From Abracadabra to Zombies
8 Jan 2015. On January 1, Johns Hopkins Medicine issued a press release entitled Bad Luck of Random Mutations Plays Predominant Role in Cancer, Study Shows --Statistical modeling links cancer risk with number of stem cell divisions. That same day The Telegraph posted an article by Sarah Knapton based on the press release entitled Most cancers are caused by bad luck not genes or lifestyle, say scientists --Scientists at Johns Hopkins University School of Medicine in the US found that the majority of cancers are not linked to environment or lifestyle.
Even a lazy, careless reader can see that the news article header makes a very different claim from the one in the press release header. The headline of the Johns Hopkins press release does not claim that most cancers are caused by bad luck, not genes or lifestyle. Before getting to what the scientists actually said, consider what journalists Bob O'Hara and GrrlScientist of The Guardian had to say about those journalists who claimed that the scientists had said that most cancers are caused by bad luck. The headline of The Guardian article reads: Bad luck, bad journalism and cancer rates.
The big science/health news story this week is about cancer rates, with news outlets splashing headlines like “Two-thirds of adult cancers largely ‘down to bad luck’ rather than genes” ... or “Most cancer types ‘just bad luck’” ... But these headlines, and the stories, are just bollocks. The work, which is very interesting, showed no such thing.
The problem with the O'Hara and GrrlScientist article is that it contains too much statistical gobbledygook to be of much value to most readers.
Bert Vogelstein, M.D, and Cristian Tomasetti, Ph.D., are the scientists/authors of a report published in Science magazine entitled "Variation in cancer risk among tissues can be explained by the number of stem cell divisions." Vogelstein is the Clayton Professor of Oncology at the Johns Hopkins University School of Medicine, co-director of the Ludwig Center at Johns Hopkins, and an investigator at the Howard Hughes Medical Institute. Tomasetti is an assistant professor of oncology at the Johns Hopkins University School of Medicine and Bloomberg School of Public Health.
The title of the report indicates that the study is about explaining the variation in cancer risk among different kinds of cancers through the number of stem cell divisions. The title of the report does not indicate that the study is about causes of different types of cancers. The abstract of the article reads (emphasis added):
Some tissue types give rise to human cancers millions of times more often than other tissue types. Although this has been recognized for more than a century, it has never been explained. Here, we show that the lifetime risk of cancers of many different types is strongly correlated (0.81) with the total number of divisions of the normal self-renewing cells maintaining that tissue’s homeostasis. These results suggest that only a third of the variation in cancer risk among tissues is attributable to environmental factors or inherited predispositions. The majority is due to “bad luck,” that is, random mutations arising during DNA replication in normal, noncancerous stem cells. This is important not only for understanding the disease but also for designing strategies to limit the mortality it causes.
The scientists found a high correlation between a tissue's stem-cell replication and lifetime risk of cancer for that type of tissue. The press release explains the claim about the results suggesting that two-thirds of the cancer risk is due to random mutation:
Using statistical theory, the pair calculated how much of the variation in cancer risk can be explained by the number of stem cell divisions, which is 0.804 squared, or, in percentage form, approximately 65 percent.
One thing the scientific report doesn't claim is that heredity and environment are unimportant factors in cancer. Nor does it claim that diet and lifestyle are not important in preventing cancer. If these scientists are right--and they don't claim they are--then the implication is that our focus should be on early detection (emphasis added):
All cancers are caused by a combination of bad luck, the environment and heredity, and we’ve created a model that may help quantify how much of these three factors contribute to cancer development.
If two-thirds of cancer incidence across tissues is explained by random DNA mutations that occur when stem cells divide, then changing our lifestyle and habits will be a huge help in preventing certain cancers, but this may not be as effective for a variety of others....We should focus more resources on finding ways to detect such cancers at early, curable stages.
This study shows that you can add to your risk of getting cancers by smoking or other poor lifestyle factors. However, many forms of cancer are due largely to the bad luck of acquiring a mutation in a cancer driver gene regardless of lifestyle and heredity factors. The best way to eradicate these cancers will be through early detection, when they are still curable by surgery.
