7 ms·

Bad luck, bad journalism and cancer rates

188 points by auton1 12y ago

toufka 12y ago
Here's the deal - we know it's 'chance' (not really luck) which part of your DNA is damaged by any given environmental effect. However, some DNA is more important than others. In general, a little damage is not a problem and is easily 'taken care of' by the cell which caries that DNA - either by literally repairing the DNA, silencing that DNA's function, killing itself, or asking other cells to help kill it. We know quite clearly that cancer is (generally) caused by a set of mutations - not necessarily in an order, but some orders are not successful. There are four or five genes which keep social order amongst the other genes. If you silence all of these, you get cancer. Chance has it's role in the roll of dice for which DNA gets damaged, but all the other parameters can be changed too - how many sided the dice are, how often the dice get rolled, and whether all the cells in the same tissue have correlated dice-rolls. Cells that deviate from what they're supposed to do are either (in order), repaired, silenced, voluntarily commit suicide, or are killed. There are proteins (genes) that are the final judges for each of these processes - and have 'go, no-go' power. Only if all of these judges are killed do you get a cell that can do anything it wants - like replicate uncontrollably to the detriment of the host ('cancer'). Thus the statistics of getting cancer roughly follow the idea that you have to get random DNA modifications of those exact 5 genes, in a single cell. Lots of things can increase your random modification rate (UV, smoke, radiation, etc). Some of these things correlate though - and again, what hurts one cell, might hurt its neighbor just as bad. They're not entirely independent events. For example, losing your DNA repair machinery (this is what HPV does - it silences your DNA repair machinery) amps up the baseline mutation rate and makes further mutations more likely (dependent correlations then arise). The Brca gene that has caused so much controversy in patent law (whether a test for its existence could be patented) and indicates whether a person might or might be susceptible to breast cancer, is the master repair technician of the cell. In people who have this gene in working order, the Brca gene signs off on whether the cell is in need of repair. But if the Brca is not it working order, cells that are in need of repair might not get it, and instead are allowed to more freely operate under non-optimal internal conditions. If you are missing or have a mutated version of Brca, you are missing one of the checkpoint processes. So again, we quite clearly know of a handful of genes which do most of the master regulation of a cell's job - and if these jobs go unfulfilled - by having their blueprints be damaged by the environment - you have fewer and fewer mechanisms to prevent that single cell from runaway growth.
- otoburb 12y ago
  This was a great overview. Thanks very much for the clear exposition!
  - auggierose 12y ago
    Yep. Exactly that.
- Lost_BiomedE 12y ago
  Also dietary magnesium and other micronutrients are an important factor here: Role of magnesium in genomic stability http://www.sciencedirect.com/science/article/pii/S0027510701000744 http://www.sciencedirect.com/science/article/pii/S0027510701... And most in the U.S. are deficient: 'Magnesium intakes for ~56% of adults in the United States are below the Estimated Average Requirement (EAR), the current measure of micronutrient inadequacy (the RDA is set at 2 standard deviations above the EAR).' This is from a Bruce Ames paper here: http://www.hindawi.com/journals/jna/2010/725071/ http://www.hindawi.com/journals/jna/2010/725071/ The Ames paper is worth the read through. Including this one on magnesium, 'In a study of 4,035 men followed for 18 years, the highest quartile with serum magnesium at baseline compared with the lowest had a 40% decrease in all-cause mortality and cardiovascular disease and a 50% decrease in cancer deaths'. He goes through other micronutrients as well.
  - IndianAstronaut 12y ago
    Interesting. I wonder if they have studied people that regularly take rolaids.
- fideloper 12y ago
  Would I be incorrect in saying that cells can replicate imperfectly without being cancer? Isn't that aging?
  - XorNot 12y ago
    No. Aging is a lot of things, but it's not generalized cellular breakdown. Aging is more associated with the gradual failure of larger systems and frameworks in the body in ways that evolution has had no incentive to effect a repair strategy for. i.e. wrinkles are associated with the loosening of the collagen support framework for the skin. This isn't cellular breakdown - it's just a system which there was never any point evolving a repair mechanism for beyond childhood.
dang 12y ago
Also https://news.ycombinator.com/item?id=8827666 https://news.ycombinator.com/item?id=8827666.
ekianjo 12y ago
Wait, is this an article from the Guardian making fun of their own article on the same study earlier on ? See http://www.theguardian.com/society/2015/jan/01/two-thirds-cancer-cases-caused-bad-luck-lifestyle-genes http://www.theguardian.com/society/2015/jan/01/two-thirds-ca... ? If that's the case, that's a novel way of doing journalism: publish crap first, then sell more paper by doing a critique of your crap.
- jonnathanson 12y ago
