Talk:Pancreatic cancer/Archive 2

Initial review by CRUK
This is a write-up of the notes made in an initial review by a CRUK specialist. The idea is to sort these points out in the article before sending the article for review by other outside specialists. Epidemiology & the all-but-missing research section were not covered - will be done with other people. By no means all points made are written up - many ticks etc, but also other points.. I'm hoping this gives the medical editing community enough to go on to start serious work on the article, but I realize it may not. Wiki CRUK John (talk) 17:09, 26 June 2014 (UTC)

Best recent papers

 * (I have p/os): (see also Stoita, used in article)


 * BRAND NEW (but paywall) Review Article Pancreatic Adenocarcinoma, David P. Ryan, M.D., Theodore S. Hong, M.D., and Nabeel Bardeesy, Ph.D., N Engl J Med 2014; 371:1039-1049September 11, 2014DOI: 10.1056/NEJMra1404198

Comments on article
Reviewed as at this diff.
 * Lead, various notes. Relative emphasis of sentence on genetic conditions seemed odd. Various smallish diffs in % figures given with what CRUK uses noted - may be UK vs global.
 * Chemotherapy section - "A Canadian-led Phase III randomised controlled trial, reported in 2005,...", also next para, old, not too relevant. MPACT Phase III study more significant. More notes, hard to read.
 * In alcohol section "chronic pancreatitis associated with alcohol consumption is less frequently a precursor for pancreatic cancer than other types of chronic pancreatitis."Ref 34 is to CRUK site - in fact that bit changed in 11/13 & that bit removed (alcohol)
 * Refs/rewrite needed for diagnosis section.
 * Neuroendocrine section not checked.
 * Research section: now 1 line!
 * Those are the main comments. Wiki CRUK John (talk) 17:09, 26 June 2014 (UTC)
 * Added that they genetic conditions are rare. Yes we are using global stats.
 * Trimmed out of place old primary source.
 * Great sources you have found. Doc James  (talk · contribs · email) (if I write on your page reply on mine) 21:11, 26 June 2014 (UTC)
 * They are what the outside specialists reccommended to the CRUK writer - one thing we need to do is pretty wholesale replacement of over 5 yo sources saying pretty much the same thing with these; they are nice and recent and will last us a while. Wiki CRUK John (talk) 09:21, 27 June 2014 (UTC)

Discussion of review

 * Add something on cysts - many more being discovered - Gastroenterology. 2013 Jun;144(6):1303-15. doi: 10.1053/j.gastro.2013.01.073. "Pancreatic cystic neoplasms: management and unanswered questions.", Farrell JJ1, Fernández-del Castillo C., also SSAT Patient Care Guidelines, Cystic Neoplasms of the Pancreas etc. Wiki CRUK John (talk) 15:47, 17 July 2014 (UTC).

Definition version cause
The first sentence of the article should be the definition of pancreatic cancer with the "cause / risk factors / mechanism" dealt with in the second paragraph IMO. Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:51, 26 June 2014 (UTC)
 * Done. 86.128.169.211 (talk) 17:35, 27 June 2014 (UTC)

"To cease" versus "the stopping of"
In this edit "to cease" is more in advice format while "the stopping of" is less in advice form. Thus why I used the latter  Doc James  (talk · contribs · email) (if I write on your page reply on mine) 07:21, 27 June 2014 (UTC)
 * Just spotted this (sorry). I didn't think of it that way, though I can see your concern. To my ears, "the stopping of" doesn't appear particularly idiomatic (though others may disagree). "Cessation of smoking" would be technically correct, but imo not stylistically indicated in the lead. Adding: In my book at least, 'giving advice' would involve saying something like, "To prevent the disease you should stop smoking." 86.128.169.211 (talk) 17:34, 27 June 2014 (UTC)

Sentence in lede needs attention of subject matter expert
This interpretation of statistical, epidemiologic data, apparently that of a layperson rather than medical/scientific professional, regards cause and effect in this cancer's etiology, and needs review by a subject matter expert with better understanding of the technical meaning of risk (see preceding sentence), of the actual roles of genetics in cancer, and of the relationship between correlation and causality:""Smoking is the cause of about 20% of cases while 10% of cases are inherited from a person's parents."" This needs to be compared to the corresponding text appearing later in the main body of the article (presumed), and then to the source cited there (presumed), to see what is actually stated, assuming that an expert, valid, verifiable secondary source appears. The sentence then needs to be corrected for the actual point of the source authors. (I guarantee, if the source is one of quality, it does not say what is stated here.) This article is at the periphery of my work scope, and I haven't the time to rigorously correct even this one mis-generalization. let to thoroughly review the article to see if such loose play with basic concepts is present elsewhere. Even if this is a one-off that results from an attempt to make the lede accessible to a broad audience, a lede should never get things so fundamentally wrong. Le Prof Leprof 7272 (talk) 14:22, 14 July 2014 (UTC)
 * I'll get on it. Wiki CRUK John (talk) 14:35, 14 July 2014 (UTC)
 * TY. In a quick glance at the main text, searching "smoking" and "gene", two things are clear: the percentages in the lede are not strictly those in the main text—where in one case an upper limit of range is chosen, and in the other, the lower limit! (Note: These data appear in different places in the main body.) Hence, the main text parallels but does not (thankfully) actually say what the lede says—the lede does, as I stated, mis-generalize/mis-report, at least, the main text if not the actual sources. Adding a couple of inline tags to call attention to the matter in the article. Le Prof Leprof 7272 (talk) 14:46, 14 July 2014 (UTC)
 * Now "Smoking is the cause of about 20% of cases while 10% of cases are linked to inherited genes." Even specialized sources do often use "caused" in a somewhat loose or simplified way. I have the source, the 2014 IARC World Cancer Report, (it was not my passage though) and that says (inter alia) "Smoking is estimated to cause 20% of pancreatic cancer". (p. 414). The 10% figure is p. 416; I agree "are inherited from a person's parents." is over-simplifying. Thanks for the comments. I'll check the 2nd one now. Wiki CRUK John (talk) 14:51, 14 July 2014 (UTC)
 * Glad you are looking also to more recent data. I would suggest, since the cancer reports are lay summaries (and so derived sources), you make the lede as technically correct as you can for a lay readership (even though the reports appear to use language a bit loosely). Something like "The … report estimates that xx% of cases can be attributed to… while yy% of cases can be attributed to…".  Also—and this is why I generally cannot take time to do such editing—make sure what you put in, both in terms of sourcing and content, is copacetic with preceding lede risk statement, and the text and sources in the main body (i.e., that there are no discrepancies). At last look, the main body gives ranges of percents for each of these risk factors, in one case citing 2007 data.  Cheers, bravo.  Le Prof  Leprof 7272 (talk) 15:14, 14 July 2014 (UTC)
 * Yes, so far the article has been lightly updated with the 2014 WCR figures (from 2012) but more is needed, and will be done - see higher up on this page. I didn't get far with your 2nd comment before being pulled away, but I will return later. Wiki CRUK John (talk) 15:51, 14 July 2014 (UTC)
 * Thanks for effort. Your "linked to" is much better than caused, for both cases. Bravo, cheers. Le Prof  Leprof 7272 (talk) 16:06, 14 July 2014 (UTC)

Research on pancreatic cancer
I've beefed up the section on research, which I've drawn mainly from the American Cancer Society (http://www.cancer.org/cancer/pancreaticcancer/detailedguide/pancreatic-cancer-new-research) and CRUK pages (http://www.cancerresearchuk.org/cancer-help/type/pancreatic-cancer/treatment/whats-new-in-pancreatic-cancer-research) on the subject. I've referenced both of these pages in the intro of the research section as they're both regularly updated and well-written, but if this contravenes any guidelines then feel free to delete, I'm a man, I can take it :) HenryScow (talk) 16:46, 17 July 2014 (UTC)

Referencing
I'm beginning to feel that the improved article would benefit from using the "sfn" template refs. The advantage of these is that they allow precise page numbers to be used for each ref, but at the cost of an extra click being required to get through to the full ref, and any online text of it. The majority of the refs here need updating, and there is an excellent set of very recent papers (several above etc). Some are long and cover large parts of the subject: eg Wolfgang et al. "Recent Progress..." is 24 pages of text, really too long to refer to without page numbers. Equally one doesn't want to clutter up the references section with the full title details up to 20 times.

I must say I've never used the sfn templates before, though I normally prefer plain short footnotes for the reasons above. See Help:Overview_of_referencing_styles for an explanation of how they work.

What do people think? Wiki CRUK John (talk) 15:25, 18 July 2014 (UTC)
 * Given technological progress I'd suggest that electronic search should supplant page search. However, if you personally prefer to use exact pages from the print/PDF/NVA version(s), rather than making a big alteration to the ref style usually used in med articles, enter ONE full copy of the full reference, use abrv format for repeat use, and append any use with suffix, where PPP is the page.
 * For example, here is wikitext for the first use:
 * Here is the second use with an rp suffix.
 * Here is what you get:
 * For example, here is first use. Here is the second use with an rp suffix.
 * Here is what you get:
 * For example, here is first use. Here is the second use with an rp suffix.


