Bad luck is primary cause of most cancer

[Greg Bernhardt]

Scientists: Random gene mutations -- 'bad luck' -- primary cause of most cancer
http://www.cnn.com/2015/01/02/health/cancer-random-mutation/index.html?hpt=hp_t2

Ever marvel at someone who smoked and still lived to be 90? Just plain good luck, researchers say. And those who live like Puritans and get cancer anyway?

That's bad luck -- and it's the primary cause of most cancer cases, says a Johns Hopkins research study.

Roughly two-thirds of cancers in adults can be attributed to random mutations in genes capable of driving cancer growth, two scientists who ran statistics on cancer cases said.

That may sound jaw-dropping. And Johns Hopkins anticipates that the study will change the way people think about cancer risk factors.

They also believe it could also lead to changes in the funding of cancer studies.

 

[Doug Huffman]

I wonder how this correlates with Pareto's Power Law Distribution. I note also my neighbor friend's book subtitle, Extinction: Bad Genes or Bad Luck? (David. M. Raup, Norton, 1992)

 

[Ygggdrasil]

Here's a link to the paper described in the CNN article:

Tomasetti and Vogelstein 2015 Variation in cancer risk among tissues can be explained by the number of stem cell divisions. Science 347: 78. doi:10.1126/science.1260825
Abstract:

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.

I'll probably have more thoughts after reading the paper more thoroughly, but the paper seem most interested in explaining the variation in cancer rates among different tissues. So mutations explain why, for example, colon cancer is more common than intestinal cancer. A more relevant statistic would be to explain the variation in cancer incidence among individuals, for example, why certain individuals get colon caner and others do not (which does not seem to be addressed by the study).

 

[Choppy]

It doesn't sound all that jaw dropping to me. In fact, it sounds more liek someone just actually bothered to do a quantitative study on something that's rather intuitive to someone working in the field. In principle, correlating lifetime risk of cancer with total numer of cell divisions would also explain why age is the single largest risk factor for developing most common cancers.

On an editotrial note, I have to express concern about use of the term "luck." There are a lot of people who believe that favourable outcomes in a random process can be influenced by all sorts of nonsense, and no lack of people who prey on those believers. I wonder if it would be better to use terminology such as "due to random, non-influenceable outcomes" ... although I doubt that would make for very catchy headlines.

 

[Ygggdrasil]

The study by Tomasetti and Vogelstein asks the question: why do certain body tissues get cancer more frequently than other body tissues? Their hypothesis is simple: tissues whose stem cells undergo more rounds of cell division are more prone to cancer because each additional round of cell division is an opportunity for cancer-causing mutations to arise (lets call this the replicative hypothesis; the authors somewhat problematically refer to this as "bad luck"). While this may seem obvious to those who study cancer, there are other plausible hypotheses, for example, that tissues with more exposure to environmental toxins (such as the digestive tract and the skin) might be more prone to cancer than sites more protected from carcinogens (like the brain). Through the quantitative analysis that the authors perform in their paper, they are able to quantify how much the replicative hypothesis explains the variation in cancer rates among tissues, and show that it is the major factor explaining these differences.

What does this study not tell us? In analyzing their data, the authors lumped data together data across different populations, so their analysis cannot (by design) tell us much about the variation in cancer incidence among different individuals. This is where some of the media coverage of the study has been problematic. For example, in the snippet of the CNN article that Greg cites, it suggests that "bad luck" is responsible for most of the variation in cancer incidence among individuals. This clearly is not true as it has been demonstrated in numerous studies that cancer rates differ greatly among different populations (e.g. smokers vs non-smokers, those with mutations in genes like BRCA1 and those without). Unless you believe that non-smokers are luckier than those who do not smoke, or that lung stem cells undergo more rounds of cell division in smokers, the "bad luck" or replicative hypotheses cannot explain why smokers get lung cancer more frequently than non-smokers. Indeed, in a critique of the media coverage surrounding the paper, writers for the Guardian take data from the paper to calculate that, in the case of lung cancer, 25% of cases in the US are due to "bad luck" and 75% of case are due to smoking.