  "If that's the case, that's a novel way of doing journalism: publish crap first, then sell more paper by doing a critique of your crap." I wouldn't assume it was an intentional strategy. Papers and magazines have never been shy about critiquing themselves. (At least, the better ones haven't.) Back in the day, it wasn't uncommon to see an editor at the New York Times take a staff reporter or columnist to task in a sidebar, or even in a followup story. And the role of ombudsman used to mean something. (Whereas today, it's a very underappreciated and little-understood role.) Second, we shouldn't necessarily presume that writers at the same paper are coordinating with each other. Say what you will about that as a business practice. But it's a fact of life at many publications, with distributed staffs, section-specific editors, shrinking numbers of full-timers, and ever-increasing numbers of independent contractors. I wouldn't attribute to malice what can better be explained by uncoordinated action. [EDIT: This comment got a downvote, and I'm guessing it's because someone assumed I'm defending the Guardian here. Just to be clear, I'm not. When I say that people aren't always coordinating, I'm not offering that as a defense of lack of coordination. I'm offering it as a possible explanation. Clearly, a lack of coordination is a bad thing.]
  - ekianjo 12y ago
    > . I wouldn't attribute to malice what can better be explained by uncoordinated action. Don't they do peer reviews before publishing articles to the eyes of millions ?
    - jonnathanson 12y ago
      In theory, this is the function of the editor -- be it a section editor, a managing editor, or someone else who bears the title "editor." In practice, editors have been very hard hit by the journalism crunch. There are still plenty of fantastic editors out there; they're just not getting hired, kept on, or appropriately compensated for their contribution anymore. Editors' heads are often the first on the guillotine when the cost-cutting rounds begin. The editors who survive the purge are often wildly overstretched. Others are extremely green--college kids, hired for a third of what an experienced editor gets paid, under the misguided assumption that anyone with a journalism degree can successfully fulfill the responsibilities of an editor. (For what it's worth: I don't know anyone on the Guardian's staff, and am making no specific accusations here.)
termostaatti 12y ago
Naturally the way one lives has an impact whether potentially he/she will get cancer. Smoking, drinking, eating certain types of food (manufactured wrong) etc. all have direct relation on cancer. It's never just luck who happens to get these horrible diseases.
- hueving 12y ago
  Please provide evidence when you say these kinds of things. Causation is rarely as intuitive as you would like to believe.
  - termostaatti 12y ago
    Well, carcinogen is one those things that has been known for decades causing cancer. And yes, human-being can make the difference whether they have those unnatural carcinogens or not.
    - michaelhoffman 12y ago
      There are plenty of "natural" carcinogens too. Avoiding carcinogens may reduce your cancer risk, restricting yourself to things that are "natural" won't.
  - IndianAstronaut 12y ago
    Waiting for a strong causal link, especially when there is a high level of statistical relationships, can mean we might be waiting too long to take serious action. If repeated studies show strong associations, like say between cigarettes and cancer, it would make sense to take some steps to prevent the potential cause.
  - TheSpiceIsLife 12y ago
    I think that's a bit unfair. Sure, we can play semantics and claim it's technically incorrect to say "smoking causes cancer" because if it did everyone who smokes would get cancer, and that isn't the case. However, we are aware of quite a few known, and many probable, carcinogens.[1][2] Among women, breast cancer comprises 60% of alcohol-attributable cancers.[3] And to support the claim that certain foods, when prepared in certain ways, can become carcinogenic take a look at the research on Heterocyclic amines,[4] and acrylamide.[5] 1. http://www.cancer.org/cancer/cancercauses/othercarcinogens/generalinformationaboutcarcinogens/known-and-probable-human-carcinogens http://www.cancer.org/cancer/cancercauses/othercarcinogens/g... 2. http://en.wikipedia.org/wiki/List_of_IARC_Group_1_carcinogens http://en.wikipedia.org/wiki/List_of_IARC_Group_1_carcinogen... 3. http://onlinelibrary.wiley.com/doi/10.1002/ijc.21903/abstract http://onlinelibrary.wiley.com/doi/10.1002/ijc.21903/abstrac... 4. http://en.wikipedia.org/wiki/Heterocyclic_amine#Heterocyclic_amines_and_cancer http://en.wikipedia.org/wiki/Heterocyclic_amine#Heterocyclic... 5. http://en.wikipedia.org/wiki/Acrylamide#Toxicity_and_carcinogenicity http://en.wikipedia.org/wiki/Acrylamide#Toxicity_and_carcino...
therealdrag0 12y ago
I want to reiterate the articles recommendation of the book "Bad Science" by Ben Goldacre [0]. I learned a lot from it and found it enjoyable--and it's available in Audiobook form. [0] http://smile.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/0865479186/ http://smile.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/...
MisterMashable 12y ago
Whatever the official cause of cancer it will never be food additives, processed or poor quality foods, prescription drugs, air pollution, contaminants in the water supply, contaminated vaccines (I'm a believer in vaccines so don't go there!), new car smells, pesticide residues in food and clothing, fire retardant chemicals, xenoestrogens, secondary cancers caused by chemotherapy or radiation... it will just be bad luck.
