, so we could actually read it.
Didnt read it, so not sure if its sufficiently understandable without the figures. Prob not.
[1] Abstract: http://www.thelancet.com/journals/lancet/article/PIIS0140-67... PDF: http://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736(...
There are plenty of problems with the original paper see
http://www.sciencemag.org/content/347/6219/224.full
The Science link is basically a retraction from the journalist who wrote the original summary article. She quotes the study authors making excuses about not having enough time to explain the subtleties under press deadlines. I've never read anything quite like it.
It seems everyone agrees that Figure 1 of the Vogelstein paper, showing the striking correlation between Log(cancer incidence) and Log(# stem cell divisions) was really a beautiful result. But then things fell apart with the further analysis and (especially) interpretation.
The Altenberg paper on Arxiv looks like exactly the sort of careful treatment that the original paper should have gotten in the review process. (MIT's Aaron Meyer raised some of these points in a blog post around the time of publication http://www.ameyer.me/science/2015/01/02/vogel.html )
I don't agree with that characterization of the Science news article at all. It's a summary of the secondary reaction to the press coverage, but I can't construe it as a retraction of the original news coverage. For example, here's what David Spiegelhalter says in it, who's entire research lately has been on accurately communicating risk:
>In this case, he felt, “the gist of the coverage is very reasonable—most cases of cancer are due to chance.”
I too feel that the coverage was quite accurate, perhaps the most accurate that I've seen of a science tidbit in quite some time. If one's entire research program is on how to minimize environmental causes of cancer, I can see how it would feel like one's research was being minimized entirely, and require a vigorous defense. However, these "defenses", in particular Aaron Meyer's, seem to be fact free. In contrast, the epidemiology world seems to agree almost exactly with the Vogelstein paper's estimate of 65% chance, 35% modifiable causes (though perhaps somewhat by chance, as their measurements are not exactly of the same thing):
http://www.thelancet.com/journals/lancet/article/PIIS0140-67...
If there's broadly accepted epidemiological data that contradicts the molecular approach from Vogelstein, I would love to see it.
Here's the section of the news article that I feel reads like the journalist walking back the central claim in her original news article. It also seemed to me that she was attempting to pin the blame for the misunderstanding on Tomasetti (by quoting from her initial interview with him, and reporting that he had vetted her initial piece):
"...[W]as the “two-thirds” figure actually referring to a fraction of cancer cases? Tomasetti had explained to Science that “if you go to the American Cancer Society website and you check what are the causes of cancer, you will find a list of either inherited or environmental things. We are saying two-thirds is neither of them.” He also confirmed the news story's language describing the study before it was published. In a follow-up interview [...] Tomasetti clarified that the study argued that bad luck explained two-thirds of the variation in cancer rates in different tissues—a subtly different claim.
Despite the confusion among reporters, Tomasetti did not feel they had been careless[...] And, he believes, he did his best to convey his findings to nonexperts. “If given enough time, or space, I can explain the subtleties of any given scientific result to anyone really,” but there were only so many hours he could spend speaking with reporters on deadline. The material is complicated even for statistical gurus, he believes. He has been busy preparing a technical report with additional details, and Johns Hopkins also sent a follow-up explainer to journalists and posted it online."
Here's a pastebin of the article in case others are interested in it: http://pastebin.com/upWV95PK
I don't agree at all with your contention that the criticisms of the original paper are 'fact-free' and based on researchers feeling threatened. There are lots of problems with this paper; the fact that their conclusions may agree with data from other fields (or may even be right) doesn't change that.
The original paper (here: http://www.sciencemag.org/content/347/6217/78.abstract -- the abstract is enough to deduce the conclusion) is a simple statistical analysis of mutation probability in certain cell types. It's basically saying that many cancers are not necessarily influenced by external factors and can be predicted by nothing more than the number of cell divisions and the probability of DNA mutation compounded.
That, in itself, is a fascinating scientific result. Arguing against the statistics here is probably not a good idea, unless there was a serious error in the math or an order of magnitude error in the probability of mutation. Certainly there are other possibilities of error in the data or methodology.
But the article in opposition (OP) is doing something odd that doesn't appear to be a scientific argument: it's railing against the language and the implications of the very concept of statistical analysis applied to the problem. It's assuming that because we discovered that, perhaps, cancer has a certain probability of occurring independent of external factors, that it will somehow slow research or cause us to throw up our hands and give up on prevention strategies or research.
I don't think that's the case. Rather than understanding the absolutely fascinating statistical analysis here, the OP article comes across as reactionary and unscientific.
Assuming the data is sound, the statistical analysis is profound. Most often, statistics such as that are profoundly misunderstood as well: people have an incredible capacity for attribution bias and data disbelief. This simply underscores the need for a better education in basic statistics as well as science across the board.