That figure that I just quoted is quite amazing, isn't it. If everyone in the US stopped smoking, 75% of the cases of lung cancer diagnosed each year would disappear. But, this is not just the case for lung cancer. Research suggests that more than 4 in 10 cancer cases could be prevented by lifestyle changes. Whereas the misleading headlines that "bad luck" suggests that we can't do anything about cancer, the fact is that over 40% of new cancer diagnoses each year are preventable.

This is one of the areas where the research done by Tomasetti and Vogelstein will be most useful. In figure 2 of their paper, they identify which tissues get cancers at rates higher than expected from the number of cell divisions the tissue undergoes. Their analysis pulls out certain cancers known to have strong influences from inherited factors (e.g. colorectal and duodenal cancers among those with familiar adenomatous polyposis) and environmental factors (e.g. lung cancer among smokers and liver cancer among those with hepatitis C infections), along with a number of other cancers. This analysis can help identify for us the types of cancers for which we should be searching for new environmental causes. Similarly, Tomasetti and Vogelstein identify cancers occurring at rates similar to or lower than expected from the number of cell divisions, so preventative measures are unlikely to help reduce the instances of these cancers.

 

[Greg Bernhardt]

tissues whose stem cells undergo more rounds of cell division are more prone to cancer because each additional round of cell division is an opportunity for cancer-causing mutations to arise

Wouldn't this put weightlifters and in general fitness people in a higher cancer risk pot?

 

[jim mcnamara]

There are tissues that undergo cell division all the time. The esophagus is lined with a kind of disposable epithelium (cells that slough off and are constantly being replaced by cell division) layer. Contrast that with hypertrophy in the deltoid muscle (weightlifter). There is a lot more to hypertrophy than cell division. There is a profound change in protein synthesis, an upswing, for example. Neuronal pathways for activating the muscle that were seldom used become much more efficient. In other words, yes, there is some cell division, but a lot of the change is due to other "stuff".

So I would say 'no', because the rates of cell division do not compare, the esophagus wins (or loses depending on your point of view) hands down.

 

[Ygggdrasil]

Wouldn't this put weightlifters and in general fitness people in a higher cancer risk pot?

Probably not, for the reasons jim mentions in his post. However, the Science paper did not include any muscle cancers in their analysis. It is worth noting that being obese or overweight is a risk factor for cancer, though this is thought to be related to effects on metabolism rather than effects on cell division.

 

[Matt G]

I am not much of a physicist or biologist, but my impression is that the uncertainty at the quantum level is rarely expressed at the non quantum level. The speed of light and gravity have little uncertainty associated with them as far as I know, so are we to understand biology to be dfferent?

 

[Shinaolord]

I am not much of a physicist or biologist, but my impression is that the uncertainty at the quantum level is rarely expressed at the non quantum level. The speed of light and gravity have little uncertainty associated with them as far as I know, so are we to understand biology to be dfferent?

I would respond to this by simply saying nonlinear dynamics/chaos theory in complex systems, especially organisms, make things much more complex than this analysis suggests.

 

[Matt G]

But complexity cuts both ways. Those arguing that no environmental or genetic cause exists for a certain number of cancers must therefore assume their modeling systems are complex enough to reflect all the variables involved.

 

[Choppy]

I am not much of a physicist or biologist, but my impression is that the uncertainty at the quantum level is rarely expressed at the non quantum level. The speed of light and gravity have little uncertainty associated with them as far as I know, so are we to understand biology to be dfferent?

But complexity cuts both ways. Those arguing that no environmental or genetic cause exists for a certain number of cancers must therefore assume their modeling systems are complex enough to reflect all the variables involved.

I think if you read the paper, you'll find that what their evidence establishes is a strong correlation between the number of stem cell divisions in a given tissue and the lifetime risk of cancer within that tissue. It does not argue AGAINST environmental or genetic causes.

