From Wikipedia, the free encyclopedia
Jump to: navigation, search
Former good article Cancer was one of the Natural sciences good articles, but it has been removed from the list. There are suggestions below for improving the article to meet the good article criteria. Once these issues have been addressed, the article can be renominated. Editors may also seek a reassessment of the decision if they believe there was a mistake.
          This article is of interest to the following WikiProjects:
WikiProject Genetics (Rated B-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Genetics, a collaborative effort to improve the coverage of Genetics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
WikiProject Molecular and Cellular Biology (Rated B-class, Top-importance)
WikiProject icon This article is within the scope of the WikiProject Molecular and Cellular Biology. To participate, visit the WikiProject for more information.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
WikiProject Medicine / Hematology-oncology / Translation / Pathology (Rated B-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Medicine, which recommends that this article follow the Manual of Style for medicine-related articles and use high-quality medical sources. Please visit the project page for details or ask questions at Wikipedia talk:WikiProject Medicine.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
Taskforce icon
This article is supported by the Hematology-oncology task force (marked as Top-importance).
Taskforce icon
This article is supported by the Translation task force (marked as High-importance).
Taskforce icon
This article is supported by the Pathology task force (marked as High-importance).
Note icon
This article was previously a Medicine Collaboration of the Month.
Wikipedia Version 1.0 Editorial Team / v0.5 / Vital / Supplemental
WikiProject icon This article has been reviewed by the Version 1.0 Editorial Team.
Taskforce icon
This article has been selected for Version 0.5 and subsequent release versions of Wikipedia.

To Maintaining Low Blood Sugar as a New Cancer Treatment[edit]

— Preceding unsigned comment added by Evilstriver (talkcontribs) 08:21, 23 September 2009‎

Semi-protected edit request on 29 December 2014[edit]

Cancer 2nd & 3rd stage purely recovery stories Subhankariam (talk) 14:27, 29 December 2014 (UTC)

Red question icon with gradient background.svg Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format. I am assuming you wish this youtube link to be added as an external link? If so, by what rationale? There are thousands of other videos of cancer patients recovering, what makes this one special? Cannolis (talk) 14:32, 29 December 2014 (UTC)

Recent study large role for 'bad luck'[edit]

The introductory paragraph suggests that Cancer can be prevented by a healthy lifestyle, which is incorrect. It should state something that the chance of developing cancer is influenced by lifestyle. However a recent study[1] showed that the number of stem cell divisions can explain why certain organ tissues are more likely to give rise to cancer, i.e. the more often the stem cells divide, the higher the chance of a mutation that leads to cancer can occur. This study claimed that 65% of cancer cases can be explained by bad luck. Just to clarify, this has nothing to do with genetic predisposition to develop certain types of cancer. Suggestions on how to implement this in the introductory paragraph? — Preceding unsigned comment added by (talk) 17:35, 6 January 2015 (UTC)