waldrews 12y ago
Oy vey. Statistician/statistical geneticist here. The second article is too harsh on the first. The point is subtle and hard to express: carcinogens set the odds, but importance of the part of the risk due to "biological luck," the part that comes from the mutation chain, means whether you individually are affected is random in a way that makes it hard to bother protecting yourself. Let's say (arbitrary numbers) having poisonium in the food supply increases the lifetime population rate of cancer of the thingamajig from 0.11% to 0.12%. In the US, banning poisonium will save 300 million * (0.12%-0.11%) = 30,000 people, very predictable and important. However, if you personally go on a poisonium-free diet, you decrease your personal risk by 0.01%. You chances go from about 1 in 1000 to about 1 in 1000... so maybe if it involves any effort, you should spend your time doing something else to improve your health. It's different if the carcinogens targeted specific vulnerable groups. If there was one gene that makes you instantly get cancer at the first whiff of poisonium, we would screen everyone for that gene. The difference between the 0.12 and 0.11 would be due that gene. Everyone with that gene would be put on poisonium-free diets. Everyone else can eat all the poisonium they want, and would still have 0.11 risk. (This is more or less the situation with the active ingredient in aspartame sweetener, though the disease is not cancer) I don't like the use of the word "luck" in these descriptions, since both genetic and mutation chain randomness are a kind of luck. Also, the studies give quality numbers across many cancers, but the basic concept has been a mainstream model of cancer risk for a long time.
- vasilipupkin 12y ago
  this makes more sense to me than the second article, thxs
- Retric 12y ago
  I think your misrepresenting the data. Absolutely nothing about the data collected related to different human populations just different types of tissue. Further, I don't see anything that relates to % of biomass of the tissue just total cell divisions. Presumably the skin, lungs, a digestive track tissues are at high risks because they compromise a lot of tissue, it divides rapidly, and it's not protected from environmental factors ex: sunlight/smoking/spicy food. However, I suspect the paper digs into things a little further to account for such factors.
vasilipupkin 12y ago
the original paper is kind of confusing itself. Why would they essentially do a 1 variable regression? why not do a multiple regression with lifetime cancer risk as y variable and both number of cell divisions and smoking as x variables?
Gatsky 12y ago
I was quite surprised to find this in one of the most prestigious journals in the world. Vogelstein is one of the biggest names in cancer research, and no doubt that was the most important factor in its publication. Some problems: 1. They leave out breast cancer and prostate cancer, two of the most common cancers out there. Most of the dots on their plot relate to a tiny proportion of cancers that people get. Furthermore, breast cancer has a very low mutation rate, which also casts doubt on their hypothesis about accummulation of somatic mutations on stem cells. 2. I ran their data using just the number of stem cells and the number of cells versus incidence. The correlations are still good, about 0.5 for just cell mass and 0.67 for number of stem cells. It's hardly a revelation that a larger more cellular organ is more likely to get cancer. 3. For glioblastoma, a type of brain tumour, their data says that there are 0 stem cell divisions, because it is thought there is no cell renewal in the brain (which isn't even true, but anyway). This contradicts the whole hypothesis, but they still include it in there. 4. The lifetime incidence of various cancers is age dependent. For example sarcomas almost always occur in the young, medulloblastoma is vanishingly rare in the elderly, testicular cancer occurs in young men. Their hypothesis ignores all this, because their model only allows for the acquisition of mutations over time, which should give a unimodal relationship between age and cancer incidence. 5. There are cancers that have almost no mutations. Gene copy number changes are just as important if not more so than mutations. There is very little evidence that copy number changes accumulate over time in stem cells in a random fashion required by this model. I don't think this warrants much attention in the end. This can join the 100,000 other unverified (or unverifiable correlations) in the cancer literature.
leg100 12y ago
What's the difference between bad luck and "we just don't know"?
- andybak 12y ago
  Possibly the difference between knowing something is random and not knowing anything. Of course 'random' is a fairly complex topic in itself but it makes sense to consider it relative to the problem domain.
- UhUhUhUh 12y ago
  Exactly. The same old debate between statistical truth and deterministic uncertainty. What bothers me is that, meanwhile, the many people who have tried to get funding for years or even generations to study deep, complex mechanisms involved in DNA replication (on histones for example) didn't get much. Of course replication will eventually, on large numbers, begin to fail. Microprocessors do too. But would we say that when instructions get contaminated it's bad luck?