Instead, it goes on to separate a number of cancers into those that are more heavily influenced by environmental and genetic factors and those that are not. This can be very important information from the point of view of public health policy and making decisions on where to allocate prevention funds. Lung cancer in smokers is an example of a cancer that is strongly influenced by the act of smoking, so it makes sense to allocate public funds towards preventing young people from taking up smoking from a public health perspective. But from what I remember (I don't have the article in front of me right now) the work suggests that with some cancers, the correlation is so strong between random mutations arising from stem cell divisions and the appearance of cancer that worrying about other factors may not provide much benefit.

It's also important to think about how statistics work. You don't need to have a complex system model to tease information out of a complex system.

 

[Matt G]

I have no objection to your description. But the phrasing of 'bad luck, which another person here objected to, seems to create a different connotation

 

[Psinter]

As we have learned more and more about cancer ... we've come to realize that a number of cancers start purely because of mutations that happen that are just unexplainable. Bad luck is, unfortunately, the right way to explain it.

That comment from Dr. Otis Brawley gave me cancer. I do believe that's the same as saying: "God's will" is, unfortunately, the right way to explain it.
I do also hope they realize the magnitude of their lack-of-intelligence remarks and fix it in the future by using proper words that are clearly defined.

Misinforming people is the same as launching a full blown assault against them.

 

[Ygggdrasil]

Science just published a number of letters from a scientists discussing flaws and limitations of the "bad luck" causes cancer study. The letters can be found here for now. Many of the letters bring up good points. For example, http://www.sciencemag.org/content/early/2015/02/04/science.aaa6799.abs point out that the study does not analyze many common cancer types with major environmental/genetic causes(e.g. stomach, breast, and cervical cancers) and oversamples many rare cancers, such that overall the cancer sites included in the study account for only 34% of the cancer cases in the US. Others point out that there is http://www.sciencemag.org/content/early/2015/02/04/science.aaa6462.abs for some of the cancers that the authors classify as caused primarily by "bad luck."

In their http://www.sciencemag.org/content/early/2015/02/04/science.aaa6592.abs to the letters, the authors of the original study stand by their result, citing the figure from Cancer Research UK that 42% of cancers are preventable, in line with their estimate that ~65% of cancers are unavoidable.

 

[Evo]

Science just published a number of letters from a scientists discussing flaws and limitations of the "bad luck" causes cancer study. The letters can be found http://www.sciencemag.org/content/early/recent[/url for now. Many of the letters bring up good points. For example, Wild et al. point out that the study does not analyze many common cancer types with major environmental/genetic causes(e.g. stomach, breast, and cervical cancers) and oversamples many rare cancers, such that overall the cancer sites included in the study account for only 34% of the cancer cases in the US. Others point out that there is strong evidence for environmental causes for some of the cancers that the authors classify as caused primarily by "bad luck."

In their response to the letters, the authors of the original study stand by their result, citing the figure from Cancer Research UK that 42% of cancers are preventable, in line with their estimate that ~65% of cancers are unavoidable.

There is too much evidence that many cancers are genetic, there are markers that have been found. Unfortunately your links (except for the UK) do not work, they may require a subscription for access. The UK site is kind of a joke with the great majority of preventable cancer being attributed to not smoking. DUH, you think? What would the percent be if you backed out just the smoking number?

 

[thankz]

in 20 years or so I may try my hand at curing cancer lol

I now know that it is possible to make drugs that target mRNA during transcription in the ribosome to create attack vectors for pre programmed cell death, the tricky part is drug delivery across cell membranes and protein folding during codon duplication.

maybe I don't know enough or I'm just talking out of my behind, but it looks easy, someone tell me what it takes to make a cancer drug.