The Science Editor's Summary of the report reads (emphasis added):
Crunching the numbers to explain cancer
Why do some tissues give rise to cancer in humans a million times more frequently than others? Tomasetti and Vogelstein conclude that these differences can be explained by the number of stem cell divisions. By plotting the lifetime incidence of various cancers against the estimated number of normal stem cell divisions in the corresponding tissues over a lifetime, they found a strong correlation extending over five orders of magnitude. This suggests that random errors occurring during DNA replication in normal stem cells are a major contributing factor in cancer development. Remarkably, this “bad luck” component explains a far greater number of cancers than do hereditary and environmental factors.
Again, there is no claim that this is the final word on the subject, much less that cancer is just a random disease or that changing your lifestyle or limiting exposure to carcinogens can do little to prevent cancer. Yet, fitness guru Gabe Mirkin, M.D., read the study to have claimed just that:
Dr. Gabe Mirkin's Fitness and Health e-Zine
January 11, 2015
Cancer Is Not Just a Random Disease
A recent study from respected researchers at Johns Hopkins claims that two-thirds of adult cancers are caused by random DNA mutations in your cells, and far fewer cancers are caused by genetics, lifestyle, or exposure to cancer-causing agents (Science, January 2, 2015). If this is so, changing your lifestyle or limiting exposure to carcinogens would do little to prevent cancer, yet the scientific literature overwhelmingly shows that lifestyle, environmental carcinogens, and genetic factors determine cancer susceptibility.
The authors of the study don't say cancer is just a random disease, nor do they say (as noted above) that lifestyle changes will do little to prevent cancer. Dr. Mirkin goes on to say:
Why I Disagree with Their Message
Scientists have known for more than 50 years that humans produce millions of cancer cells every day. Virtually all cancer cells are searched out by a person’s immunity and destroyed before they can cause cancer. Why wouldn’t it be reasonable to conclude that factors in the environment impair immunity to allow the cancer cells to live, or make cancer cells more resistant to a person’s immunity?
The authors of the study did not claim that environmental factors don't impair the ability of the immune system to destroy cancer cells. Nor do they claim that environmental factors don't make cancer cells more resistant to the immune system.
Dr. Mirkin says:
The authors support their theory of random genetic mutations causing cancer by giving the example that the colon has a lifetime cancer risk of 4.8 percent that is 24 times higher than in the small intestine, where it is 0.2 percent. The scientists found that the large intestine has many more stem cells than the small intestine, and that they divide more often: 73 times a year, compared with 24 times.
Actually, what the authors found was that a change in the number of stem cell divisions in a tissue type is highly correlated with a change in the incidence of cancer in that same tissue. The example they used compares colon tissue and small intestine tissue (four times more stem cell divisions in colon tissue) to colon cancer and small intestine cancer (cancer is much more prevalent in the colon than in the small intestine). The authors note: “You could argue that the colon is exposed to more environmental factors than the small intestine, which increases the potential rate of acquired mutations,” says Tomasetti. However, the scientists saw the opposite finding in mouse colons, which had a lower number of stem cell divisions than in their small intestines, and, in mice, cancer incidence is lower in the colon than in the small intestine. They say this supports the key role of the total number of stem cell divisions in the development of cancer. I think they're right; these facts do support the notion that stem cell divisions play a key role in the development of cancer. But this in no way rules out the importance of environmental factors, such as bacteria, in colon cancers, though Dr. Mirkin thinks it does.
Yet a recent study gives a far more plausible explanation that is linked to diet, an environmental factor (see Bacteria May Cause Colon Cancer). This study shows that bacteria associated with colon cancer live in your colon and manufacture biofilms that they use to prevent your immunity from killing them. These biofilms are associated with increased risk for colon cancer and are not regularly found in the small intestine. We do not yet know all of the reasons why some cancer cells are destroyed by your own body and others go on to become cancers, but I think it is more reasonable to state that bacterial biofilms in the colon cause colon cancer than it is to state that the high number of random mutations in the colon cause colon cancer. The random mutation theory also does not explain why mice, which have the highest rate of cell reproduction, do not have the highest rate of cancers or why blue whales, which have the largest number of cells of any organism, have a low cancer rate.
Tomasetti and Vogelstein do not claim that the two explanations (random mutations vs. heredity and environment) are mutually exclusive. As I understand their argument, they are not claiming that the more cells you have, the greater the chance of having cancer; nor are they claiming that the greater your rate of cell reproduction (as a species), the greater your cancer rate.