 * FeatherPluma (talk) 01:25, 16 November 2014 (UTC)

Thank You
I just wanted to drop a line and thank everyone who contributed to this article. Someone close to me was just diagnosed with pancreatic cancer, and naturally this was my very first stop. It makes for some damned sober reading but I am one of those people who likes my bad news the same way I like my bourbon, straight. Better to know the worst than wonder or be trapped in false hope. Thanks again... -Ad Orientem (talk) 20:23, 1 August 2014 (UTC)
 * Glad it helped some and sorry to hear the bad news. Doc James  (talk · contribs · email) (if I write on your page reply on mine) 22:55, 1 August 2014 (UTC)

NEJM review
10.1056/NEJMra1404198 JFW &#124; T@lk  22:01, 13 September 2014 (UTC)
 * Another new review from Oncology, all online at: An Update on Gastroenteropancreatic Neuroendocrine Tumors, Review Article | September 15, 2014 | Oncology Journal, Gastrointestinal Cancer, Neuroendocrine Tumors. By Mauro Cives, MD and Jonathan Strosberg, MD Wiki CRUK John (talk) 17:10, 2 October 2014 (UTC)

Prognosis figures - presented in a show/hide box
[from Med talk]Some of you will remember we discussed this at CRUK before Wikimania. Just as a matter of presentation (I haven't edited the actual content yet), what do people think of the approach I've tried at Pancreatic_cancer? Again the wording probably needs work. Does anyone object to this approach as such? The idea would be to have overall survival rate information elsewhere, but to treat prognosis stats by stage this way. Comments please Wiki CRUK John (talk) 17:19, 23 September 2014 (UTC)


 * NOTE - I reverted the addition of the collapse box, but you can see it here in the former version of the article. Jytdog (talk) 18:06, 23 September 2014 (UTC)
 * That was unhelpful and unnecessary - this is not a censorship issue, and a variety of strategies for cautioning for content have been accepted by the community. Wiki CRUK John (talk) 10:50, 24 September 2014 (UTC)
 * Yes, I object. Hiding information from readers is a violation of MOS:COLLAPSE.  Some people will be unable to access the information that way, and for others (e.g., people with RSIs) accessing it will be needlessly complicated or painful.  WhatamIdoing (talk) 17:47, 23 September 2014 (UTC)


 * Oppose in addition to what WAID writes about logistics, Wikipedia is not censored (that is policy).  The section of this article where you created the hat is called "prognosis" and if somebody doesn't want to read about prognosis they can skip that section. Finally, there is no consensus to have trigger warnings in WP articles that I am aware of. I am not in favor of them. Jytdog (talk) 18:01, 23 September 2014 (UTC)
 * I find the information very clear. Even if poor, I think it's important to include the information on 5 year survival, which is very important. I also like how you provide information by stage and type, which seems obvious, but is not always done well. I would not support putting it in a collapse box, I think there are usability issues and there isn't a real need for it. --Tom (LT) (talk) 21:11, 23 September 2014 (UTC)


 * I agree with WhatamIdoing, Jytdog & LT910001 regarding the collapsible box.


 * The "Prognosis" section has other problems. It states that "Exocrine pancreatic cancer... typically has a poor prognosis". While I agree with the sentiment, it requires an inline citation. Why is a list of symptoms included in this section? Why are benign tumours discussed here? Axl  ¤  [Talk]  21:29, 23 September 2014 (UTC)
 * John only added the collapse box to test out the concept (Jytdog just removed it, so people who want to see what it looked like will have to check the older versions). He hasn't changed any of the content in that section.  WhatamIdoing (talk) 22:21, 23 September 2014 (UTC)


 * NOTE - I reverted the addition of the collapse box, but you can see it here in the former version of the article. Jytdog (talk) 18:06, 23 September 2014 (UTC)

Yeah agree on removal of collapse box. Pancreatic cancer has an awful prognosis and it's no secret. Cas Liber (talk · contribs) 05:29, 24 September 2014 (UTC)

I find the format to be fine, being to the point without necessarily stating too much about what the numbers mean. I'm inclined to think that if someone is looking for more information about these numbers they should do so on their own, and we don't need to explain it in every article.

As for the collapsed window I think it's relevant, though maybe the implementation was sub-optimal. The alternative is to try and make the information less "in your face". This could be done through use of infoboxes or sideboxes. I tried to make a mock-up here:

This is a quick mock-up, and could be improved and made to have a general format for all diseases. The benefit of this is also that we could use Wikidata to update the tables automatically. -- CFCF  🍌 (email) 06:50, 24 September 2014 (UTC)


 * I don't consider the current style to be "in your face". It is the "Prognosis" section of "Pancreatic cancer". What do you expect? Ironically, I think that putting it into a separate table/sidebar makes it more "in your face". I don't know how the Wikidata system works, but if it really provides a reliable way of updating information when new sources are published, that is worth considering. Axl  ¤  [Talk]  09:10, 24 September 2014 (UTC)


 * This doesn't seem a hit with the community! While it may be "no secret" that PC prognosis is poor, for non-medical readers the answer to "It is the "Prognosis" section of "Pancreatic cancer". What do you expect?" is probably mostly "nothing like as bad as the actual figures". WP:CENSORED is not the issue here - the community has approved a variety of ways of cautioning people who might not see various types of content, for examples around images of Muhammad. Does nobody feel there is any problem just giving these daunting stats without any warning? That is obviously appropriate for a professional medical audience, but general/patient directed sites such as CRUK & ACS generally take a cautioning approach, where they give prognosis by stage at all. I'll reiterate that the text is not the issue here, it is just as I found it. Currently it doesn't even clarify that the figures are for pancreatic adenocarcinoma (aren't they?). Nor does CFCF's box here. The figures for some other "pancreatic cancers" are much better. Wiki CRUK John (talk) 10:50, 24 September 2014 (UTC)


 * "for non-medical readers the answer ... is probably mostly "nothing like as bad as the actual figures"." I have no idea what you mean by that.


 * Regarding the Muhammad images, I am pretty sure that people have complained about them. Despite this, neither "Muhammad" nor "Depictions of Muhammad" seem to have any warning. Has anyone actually complained about the presence of the pancreatic cancer survival figures?


 * In my opinion, anyone who reads the "Prognosis" section should expect to see the figures without any warning or other mollycoddling. If the reader is a patient/relative and is upset by the figures, so be it. I think that trying to hide the figures is a disservice to readers. Axl  ¤  [Talk]  12:01, 24 September 2014 (UTC)


 * Oppose hide show boxes as they decrease the usability of our articles by screen readers. Doc James  (talk · contribs · email) (if I write on your page reply on mine) 16:46, 24 September 2014 (UTC)
 * Also oppose using a collapsible box. Otherwise, care has been used to portray the information about prognosis. I think that the article as currently worded does a good job of discussing the prognosis in an accurate but sensitive way. I'll use it as an example of how to discuss prognosis in often fatal diseases. Sydney Poore/FloNight&#9829;&#9829;&#9829;&#9829; 04:18, 26 September 2014 (UTC)

Main image
I'd like to propose a new main image for the article. I'm not exactly sure which image we should use, but I feel the current image is poor in that it does not explain the position of the pancreas relative to other organs. I've already added some images, but I wouldn't use these as the main image as they are wide and short, which tends to do poorly in infoboxes which have width limits. (Will also be adding some more images below that might be relevant.) -- CFCF  🍌 (email) 07:22, 24 September 2014 (UTC)

Comments
I quite like it as it is now, with the old main one, but a complicated one below. What I need is one with the head and tail labelled, as these are mentioned several times in the text. I like this one, which I will use lower down. Wiki CRUK John (talk) 17:06, 29 September 2014 (UTC)
 * Okay, I guess it has to do with who you're targeting with the image. I still think a broad topographic image that acutally shows how far behind everything the pancreas lies would be good. Normally I find the Blausen images slightly glossy, which decreases usability, but that one is decent. -- CFCF  🍌 (email) 10:46, 30 September 2014 (UTC)

Old/primary refs I'm cutting
Put here for convenience, if anyone needs them. Not a full list of those removed today. Wiki CRUK John (talk) 19:54, 4 November 2014 (UTC)
 * Causes/risks:
 * Removed as cause - see CRUK & NCI; no more recent studies & not in latest sources. " Helicobacter pylori infection "
 * diets low in vegetables and fruits,
 * In 2006, a large prospective cohort study of over 80,000 subjects failed to prove a definite association. The evidence in support of this lies mostly in small case-control studies.
 * These removed today Wiki CRUK John (talk) 18:10, 12 November 2014 (UTC)
 * These removed today Wiki CRUK John (talk) 18:10, 12 November 2014 (UTC)


 * In the United States, as of 2006, the economic costs of pancreatic cancer are estimated at $8.6 billion.
 * updated; diabetes as symptom.
 * updated- heretitary syndromes
 * Risk factors:
 * Gingivitis or periodontal disease
 * Partial gastrectomy
 * Partial gastrectomy

Links mentioned with diet
Thus added "A diets high in red meat or processed meat has been linked to pancreatic cancer in men. As have diets high in sugar-sweetened drinks (soft drinks). " Doc James  (talk · contribs · email) 16:48, 8 November 2014 (UTC)

Risk Factors | Diet
There has been some debate with regard to adding diet as a 'risk factor.' The current version suggests (per Johns Hopkins) that diet, in general times, 'may' be a risk factor. There are then specific reference to direct dietary links and references to the clear diet-diabetes-pancreatic-cancer-link and the clear diet-obesity-pancreatic-cancer-link. Both diabetes and obesity have been left in their own bullets as risk factors because while both are directly linked to diet, not all diabetes or all obesity is dietary.

An earlier version suggested that 'diet was not generally considered a risk factor' and Zaurus correctly added a 'by whom' tag which was then deleted. Since then there has been some resistance to including diet as a risk factor.SpecialFXavier (talk) 14:08, 9 November 2014 (UTC)
 * The most current MEDRS sources generally discount or are very cautious about diet, even red meat, as a direct factor, which the original reference covered. I shall return to this next week. Johnbod (talk) 14:29, 9 November 2014 (UTC)


 * The trouble with any one 'study' is that it might be very narrow in its approach to direct causality or relationship. The dietary link to pancreatic cancer is, however, crystal clear through its relationship with TYPE II Diabetes and Obesity. So while it might be hard to draw a straight line from, say, sugary drinks to the development of actual cancer cells in the pancreas; it is absolutely clear that sugary drinks contribute to both TYPE II Diabetes and Obesity and, therefore, to pancreatic cancer. It would concern me to have an article on Wikipedia that suggested that diet was not a risk factor when it clearly is, in specific terms, and probably is in more general terms. It would be irresponsible to say otherwise. When I arrived on this page the statement was the diet was 'generally accepted' NOT to be a risk factor. Of course, there was no ref for this. SpecialFXavier (talk) 18:16, 9 November 2014 (UTC)
 * As I keep telling you, there was a ref, which you find in the history. And the statement is correct. What should perhaps be made clearer is that "diet" means directly what you eat, not the 2nd & 3rd degrees of consequences of eating too much of it (for some people). Diabetes & I think obesity were already dealt with seperately, and should be kept distinct. Johnbod (talk) 01:27, 10 November 2014 (UTC)