 

[Uollises]

So if you get cancer or not, it is a simply random, like dice? )

 

[audire]

This sprung up to mind when I read something like this.

http://en.wikipedia.org/wiki/Moral_luck

 

[gleem]

Genetics plays a big part in cancer incidence both directly and indirectly, there is no doubt about that. If mitotic activity is a key factor in cancer incidence then why aren't leukemia and other hematopoetic cancers much more common. Leukemia is more prevalent in children but much rarer in adults. Skin has very high mitotic activity but UV exposure is definitely linked to its incidence as they noted. Colon cancer is another question. Although higher mitotic activity than some other tissues it in addition is more exposed to carcinogen because the feces sits in the colon perhaps letting carcinogenic agent work on the tissue for a longer time. It has been assumed for decades that mitotic activity is key in cancer incidence because any genetic changes caused by endogenous or exogenous factors can be amplified through this process. So yes there are more cancers in these tissues. But how did they tell the difference between those cause by inherited genetic defects and those cause by carcinogenic agents which caused the mutations in the first place. This information is hardly available for analysis. So assumptions have to be made. I am very skeptical. And the term "luck" is a poor choice better chance. It seems that their hypothesis Is that the cancer rate should track with the mitotic rate except when it is higher it must be due to exogenous factors. This would still be true if exogenous factor where important in those cancers that tracked the mitotic rate according to accepted theory. Why not conclude that those tissues that had a higher incidence of cancer where somehow more sensitive to exogenous factors. We need another independent study I think.

 

[mathwonk]

As my mom liked to point out: 100% of smokers die of lung cancer - unless they die of something else first.

 

[Laroxe]

I think a problem in the discussion is the idea that cancer represents a single disease. there certainly are cancers that occur in the young, some are clearly linked with infections, and we may inherit certain predispositions. However generally cancer is a disease associated with ageing and it makes sense that the potential for a mistake in copying a cells DNA increases every time a cell divides, most such mistakes are irrelevant and others are removed by our immune system so it needs a combination of things to happen for cancer to develop. This doesn't mean that our environment has no part in this its bound to really, there are certain chemicals that we know directly effect the way DNA is copied, the tar of cigarette smoke contain several, some forms of radiation, most notably the UV in sunlight damages DNA, there are things in our diet that might increase or decrease risk, the efficiency of our immune system, which again can be influenced by a lot of things. So if your a betting person and like the idea of luck, most people will start off with fairly similar levels of risk but there are lots of things that might alter the odds, generally in a small way (smoking being an exception) one way or another.

 

[Ygggdrasil]

New research, published online yesterday in the journal Nature challenges the conclusion that bad luck causes the majority of cancers. Here's the paper:
Wu, Powers, Zhu & Hannun, 2015. Substantial contribution of extrinsic risk factors to cancer development. Nature, published online 16 Dec 2015. http://dx.doi.org/10.1038/nature16166

And here's my Insights article summarizing its findings:
https://www.physicsforums.com/insights/causes-cancer-bad-luck-bad-lifestyles/

 

[PeroK]

As someone with no medical knowledge but an interest in probability:

1) Unless you inject yourself with a 100% guaranteed cancer-causing chemical, then it's always bad luck if you get it. Increasing or decreasing your odds through lifestyle change doesn't mean it's not bad luck if you actually get cancer.

2) The more people who improve their lifestyle to avoid cancer, the higher will become the percentage of unavoidable cancer. If, eventually, everyone avoids the bad lifestyle choices, then 0% of cancers will be due to bad lifestyle. Which you can't then interpret as "lifestyle makes no difference, it's all down to luck".

3) The age you get cancer must be the critical concern. If the data showed that the average age of the "bad lifestyle" cancer victim was 55, say, and the average age of the "unlucky" cancer victim was 75, say, then there's still a major motivation to improve your lifestyle.

4) I can imagine that a good lifestyle may even increase your chances of getting cancer, but at a much later age, because you've avoided dying younger of all the avoidable causes.

The key question for me (with apologies to those focused on studying cancer) is:

If you smoke, drink and eat processed foods to excess and take no exercise, what is your life expectancy? Likewise, if you do the opposite? The percentage chance that it's cancer that gets you (eventually) is almost irrelevant. Age is everything.