  1. ^ Science 2 January 2015: Vol. 347 no. 6217 pp. 78-81 DOI: 10.1126/science.1260825
I heard that story too and I'd love to see it mentioned somehow. Problem is it doesn't quite meet WP:MEDRS since its primary research. BakerStMD T|C 20:34, 17 January 2015 (UTC)
The paper is not primary research, since it is a correlation study based on collected patient data in the entire US and estimated 'division rates' in various tissues obtained from literature — Preceding unsigned comment added by 2A02:908:FD51:AA60:E9AC:4A93:6E0A:695E (talk) 01:34, 5 February 2015 (UTC)
I can see nothing in the lead that "suggests that Cancer can be prevented by a healthy lifestyle", which is indeed incorrect. Johnbod (talk) 20:37, 17 January 2015 (UTC)
Please note that disease prevention does not only mean *eliminating* (eradicating) risk of disease, but also significantly *reducing* the risk. A healthy lifestyle does not mean that you won't get cancer, but that you'll be less likely to get it. (talk) 18:55, 21 January 2015 (UTC)
Need a secondary source. This is primary [1] Doc James (talk · contribs · email) 00:30, 18 January 2015 (UTC)
That study is limited by presenting data for only most cancers, but not for two of the most common (breast and prostate). "65% of some cancers" is not as interesting as "65% of all cancers".
Johnbod, a great deal of the lead talks about lifestyle, without coming right out and saying "lifestyle causes almost all cancers". See "Many cancers can be prevented by [long list of healthful lifestyle choices]". The lead completely omits the fact that most invasive cancers in non-smokers, i.e., more than 50% of cancer cases, cannot be prevented through individual action ("lifestyle choices"), and that many cannot be prevented through collective action (like curbing polluting industries), either.
However, there is significant overlap here. How does drinking scalding hot tea cause esophageal cancer? Because it increases the rate of cell division. How does alcohol abuse cause liver cancer? Because it increases the rate of cell division. Do we attribute those cases to "bad luck" or to "lifestyle choices"? It's kind of both. "Bad luck" explains why some people who engage in the risky behavior get cancer, and "good luck" explains why some people who engage in the risky behavior do not.
Looking over the list, some of the other items may or may not be in the list of "lifestyle choices": The main sources of infectious caner are sexually transmitted infections. Neither this article nor the main article on infectious causes of cancer even mention how the typical person acquires these infections. Having a lot of sexual partners is usually an individual lifestyle choice (we all hope, right?), but whether to get multiple expensive vaccines isn't always a choice, and even lifelong monogamy is not a guarantee. I also wonder if "getting vaccinated" is the right way to present the infectious disease component in the lead. It is a rather Western-wealthy-country-centric approach, and there aren't vaccines for all of the infectious agents. "Avoid getting certain sexually transmitted viruses" (regardless of the means you employ to avoid getting infected) is more accurate. I expect that the Wikipedia community would be a bit squeamish about non-drug approaches to STI control, though. WhatamIdoing (talk) 18:09, 18 January 2015 (UTC)
It doesn't say anything at all like "lifestyle causes almost all cancers"! The main bit, before what you quote, is "Tobacco use is the cause of about 22% of cancer deaths.[1] Another 10% is due to obesity, a poor diet, lack of physical activity, and drinking alcohol.[1] Other factors include certain infections, exposure to ionizing radiation, and environmental pollutants.[4] In the developing world nearly 20% of cancers are due to infections such as hepatitis B, hepatitis C, and human papillomavirus.[1]", then 10% for inherited genetics. So the main individual "lifestyle" factors account for 32%, which clearly isn't even "most", never mind "almost all". Adjust the wording if you like - the headline from the paper was nothing new & that most cancers are sporadic can be referenced easily enough. Wiki CRUK John (talk) 18:41, 18 January 2015 (UTC)
John, don't do the arithmetic. Most of our readers won't. Instead, count up the number of things that are (or could be) counted as "choices" that cause cancer, and compare it to the amount of space given to things completely outside the patient's control. I'll save you some trouble. Here's the relevant parts of the lead text. I've color-coded it.

Tobacco use is the cause of about 22% of cancer deaths.[1] Another 10% is due to obesity, a poor diet, lack of physical activity, and drinking alcohol.[1] Other factors include certain infections, exposure to ionizing radiation, and environmental pollutants.[4] In the developing world nearly 20% of cancers are due to infections such as hepatitis B, hepatitis C, and human papillomavirus.[1] These factors act, at least partly, by changing the genes of a cell.[5] Typically many such genetic changes are required before cancer develops.[5] Approximately 5–10% of cancers are due to genetic defects inherited from a person's parents.[6]

Many cancers can be prevented by not smoking, maintaining a healthy weight, not drinking too much alcohol, eating plenty of vegetables, fruits and whole grains, being vaccinated against certain infectious diseases, not eating too much red meat, and avoiding too much exposure to sunlight.[8][9]