Why did we duplicate all this content twice? Also we need to stay away from primary sources. Doc James (talk · contribs · email) 04:15, 10 November 2014 (UTC)
 * I've gone to "Diet is not generally accepted as a direct risk factor (though see the sections on obesity and alchohol),[3][47] although some individual studies have found dietary factors such as diets high in red meat or processed meat,[47] sugar-sweetened drinks (soft drinks),[48] In particular, limited epidemiological studies link the common soft drink sweetener fructose with growth of pancreatic cancer cells.[49]". This has the advantage of being true, and in agreement with the latest RS. We should not conflate diet and obesity in these sections, let alone diabetes - there the true symptom/cause picture remains unclear anyway. Risk factor could does with some rewriting and explaining to clarify how these work - you can only take each step in a possible progression as its own factor. The reader will have seen the sections on obesity and diabetes before reaching this in any case. Most sources, such as Ryan and the IARC just don't regard diet itself, even red meat, as a confirmed risk factor, cautious though that may be. Wiki CRUK John (talk) 19:13, 10 November 2014 (UTC)
 * Why do you keep adding primary sources? We should be using secondary source.
 * And why did you remove the secondary source I replaced for the primary source?
 * Can we at least agree to use secondary sources? Doc James  (talk · contribs · email) 07:46, 11 November 2014 (UTC)
 * Which ones are you talking about here? There are so many changes. Wiki CRUK John (talk) 12:13, 11 November 2014 (UTC)
 * Ok, sweet drinks - a mistake, sorry. Wiki CRUK John (talk) 15:32, 11 November 2014 (UTC)
 * Good as long as we all agree to use secondary sources :-) Doc James  (talk · contribs · email) 05:57, 12 November 2014 (UTC)


 * Which ref says "Diet is not generally accepted as a direct risk factor"? Doc James  (talk · contribs · email) 08:01, 11 November 2014 (UTC)
 * Bizarrely enough, the one right after the statement, that has always been there, is freely available online, but apparently completely invisible to every one except me!!!! Larssen, 2012, immediately after the Abstract:"Dietary factors could conceivably influence the risk of developing pancreatic cancer, although no dietary factor has been convincingly associated with pancreatic cancer risk". There is also a lot of evidence by omission - Ryan has a long Table 1 listing risk factors, which has no dietary ones, but obviously obesity and diabetes. I haven't cited the ACS and CRUK pages, which both discuss the issue and confirm this uncertainty, or the IARC lists, which just omit it. And so on. Wiki CRUK John (talk) 12:13, 11 November 2014 (UTC)
 * This is the ACS:

"Factors with unclear effect on risk Diet Some studies linked pancreatic cancer and diets that include a lot of red meat, pork, and processed meat (such as sausage and bacon). Others have found that diets high in fruits and vegetables may help reduce the risk of pancreatic cancer. But not all studies have found such links, and the exact role of diet in relation to pancreatic cancer is still being studied." -

See also the WCR 2014, which has "convincing evidence" for obesity, but only "limited" evidence for red meat, and doesn't mention even fructose etc, as most recent reviews don't. We are pretty much back where we started because that text was based on careful examination of a wide range of current sources, far more than were actually referenced. I'll add the WCR ref as belt and braces. I was already thinking that removing the soft drinks/fructose altogether might well be best, but will see what others think. Wiki CRUK John (talk) 17:43, 11 November 2014 (UTC) Wiki CRUK John (talk) 17:30, 11 November 2014 (UTC)
 * This bit "Dietary factors could conceivably influence the risk of developing pancreatic cancer, although no dietary factor has been convincingly associated with pancreatic cancer risk (2007)." is providing justification for their review with this sentence being followed by "We therefore conducted a dose–response meta-analysis of prospective studies to examine the associations of red and processed meat consumption with pancreatic cancer risk." Should we use paste tense? ie "Diet has not generally been accepted as a direct risk factor" Doc James  (talk · contribs · email) 05:57, 12 November 2014 (UTC)
 * I think not, given their conclusions, and the pick-up from in other sources so far. That might change, but I think we should stick with what the sources say so far. The ACS page was revised this June, CRUK around the same time, and so on, and they keep the cautious language despite this 2012 study (in the journal CRUK own btw). Wiki CRUK John (talk) 11:24, 12 November 2014 (UTC)
 * This version is irresponsible and a game of ridiculous semantics. First of all, one reference that did not find a direct link to specific cancer cel growth does not suggest that "Diet is (or was) generally not accepted" but rather that one study made a statement loosely resembling that. Most cancer research and treatment centers acknowledge that diet (indirectly through obesity and diabetes) is a strong risk factor and that diet (directly) may be a risk factor. So for Wikipedia to suggest that it is 'generally' accepted that his is not true is ridiculous. And irresponsible. And feels like lobbying. Lets get clear: Diet IS a risk factor. Of that, there is no question. SpecialFXavier (talk) 15:35, 12 November 2014 (UTC)
 * Sadly none of the sources, many of which are open access online, agree with you. "Risk factors" must be direct, and the factors further up the progression have always been included in their own rights, as they should be. That is the standard terminology. Not everyone who eats a lot gets obese and not everyone who gets obese gets diabetes. Wiki CRUK John (talk) 17:04, 12 November 2014 (UTC)

The World Cancer Report says "convincing evidence that body fatness increases risk and probable evidence in relation to abdominal fatness, whereas evidence is limited for red meat consumption; coffee drinking is unlikely to affect risk" Doc James  (talk · contribs · email) 23:31, 12 November 2014 (UTC)
 * Also the ref is about dietary factors not diet generally. Altered some to "Evidence for specific dietary factors is poor " Doc James  (talk · contribs · email) 23:35, 12 November 2014 (UTC)

what about this now? http://edition.cnn.com/2015/01/02/health/cancer-random-mutation/index.html?hpt=hp_c3 — Preceding unsigned comment added by 85.253.102.86 (talk) 18:07, 2 January 2015 (UTC)
 * That wasn't actually news, for pancreatic and other cancers, as the article (plus a little arithmetic) shows. Wiki CRUK John (talk) 20:56, 4 January 2015 (UTC)

WP:LEAD
We should keep the lead of the article to 4 paragraphs. Doc James (talk · contribs · email) 21:35, 12 November 2014 (UTC)
 * User:Wiki CRUK John are you going to discuss? Doc James  (talk · contribs · email) 23:41, 12 November 2014 (UTC)
 * (ec) It was 4 paras until you made further changes. Wiki CRUK John (talk) 23:48, 12 November 2014 (UTC)
 * John take a little more time to look at things. You were the one that changed it from that diff not me. I am counting 5 here  Doc James  (talk · contribs · email) 00:01, 13 November 2014 (UTC)
 * I took all the time I needed. It was at 4 paras, after you raised the point, and would have remained there if you had not made further, and imo largely unnecessary, changes. The main thing we must do is to avoid 2 people editing the same section at the same time, which just causes confusion. As here (note the timings) Wiki CRUK John (talk) 15:29, 13 November 2014 (UTC)
 * Just my 2p's worth: WP:LEAD says articles should 'ideally' be no more than 4 paras - I wasn't aware that this was a hard and fast rule (and something something 5th pillar something). Wouldn't it be infnitely more helpful to readers to divide up the paragraphs a little more, as it's a very dense piece, rather than squishing it together as it is at the moment? It is currently causing a numer of inadvertant non-sequiturs and repetitions, for example:


 * There are also a number of other types of pancreatic cancer. Pancreatic cancer is usually diagnosed by a combination of medical imaging...


 * Which actually caused me to pause and re-read just now. HenryScow (talk) 14:45, 17 November 2014 (UTC)


 * Agree and also support WP:IAR when obligatory run-ons make for confusing reading, as here. Obviously slavish adherence to the general LEAD guideline wouldn't serve the purpose here (and I think elsewhere - I seem to remember backing down in the past on this issue due to FA considerations, and I wouldn't venture to interpret that "ideally" from an FA rule-abiding perspective). If the wording in the general style manual effectively discourages having enough paragraph breaks for clear presentation in the lead of the different topic areas that are mandatory for articles on disease, then surely we need to ignore the rule? Either that or specify in MEMOS exceptions for cases such as this? 109.157.83.50 (talk) 10:14, 18 November 2014 (UTC)
 * I'm happy with that if others are, though it might cause rumbles at FAC I admit. I always think WP (including medical articles) has far more too-short leads than too-long ones, though 4 paras is a decent guideline for the majority of articles. Wiki CRUK John (talk) 11:42, 18 November 2014 (UTC)
 * If the proposal is to have all medical articles be 5 or 6 paragraphs than we should request wider input. Doc James  (talk · contribs · email) 18:32, 18 November 2014 (UTC)


 * I'm not suggesting this as a global change at all, just commenting that in this case, with the content as written, it needs more space to mitigate against confusion. I definitely think that for most cases, 4 paras will suffice, but in this case I think WP:IAR is worth invoking.
 * Basically all medical articles are similar with respect to the broad scope we are trying to get into 4 paragraphs. Thus WP:IAR is not suitable IMO. Doc James  (talk · contribs · email) 18:37, 18 November 2014 (UTC)

Repeated references clean-up
User: FeatherPluma is doing a great - and irksome - job cleaning up repeated references using shortened footnote templates. However, in in his citations the previous editor had been including the specific page where the cited information was found. Isn't it a better idea to use the shortened footnote template and still mentioning the specific page? NikosGouliaros (talk) 23:33, 15 November 2014 (UTC)
 * I agree that it is extremely easy to add the specific page to these shortened templates, using a reference page (rp) suffix. In principle, I considered doing so at the time I started going through the article. In broad terms, the modern compelling issue is that the rp information is not really adding much, for several reasons. Not the least of these is that electronic versions can be searched very rapidly for the specific component of interest. A lesser issue is that when I tested a quick subset of usage by electronic search and then cross-referenced to PDF format I found errors. FeatherPluma (talk) 23:56, 15 November 2014 (UTC)
 * Point taken. (To be honest, I only mention specific pages only within chapters from books or very long articles). On the other hand, I fear the result of the same reference being cited dozens of times might be impractical and unappealing to the eye. NikosGouliaros (talk) 00:11, 16 November 2014 (UTC)
 * A question was raised earlier here about adding page numbers. I added one wikitext option example. As you say, specific pages need only be given from books or very long articles, if at all. FeatherPluma (talk) 01:54, 16 November 2014 (UTC)

Pancreatic Cancer vs Pancreatic Adenocarcinoma
Hi! So this article is titled "Pancreatic Cancer" but the majority of info is of "Pancreatic Adenocarcinoma" meaning that "Signs and Symptoms" and "Risk factors" are all pertaining to panc. adenocarcinoma not the umbrella term "pancreatic cancer". Insulinom/Glucagonoma have separate articles. Is there any way we can make this article solely about "Pancreatic Adenocarcinoma"? and perhaps make another article about "Pancreatic Tumors" listing all the different tumors of pancreas?
 * We state in the lead that pancreatic cancer is sometimes used to mean pancreatic adenocarcinoma. We do discuss the other types briefly. As adenocarcinoma is the most common type it is also not unreasonable that much of the article covers that type. Feel free to add a bit more on the other types here. Doc James  (talk · contribs · email) 21:36, 16 November 2014 (UTC)
 * Yes, it is noticable how many of the MEDRS sources are called "Pancreatic cancer" but explain in para 1 they are actually all or nearly all about "Pancreatic Adenocarcinoma" (eg those with Wolfgang, Syl De La Cruz, & Wolfgang as the lead authors). "Pancreatic cancer" can be regarded as the WP:COMMONNAME of "Pancreatic Adenocarcinoma".  The next most common types all share many features with it in terms of management etc. "Pancreatic cancer" will certainly get by far the most traffic (little comes through the PA redirect page).  Wiki CRUK John (talk) 11:21, 17 November 2014 (UTC)

Ok thank you.