188 points by auton1 12y ago

Here's the deal - we know it's 'chance' (not really luck) which part of your DNA is damaged by any given environmental effect. However, some DNA is more important than others. In general, a little damage is not a problem and is easily 'taken care of' by the cell which caries that DNA - either by literally repairing the DNA, silencing that DNA's function, killing itself, or asking other cells to help kill it. We know quite clearly that cancer is (generally) caused by a set of mutations - not necessarily in an order, but some orders are not successful. There are four or five genes which keep social order amongst the other genes. If you silence all of these, you get cancer. Chance has it's role in the roll of dice for which DNA gets damaged, but all the other parameters can be changed too - how many sided the dice are, how often the dice get rolled, and whether all the cells in the same tissue have correlated dice-rolls. Cells that deviate from what they're supposed to do are either (in order), repaired, silenced, voluntarily commit suicide, or are killed. There are proteins (genes) that are the final judges for each of these processes - and have 'go, no-go' power. Only if all of these judges are killed do you get a cell that can do anything it wants - like replicate uncontrollably to the detriment of the host ('cancer'). Thus the statistics of getting cancer roughly follow the idea that you have to get random DNA modifications of those exact 5 genes, in a single cell. Lots of things can increase your random modification rate (UV, smoke, radiation, etc). Some of these things correlate though - and again, what hurts one cell, might hurt its neighbor just as bad. They're not entirely independent events. For example, losing your DNA repair machinery (this is what HPV does - it silences your DNA repair machinery) amps up the baseline mutation rate and makes further mutations more likely (dependent correlations then arise). The Brca gene that has caused so much controversy in patent law (whether a test for its existence could be patented) and indicates whether a person might or might be susceptible to breast cancer, is the master repair technician of the cell. In people who have this gene in working order, the Brca gene signs off on whether the cell is in need of repair. But if the Brca is not it working order, cells that are in need of repair might not get it, and instead are allowed to more freely operate under non-optimal internal conditions. If you are missing or have a mutated version of Brca, you are missing one of the checkpoint processes. So again, we quite clearly know of a handful of genes which do most of the master regulation of a cell's job - and if these jobs go unfulfilled - by having their blueprints be damaged by the environment - you have fewer and fewer mechanisms to prevent that single cell from runaway growth.

