The stories first started showing up about a week ago, and now they're metastasizing: “Statins linked to dramatic reduction in cancer death rates” blared The Independent. “Patients taking cholesterol-busting drugs are 'half as likely to die'” claimed the Daily Mail. It started in the British tabloids, then spread to India and Pakistan. Now it's hitting US health and medical websites. It's amazing to track its spread, like watching the spread of an annoying virus, just to see the amount of cutting and pasting that tries to pass itself off as journalism. The stories are all close to identical. They all come from the same source. And they're all wrong.
The real headline here is “More Crap Medical News!” The story suffers from a simple problem: a lack of evidence to back up the headline. A paucity of fact. An unmooring from reality. This, for a supposed medical science story, is fatal.
Track it back and you find, unsurprisingly, that it all began as a press release (this is warning sign #1: Press releases by their very nature constitute hype; anyone interested in the medical news should know that anything that starts as a PR release should be viewed skeptically). This one was based on a presentation at a medical conference (warning sign #2; like most conference presentations, this one was unpublished in a journal and so lacked formal peer-review; there is no way to judge the quality of the work). And the title of the conference presentation did not include the key word statins (warning sign #3, news headline does not correlate with study).
The study in question came from a number-crunching group in Britain, which had analyzed close to a million medical files and found that patients with a diagnosis of high cholesterol showed dramatically improved cancer survival rates over those with lower cholesterol levels. Note: They did not track statin use at all.
Hmm. . . . high cholesterol, low cancer mortality. There's actually quite a bit of earlier evidence hinting that low cholesterol might help spur cancer, so it's not too much of a reach to think that higher cholesterol might help fight it. But that wasn't the angle the PR people working to promote the conference went for. They crafted this headline for their press release: “Statins may be associated with reduced mortality in four common cancers.” Odd, no mention of cholesterol, the only thing the researchers tracked for this study. And the inclusion of statins, which the study did not track.
So where did statins come from, if they weren't in the study? From quotes by two of the authors, apparently in interviews with the PR people. The researchers opined that since just about everybody with high cholesterol was on statins, perhaps statins were the reason for the sizable decline in cancer deaths in high-cholesterol patients. They had no direct new evidence to back up their idea. It was a guess. And the guess got the news.
Well, that's the media. But it's not science.
Which got us working. We dug a bit to find out what scientists who actually have studied statins and cancer say. And we found this:
In summary: statins might trigger cancer. Or they might prevent it. Or they might do nothing. They might improve survival. Or they might not. It sort of depends on whether you're looking at men or women. Or how old they are. Or else these things make little difference.
In other words, we don't know. And that is very depressing. There has actually been quite a bit of work done on cancer and statins. The fact that we don't know after decades of research and hundreds of published studies says to us that if there is a link, it is weak and variable. The mechanisms are complex and unclear.
And the same goes for studies looking at the relationship between serum cholesterol and cancer. Some types of cancer seem to be more common in high-cholesterol patients, others appear to be more common in low-cholesterol patients. And in others it doesn't seem to make any difference. Results are all over the map.
So we're left agreeing with recent reviewers that more or less throw up their hands and call for, you guessed it, more study. More experiments. More clinical trials.
Unfortunately, those trials are really hard to run. In part it's because there are more than 100 kinds of cancer, and it appears that they all don't react the same ways to statins. In part it's because statins have multifold effects. In part it's because both heart disease and cancer take many years to develop, and can arise from multiple causes. In part it's because so many people are taking statins (or have gone on and off them) that it's difficult to put together a clean long-term study.
Final thoughts: 1) There's a lot of bad medical reporting out there; 2) conference press releases are evil; and 3) take everything you read about the statin-cancer link with a giant grain of salt.