I've highlighted "all your fault" in orange, things that the typical person might believe could go either way in yellow (people might disagree, for example, on whether working as a radiology tech or living in a smoggy city is a "lifestyle choice"), and things that are "not your fault" in green. Here are the questions you need to answer:
  1. Does that seem about right to you? Thinking about the perception of the median patient, with median education attainment (in the US, that's a high school diploma), would you put anything that I've marked as orange into the green category?
  2. Which color do you see the most of? WhatamIdoing (talk) 04:25, 19 January 2015 (UTC)
Of course. But what do you do? You can add more green, but that doesn't take long at this level of simplicity. Then do you cut the orange and yellow to reflect a "correct" balance? At the end of the day the orange is relatively complex information that you can do something about and the green isn't either of those, which is why it tends to get predominance in coverage. The 3rd world infectious causes count as "not your fault"/nothing to worry about for Western readers surely? Wiki CRUK John (talk) 09:03, 19 January 2015 (UTC)
Worth remembering perhaps that lifestyle choicesindividual fault (culture is a key player). (talk) 19:48, 21 January 2015 (UTC)
86.134, you're not an American, are you? Face-wink.svg
John, I don't think that it can be usefully solved. It simply takes more words to write a basic explanation for the "what you can do" part than for the "you're helpless about the rest" part. But the result is that readers will feel like cancer is mostly under individual control. We could maybe add another sentence that says something like, "Most cancers cannot be prevented solely through individual lifestyle choices", and perhaps put that first instead of last, but I don't think that the problem can be easily solved. WhatamIdoing (talk) 18:03, 22 January 2015 (UTC)
I agree (about the last point). Wiki CRUK John/ Johnbod (talk) 03:14, 24 January 2015 (UTC)
That's a healthy read :) (talk) 20:00, 21 January 2015 (UTC)

On a related matter...[edit]

One of the opening sentences of ==Causes== currently reads [2]:

Common environmental factors that contribute to cancer death include tobacco (25–30%), diet and obesity (30–35%), infections (15–20%), radiation (both ionizing and non-ionizing, up to 10%), stress, lack of physical activity, and environmental pollutants.4

I've requested clarification [3] on the Causes of cancer page(the present page is semi-protected) as to how "obesity" can be listed/counted as "environmental" – when it isn't, given that it's not an external factor. A somewhat similar consideration could be made for "stress" (ie the organism's response, as distinct from environmental stressors).

This 'review' source (PMID 18626751) from 2008 is actually *proposing* a "unifying hypothesis". I feel PMID 24351322paywalled and PMID 23514316paywalled might be useful recent MEDRS.

@HenryScow: Perhaps WikiProject CRUK could somehow help get this key cancer-related content right? (talk) 17:07, 22 January 2015 (UTC)

If we're doing it here, I'll repeat here the comment I made on that talk page:

I'm not happy about this edit [4], which leaves the start of the article not making much sense. The previous version, I think adjusted by me following a discussion elsewhere (in which 86 participated as 109), defined "environmental", as used by epidemiologists, as anything not caused by inherited genetics, so including obesity, as well as sporadic "bad luck" cancers. This was supported by the 1st ref (although it comes after that ref, which is of course too old (2008) and I think rather tendentious in implying that all such cancers are preventable), whose use of "environmental" explicitly covers obesity. Now the sum using that definition (90-95 + 5-10 = 100) remains right at the start ("The great majority of cancers, some 90–95% of cases, are due to environmental factors. The remaining 5–10% are due to inherited genetics"), but the definition of "environmental" has been significantly altered, leaving no place in that implied sum for "internal" environmental factors, and "bad luck" ones, which actually represent the majority of cases! We could do with better sources, of course, but either the whole para should be changed, or left.

User:WhatamIdoing, what do you think? Wiki CRUK John (talk) 12:46, 22 January 2015 (UTC)