otoburb 12y ago
This was a great overview. Thanks very much for the clear exposition!
- auggierose 12y ago
  Yep. Exactly that.
Lost_BiomedE 12y ago
Also dietary magnesium and other micronutrients are an important factor here: Role of magnesium in genomic stability http://www.sciencedirect.com/science/article/pii/S0027510701000744 http://www.sciencedirect.com/science/article/pii/S0027510701... And most in the U.S. are deficient: 'Magnesium intakes for ~56% of adults in the United States are below the Estimated Average Requirement (EAR), the current measure of micronutrient inadequacy (the RDA is set at 2 standard deviations above the EAR).' This is from a Bruce Ames paper here: http://www.hindawi.com/journals/jna/2010/725071/ http://www.hindawi.com/journals/jna/2010/725071/ The Ames paper is worth the read through. Including this one on magnesium, 'In a study of 4,035 men followed for 18 years, the highest quartile with serum magnesium at baseline compared with the lowest had a 40% decrease in all-cause mortality and cardiovascular disease and a 50% decrease in cancer deaths'. He goes through other micronutrients as well.
- IndianAstronaut 12y ago
  Interesting. I wonder if they have studied people that regularly take rolaids.
fideloper 12y ago
Would I be incorrect in saying that cells can replicate imperfectly without being cancer? Isn't that aging?
- XorNot 12y ago
  No. Aging is a lot of things, but it's not generalized cellular breakdown. Aging is more associated with the gradual failure of larger systems and frameworks in the body in ways that evolution has had no incentive to effect a repair strategy for. i.e. wrinkles are associated with the loosening of the collagen support framework for the skin. This isn't cellular breakdown - it's just a system which there was never any point evolving a repair mechanism for beyond childhood.

dang 12y ago

Also https://news.ycombinator.com/item?id=8827666 https://news.ycombinator.com/item?id=8827666.

ekianjo 12y ago

Wait, is this an article from the Guardian making fun of their own article on the same study earlier on ? See http://www.theguardian.com/society/2015/jan/01/two-thirds-cancer-cases-caused-bad-luck-lifestyle-genes http://www.theguardian.com/society/2015/jan/01/two-thirds-ca... ? If that's the case, that's a novel way of doing journalism: publish crap first, then sell more paper by doing a critique of your crap.

jonnathanson 12y ago
"If that's the case, that's a novel way of doing journalism: publish crap first, then sell more paper by doing a critique of your crap." I wouldn't assume it was an intentional strategy. Papers and magazines have never been shy about critiquing themselves. (At least, the better ones haven't.) Back in the day, it wasn't uncommon to see an editor at the New York Times take a staff reporter or columnist to task in a sidebar, or even in a followup story. And the role of ombudsman used to mean something. (Whereas today, it's a very underappreciated and little-understood role.) Second, we shouldn't necessarily presume that writers at the same paper are coordinating with each other. Say what you will about that as a business practice. But it's a fact of life at many publications, with distributed staffs, section-specific editors, shrinking numbers of full-timers, and ever-increasing numbers of independent contractors. I wouldn't attribute to malice what can better be explained by uncoordinated action. [EDIT: This comment got a downvote, and I'm guessing it's because someone assumed I'm defending the Guardian here. Just to be clear, I'm not. When I say that people aren't always coordinating, I'm not offering that as a defense of lack of coordination. I'm offering it as a possible explanation. Clearly, a lack of coordination is a bad thing.]
- ekianjo 12y ago
  > . I wouldn't attribute to malice what can better be explained by uncoordinated action. Don't they do peer reviews before publishing articles to the eyes of millions ?
  - jonnathanson 12y ago
    In theory, this is the function of the editor -- be it a section editor, a managing editor, or someone else who bears the title "editor." In practice, editors have been very hard hit by the journalism crunch. There are still plenty of fantastic editors out there; they're just not getting hired, kept on, or appropriately compensated for their contribution anymore. Editors' heads are often the first on the guillotine when the cost-cutting rounds begin. The editors who survive the purge are often wildly overstretched. Others are extremely green--college kids, hired for a third of what an experienced editor gets paid, under the misguided assumption that anyone with a journalism degree can successfully fulfill the responsibilities of an editor. (For what it's worth: I don't know anyone on the Guardian's staff, and am making no specific accusations here.)