Dr. Mirkin also says:
They did not study people and they did not study environmental or genetic causes of cancers. To show that cancer is caused by random chances of gene mutation, they would have had to study how environmental and genetic factors affect gene mutations. Since they did not do this, they have to state only that their conclusions are their opinions and are not supported by adequate data on lifestyle and genetic factors. They have shown what we already know, that SOME cases of cancer are caused by random gene mutations. I believe that we will find that the majority of cases of cancer are associated with genetic cancers that run in families, unhealthful lifestyles, and exposure to factors that cause cancer, and that the more of these factors a person accumulates over a lifetime, the greater the cancer risk.
Since Tomasetti and Vogelstein do not claim to be proving that "cancer is caused by random chances of gene mutation," Mirkin's assertion that they have shown what we already know is a bit off-base. There is no reason for Tomasetti and Vogelstein to study "how environmental and genetic factors affect gene mutations," as Mirkin claims. The study created a statistical model that measures the proportion of cancer incidence, across many tissue types, caused mainly by random mutations that occur when stem cells divide. Mirkin states his opinion regarding the causes of cancer boldly and cautiously. He is cautious and uses the words 'associated with' rather than 'causes,' when listing what he thinks causes the majority of cancers.
Finally, Dr. Mirkin misrepresents the authors when he claims: "The authors did not include data on thyroid, breast, or prostate cancers because they did not fit into their model." It is true that the authors did not include data on all types of cancers, but it is not true that they intentionally did not include data that didn't fit into their model. "The scientists note that some cancers, such as breast and prostate cancer, were not included in the report because of their inability to find reliable stem cell division rates in the scientific literature. They hope that other scientists will help refine their statistical model by finding more precise stem cell division rates."
If someone as astute and knowledgeable as Dr. Gabe Mirkin can misread this report on cancer risk among tissues, it should be no surprise that journalists with little background in medicine or statistics, and with editors who know that the more a story is hyped the more likely it is that people will read it, would also misrepresent the scientists' work. The opening lines of Sarah Knapton's article are typical of how journalists hyped the report:
The majority of cancers are the result of bad luck rather than unhealthy lifestyles or inherited genetic faults, scientists have discovered.
For years health experts have warned that tumours are driven by a bad diet, lack of exercise, or gene errors passed down from parents.
The government even set up its ‘100,000 Genomes Project’ to try and find the genetic causes of many rare diseases and cancers.
But now a study has shown that most cancers are primarily caused by bad luck rather than poor lifestyle choices or defective DNA.
Researchers found that more than two thirds of cancers are driven by random mistakes in cell division which are completely outside of our control.
What the scientists discovered by statistical analysis suggests that random mutations play a large role in the development of cancer. They did not discover, nor did they claim to discover, that more than two-thirds of cancers are caused by random mutations. Nothing they discovered suggests that unhealthy lifestyles or inherited genetic faults do not result in cancer. The scientists write about what would be reasonable "if " two-thirds of cancer incidence across tissues is explained by random DNA mutations that occur when stem cells divide. The word 'if' seems to have no meaning to Knapton, for she writes (immediately before quoting Tomasetti): "The scientists claim that because it is impossible to prevent most types of cancer through behavioral changes or genetic screening, more should be done to speed up diagnosis so they can be spotted as early as possible."
Knapton writes as if she has discovered something Tomasetti and Vogelstein missed:
However health experts said the study demonstrates how important it is to help lower the risk of certain cancers by eating healthily, exercising and giving up smoking.
“While some genetic mistakes are due to bad luck, we know that our cancer risk depends on a combination of our genes, our environment and other aspects of our lives, many of which we can control,” said Dr. Emma Smith, senior science information officer at Cancer Research UK.
“We estimate that more than four in 10 cancers could be prevented by lifestyle changes, like not smoking, keeping a healthy weight, eating a healthy diet and cutting back on alcohol. “
Making these changes is not a guarantee against cancer, but it stacks the odds in our favour.
I'm sure Vogelstein and Tomasetti would agree. Nothing said by Dr. Smith in any way conflicts with their report.
To clarify their work, a week after the Johns Hopkins press release went out Tomasetti and Vogelstein added these comments:
Some have misunderstood our research to say that two-thirds of cancer cases are due to bad luck. We want to stress that cancer is caused by a combination of many factors..
...cancer is caused by a combination of factors—random DNA changes made during stem cell divisions that are not within our control, environmental exposures, and inherited gene mutations. As a result, there are many opportunities for cancer prevention. The best way to prevent some cancer types is by eliminating environmental factors and by changing lifestyles. This is known as primary prevention. Quitting smoking is one valuable example of primary prevention.