New onset diabetes and prevention research?
Pinging as this might, I suspect, be a pertinent topic  for the Research section he's kindly been compiling. (Though I don't think I'd support including it in the Prevention section, as it isn't currently used for screening, afaik.)

NB Perhaps worth bearing in mind that there currently seems to be relatively little mention on Wikipedia of pancreatogenic diabetes (see ), aka "type 3c diabetes mellitus". 109.157.83.50 (talk) 14:13, 17 November 2014 (UTC)


 * Thanks! Good ref, have added a couple of mentions in the Research section. HenryScow (talk) 18:17, 18 November 2014 (UTC)


 * Cheers - I've rephrased one of the sentences to try to provide a specific focus on screening, though my wording can almost certainly be improved (not altogether straightforward to explain succinctly for lay readers, who may not have much background on harm issues in screening). 109.157.83.50 (talk) 19:11, 18 November 2014 (UTC)

Opening line
Hi all. Am a little concerned that there's some circularity/lack of clarity/precision in the opening line of the lead...


 * 'Pancreatic cancer occurs when cancer cells develop in the pancreas'...

As it stands, this a) doesn't actually explain what cancer cells are (and directs people to the rather sub-optimal [Cancer] page for a definition), and b) implies cancer cells are conceptually some sort of uniform external 'other' when, as we know, it's a lot more complex than that and includes disregulated immune cells, fibroblasts etc. I would be much more comfortable talking about the disease as a condition arising from uncontrolled growth and regulation of cells or similar. How about either of these:


 * Pancreatic cancer arises when the cells that make up the pancreas - a glandular organ located behind the stomach - begin to multiply out of control.


 * Pancreatic cancer is caused by the uncontrolled growth of the cells that make up the pancreas - a glandular organ located behind the stomach.

Thoughts? HenryScow (talk) 14:56, 17 November 2014 (UTC)
 * Okay how about "Pancreatic cancer arises when there is uncontrolled growth of the cells that make up the pancreas - a glandular organ located behind the stomach."? Do not wish to refer to the mechanism as the cause. Also wish to define the disease before defining what the pancreas is. Doc James  (talk · contribs · email) 00:03, 18 November 2014 (UTC)
 * I like having a link to cancer pretty early. At one stage I don't even think it was in para 1. Wiki CRUK John (talk) 01:08, 18 November 2014 (UTC)
 * Same. I have moved the link to the second sentence. Doc James  (talk · contribs · email) 01:16, 18 November 2014 (UTC)
 * Much better! If I'm being really pedantic, I prefer 'growing' (a verb) to 'growth' (a noun) as it makes it more active and comprehensible. But basically happy as is.HenryScow (talk) 17:56, 18 November 2014 (UTC)
 * Sure sounds good to me. Doc James  (talk · contribs · email) 18:33, 18 November 2014 (UTC)

and The current version suggests that we are dealing with a diffuse process (hyperplasia), which is incorrect. I like Henry's first suggestion. (See also the FAC page.) JFW &#124; T@lk  00:40, 19 December 2014 (UTC)
 * Yes agree not ideal right now. How about as two sentences? Doc James  (talk · contribs · email) 01:01, 19 December 2014 (UTC)

Imo, the current version is pretty damn good. Just maybe (though really not sure... ...) more reader-friendly to use brackets for the gloss, like this: Pancreatic cancer arises when cells in the pancreas (a glandular organ behind the stomach) begin to multiply out of control and form a mass. 109.158.8.201 (talk) 13:08, 23 December 2014 (UTC)

Peer review closed
From the peer review: Closing comments I'm closing this now, as it has run 3 weeks or so, and gone quiet. Very many thanks to all who contributed here, or just by editing the article, as several did. I think all points have been addressed, or by-passed by other changes, except for the image question in the last section. I'll copy that to the article talk page, and if anyone has further comments on other matters below, please continue the discussion there. I will take the article to FAC shortly. Thanks again for a great response! Wiki CRUK John (talk) 10:33, 1 December 2014 (UTC)
 * Thank you for doing good work together. How about a good article nomination? NikosGouliaros (talk) 10:55, 1 December 2014 (UTC)
 * Straight to FAC I think; this peer review has been more thorough than most GANs, with the advantage of several reviewers. Wiki CRUK John (talk) 12:00, 1 December 2014 (UTC)

Image error
I just noticed that the first image in the signs & symptoms section (File:1820 The Pancreas.jpg) erroneously shows pancreatic hormones flowing from the pancreas into the splenic artery (even in the wrong direction, against the bloodflow). The correct image should show hormones flowing into the splenic vein and the pancreaticoduodenal veins, which then drain into the portal vein. --WS (talk) 20:06, 23 November 2014 (UTC)
 * Damm! There doesn't seem to be another image as good. Can anyone else confirm this, though I'm sure WS is correct? Wiki CRUK John (talk) 09:05, 24 November 2014 (UTC)
 * Wouterstomp is right. The blood vessel below the splenic artery appears to be the splenic vein&mdash;at least it is in the right position, just behind the upper part of the pancreas. By anatomical convention, the splenic vein, like other systemic veins, is often coloured blue on diagrams. Axl ¤ [Talk] 12:26, 27 November 2014 (UTC)
 * Thanks. Oh well, it will have to go. Maybe the producers can correct it. Unfortunately the Commons images are not well categorized at all, but I can't see a decent substitute. Wiki CRUK John (talk) 13:22, 27 November 2014 (UTC).
 * How about this one? A wider scope, and less detail, but the best I can see for now. There is also room for this (COI declared), with more detail on the ducts. Wiki CRUK John (talk) 18:53, 30 November 2014 (UTC)

We can request the image to be fixed by filling in an errata report to CNX. I will do so, but until that has been done maybe we should use another image. -- CFCF  🍌 (email) 14:42, 1 December 2014 (UTC)
 * We could potentialy use this image instead. -- CFCF  🍌 (email) 14:47, 1 December 2014 (UTC)
 * Thanks. If we used that, I think it should replace the 2 images now at "Signs and symptoms". I presume that can be loaded in terms of licensing etc? Wiki CRUK John (talk) 15:29, 1 December 2014 (UTC)

Anything else outstanding?
Please add here, if so. Wiki CRUK John (talk) 10:50, 1 December 2014 (UTC)


 * RE the PR discussion about the repetition of symptoms (or signs?) opening of the "Diagnosis" section: I think  raises a serious editorial point here (John, sorry not to have got onto this earlier!) . If some combination of signs existed that can signal a specific diagnostic suspicion ('red flag') of this disease I would certainly support stating that. But here, apart from Courvoisier sign and early symptoms in functioning PanNETs, the only other clinical red flags I can see might conceivably be new onset of diabetes or pancreatitis. So it seems to me unhelpful to return to a generalized account of "symptoms" (ie experiences of the patient, as distinct from the range of "signs" available for a physician to interpret) which are really not diagnostic. The one pertinent part, imo, could be the opening sentence about the lack of clinical *signs* - after some rephrasing for focus. I think it should be possible to merge the other content on symptoms into the "Signs and symptoms" section in a way that is still friendly to lay readers. 2c, 86.164.164.29 (talk) 16:46, 2 December 2014 (UTC)


 * Any progress on this? The current text is potentially misleading to readers, imo (including perhaps to ones with a naive but probably inevitable gf tendency to "self-diagnose"). Not sure how to put it in WP:... terms, but I'll have a go... Is all the content you're repeating/presenting here actually sourced to specific passages of text directly discussing Diagnosis as such? If not, then I find it hard to see how it can meet the strict FA criteria (though I don't pretend to understand the process) . In non-WP terms, I think a perfectly acceptable (imo) alternative editorial approach could be to reframe the three opening paragraphs of the section (that currently starts with "Pancreatic adenocarcinoma has a number of symptoms...") so as to focus directly on questions of diagnostic suspicion (rather than providing unfocussed content on "symptoms at diagnosis" and other general background information). Hope this helps. Fwiw, I am very supportive of the effort to increase reader-friendliness. 86.164.164.29 (talk) 10:37, 15 December 2014 (UTC)
 * From the Bond-Smith et al review in the BMJ: "'... Development of any of the above symptoms in the presence of late onset diabetes should strongly alert the physician to the possibility of pancreatic cancer... The clinician should be alert to a potential diagnosis of pancreatic cancer with patients over 50 years old who present with unexplained weight loss, persistent abdominal or back pain, dyspepsia, vomiting, or change of bowel function...'" 109.158.8.201 (talk) 10:26, 28 December 2014 (UTC) previously 86.164.164.29