termostaatti 12y ago

Naturally the way one lives has an impact whether potentially he/she will get cancer. Smoking, drinking, eating certain types of food (manufactured wrong) etc. all have direct relation on cancer. It's never just luck who happens to get these horrible diseases.

hueving 12y ago
Please provide evidence when you say these kinds of things. Causation is rarely as intuitive as you would like to believe.
- termostaatti 12y ago
  Well, carcinogen is one those things that has been known for decades causing cancer. And yes, human-being can make the difference whether they have those unnatural carcinogens or not.
  - michaelhoffman 12y ago
    There are plenty of "natural" carcinogens too. Avoiding carcinogens may reduce your cancer risk, restricting yourself to things that are "natural" won't.
- IndianAstronaut 12y ago
  Waiting for a strong causal link, especially when there is a high level of statistical relationships, can mean we might be waiting too long to take serious action. If repeated studies show strong associations, like say between cigarettes and cancer, it would make sense to take some steps to prevent the potential cause.
- TheSpiceIsLife 12y ago
  I think that's a bit unfair. Sure, we can play semantics and claim it's technically incorrect to say "smoking causes cancer" because if it did everyone who smokes would get cancer, and that isn't the case. However, we are aware of quite a few known, and many probable, carcinogens.[1][2] Among women, breast cancer comprises 60% of alcohol-attributable cancers.[3] And to support the claim that certain foods, when prepared in certain ways, can become carcinogenic take a look at the research on Heterocyclic amines,[4] and acrylamide.[5] 1. http://www.cancer.org/cancer/cancercauses/othercarcinogens/generalinformationaboutcarcinogens/known-and-probable-human-carcinogens http://www.cancer.org/cancer/cancercauses/othercarcinogens/g... 2. http://en.wikipedia.org/wiki/List_of_IARC_Group_1_carcinogens http://en.wikipedia.org/wiki/List_of_IARC_Group_1_carcinogen... 3. http://onlinelibrary.wiley.com/doi/10.1002/ijc.21903/abstract http://onlinelibrary.wiley.com/doi/10.1002/ijc.21903/abstrac... 4. http://en.wikipedia.org/wiki/Heterocyclic_amine#Heterocyclic_amines_and_cancer http://en.wikipedia.org/wiki/Heterocyclic_amine#Heterocyclic... 5. http://en.wikipedia.org/wiki/Acrylamide#Toxicity_and_carcinogenicity http://en.wikipedia.org/wiki/Acrylamide#Toxicity_and_carcino...

therealdrag0 12y ago

I want to reiterate the articles recommendation of the book "Bad Science" by Ben Goldacre [0]. I learned a lot from it and found it enjoyable--and it's available in Audiobook form. [0] http://smile.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/0865479186/ http://smile.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/...

MisterMashable 12y ago

Whatever the official cause of cancer it will never be food additives, processed or poor quality foods, prescription drugs, air pollution, contaminants in the water supply, contaminated vaccines (I'm a believer in vaccines so don't go there!), new car smells, pesticide residues in food and clothing, fire retardant chemicals, xenoestrogens, secondary cancers caused by chemotherapy or radiation... it will just be bad luck.