The best way to prevent deaths from other cancer types is to detect them and treat them early, while they are still curable. This is called secondary prevention. One of the important aspects of our research was to further highlight cancer types that could be best impacted by primary prevention versus secondary prevention.
It seems likely that much of the opposition to the Vogelstein and Tomasetti report is resistance to the idea that many, if not most, cancers are not preventable and are beyond our control. Dr. Mirkin's response seems clearly driven by his reluctance to admit that lifestyle and dietary changes cannot prevent most cancers. There is another side to consider, however, and the authors of the report bring it up in their addenda to the original press release:
What do you say to those who have been discouraged by your findings?
We are aware that the idea that a major contributing factor to cancer is beyond anyone's control can be jarring. This doesn't mean that cancer research should be stalled in any way. Quite the opposite—our research emphasizes the likelihood that more cancers will appear in the future simply because aging increases the number of stem cell divisions. Research on primary and secondary prevention, cancer treatment, and the biology of the disease is more important than ever.
By the same token, many people have found relief in this research. Cancer has a long history of stigmatization. Patients and family members frequently blame themselves, believing there was something they could have done to prevent their or their family member’s cancer. We have heard from many of these families and are pleased that our analysis could bring comfort and even lift the burden of guilt in those who have suffered the physical and emotional consequences of cancer.
Personally, it doesn't matter if anyone thinks I got cancer because I didn't eat enough turmeric or drink enough green tea with lemon or didn't exercise or meditate enough or take enough supplements. I don't feel stigmatized and have nothing but disdain for those who think people with cancer deserve it because of their sins or their failure to engage in enough primary prevention. Such people are just ignorant and their opinions are of no interest to me.
The claim that bothered me the most about the press release was the characterization of cancer cells as cells that grow unchecked. The implication is that these cells keep growing and growing, multiplying and multiplying until they destroy the lungs or liver or some other organ, thereby leading to the death of the host. This may be true in many cases but the most important quality of these cancer cells is that they don't die a natural death (apoptosis). If not killed, they will live forever. Some cancers grow very slowly and their hosts can live a "normal" life for many years even though they have cancer. Many people die without knowing they even had cancer. While I have no quarrel with those who recommend a healthy diet and exercise for those who have cancer, I know that sometimes the immune system needs a little help from the surgeon or the pharmacist or the nuclear radiologist.
For me, the most important take-away from the report is not that cancer prevention actions are a waste of time but that there is an urgent need to focus on early detection of cancer. Obviously, the earlier a cancer is detected, the more likely it is that it can be cured. What may not be obvious, however, is that many people suffer for years with a variety of symptoms that are misdiagnosed as non-cancerous, when in fact they are suffering because of some type of neuroendocrine cancer. Focusing on early detection, however, does not mean people should start smoking, working with loose asbestos, eat unhealthy foods until they are obese, and never exercise.
I guess another take-away might be that reading news articles by science journalists may be hazardous to your health.
Some readers who looked at the scientists' clarification added to the Johns Hopkins' press release may wonder why I make no mention of the analogy used: Getting cancer could be compared to getting into a car accident. I didn't make any comments about the analogy because analogies often confuse the very people they are aimed at helping understand your point. Analogies are used to illustrate points; they are also used to argue for points. Many readers confuse the two. Tomasetti and Vogelstein’s comparison of getting cancer to getting into a car accident is meant to illustrate their point about the roles of random mutations, heredity, and environment in cancer. Yet, I would bet the house that many readers will take their illustration as an argument and waste everybody's time by pointing out the many ways that getting cancer is not like getting into a car accident.
To those who, after reading the clarification, continue chanting the mantra correlation does not prove causation I can only say "yes, you are right, but remember: without correlation there is no causation."
Finally, I'm sorry if it seems like I singled out Sarah Knapton for criticism. Her article happened to be the first one I read that turned 'if' into 'is' and 'suggests' into 'causes'. Others include Sarah Bosely of The Guardian, James Gallagher of the BBC, Ben Brumfield of CNN, Helen Regan of Time, and many others. Geoffrey Kabot's article in Forbes is one of the better articles on the subject. Dr. David Gorski's article, as expected, is very thorough and the best I've read. I don't reference Gorski because, unfortunately, I wrote my piece before reading his.