The release focuses on a recent study published in Cell, which finds that a protein called AIM2 may play a significant role in preventing or slowing the progression of colorectal cancer in mice. Specifically, the release says researchers found that AIM2 inhibits the proliferation of intestinal stem cells — slowing the progression of cancer. According to the release, the study also found that AIM2 also contributes to healthy populations of “good” bacteria in the gut, which can help protect against colon cancer. However, the release offers little information to support the headline, which says “discovery promises new treatments to thwart colon cancer” — at least for cancers in human patients.
News releases should inform, not tease. According to the National Cancer Institute, more than 132,000 people in the United States will likely be diagnosed with colon or rectum cancer in 2015 — and an estimated 49,700 people will die from the disease this year. Any one of those patients, as well as family and friends, would be thrilled by the headline and lead paragraphs of this release, only to be crushed by the late revelation that the study involved only mice with induced cancers, not people with real disease. What’s more, the release completely fails to tell readers that the study design only showed what goes wrong when the AIM2 protein is hobbled, which is not at all like testing whether treatments that boost AIM2 can make things better.
The release does not address cost at all. This is not surprising, given that this is an early study, in an animal model, which identifies the role of a particular protein in preventing and, possibly, slowing the progression of colorectal cancer. However, the release explicitly discusses specific therapeutic possibilities, such as interferon therapy to increase AIM2 activity and transferring healthy bacteria into colorectal cancer patients. If you’re going to discuss possible treatment options (even in general terms), it’s worth mentioning what they may cost (even in general terms).
The release is overrun with general statements and specificity is mostly missing.The release says that AIM2 suppresses “abnormal expansion of intestinal stem cell populations,” but it doesn’t quantify what that means. How much more proliferation of these stem cell populations was there in mice that had a “malfunction of AIM2”? And how does stem cell proliferation translate to cancer growth? Similarly, the release noted that when “normal” mice and AIM2-deficient mice had the opportunity to exchange gut bacteria, “scientists found a striking reduction in colon tumors in the AIM2-deficient mice and an increase in tumors in the normal mice.” But it doesn’t give readers any numbers that tell us how different those tumor numbers were or what that meant in terms of survival rate.
Also, importantly, the release does not make clear how much can be inferred about the behavior of AIM2 in humans from the behavior of AIM2 in mice. What level of uncertainty is there? What are “good bacteria” and how are they transferred to experimental mice? How would that happen in humans? The release speaks of AIM2 findings and suggests possible meanings but fails to provide conclusive evidence to support the researchers’ conjecture.
This work is so far removed from clinical use in human subjects that it may be difficult to adequately assess potential risks for human patients. However, as noted above, the release does mention interferon treatment and transferring “good” bacteria populations into human patients. Every treatment carries some risks, and those are worth discussing.
The human digestive system is awash in colonies of different bacteria, all striving to maintain dominance in their environment. An imbalance between these different strains can cause a host of intestinal problems ranging from simple discomfort to life-threatening conditions. The release’s cavalier mention of possibly transferring “good bacteria” does readers an injustice by ignoring risks. While fecal transplants have shown some success in early research, they do carry substantial risks, and while some colon cancer patients may be willing to accept that risk, the release should at least mention that it exists.
The release seriously mischaracterizes the research. The first mention that the study was done on mice, not men, is buried deep in the release. Readers are also misled about the design and results of the study. The researchers revealed details about how hobbling an immune system protein, AIM2, shortened the lives of mice with colorectal cancer. But the release (perhaps encouraged by a few suggestive comments in the research article) pushes the hope that boosting AIM2 would fight cancer, something that was not even studied.
The release doesn’t give readers much quantitative information about the study. How many animals were involved? How many time points were assessed? In short, how much data did the researchers have in order to reach these findings? It’s not clear.
Colorectal cancer is a serious problem and more effective treatments are needed. The release did a fine job of explaining that AIM2 mutations were often associated with colorectal cancers, while still distinguishing those mutations from the cancer itself.
The release identifies the primary funding sources as the National Institutes of Health and ALSAC (though readers have to do their own searches to learn that ALSAC stands for the American Lebanese Syrian Associated Charities). However, the release fails to mention that Genentech supplied the Aim2-deficient mice. And neither the release nor the original article in Cell tell readers whether or not there are any financial or other ties between the researchers and Genentech… or whether any of the researchers hold any patent or royalty rights that are relevant to this work.
We’ll give the benefit of the doubt since we did not see any evidence of conflicts of interest, although we would encourage all news releases to make a clear statement regarding the presence or absence of such conflicts. Readers should not be left in the dark to guess about potential conflicts.
The release does not address any other treatments for colorectal cancer, nor does it mention screening or any of the steps that people can take to lower the risk of developing colorectal cancer — such as changes in diet or exercise. Since much of the release addresses the role of AIM2 in preventing cancer, this is problematic.
The release quotes one story author on this point: “‘We have only scratched the surface of the role of AIM2 in controlling stem cell proliferation and the maintenance of a healthy gut microbiota,’ Kanneganti said. ‘How exactly AIM2 does both of these functions is an exciting research area to pursue.'” Unfortunately, that’s the last paragraph of the release. And the first seven paragraphs talk about research as if it were done in a human model. It’s not until the eighth paragraph that we learn the study was done using mice. In short, this work is far, far removed from any potential clinical intervention for cancer patients. That should be said explicitly somewhere in the text, and the fact that the work was done in an animal model should have been made clear at the very top.
The release basically reports on findings about one protein that plays a role in immunity and therefore, may have some significance in the progress of colon cancer. But the release couches all its information in the context of that disease, lending support that may be unwarranted to their hypothesis.
While we’re concerned about this characterization, the release does make clear that this work fits into a larger body of work on the role of AIM2 in the immune system and the role of gut bacteria in protecting against colorectal cancer. So we’ll award a Satisfactory on that basis. However, it would have been great to get a little more information — for example, on how this fits into other recent work indicating that the gut microbiome could be used to screen for colorectal cancer or pre-cancerous polyps.
This release is unsatisfactory for two reasons. First, the headline is wildly misleading. Cancer patients expecting to read about new treatments will be brought up short, because the findings here (however promising) are still only findings from an animal model. It’s not clear when or if there will even be the clinical studies that can reveal whether or not treatments based on boosting AIM2 have any benefits. (Remember, this study only looked at the problems that follow impairment of AIM2.) Second, the release waits until the eighth paragraph (the second page, if you were to print it out) to tell readers that the work was done in an animal model. That’s a crucial piece of information, and should have been made clear in the headline and lead.