waldrews 12y ago

Oy vey. Statistician/statistical geneticist here. The second article is too harsh on the first. The point is subtle and hard to express: carcinogens set the odds, but importance of the part of the risk due to "biological luck," the part that comes from the mutation chain, means whether you individually are affected is random in a way that makes it hard to bother protecting yourself. Let's say (arbitrary numbers) having poisonium in the food supply increases the lifetime population rate of cancer of the thingamajig from 0.11% to 0.12%. In the US, banning poisonium will save 300 million * (0.12%-0.11%) = 30,000 people, very predictable and important. However, if you personally go on a poisonium-free diet, you decrease your personal risk by 0.01%. You chances go from about 1 in 1000 to about 1 in 1000... so maybe if it involves any effort, you should spend your time doing something else to improve your health. It's different if the carcinogens targeted specific vulnerable groups. If there was one gene that makes you instantly get cancer at the first whiff of poisonium, we would screen everyone for that gene. The difference between the 0.12 and 0.11 would be due that gene. Everyone with that gene would be put on poisonium-free diets. Everyone else can eat all the poisonium they want, and would still have 0.11 risk. (This is more or less the situation with the active ingredient in aspartame sweetener, though the disease is not cancer) I don't like the use of the word "luck" in these descriptions, since both genetic and mutation chain randomness are a kind of luck. Also, the studies give quality numbers across many cancers, but the basic concept has been a mainstream model of cancer risk for a long time.

vasilipupkin 12y ago
this makes more sense to me than the second article, thxs
Retric 12y ago
I think your misrepresenting the data. Absolutely nothing about the data collected related to different human populations just different types of tissue. Further, I don't see anything that relates to % of biomass of the tissue just total cell divisions. Presumably the skin, lungs, a digestive track tissues are at high risks because they compromise a lot of tissue, it divides rapidly, and it's not protected from environmental factors ex: sunlight/smoking/spicy food. However, I suspect the paper digs into things a little further to account for such factors.

vasilipupkin 12y ago

the original paper is kind of confusing itself. Why would they essentially do a 1 variable regression? why not do a multiple regression with lifetime cancer risk as y variable and both number of cell divisions and smoking as x variables?

Gatsky 12y ago

I was quite surprised to find this in one of the most prestigious journals in the world. Vogelstein is one of the biggest names in cancer research, and no doubt that was the most important factor in its publication. Some problems: 1. They leave out breast cancer and prostate cancer, two of the most common cancers out there. Most of the dots on their plot relate to a tiny proportion of cancers that people get. Furthermore, breast cancer has a very low mutation rate, which also casts doubt on their hypothesis about accummulation of somatic mutations on stem cells. 2. I ran their data using just the number of stem cells and the number of cells versus incidence. The correlations are still good, about 0.5 for just cell mass and 0.67 for number of stem cells. It's hardly a revelation that a larger more cellular organ is more likely to get cancer. 3. For glioblastoma, a type of brain tumour, their data says that there are 0 stem cell divisions, because it is thought there is no cell renewal in the brain (which isn't even true, but anyway). This contradicts the whole hypothesis, but they still include it in there. 4. The lifetime incidence of various cancers is age dependent. For example sarcomas almost always occur in the young, medulloblastoma is vanishingly rare in the elderly, testicular cancer occurs in young men. Their hypothesis ignores all this, because their model only allows for the acquisition of mutations over time, which should give a unimodal relationship between age and cancer incidence. 5. There are cancers that have almost no mutations. Gene copy number changes are just as important if not more so than mutations. There is very little evidence that copy number changes accumulate over time in stem cells in a random fashion required by this model. I don't think this warrants much attention in the end. This can join the 100,000 other unverified (or unverifiable correlations) in the cancer literature.

leg100 12y ago

What's the difference between bad luck and "we just don't know"?

andybak 12y ago
Possibly the difference between knowing something is random and not knowing anything. Of course 'random' is a fairly complex topic in itself but it makes sense to consider it relative to the problem domain.
UhUhUhUh 12y ago
Exactly. The same old debate between statistical truth and deterministic uncertainty. What bothers me is that, meanwhile, the many people who have tried to get funding for years or even generations to study deep, complex mechanisms involved in DNA replication (on histones for example) didn't get much. Of course replication will eventually, on large numbers, begin to fail. Microprocessors do too. But would we say that when instructions get contaminated it's bad luck?