I feel that some sort of response to this issue is eventually necessary (at least before the FAC process closes). These sorts of questions have also been raised by several reviewers at FAC (I posted here so as not to clutter the FAC page, but do feel free to move these posts). As I said above, my own take on this is sympathetic with inclusion of clinical signs in the Diagnosis section, but not with apparently arbitrary/unfocused content/repetition regarding symptoms that are present at the time of diagnosis. Imo, the blockquote above (from the Bond-Smith et al review) provides a helpful illustration of how this sort of content could be refocused. Adopting this sort of approach, I think it might be editorially appropriate to retitle/reframe ===Signs and symptoms=== as ===Symptoms=== (as WhatamIdoing has clarified, I think, at WT:MED, here, the headings suggested at MEDMOS are only *suggestions* and are by no means mandatory). 86.181.67.166 (talk) 10:01, 8 January 2015 (UTC), previously 109.158.8.201, 86.164.164.29 etc
 * Most of this was removed a while back, but I'm still convinced it's best to keep the more complicated relationship between symptoms and location of the tumour on the pancreas in the diagnosis section. Otherwise, there is a briefer list of symptoms with more %s and a little on differentiation and so on. I really don't want to put tiredness and a dry mouth into the main symptoms section, & would rather cut these minor and very non-specific symptoms completely (they are'nt in many or most sources) than move them there, which goes for all of this I think. The section is called diagnosis, and I don't see the difficulty with having information essential for diagnosis in there, even at the price of some repetition. Johnbod (talk) 14:26, 11 January 2015 (UTC)
 * John, the whole point is that at the moment there's a whole lot of information in there that not only is not "essential for diagnosis", but has little or no obvious relevance to this specific Diagnosis (other than being commonly present at the time of clinical presentation[and, perhaps, being relevant to a 'nursing diagnosis' – something not currently mentioned]). Your example of tiredness and a dry mouth actually highlights the issue rather well. Given that  tiredness and a dry mouth are listed as "non-specific symptoms", isn't the real editorial choice here either: a) to remove them altogether (as you helpfully suggest); or b) include them in the symptoms section (hum, maybe not...)? But not: c) to leave them under ==Diagnosis== where (as "non-specific symptoms") they have no clear relevance... This seems to me to be another false dilemma. Like the larger one of accessibility vs. clarity. If we can get beyond these false choices (eg by using sourcing like this), I think the article could benefit in both clarity and accessibility. 86.181.67.166 (talk) 16:58, 11 January 2015 (UTC)


 * tl;dr– There is an obvious editorial rationale for starting ==Diagnosis== with diagnostic red flags. On the other hand, listing non-specific symptoms under ==Diagnosis== instead of ==Signs and symptoms== (or ==Symptoms==?) seems sloppy and not what we're trying to do here for accessibility to lay readers, and clarity for everyone. I think we should agree to fix this mismatch before the page becomes an (exemplary) FA. 86.181.67.166 (talk) 11:11, 12 January 2015 (UTC)

Management of PanNETs
This subsection] still contains several statements that are apparently unsourced. (The reader is also referred to a "section of the main article on neuroendocrine tumors", but I'm not sure how appropriately.) 109.158.8.201 (talk) 10:57, 21 December 2014 (UTC)

Citation work and other MOS stuff
I would like to put the article Template:In use during a quiet moment of editing to standardize the citation formatting (see crit 2c). Editing seems to have slowed down, but there has been no response on the FAC to my query about which format to use for dates and author names. It looks like the preferred date format is not ISO, rather spelled out as day month year, but the author name format is still all over the place. I prefer the Diberri/BogHog format, because its appearance is so clean: and because you can generate a template from a PMID in that format by using the Diberri/BogHog PMID filler. I could switch all to that unless others prefer something else. The other format in the article now (several formats are present now) is less clean-looking: that is, more cluttered with extra punctuation. Please let me know, because doing this work is quite tedious, and getting edit conflicts or reverted afterwards stinks! And there are many incomplete citations. I can complete those when I'm in there. And do the WP:NUMERAL corrections. And WP:CAPFRAG. And WP:DASH (there are hyphens that should be WP:ENDASHes). I like to get the MOS-y stuff out of the way before reviewing content. Today is good for me to do this tedious stuff-- weekend, less likely. Sandy Georgia (Talk) 21:25, 5 December 2014 (UTC)
 * Jones AB, Smith CD, Miller EF
 * Jones, AB; Smith, CD; Miller, EF
 * Struck some done, but I wasn't checking comprehensively-- will look for more of those when I get into actually reading ... so far, just MOS flyover. Sandy Georgia  (Talk) 03:18, 6 December 2014 (UTC)


 * So, at the moment the article has, for one much-used source (this I think is as changed to by User:FeatherPluma):


 * which using the edit window template (as it was originally put in by me) is:
 * which using the edit window template (as it was originally put in by me) is:


 * and using the Diberri/BogHog PMID filler is:
 * and using the Diberri/BogHog PMID filler is:



- FeatherPluma's style is similar but not identical to the Diberri/BogHog. He has commented in a section above, & that might be relevant. I won't even bother giving my own preference.

There was a big edit in May 2014 by Boghog, just after the CRUK project began but nothing to do with it, this adjusted things in accordance with a Med project consensus on Doi's & such like. From that diff you can see the previous style, which also seems somewhat mixed, but perhaps mostly in the style that User:FeatherPluma recently adjusted many refs to. I must admit that when I began editing I didn't worry too much about the existing style, nor I am I quite sure what the existing project consensus is, or if that affects things here.

For example the May edit changed from:
 * - to-
 * - ok those look the same.
 * Sandy, I think there is probably no established style or it is Diberri/BogHog, and it has not been much enforced. So no one should object if you convert to D/B. I would certainly be very grateful. Wiki CRUK John (talk) 23:04, 5 December 2014 (UTC)
 * We missed my afternoon free-time editing window, but I may be able to start cleanup after dinner. Maybe what I'll do is all the other little stuff first, and leave author formats for last.  That way, if someone hates it, they can revert just that portion.  Which would be a bummer because it's so time-consuming, but I can live with it ;)  Sandy Georgia  (Talk) 23:16, 5 December 2014 (UTC)
 * Cool! If I am editing this page, tell me & I'll stop. If anyone else has views on the citation style, please speak now or .... Wiki CRUK John (talk) 23:25, 5 December 2014 (UTC)
 * I'm going to fix dinner, and start work in about an hour. Sandy Georgia  (Talk) 23:35, 5 December 2014 (UTC)

I don't know what this is ... we have to give readers enough info to be able to verify, that is, find the information. Where do they look for this? Is it a journal, a website, Pubmed indexed, a book, what? Sandy Georgia (Talk) 04:40, 6 December 2014 (UTC)
 * Ryan DP. Chemotherapy for advanced exocrine pancreatic cancer. UpToDate. Review current through October 2014: Topic 2475, Version 46.0

Diff of work done so far-- nowhere near complete, too tired to continue. I found and inline noted a couple of citations that I have no idea what they are, a book whose citation I can't fix, found many missing PMIDs and PMCs, moved quotes in to quote parameter, more different styles than I thought, haven't even begun to look at prose or suitability of sources. That was much more work than I expected, pls check my work. Sandy Georgia (Talk) 05:28, 6 December 2014 (UTC)

Uncited text
I see lots of uncited text in Diagnosis??? Sandy Georgia (Talk) 05:28, 6 December 2014 (UTC)
 * gone Wiki CRUK John (talk) 20:31, 6 January 2015 (UTC)

NORD
I was just beginning to look at how to standardize the cite web templates, when I encountered this: NORD has them listed in their database of rare diseases, but the problem I have is that the NORD database is demonstrably inaccurate. Can a better source for this information be found? As an example of why NORD is not a good source, they state that a disease is rare if it affects fewer than 200,000 Americans. Yet they *still* list Tourette syndrome as a "rare disease", based on decades old prevalence estimates, when it has been acknowledged for a very long time that TS is not rare. (They don't remove things from their database when they are no longer considered rare.) Are NETs considered rare by someone other than NORD? Do we have a US prevalence number? So, I popped down to the Epidemiology section of this article to try to answer that, and am unable to find a satisfactory answer to the "fewer than 200,000 Americans" according to NORD. Can the "rare" issue here be either better sourced, or rephrased, or actual prevalence numbers given? Back to citations ... I got stalled on that. Sandy Georgia (Talk) 01:08, 6 December 2014 (UTC)
 * NETs can occur in most organs of the body, including the pancreas, where the various malignant types are all considered to be rare.
 * I've removed the ref, which was not added by me. Previously went to the next ref, Burns & Edil, whose abstract begins ""Pancreatic neuroendocrine tumors are a diverse group of rare neoplasms." though they begin the "epidemiology" section with "Initially believed to be exceedingly rare, neuroendocrine tumors are likely more common than previously appreciated.", going on cite SEER US incidence for 2004 at 5.25 per 100,000, up x5 from 1973 (this is all NETs, not just PanNETs), and then further muddying the waters with autopsy studies. Very strictly it might be better to say "clinically-relevant" PanNETs are rare (which I don't propose to do). Burns & Edil are both from John Hopkins.  This is enough I think, though there are issues identifying small tumours, and defining "rare", as Rare disease also points out. But as there are several types, with none predominating, the wording used meets all definitions, I think. I also think it's best not to get too specific on prevalence under the circumstances, though I could add something if people think I should.  "Fewer than 200,000 Americans" is the definition per the US Rare Diseases Act of 2002, roughly = 15 per 100,000 (in the UK press size tends to measured in London buses or football pitches, but I don't think that's a good idea either). Thanks for the great work! Wiki CRUK John (talk) 01:37, 6 December 2014 (UTC)
 * PS, I see NORD also list HIV/AIDS (under A)- if only! Wiki CRUK John (talk) 01:59, 6 December 2014 (UTC)