 

[Ygggdrasil]

As someone with no medical knowledge but an interest in probability:

1) Unless you inject yourself with a 100% guaranteed cancer-causing chemical, then it's always bad luck if you get it. Increasing or decreasing your odds through lifestyle change doesn't mean it's not bad luck if you actually get cancer.

2) The more people who improve their lifestyle to avoid cancer, the higher will become the percentage of unavoidable cancer. If, eventually, everyone avoids the bad lifestyle choices, then 0% of cancers will be due to bad lifestyle. Which you can't then interpret as "lifestyle makes no difference, it's all down to luck".

3) The age you get cancer must be the critical concern. If the data showed that the average age of the "bad lifestyle" cancer victim was 55, say, and the average age of the "unlucky" cancer victim was 75, say, then there's still a major motivation to improve your lifestyle.

4) I can imagine that a good lifestyle may even increase your chances of getting cancer, but at a much later age, because you've avoided dying younger of all the avoidable causes.

The key question for me (with apologies to those focused on studying cancer) is:

If you smoke, drink and eat processed foods to excess and take no exercise, what is your life expectancy? Likewise, if you do the opposite? The percentage chance that it's cancer that gets you (eventually) is almost irrelevant. Age is everything.

All good points. I agree that the age at which you get cancer is the critical factor. At the end of the article, I link to a blog post from Cancer Research UK that gets to the question you asked. The post covers a publication in PLOS Medicine arising from the the European Prospective Investigation into Cancer and Nutrition (EPIC) study, a large, Europe-wide study of the effects on diet on cancer. In the study, they follow 20,000 healthy individuals over the course of 11 years. They assign each participant in the study a health score of zero to four based on whether or not they smoked, kept physically active, had a moderate alcohol intake and ate at least five portions of fruit and vegetables per day (as assessed by a blood test for vitamin C). Here's what they found:

Compared to people who ticked all four healthy boxes, those who scored 0 were four times more likely to have died within the 11-year period. They had a higher risk of dying from cancer and an even higher risk of dying from heart disease. Even people who scored 2 out of 4 were twice as likely to have died.

What’s more, the study found that people with scores of 0 had the same risk of dying by the end of the study as people with scores of 4 who were 14 years older. By making unhealthy choices, they had effectively lost a substantial 14 years of life.

While this doesn't directly answer the question of life expectancy, it does support the notion that a healthy lifestyle can delay mortality and the onset of diseases like cancer or heart disease.

 

[Laroxe]

I suppose the issue becomes, what constitutes a healthy lifestyle, and this is no easy question to answer. Within the population there is considerable variation in people which is a product of their genes and the earlier environment, the process of living exposes us to a wide range things which impact on us. There are a number of things which can clearly be identified as carcinogenic but even in these the degree of risk is highly variable, however it makes good sense to avoid things like smoking and asbestos. The trouble is that once you get beyond the relatively few agents that present the highest risks trying to generalise lifestyle choices is difficult. Virtually everything we eat has a potential for harm, the people who believe that natural foods or plant foods are all wholesome and healthy don't appear to understand that plants don't like being eaten either and they are engaged in long term chemical warfare with their predators, us included. The main reason that our liver is so big is that it has to constantly deal with the poisons that plants are trying to kill us with. You only have to look at dietary advice to notice their is a problem, saturated fats are no worse than others and may be better than some, certain food preservation techniques increase carcinogens, don't preserve and you die of infections, dietary fibre while it keeps you regular protects you from nothing and you can kill yourself with vitamin's. Even the bogeyman of stress, in reality, seems irrelevant. People have been predicting an apocalypses of heart disease based on lifestyle & obesity, instead, the incidence of heart disease is dropping.
We simply don't really have the information about the relationship between individuals and their environment to offer reliable advice about healthy living and the quality of the evidence hasn't really moved on from the seven deadly sins.