- It all goes back to an earlier discussion on this page, now archived (link in quote). The "ref 1" at causes is this, which is very happy to describe both obesity and sporadic "bad luck" as "environmental factors". Wiki CRUK John (talk) 17:30, 22 January 2015 (UTC)
I think it "all goes back" to whether we're actually getting it right or not... Restarting with more appropriate sources might help, imo. I agree that environmental factors + inherited genetic factors = 100% (in terms of factors that are *preventable* – in the broadest possible sense). Other individual factors such as obesity and stress are 'downstream' of that equation. Btw, in the present (ie epidemiological) context there's no such thing as "internal" environmental factors. (talk) 18:04, 22 January 2015 (UTC)
I think that the problematic edit is this one, which uses a completely different (and minority) definition and then claims that it applies to the stats produced using the original definition. That is not exactly an example of "getting it right".
To give you an idea of how significant this difference is, the definition that was added declares that the number of children a woman has, the age she starts having children, and whether and how long she breastfeeds—which are all significant biological factors in sporadic breast cancer due to "inside the body" hormones—are entirely non-environmental. Back here in the real world, those are seriously cultural factors, and human culture is definitely environmental. Worse, since this paragraph sets up a binary system, it effectively implied that these are therefore somehow hereditary genetic factors, which is just nonsense. (Obesity, by the way, is the cancerous tissue's environment, so that's not even as wrong as it sounds when you first hear it.)
I've reverted for now. I think we need improvements, but this attempt wasn't working. My general recommendation is that when we're presenting definitions and statistics, then we need to have the definition match the one used by the people producing the statistics. Perhaps the first place to start is with a proper lead sentence, like "There are many causes of cancer. Most of them cannot be easily controlled by individuals." It might also help to put the "hereditary genetics" first, so that the category of "non-hereditary genetics" seems more natural. WhatamIdoing (talk) 18:18, 22 January 2015 (UTC)
WAID, could you perhaps look at that again? In particular, on what basis do you say the definition I paraphrased is "completely different" (and minority)"? It's supported by highly authoritative sources,[1][2] as well as the "original" source[3] (which was/is being misrepresented here]). No-one is disputing that obesity is *related* to environmental factors – just that it's NOT itself an environmental factor.
  1. ^ A Dictionary of Epidemiology. Oxford University Press. 2014. p. 93. ISBN 978-0-19-997673-7. ENVIRONMENT: All that which is external to the individual human host 
  2. ^ Cancer and the Environment: What you Need to Know, What You Can Do. NIH Publication No. 03-2039: National Institutes of Health. 2003. Cancer develops over several years and has many causes. Several factors both inside and outside the body contribute to the development of cancer. In this context, scientists refer to everything outside the body that interacts with humans as 'environmental'. 
  3. ^ Kravchenko J, Akushevich I, Manton, KG (2009). Cancer mortality and morbidity patterns from the U. S. population: an interdisciplinary approach. Berlin: Springer. ISBN 0-387-78192-7. The term environment refers not only to air, water, and soil but also to substances and conditions at home and at the workplace, including diet, smoking, alcohol, drugs, exposure to chemicals, sunlight, ionizing radiation, electromagnetic fields, infectious agents, etc. Lifestyle, economic and behavioral factors are all aspects of our environment. (talk) 18:44, 22 January 2015 (UTC)
Adding: I've been trying to understand WAID's objection above (stemming from a simple misreading of this version maybe?). Perhaps it regards the choice of the NIH source (ref 2, above), with its rather specific focus on environmental "substances"? Having some professional background in epidemiological / environmental health research myself, I didn't (and don't) see that brief definition as conflicting in any real way with the more extensive definition provided by Kravchenko et al in ref 3. However, I'm not wedded to use of any one particular source. Fwiw, I do agree that further elucidation is needed (which is why I pinged User:HenryScow at CRUK, who clearly has a good overall grasp of these causality questions). (talk) 23:03, 22 January 2015 (UTC)
If we are saying that "environmental factors" is restricted to those arising outside the body then the opening sentence "The great majority of cancers, some 90–95% of cases, are due to environmental factors" just isn't true, surely? Wiki CRUK John (talk) 23:39, 22 January 2015 (UTC)
That is exactly the problem. WhatamIdoing (talk) 23:45, 22 January 2015 (UTC)
Sorry, I hadn't seen you've addressed that in the article. Wiki CRUK John (talk) 23:48, 22 January 2015 (UTC)
(ec) adding: Talking of the US, one of the PMID links above, "Current perspective on the global and United States cancer burden attributable to lifestyle and environmental risk factors", by Schottenfeld D1, Beebe-Dimmer JL, Buffler PA, Omenn GS, gives (in the abstract) a maximum of 60% of US cancers to the big external environmental factors plus infections. The big UK survey of a year or two back gave around 45% to the big external environmental factors as I recall. You only get to 90-95% by adding sporadic unexplained mutations, as the first ref used does. Wiki CRUK John (talk) 23:46, 22 January 2015 (UTC)


86.134, perhaps it will be easier if you try to explain what pregnancy-related hormones are. Are they "outside the body", or not? Well, no, obviously hormones are inside the body. And if they're not "outside the body", then are they "inherited genetics"? Well, no, the are obviously not inherited genetics. So what are they?

This paragraph is using a strict binary system: All causes of cancer, without exception, are either A or not-A. All causes of cancer, without exception, are either in the group that causes 90–95% of cancers, or in the group that causes 5–10% of cancers. This simple pie graph shows the central point that we're talking about. There are only two pieces in the pie. All causes are either one or the other. One piece is labeled "Genes" and the other is labeled "Environment". We want to tell people what those two numbers are. We want them to know that one of those numbers involves way the heck more than pollution.

