Fox News, Keith Olbermann–Cleveland Clinic Cures Breast Cancer (Not).
Gary Schwitzer is so mad he’s starting to sound like MSNBC‘s Keith Olbermann.
Schwitzer, on his Health News Review blog, does his own version of Olbermann’s worse-worser-worst segment, and, as is often true with Olbermann, finds that Fox News is the villain. “We’ve seen a lot of bad health news stories,” Schwitzer writes, “but this is one of the worst: Fox News posted a story under the headline, “U.S.-Developed Vaccine ‘Could Eliminate’ Breast Cancer.” Fox went on to say that the vaccine ”could prevent breast cancer and save the lives of millions of women.”
The story concerned a study in this week’s Nature Medicine, in which researchers reported essentially what Fox said: The vaccine “may provide safe and effective protection against the development of breast cancer for women in their post-child-bearing, premenopausal years…”
Speaking of Olbermann, maybe Schwitzer could have included him, too. On last night’s show, Olbermann, who occasionally detours from politics to explore a science topic, concluded with a lengthy interview with the study’s senior author, Dr. Vincent K. Tuohy of Cleveland Clinic. If Fox was the worst, Olbermann was close. This is undoubtedly the most in common that Olbermann and Fox News have ever had–or will ever have.
Schwitzer is right to be mad about the coverage (and mad about my comparing him to Olbermann). Schwitzer mentions a few others who hyped the story, and concludes with this acid observation: “To talk about cures and prevention after an experiment in six mice is not sound health journalism.”
I agree.
On the other hand…
What was it that the abstract said? The vaccine “may provide safe and effective protection” against breast cancer. That’s pretty clear. And that was written by the people we pay to do this stuff. Literally, in this case–this study was done with a taxpayer-supported National Institutes of Health grant. That also means NIH thought these researchers were hot stuff, too, deserving of a grant when many other scientists who asked for one came up empty. The experts–our experts–say the vaccine could protect against breast cancer. Why blame Fox for repeating that?
If the abstract doesn’t convince you, check out the press release written by press experts at Cleveland Clinic. “Research Could Lead to First Vaccine to Prevent Breast Cancer Formation in Women over Age 40 and Women at High Risk,” screams the headline. Tuohy, the principal investigator, goes even further, removing the conditional. “We believe that this vaccine will someday be used to prevent breast cancer in adult women in the same way that vaccines prevent polio and measles in children,” he says in the release.
O.K. Now I’m thinking maybe Fox underplayed this thing.
Before we get carried away, however, Chris Morrison at The In Vivo Blog throws a bucket of cold water on all of us. ”ASCO approaches, and the season of cancer vaccine hype is upon us,” he writes. (The American Society of Clinical Oncology meets this weekend in Chicago.) Ya gotta love In Vivo’s headline: “Turn Out the Lights and Go Home–The Cleveland Clinic Has Cured Cancer, Convinced LeBron to Stay.”
A few years ago, an editor from Wired, speaking at a pitch slam at the annual ScienceWriters meeting, dismissed a pitch with the riposte, “If it’s in mice, it’s not in Wired.” Funny, but not well considered, in my view. Mice and humans share 90 percent of their genes. Many, many medical findings were made in mice, rats and even flatworms and fruit flies before they were made in humans.
A successful vaccine trial in mice most certainly does not mean that breast cancer has been cured in women. But it also doesn’t mean nothing, as the critics sometimes seem to be saying. It means something, and it ought to be reported. Why would we shield our readers from this sort of thing? Because they’re not as smart as us and can’t understand the difference between mice and people?
I’m not charging Schwitzer with being overly critical; his criticism is appropriate. But we can’t dismiss mice entirely. They teach us an enormous amount, and we should share that with our readers and listeners.