Other sourcing

 * I will catch up with posts here when I come up for air ... doing citation work requires full focus :) I also re-check my work for mistakes when done. Meanwhile ... Please convince me why this is the kind of source we want to use for an FA? It looks like an advocacy organization??  Sandy Georgia  (Talk) 02:01, 6 December 2014 (UTC)
 * Well I suppose it is, though I'm not sure why you pick on CRUK rather than the adjacent ref to the American Cancer Society, which actually seems to spend a lower proportion of its revenue on research and a higher proportion on things you might call "advocacy". Both links go to a very up to date tour d'horizon of areas of current research activity, of the sort that classic MEDRS sources don't generally cover in one sweep (the statement referenced is "There are several fundamental unanswered questions"). At least we've not found such a source. Obviously I have a WP:COI with CRUK, as does User:HenryScow who added both refs back on 17 July. Wiki CRUK John (talk) 02:34, 6 December 2014 (UTC)
 * oopsie !!! Sorry, John, as I was editing it did not even occur to me that Cancer Research UK is you all  ... not thinking CRUK, and since I'm from the other side of the pond, it's not on my brain.  oh my.  Well, OK, the problem I have with it is that it seems akin to, in my "world", citing Tourette syndrome to literature from the Tourette Syndrome Association, rather than journal-published research.  Their agenda, for want of a better word, is more oriented to advocacy.  I picked on CRUK because I had never heard of it (in its full name version, not realizing that was you all), but I actually do have the same problem with the American Cancer Society.  I don't think they are the highest quality sources ... at least, in the realm I edit, citing TS to the TSA would not be the best sourcing.  In fact, as discussed at WT:MED and WT:MEDRS several times, I don't even like citing to the NIH or Mayo (and this article cites Johns Hopkins)... the NIH has errors in its TS literature.  So, yes, it is a broader, general question, and I only grabbed CRUK because it was the first I encountered that I didn't know.  I hadn't yet gotten to the level of checking what text was being cited to these kinds of sources ... was just running through the cite web templates and hit one I didn't recognize :) I did cite web first because I want to be able to check everything cited to websites closely once I start reading ... which I haven't done yet.  Back now, still have to clean up cite journal templates, and probably won't get to any reading until tomorrow.  Just catching what I notice on flyover wrt prose.  Sandy Georgia  (Talk) 03:24, 6 December 2014 (UTC)
 * oopsie !!! Sorry, John, as I was editing it did not even occur to me that Cancer Research UK is you all  ... not thinking CRUK, and since I'm from the other side of the pond, it's not on my brain.  oh my.  Well, OK, the problem I have with it is that it seems akin to, in my "world", citing Tourette syndrome to literature from the Tourette Syndrome Association, rather than journal-published research.  Their agenda, for want of a better word, is more oriented to advocacy.  I picked on CRUK because I had never heard of it (in its full name version, not realizing that was you all), but I actually do have the same problem with the American Cancer Society.  I don't think they are the highest quality sources ... at least, in the realm I edit, citing TS to the TSA would not be the best sourcing.  In fact, as discussed at WT:MED and WT:MEDRS several times, I don't even like citing to the NIH or Mayo (and this article cites Johns Hopkins)... the NIH has errors in its TS literature.  So, yes, it is a broader, general question, and I only grabbed CRUK because it was the first I encountered that I didn't know.  I hadn't yet gotten to the level of checking what text was being cited to these kinds of sources ... was just running through the cite web templates and hit one I didn't recognize :) I did cite web first because I want to be able to check everything cited to websites closely once I start reading ... which I haven't done yet.  Back now, still have to clean up cite journal templates, and probably won't get to any reading until tomorrow.  Just catching what I notice on flyover wrt prose.  Sandy Georgia  (Talk) 03:24, 6 December 2014 (UTC)


 * Don't worry, no reason you should. But there are some things charity websites can do very well - state of research & stats (for non-specialists) are 2 examples, both for us and ACS. Both are often cited on such things by the classic MEDRS papers, btw. The alternatives are primary papers, mostly in this case ones which won't be published for at least a year or so, or typically reviews etc (eg Wolfgang et al.) which are very good on depth, but tend to have a narrower focus when looking at research directions. Wiki CRUK John (talk) 03:20, 6 December 2014 (UTC)


 * Back to work for a bit. Best, Sandy Georgia  (Talk) 03:24, 6 December 2014 (UTC)

Prose and sourcing concerns
Things I see as I'm checking citations (but please fix after I'm done so we don't edit conflict !!)


 * The nature of the changes that lead to the disease are being intensely scrutinized, such as the role played by genes such as KRAS and p53.
 * Repetitive prose ... such as ... such as ... and isn't gene linked sooner in the article?  Nature ... are ??  Think the whole sentence needs to be rejigged.  Sandy Georgia  (Talk) 02:10, 6 December 2014 (UTC)
 * ✅. Sandy Georgia  (Talk) 16:11, 15 December 2014 (UTC)
 * Ok, sorry if I caused an ec earlier. I'm off to bed now. Wiki CRUK John (talk) 02:35, 6 December 2014 (UTC)
 * Hang on ... I'll go take my bubble bath in case you want to squeeze in any work now ... Sandy Georgia (Talk) 02:38, 6 December 2014 (UTC)

Personal prose preference: "...  that arise from neuroendocrine cells (i.e. cells responsible for integrating ... " I hate seeing i.e. in articles, doesn't seem very user friendly (some folks may not know what it is).  Change only if you agree.  Off for a moment, back later.  Sandy Georgia  (Talk) 02:38, 6 December 2014 (UTC)
 * The 2 "i.e."s gone, and a purge of "such as" . Wiki CRUK John (talk) 15:15, 15 December 2014 (UTC)
 * ✅ Sandy Georgia  (Talk) 16:11, 15 December 2014 (UTC)
 * I'll aim for a good run through over the W/e. Wiki CRUK John (talk) 03:09, 6 December 2014 (UTC)

✅ WP:RED is a good thing (it helps build the website), but: the reader is left with no idea what a hepatic artery embolization is. That can be fixed by either writing a stub at the article, or adding a parenthetical explanation. As of now, we have to exit Wikipedia and go to google to figure out what it is. Sandy Georgia (Talk) 02:51, 6 December 2014 (UTC)
 * Radiofrequency ablation (RFA), cryoablation, and hepatic artery embolization may also be used.
 * Yes, I think that needs to go anyway, as I can't find a ref. Wiki CRUK John (talk) 03:09, 6 December 2014 (UTC)
 * There are abundant sources for hepatic artery embolization, including recent reviews. See note below, entered chronologically (time stamp 00:57, 10 December 2014) FeatherPluma (talk) 01:01, 10 December 2014 (UTC)
 * ✅ Sandy Georgia  (Talk) 16:12, 15 December 2014 (UTC)

I haven't started really reading yet, but just in there working on citations and MOS, I'm seeing rough prose everywhere I glance ... More than 90% of all grades carry a faulty KRAS gene ... grades carry genes? Or people with cancer at a certain grade carry a gene ? Sandy Georgia (Talk) 05:34, 6 December 2014 (UTC)
 * The latter: "cases at" added. Wiki CRUK John (talk) 20:20, 17 December 2014 (UTC)

Change in tense, change in number style, and not clear if the 46 is among that 2/3 or among entire population:
 * In up to two-thirds of people abdominal pain is the main symptom, for 46% accompanied by jaundice, and 13% have jaundice without pain.
 * The latter, rephrased. Wiki CRUK John (talk) 20:20, 17 December 2014 (UTC)

This list changes in tense/style: The first three start with sentence fragments; the last two are full sentences/statements.
 * Bearing in mind that the disease is rarely diagnosed before the age of 40, common symptoms of pancreatic adenocarcinoma occurring before diagnosis include:
 * Yes, #4 includes various things in a sequence, and it would be misleading to start #5 with "diabetes" as it's more complicated than that. If consistency was absolutely vital I could scramble up 1-3 so they turn into sentences, but I don't think this would be an improvement. I'm inclined to leave it as is. Wiki CRUK John (talk) 22:34, 17 December 2014 (UTC)

Excessive wordiness:
 * The undigested fat leads to fatty feces that are foul-smelling and difficult to flush away.
 * The undigested fat leads to foul-smelling, fatty feces that are difficult to flush.
 * Done, keeping "away". Wiki CRUK John (talk) 21:59, 17 December 2014 (UTC)

Parentheticals have the reader bouncing around the article: ugh ... don't bounce the reader around, and why the need to repeat?
 * cancer (see Risk factors), the
 * and, as noted in Signs and symptoms, ...
 * Evidence for specific dietary factors is poor (though see the sections on obesity and alcohol).

What is meant by "usual" here? Relative to all people with the disease, or what population?
 * People over 50 who develop diabetes have eight times the usual risk of developing pancreatic adenocarcinoma within three years, after which the relative risk declines.
 * "compared to the general population" the source says, but that might be the general population of that age. to check his cited source. Wiki CRUK John (talk) 22:34, 17 December 2014 (UTC)
 * Yup: "In this cohort, we determined the likelihood of pancreatic cancer diagnosis within 3 years of meeting criteria for diabetes and compared observed rates with those expected for persons of similar age and sex distribution." - PMID 16083707. Is it enough to add this to the reference? The 2nd level ref is 2005, and I don't like to go beyond what the first ref actually says without the further reference. Wiki CRUK John (talk) 22:42, 17 December 2014 (UTC)

Is "some" really needed?
 * with pancreatic cancer in some 10–20% of cases,
 * It helps to indicate that this is a very imprecise number, which it is. Wiki CRUK John (talk) 22:34, 17 December 2014 (UTC)

Awkward phrasing: Or is perhaps? Or may be a result of? Why isn't final sentence merged in to first sentence?
 * The depression may be there before the cancer is diagnosed, and is perhaps caused by the cancer. The mechanism for this association is not known.
 * Changed to "or". Short sentences are good. Wiki CRUK John (talk) 22:34, 17 December 2014 (UTC)

Wordiness? Unsure here: The primary pancreatic cancer may metasti (not sure if you are using british or english spelling) to other organs? Why so much wordiness? It reads as if the wordiness may be the result of trying to work in the wikilink ??
 * There may be symptoms of pancreatic cancer metastasis, where the primary pancreatic cancer has spread to other organs.
 * I don't think it is - just of trying to be accessible. Wiki CRUK John (talk) 22:34, 17 December 2014 (UTC)

Another list that doesn't have a consistent style: Some items in the list are bullet, fragments, while others are complete sentences. Some bullets end in periods, other in colons.
 * Risk factors for pancreatic adenocarcinoma include:
 * Tidied to sentences, with semi-colons after bullets as appropriate. Wiki CRUK John (talk) 22:51, 17 December 2014 (UTC)

Another "some": Average? Up to? Why the vague "some"?
 * taking some 20 years to return ...
 * No, some, about, around, more or less! Please think what "average" might mean in this situation, and you'll see why there isn't an average, and why it is vague. There's a clamour for unattainable precision from various reviewers here. Wiki CRUK John (talk) 22:06, 17 December 2014 (UTC)
 * Further, from the source: "... risk decreased with increasing time since cigarette cessation, the OR being 0.98 after 20 years. ... A 20% excess risk of pancreatic cancer was found among former smokers, which declines with time since quitting, and reached the level of never cigarette smokers ∼20 years after quitting.... ... More specifically, our large population and detailed data on smoking allowed us to confirm that after 20 years of smoking cessation, risk of pancreatic cancer approaches that of never smokers", and in another study the same source cites " ...the RR for those who had quit smoking for >15 years was similar to that of never smokers". Wiki CRUK John (talk) 23:00, 17 December 2014 (UTC)