You have just (accidentally) re-defined one of the terms so that we now have logical nonsense. You have re-defined it so that we have 90–95% caused by "things outside the body", 5–10% caused by "inherited genetics", and 0% leftover for the very sizable contribution of things inside the body that are not inherited genetics (for example, pregnancy-related hormones).

You seem to want to create a trinary system:

  1. things outside the body that cause cancer (e.g., pollution),
  2. things inside the body that are not hereditary genetics and cause cancer (hormones), and
  3. inherited genetics

—but if we do that, then we need a different statistics (i.e., statistics that nobody has been able to find, because good sources don't make that particular three-way distinction).

If we could find stats, we could use the other binary system:

  1. Internal factors (both inherited genes and hormones)
  2. External factors (both pollution and sunlight)

but I haven't seen recent reliable sources with statistics for this system.

As I see it, the immediate realistic options are either:

  • Remove the definition that is not used by the people who created these statistics, or
  • Remove the statistics that are not following the definition posted on the unrelated website you cited.

What we cannot do is say that "A" is defined this way, and then say that "A" is 95%, when we know perfectly well that the people who said "95%" used a noticeably different definition for "A". It would be like saying that "Luxury cars are defined by source #1 as any automobile with air conditioning" and then saying that luxury cars cost an average of US $90,000—according to source #2, whose definition of "luxury cars" had very little to do with the presence of air conditioning. That's what we've got right now, and it needs to be fixed right now. The definition and the stats need to be congruent. That means telling the reader that this 95% "environment" group includes far more than "substances outside the body". WhatamIdoing (talk) 23:49, 22 January 2015 (UTC)

@WhatamIdoing: This conversation seems to me to be going off-track. Fwiw, as I indicated in my comment above, in the field of environmental epidemiology, individual factors like obesity, stress, pregnancy-related hormones are understood as being *related* to environmental factors (such as diet, physical activity, sleep etc; stressors; and the ones you yourself mention above), but they aren't considered, in themselves, to be environmental factors. Since (with genuine respect) I don't think it's altogether reasonable to require a Wikipedia editor to expound in any depth on the methodological relevance and implications of such matters, instead I'll try to reframe my genuine editorial concerns here in purely WP terms.
(Btw, what you call a "simple pie graph" I do not look upon in any way as being a *simple* pie chart – though as far as I can see the information (confusingly) presented in it appears to be unsourced and, presumably, represents original research work conducted by the authors themselves, using unspecified methods.) (talk) 10:13, 23 January 2015 (UTC)
That's a deep link to a diagram in a review article. If you want to read the methods, then you need to read the rest of the paper.
(I wonder why "diet, physical activity, sleep etc" are not considered "individual factors". Surely they are not group factors, and the mechanism for sleep, at least, ought to be in the same general class as the mechanism for obesity.) WhatamIdoing (talk) 15:56, 23 January 2015 (UTC)
@WhatamIdoing: RE If you want to read the methods, then you need to read the rest of the paper. Erm, what "methods", where? The only explanation of methods that I've been able to find is in the caption to the figure. In particular: A provides no indication of how or where the 5–10% / 90–95% ratio was obtained (and the relevant paragraph at the start of the 'Prevention of cancer' section of the main text is completely unsourced); although B at least provides a bit more information about what the figures are intended to represent, it also says "data shown here is taken from a study conducted in Utah..." (hmm... what study on what population, using what methods? have those data even been peer-reviewed?); C again gives some information on what the figures represent ("the attributable-fraction of cancer deaths due to the specified environmental risk factor") but doesn't explain how those estimates were extracted, or how the authors avoided pitfalls such as counting the same data twice in different categories. That's not what I would call an acceptable description of methods.

But, like I said, my understanding of MEDRS is that a Wikipedia editor shouldn't be expected to provide a detailed critical appraisal ("review") of a paper's methodology to question the editorial weight given in Wikipedia to the data it reports. Anyway, per WP:MEDASSESS, I've seriously questioned (I think) "the quality of the study". I've also pointed out that the content of the paper was "outside the journal's normal scope" (not altogether surprisingly, imo).