The point is to do it correctly. Schwitzer points to Liz Szabo‘s story in USA Today, which is a model of reporting, both engaging and circumspect. I hope Szabo (or her publisher) won’t mind if I quote her at some length:
An experimental vaccine prevented breast cancer in genetically engineered mice, according to a preliminary study in the June 10 issue of Nature Medicine. The vaccine has not been tested in humans.
Though the approach is intriguing, it is far too early to know whether a vaccine could also help women avoid breast cancer, says Massimo Cristofanilli, chair of medical oncology at Philadelphia’s Fox Chase Cancer Center, who wasn’t involved in the experiment.
Many drugs appear promising in mice, but very few succeed in humans, Cristofanilli says.
On average, only one out of every 250 drugs in lab studies or animal models get approved, according to Pharmaceutical Research and Manufacturers of America.
That’s how it’s done. Excitement (“intriguing”), with caution (“very few succeed in humans”). She then goes on to describe the experiment.
I think I’ve offended everyone I’ve mentioned in this post except Szabo. I should probably throw in a gratuitous shot at her, too, just to show I’m capable of equal treatment for all. But, I can’t. She did too good a job.
- Paul Raeburn
June 2nd, 2010 at 1:14 pm
If only I had Olbermann’s salary or his suits.
June 2nd, 2010 at 3:40 pm
This is a great post. I will say that it’s not so surprising that the PR people – and even the researchers themselves – would over-hype the study. There are researchers who exercise caution when explaining their results, and then there are those who get a little carried away, perhaps in the belief that media exposure will help in fundraising efforts.
June 2nd, 2010 at 5:09 pm
I saw the Amazing Breakthrough segment on Olbermann, and recall (my ears were open even as my eyes were rolling) that the researcher *did* say that he needed more funding. The average medical writer would see that as a second tip-off after the sample of a measly six mice.
You gotta wonder why the researcher submitted and Nature accepted a paper based on only six mice. The Cleveland Clinic must be on very hard times if it couldn’t stake its lab to more mice, if only to keep the CC from looking so bad to professional medical writers.
Or could it be that the paper reports on the first six mice and that a subsequent larger sample didn’t yield results as likely to shake the money tree?
June 3rd, 2010 at 11:09 am
I took a look at the paper available for free online as I could just not believe that a cancer researcher would use only six mice, or that Nature Medicine would overlook such weak methods.
I could be reading Figure 2 incorrectly, but what I see indicates that the in vivo innoculation experiments, which certainly were not the only experiments, entailed at least 12+16+12+16+16+15 = 87 mice. Even if you subtract the controls, that’s far cry from six.
June 3rd, 2010 at 12:25 pm
Ah, the number of mice…interesting question.
You’ll notice I quietly sidestepped that issue in the post, aside from quoting Schwitzer, because it was difficult to figure out.
Liz Szabo reports that 50 mice were vaccinated.
Jenny’s math looks correct if one adds up all the “n’s” in the grafs in Figure 2. But it’s not clear to me if those all represent different mice; the caption is dense.
Certainly anyone reporting this story should have nailed that down–how many experimental mouse subjects were used, and how many controls.
Thanks to Tom for highlighting the highlighting the blame I put on the researcher and the PIOs.
And–should we chip in and buy Schwitzer a suit?
June 3rd, 2010 at 3:36 pm
Trying to get to the bottom of the mouse question. The Cleveland Clinic told me in an email that,”More than 100 mice were used in three studies … half received the vaccine containing alpha-lactalbumin, half received a vaccine that did not contain the antigen.”
June 3rd, 2010 at 4:04 pm
Liz,
Thanks for digging…The numbers you were given don’t quite seem to square with the math in figure 2, which might just mean that the paper itself wasn’t clear enough on this. One would think the reviewers would have insisted on a little more clarity.
The copy editors at the AP and BusinessWeek certainly would have. We always seemed to have at least one whose great joy in life was finding numbers in a story that didn’t add up. Copy editors have, you know, their own particular ways of finding excitement. And I’m glad they do; I’d hate to count how many times they’ve saved me.