WP:NUMERAL should be three-fold, but there is a 1.5-fold later, so I left it for consistency.
 * an almost 3-fold increase
 * All "fold" numbers gone; I suspect they are not understood by many (more common in US English these days perhaps). Wiki CRUK John (talk) 00:47, 6 January 2015 (UTC)

Reduce wordiness? Alcohol consumption and abuse is associated with pancreatitis, but have not been established as risk factors for pancreatic cancer; most studies have found no association. Or something to that effect, less wordy?
 * While the association between alcohol abuse and pancreatitis is well established, considerable research has failed to firmly establish alcohol consumption as a risk factor for pancreatic cancer. Overall, the association is consistently weak and the majority of studies have found no association.
 * No, sorry. I think the "wordy" version conveys better that this is a much researched area that has failed to prove anything. Wiki CRUK John (talk) 00:39, 6 January 2015 (UTC)
 * I've added the 2014 diet review ref, & removed the old primary paper. Wiki CRUK John (talk) 04:05, 6 January 2015 (UTC)

More on sourcing
NOW, there I stop because the sources are not adequate to the statement (and I haven't even begun to look at sourcing throughout the article-- just saw that on my prose check). One source is an extremely old press release from an advocacy group, and the other source is a very old primary study. And the weakest kind of study (a questionnaire). So, having cleaned up the citations to the point of being able to see what the sourcing is, I'm concerned that this article needs work not only on prose, but more seriously, on sourcing. Upgrading all web sources to recent high-quality secondary journal reviews will help.

So, those are a few samples from just a few sections (it doesn't make sense at this point to list every problem). I'm finding (as I found five different citation styles and mechanisms in the article) that (perhaps because it was built by many editors??) the article does not have a smooth voice, and could benefit from an overall prose check and smoothing by one copyeditor. At the same time, it would be optimal if all text cited to websites (individual hospital programs or advocacy groups) could be reviewed and cited where possible to journal sources. Because the alcohol information was not sourced to a secondary review, I'm concerned that every citation should be checked to make sure they are high quality, recent, secondary reviews. Someone (can't remember who?) at WT:MED has a script that can flag on talk which sources are primary and which are secondary. Perhaps ask is he knows? If you ping me when a copyedit has been done, and sourcing has been tightened, I will next read through the entire article for jargon, layperson readability, etc. Best, Sandy Georgia  (Talk) 14:08, 6 December 2014 (UTC)
 * Thanks enormously for this. A few parts of the article are pre-2014 survivals, but also the PR & start of the FAC generated loads of direct edits by several editors (over 700 edits in November I think), on prose as well as references, which have varied the "voice".  The problem with alcohol is that recent reviews (and clinical reviews deal pretty briskly with well-established risk factors) simply don't mention alcohol - eg the Ryan & Wolfang ones, which both have tables of RFs, nor Vincent (against the grain, Bond-Smith does include it in a list in his BMJ review, but gives no specific source). Yet it was an issue some way back, and I think the lack of association should be stated (whereas the similar case of coffee doesn't need to be).  This is similar to Breast cancer and smoking. I'll see if textbook sources can give better evidence of absence, but I'm not back in the office until Wednesday now. It certainly needs an all-over prose check. Wiki CRUK John (talk) 15:00, 6 December 2014 (UTC)
 * I was afraid you would say you regret knowing the former FAC delegate :) :) Sandy Georgia  (Talk) 15:03, 6 December 2014 (UTC)

Checking back in, I see that did a nice copyedit, but most of the above has not yet been corrected. Sandy Georgia (Talk) 14:58, 8 December 2014 (UTC)

Comprehensive, 1b; well-researched, 1c
I see suggested  (a recent secondary review that has free full text available). Regards, Sandy Georgia (Talk) 15:03, 6 December 2014 (UTC)
 * Included in three sections now. Wiki CRUK John (talk) 17:48, 6 January 2015 (UTC)
 * Now removed from risk by 86, with the edit summary "trim an uncritical review of the evidence (redundant vs. cited SR with meta-analysis)". Wiki CRUK John (talk) 15:15, 8 January 2015 (UTC)
 * A new review on screening:  and  Sandy Georgia  (Talk) 00:48, 9 December 2014 (UTC)
 * I'm not sure about these; both are calls for changes in practice/policy, and I suspect it would be premature/undue to include them in the screening section. I don't hear that this call has gained much traction at present. We have 2 2014 refs already on this, but I daresay things move slowly. I've added them to research directions. Wiki CRUK John (talk) 00:17, 6 January 2015 (UTC)
 * A new review on treatment:  Sandy Georgia  (Talk) 00:48, 9 December 2014 (UTC)
 * Thanks, but I don't think this is needed; "emerging" treatment in East Asia. Not mentioned in the several other general reviews from the last 12 months. Wiki CRUK John (talk) 02:44, 10 December 2014 (UTC)
 * OK ... I see Feather is hard at work, will look in tomorrow. Sandy Georgia  (Talk) 03:17, 10 December 2014 (UTC)


 * There is extensive available sourcing for hepatic artery embolization and the other arterial approaches mentioned in the endocrine tumor section. To accommodate the above input I have put together a quick Wikipedia article so it's no longer WP:RED. I won't have time to work much on it right now but it's a start.FeatherPluma (talk) 00:57, 10 December 2014 (UTC)

Pingie pongie
I'm wondering if/when my outstanding points will be addressed? I'd like to unwatch sometime in this lifetime :) Sandy Georgia  (Talk) 17:26, 17 December 2014 (UTC)
 * I hope they are all addressed now. Sorry for the wait! Wiki CRUK John (talk) 20:33, 6 January 2015 (UTC)

Comments from Doc James
, I've merged all of your comments to one thread for easier tracking of what has been completed for the FAC; please revert me if you dislike this :) Sandy Georgia (Talk) 15:00, 8 December 2014 (UTC)
 * Thanks for doing this Sandy.  Doc James  (talk · contribs · email) 21:58, 8 December 2014 (UTC)

Predictions
I am not convinced this belongs in the lead "Due to lack of progress with treatment, it is predicted to be the second leading cause within a decade."check to WCR ref"" I generally think predictions are not great as they depend on so many factors.

Also I am looking through the world cancer report and not seeing it mentioned. And what does within a decade mean? By 2024? Doc James (talk · contribs · email) 17:46, 7 December 2014 (UTC)
 * So you've removed. This was one of a few additions suggested by the top clinician outside reviewer. Yes 2024. I think, with a ref, it is useful in the distribution section rather than the lead. Wiki CRUK John (talk) 15:06, 15 December 2014 (UTC)
 * Were in the ref does it support this? Doc James  (talk · contribs · email) 23:10, 17 December 2014 (UTC)

We appear to have some issues
While I updated the lead of the article a while ago these updates do not appear to have taken place consistently in the body of the text.

We stated that "Pancreatic cancer has one of the highest fatality rates of all cancers, and is the fourth-highest cancer killer among both men and women worldwide. "

It is referred to cancer.gov which usually only provides states for the States. It however has been like that for a long time. Have updated / fixed to represent actual global stats. Doc James (talk · contribs · email) 18:01, 7 December 2014 (UTC)

Article needs some simplification
Terms such as: Doc James (talk · contribs · email) 18:05, 7 December 2014 (UTC)
 * "Definitional migration" have changed to "Changes in definitions over time"
 * etc
 * Now: "The definition and classification of PanNETs has changed over time[74] which adversely affects what is known about the epidemiology and outcomes without treatment for these tumors.[74]". No "etc"s left. Wiki CRUK John (talk) 15:03, 15 December 2014 (UTC)


 * suggest: "...the epidemiology and natural history of these tumors" (fwiw, I had to read the sentence twice to understand the phrase "...outcomes without treatment for these tumors"). 86.164.164.29 (talk) 09:27, 16 December 2014 (UTC)
 * Epidemiology and natural history are not simple English but technical terms. "outcomes without treatment" are at least common words. Doc James  (talk · contribs · email) 22:12, 17 December 2014 (UTC)
 * Now fixed, I think. 109.158.8.201 (talk) 10:35, 28 December 2014 (UTC), previously 86.164.164.29

Aspirin
Concerned with this "Taking a daily low-dose aspirin regimen for more than five years may decrease the risk of developing pancreatic adenocarcinoma (ductal pancreatic cancer) by as much as 75%. " which text supports this from the ref?

We also have which did not find an effect from low dose ASA.

This was also much more reserved

Doc James (talk · contribs · email) 05:38, 12 December 2014 (UTC)
 * And there's this, which doesn't mention those two. I didn't add this bit & haven't seen the Rothwell. "low" seems clearly wrong. There are also the complicating effects of other potential beneficial & harmful effects. I doubt anyone will or should take a decision on taking regular aspirin based mainly on pancreatic cancer. I'll cut for now, but if anyone disagrees please say so. Wiki CRUK John (talk) 14:42, 15 December 2014 (UTC)

Risk factors removed
I've cut these, previously commented out, as not typically found in recent reviews, and with older refs: Wiki CRUK John (talk) 14:18, 15 December 2014 (UTC)
 * Gingivitis or periodontal disease
 * Partial gastrectomy PanNets?