Disclaimer: I have absolutely no POV to push – I just don't think we should be giving such a strong encyclopedic voice to these figures. I believe it's unjustified and, ultimately, mistaken and unhelpful. Rather, we should also be considering how other potential MEDRS[including "ideal" ones] approach these matters (see below). (talk) 17:55, 23 January 2015 (UTC)

Arbitrary break: some MEDRS considerations[edit]

Although the paper in question is presented and indexed as a review, it claims (at the start of the Conclusions) to "propose a unifying hypothesis". On purely WP editorial grounds, I think we're conferring undue weight to one particular possible MEDRS (one which also happens no longer to satisfy WP:MEDDATE). WP:MEDRS specifically warns about using articles whose "content [is] outside the journal's normal scope", as is the case I think of this environmental epidemiology paper published in an off-topic journal called Pharmaceutical Research

I've already proposed two alternative potential MEDRS (both unfortunately paywalled), which I think could be considered here: PMID 24351322 and PMID 23514316.[1][2] (talk) 10:13, 23 January 2015 (UTC)

  • The 'Etiology' section in the WHO 2014 World Cancer Report begins:

    Most cancers are associated with risks from environmental, lifestyle, or behavioural exposures.

    I think this could be a reasonable starting point (something like this perhaps?). As far as I can see, the report doesn't attempt to provide any more specific estimates regarding proportions of cases of this heterogeneous class of diseases that can be attributed to these overlapping categories of exposure (something I don't find altogether surprising). (talk) 11:02, 23 January 2015 (UTC)
  • To avoid this problem, I have previously proposed describing these two categories as "genetic and non-genetic", but that disturbs people who say that all cancer is genetic (which is not unreasonable). (That's when we fell back to the approach of telling readers that the source was using specialist jargon.) WhatamIdoing (talk) 15:56, 23 January 2015 (UTC)
I only have a short break, away from any desk, but there is a large recent UK study, referred to above. In Nature or the BMJ I think. I tried to reference that somewhere a while back, but it was removed by Doc James as primary which slightly puzzled me at the time. The WCR is indeed remarkably unforthcoming about the overall picture, and their factsheet, used as a ref later in the lead, is a far from ideal source, with a string of large % figures, which we use, shortly followed by saying (something like) "30% of cancers are preventable", which we don't use. In general we should identify 1 or 2 big detailed sources & follow them consistently. I'm back in the office on Monday & will try to consult. Wiki CRUK John (talk) 16:29, 23 January 2015 (UTC)
@Wiki CRUK John: Was that large recent UK study perhaps PMID 22158327 [5] (added here). If so , yes, that is indeed technically a primary source, per the MEDRS definition. Personally, as I hinted above, I'm not the surprised that the WCR avoids getting caught up in this methodological conundrum, and on both scientific and editorial grounds I can't help wondering whether we shouldn't be avoiding it too. But I suppose that isn't really the usual WP way... (as in... why waste a chance to bash out a good controversy?) Imo, an impartial MEDRS-based editorial indication might come from perusing the content of PMID 24351322 and PMID 23514316 (already suggested above), alongside the reticent WCR summary, and any other potentially appropriate, recent secondary sources. 2c / 2p (or whatever...), (talk) 19:17, 23 January 2015 (UTC)
Yes, thanks, that's the one. I know what you mean, but it surely is not unreasonable for the reader to expect some quantification, however caveated and vague, on these points - meaning both readers of our Causes article and the 600 page-long WCR. If the IARC don't feel confident to attempt an overall synthesis of the data in that (& I too have looked, & failed to find this), then they should not put out very short fact sheets that do, with obvious problems & Wikipedia-like internal contradictions. At least we say where our numbers came from, allowing debates like this one. My opinion anyway. Wiki CRUK John (talk) 15:41, 25 January 2015 (UTC)
Agree. Let's take a look, shall we? (@Wiki CRUK John: could you help me access PMID 24351322 and PMID 23514316? tks) (talk) 20:12, 25 January 2015 (UTC)
This review[when published?] also seems pertinent.
From a wp:weight perspective, I feel we need to bear in mind that these estimates generally reflect *all* cancer deaths, lumped together, without distinguishing in any way between people with a good life-expectancy cut down in the prime of their life by fulminating diseases and, say, very elderly people with significant comorbidities that were already affecting their quality of life. So, on encyclopedic grounds, I feel we would do well not to overemphasize the question, especially given the limited amount of reliable information that is available on this topic, which (apart from theoretical considerations regarding prevention) also has subtle human implications. 2c, (talk) 09:53, 26 January 2015 (UTC)
I'm puzzled why you think the Guardian editorial supports what you are arguing! Wiki CRUK John (talk) 12:14, 26 January 2015 (UTC)
Oh... then, perhaps I hadn't noticed *what* I was actually *arguing*[!?] One of the things I was trying to point out was that the way Wikipedia presents this question may not be humanly trivial for our readers. I think the Guardian editorial (disclaimer: which represents the newspaper's opinion, not mine) illustrates that concept. (talk) 12:51, 26 January 2015 (UTC)

Arbitrary break: pertinent MEDRS[edit]

Here are the pertinent passages, I think, from the main text of the two paywalled MEDRS suggested above.

  • From the Lancet 2014 review:[1]

Primary prevention

An overall estimate of preventable cancers

Genetic variants are unlikely to account for a major proportion of cancer cases; 5–10% are attributable to highly penetrant mutations such as in BRCA1 or mismatch repair genes, and an unknown proportion to the interaction between low-penetrant variants and external risk factors. For a long time, since the publication of Doll and Peto’s The causes of cancer83: PMID 7017215 (1981) containing estimates for the USA, the proportion of cancers in a population that is attributable to known risk factors has been a controversial issue. The most recent estimates have been provided by Parkin and colleagues85 (the big primary study suggested above by John - PMID 22158327) for the UK. Parkin and colleagues’ report85 has several merits: past exposures are considered in relation to present cancers (to take latency into account); risks are estimated with attainable levels of exposure as reference categories; and 14 risk factors and 18 cancer sites are considered. The conclusion is that 45% of cancers in men and 40% in women could have been prevented had risk factors been reduced to the optimum levels or eliminated (eg, tobacco).

Estimates of the proportion of cancers that can be prevented will differ substantially geographically, showing the prevalence of different risk factors, hence the need to set prevention priorities at a local and regional level.1 This need is best characterised within the categories of the low-HDI and medium-HDI countries, where the estimates from the USA and the UK cannot be directly applied. In simplification of messages for political effect, it is important to avoid a failure to adapt and support solutions that will bring the most benefits for public health to particular settings.

  • From the 2013 paper by Schottenfeld et al:[2]


In their 1981 publication, Doll & Peto concluded that 75–80% of cancer deaths in the United States could have been avoided. The overall estimates reflected uncertainties about “diet” (PAF = 35%, range: 10–70%), but without estimating attributable fractions for obe- sity or physical inactivity, and uncertainty about attributable risks for various infectious agents (27). Doll & Peto defined “diet” as “all materials that occur in natural foods, are produced during processes of storage, cooking, and digestion, or added as preservatives, or giving foods color, flavor or consistency” (p. 1226). Our current perspective for industrialized countries is summarized for the United Kingdom, France, and the United States (Table 3). The data show contrasting estimates for the three countries. In the review by Parkin, 14 lifestyle and environ- mental risk factors were responsible for 43% of cancer cases (45% in men, 40% in women) and for ∼50% of cancer deaths in 2010. In addition to the risk factors reviewed in detail above, Parkin assessed four dietary factors— low consumption of fruits and vegetables [PAF (combined men and women) = 4.7%], red and processed meat consumption, low dietary fiber, and salt—and oral contraceptives, hormone replacement therapy, and reproductive factors. In their review of biologic agents, lifestyle behavioral patterns, and physical environmental factors that are established determinants of cancer incidence and mortality, Colditz & Wei (20) concluded that 50–60% of cancer deaths and more than 60% of cancer cases in the United States were potentially avoidable. Similarly in our analysis we suggest that 60% of cancer deaths in the United States may be attributable to eight risk factors. Because we may assume some degree of overlap in the distribution of such combinations of risk factors as tobacco and alcohol, and obesity and physical inactivity, our estimate of 60% may represent a maximal upper limit. The contrast between France and the United States is apparent for obesity and may be explained by the lower estimate for the prevalence of obesity of ∼7% in French women (PAF: all cancers = 3.3%) and of ∼6% in men (PAF: all cancers = 1.4%) (99).


3. The epidemiology and pathogenesis of eight lifestyle risk factors are estimated to be determinants of ∼60% of cancer mortality in the United States. These risk factors include tobacco, alcohol, ionizing and solar radiations, occupations, biologic agents, obesity, and physical inactivity.

Thoughts? (talk) 11:11, 28 January 2015 (UTC)

I've now read both of these (skipping some risk sections in Schottenfeld), and they are both highly useful refs, for many articles (including the uncreated Obesity and cancer). Thanks for finding them. Among other things the detailed coverage of Parkin et al.'s UK study, in the passage quoted above and elsewhere, allows this stuff to pass MEDRS. I think we should replace nearly all the existing lead and sources on these points in the "Causes" article with a rewrite using "preventable" rather than environmental, with explanations, then inherited genetics, then mentioning the existence of a remainder (these two are not so helpful here, & new sources might be needed). The Schottenfeld also has a very nice bit on the overall mechanism for causes. Note that here at Cancer the start of the causes lead is repeated. Wiki CRUK John (talk) 13:13, 28 January 2015 (UTC)
Both Wikipedia articles rely heavily for causes/risks on: Anand P, Kunnumakkara AB, Kunnumakara AB, Sundaram C, Harikumar KB, Tharakan ST, Lai OS, Sung B, Aggarwal BB (September 2008). "Cancer is a preventable disease that requires major lifestyle changes". Pharm. Res. 25 (9): 2097–116. doi:10.1007/s11095-008-9661-9. PMC 2515569. PMID 18626751. - from 2008, which now should be replaced. Wiki CRUK John (talk) 14:47, 28 January 2015 (UTC)
  1. ^ a b Vineis P, Wild CP (2014). "Global cancer patterns: causes and prevention". Lancet 383 (9916): 549–57. doi:10.1016/S0140-6736(13)62224-2. PMID 24351322. 
  2. ^ a b Schottenfeld D, Beebe-Dimmer JL, Buffler PA, Omenn GS (2013). "Current perspective on the global and United States cancer burden attributable to lifestyle and environmental risk factors". Annu Rev Public Health 34: 97–117. doi:10.1146/annurev-publhealth-031912-114350. PMID 23514316. 


I'd like to add the possibility of exercise as adjunct therapy, and the role of p27 in colorectal cancer. Can I have access edit this page? Thank you. — Preceding unsigned comment added by Edward Tong (talkcontribs) 16:48, 5 February 2015 (UTC)

Inducing cancer through transplant:Cornelius Rhoads[edit]

In Principles of Research Methodology: A Guide for Clinical Investigators, on page 235, it mentions that Dr. Cornelius Rhoads was able to successfully transplant cancer cells that resulted in the deaths of 8 patients. Most people do not think cancer is something that can be spread through transplant, this information is worth mentioning in this article as it applies to all cancers. Bullets and Bracelets (talk) 15:19, 22 March 2015 (UTC)

Wow. The guy was an asshole, evidently, but it seems that "successfully transplant" is not true, based on the information in the article on him. In the quoted part of his letter (Cornelius P. Rhoads § Scandal), he claims "credit" for killing 8 patients (presumably by other malpractice than cancer transplant), but he indicates that his attempts to transplant cancer were not successful so far; and in the 2002-2003 section (Cornelius P. Rhoads § Revival of controversy), it says that no one could find evidence that the cancer transplants worked, but that the guy was a dick for sure. Quercus solaris (talk) 17:10, 22 March 2015 (UTC)
It happens in Tasmanian devils with concerns it may lead to there extinction. Doc James (talk · contribs · email) 23:57, 22 March 2015 (UTC)
Wow again. I never knew that clonally transmissible cancer existed, but I just looked over the WP article. Thanks for the tip. Quercus solaris (talk) 02:04, 23 March 2015 (UTC)
Did Dr. Rhoads experiment doing the cancer transplant with different transplant rejection drugs; did he, (or anyone else since than), try cordyceps? Has their been research on cancer transplants? Bullets and Bracelets (talk) 17:23, 23 March 2015 (UTC)
There's more on this subject from Scientific American. It appears there are cases in medical literature where cancer has been transmitted from a patient onto a surgeons cut, and between two genetically similar people. Also, a mother can pass cancer on to a fetus—for example, things like melanoma have been passed from mother to fetus, and during organ transplantation, if the organ donor has cancer, it is possible to transmit cancer. [6] Bullets and Bracelets (talk) 03:12, 28 March 2015 (UTC)
I think we have enough on this here, given how rare it is, but the other articles might be added to - Clonally transmissible cancer etc. Johnbod (talk) 14:00, 28 March 2015 (UTC)