"fatality rates"

 * In the Distribution section: Pancreatic cancer has one of the highest fatality rates of all cancers, and is the seventh-highest fatal cancer among both men and women worldwide.[ref WCR-2014, chapter 5.7] Pancreatic cancer has one of the highest fatality rates of all cancers, and is the seventh-highest fatal cancer among both men and women worldwide. I assumed the intended meaning here was "Pancreatic cancer has one of the highest fatality rates of all cancers—the seventh(??) highest among both men and women worldwide", but when I consulted the 2014 World Cancer Report to check I couldn't find a corresponding statement regarding fatality rates as such. Was this statement perhaps sourced to the bullet point in the "Summary" (I couldn't see any more details provided in the rest of the chapter) "Pancreatic cancer is the seventh most common cause of cancer death worldwide..." If so, that refers to the absolute numbers of deaths, rather than a "fatality rate". 109.158.8.201 (talk) 12:36, 17 December 2014 (UTC) - previously 86.164.164.29 etc, etc
 * This text "Pancreatic cancer has one of the highest fatality rates of all cancers" exactly how I copied and pasted it here used to be in the epidemiology section. 109 I do not know to what you refer.
 * If one ranks all cancers by the number of people they kill pancreatic comes 7th. Doc James  (talk · contribs · email) 17:12, 17 December 2014 (UTC)
 * James, where did you copy and paste "Pancreatic cancer has one of the highest fatality rates of all cancers" from? (Not from WCR-2014 it would seem.)
 * Sure, "If one ranks all cancers by the number of people they kill pancreatic comes 7th". But "the number of people they kill" isn't the same thing as a "case fatality rate", which is a specific epidemiological measure (in fact, I'd be somewhat surprised if pancreatic cancer only ranked 7th on that particular scale). 109.158.8.201 (talk) 17:33, 17 December 2014 (UTC)
 * I've readjusted the wording following your removal of the "fatality rate" phrase. I hope this is now sorted. 109.158.8.201 (talk) 19:03, 17 December 2014 (UTC)


 * A related request for clarification: The number of cases that occur each year vary greatly by country, with those in the developed world generally having higher rates. "Rates" of what? (The statement appears to be unsourced, so I wasn't able to clarify). Adding: A possible source here could be the opening WCR-2014 bullet point: "A majority of pancreatic cancers occur in countries with high or very high levels of human development." This statement appears to refer to absolute numbers of cases rather than "rates". 109.158.8.201 (talk) 12:47, 17 December 2014 (UTC)
 * I've now sourced the sentence, per above, and reworded it accordingly . 109.158.8.201 (talk) 19:15, 17 December 2014 (UTC)


 * RE PanNETS: However, studies of autopsies have incidentally found small PanNETs at much higher rates, suggesting that the prevalence of tumors that remain inert and asymptomatic may be relatively high. I think this sentence could be phrased more clearly, and that some sort of a link like subclinical/asymptomatic (as well as to 'Prevalence') could be useful. Perhaps something like: "Small PanNETs can be uncovered relatively frequently as an incidental finding at autopsy, suggesting that the prevalence of subclinical PanNETS that remain inert and asymptomatic may be considerably higher." 109.158.8.201 (talk) 13:17, 17 December 2014 (UTC)
 * That is actually a significantly different statement, though no doubt also true. I'm not seeing the extra clarity, to be honest. 13:24, 17 December 2014 (UTC)
 * Well, the current statement wouldn't be crystal clear to me if I wasn't already aware of the intended meaning (though, admittedly, I don't have access to the cited source). The only significant change I reckon I've made to the intended meaning regards avoiding the reference to "studies of autopsies" (based on the consideration that many of out readers aren't likely to be interested in the actual research methodology). I think you'll find that the second clause (beginning "suggesting that...) retains the original meaning ("considerably"[?] might be an editorial scruple, though that's ultimately the epidemiological message I think). 109.158.8.201 (talk) 13:37, 17 December 2014 (UTC)
 * This is prognosis "Pancreatic cancer has one of the highest fatality rates of all cancers," and is dealt with there. Doc James  (talk · contribs · email) 16:28, 17 December 2014 (UTC)
 * Erm, actually no... (I think you must have misread this mislocated your last post James). This is about particular concern (bullet-point) regards small subclinical neuroendocrine tumors of the pancreas that are only diagnosed as incidental findings (or at some form of screening - e.g. those insurance/corporate programmes that are sometimes offered to employees). The potential clinical significance of such lesions is quite unclear, but we know from autopsy studies many exist that never progress during people's lifespan (as also seems to occur in some of the subclinical lesions uncovered, for example, by prostate cancer screening). This is all just by way of explanation – obviously, I'm not suggesting we mention screening issues under ==Distribution== 109.158.8.201 (talk) 17:07, 17 December 2014 (UTC)
 * My comment should have gone above. Doc James  (talk · contribs · email) 19:47, 17 December 2014 (UTC)

Arbitrary break: Prognosis of PanNETs
At present there's nothing ,whereas there's a sentence dedicated to prognosis under ==Distribution== (with 3 sources, of which one is primary and the other two are a bit old). I realize that succinct discussion of this highly diverse class of tumors isn't exactly straightforward. But I think we can find a slightly better solution (eg here?) Thoughts? 109.158.8.201 (talk) 20:26, 19 December 2014 (UTC)
 * I've now boldly removed the sentence in question . 109.158.8.201 (talk) 17:48, 20 December 2014 (UTC)

History section
Would (pdf) be useful here? 109.158.8.201 (talk) 11:16, 18 December 2014 (UTC)
 * Possibly, if you read Spanish. But it's mainly about pancreatitis it seems. Thanks for looking anyway. Wiki CRUK John (talk) 11:30, 18 December 2014 (UTC)
 * De nada... But why Spanish? The paper seems to be written in perfectly decent English, and there's a substantial section on "Pancreatic cancer" (pp. 544–546), as well as another on "Islet-cells tumors" (pp. 543–544).


 * Despite its age, that source looked to me potentially more useful than, though admittedly I wasn't actually able to access that one. (The title of looked promising, but it turns out that the content really is just about pancreatitis .) 109.158.8.201 (talk) 12:52, 18 December 2014 (UTC)


 * How about something like this...


 * The earliest recognition of pancreatic cancer has been attributed to the 18th-century Italian scientist Giovanni Battista Morgagni, the historical father of modern-day anatomic pathology, who claimed to have traced several cases of cancer in the pancreas. Many 18th and 19th-century physicians were skeptical about the existence of the disease, given the similar appearance of pancreatitis. Some case reports were published in the 1820s and 1830s, and a genuine histopathologic diagnosis was eventually recorded by the American clinician Jacob Mendes Da Costa, who also doubted the reliability of Morgagni's interpretations. By the start of the 20th century, cancer of the head of the pancreas had become a well-established diagnosis.


 * Regarding the history of the recognition of PanNETs, the possibility of cancer of the islet cells was initially suggested in 1888. The first case of hyperinsulinism due to a tumor of this type was reported in 1927. Recognition of a non-insulin-secreting type of PanNET is generally ascribed to the American surgeons, R.M. Zollinger and E. H. Ellison, who gave their names to Zollinger–Ellison syndrome, after postulating the existence of this disease related to a gastrin-secreting PanNET in a report of two cases of unusually severe peptic ulcers published in 1955. In 2010, the WHO recommended that PanNETs be referred to as "neuroendocrine" rather than "endocrine" tumors.undefined


 * Hope this helps, 109.158.8.201 (talk) 15:35, 18 December 2014 (UTC)

Style

 * There's an unexplained quotation (following a semicolon): "For the next decade, little attention was paid to this report; however, over the subsequent 15 years, there has been a virtual explosion in the recognition of this tumor."[38] Without consulting the reference this juxtaposition makes no sense. I'd also question the explosive tone, which seems to jar with the more encyclopedic prose. Couldn't the quotation be inserted in the reference? (Note: I'm also unsure about the reliance on primary sources in the surrounding text - can't we have secondary source/s [ ? ] alongside?) 109.158.8.201 (talk) 20:48, 19 December 2014 (UTC)
 * There are already two FAC points on this; can we try to keep everything together? Wiki CRUK John (talk) 00:53, 20 December 2014 (UTC)
 * Done (sorry). 109.158.8.201 (talk) 14:15, 20 December 2014 (UTC)

Citations still needed
I'm cutting to here the following uncited bits, from the old text. I'd still like to include them (assuming they are correct, which I think they are) but need MEDRS refs:

About 80% to 85% of people with locally advanced or advanced metastatic disease experience pain, which is usually felt in the upper abdomen as a dull ache that radiates straight through to the back. [It may be intermittent and made worse by eating. - less bothered about this bit] Painful jaundice is present in approximately one-half of people for whom curative surgery is not an option, while painless jaundice is present in approximately one-half of people with a cancer that is potentially curable by surgery.

Pancreatic cancer has an immunohistochemical profile that is similar to hepatobiliary cancers such as cholangiocarcinoma, and some stomach cancers; thus, it may not always be possible to be certain that a tumor found in the pancreas arose from it.
 * The immunohistochemical profile seems to be relatively unhelpful in terms of both diagnostic and prognostic  markers (the latter in current clinical practice at least ). I haven't actually verified the unsourced statement. Fwiw, thus, it may not always be possible to be certain that a tumor found in the pancreas arose from it sounds to me like a surprisingly strong general claim to make based on immunohistochemistry alone. Though in the developing world that could certainly be the case  (video ). 86.181.67.166 (talk) 17:34, 9 January 2015 (UTC)
 * Yes, thanks. Wolfgang, with a lot on thias area, doesn't say this, nor my other main sources, so while it might be true, I don't think it can be important. Wiki CRUK John (talk) 17:54, 9 January 2015 (UTC)
 * Agree. 86.181.67.166 (talk) 18:00, 9 January 2015 (UTC)

More fiddly: "Together with lung cancer, pancreatic cancer as a whole has shown the least improvement in US survival rates of all common cancers over the three decades to 2013," - not in latest version of source, and needs too much definition/qualification for "least improvement", "common" etc to use really, but some recently sourced statement along these lines would be good, since this is fundamentally the case. Wiki CRUK John (talk) 17:18, 5 January 2015 (UTC)
 * This might be it - need access Wiki CRUK John (talk) 14:58, 13 January 2015 (UTC)
 * Nope, doesn't cover pancreas. Wiki CRUK John (talk) 16:12, 13 January 2015 (UTC)

Does anyone have refs for these? Wiki CRUK John (talk) 16:15, 5 January 2015 (UTC)

To think about

 * Deaths from pancreatic cancer have changed little over time. ref name=WCR2014/>. Needs expanding if to keep. Wiki CRUK John (talk) 03:50, 6 January 2015 (UTC)
 * Deaths from many concerns are decreasing but not from pancreatic ca. Thus important IMO. Doc James  (talk · contribs · email) 18:18, 6 January 2015 (UTC)
 * The actual WCR (last bullet, box on p.414) says "Trends in incidence and mortality rates in both sexes tend to be rather stable over time" (I presume this was the place cited) which isn't quite the same thing - a stable trend is not the same as a stable figure, and this is death rates not deaths. Most of the national figures in the their graphic on p. 415 are stable, but some are not (these are age-standardized incidence rates). The point is worth making, as I think was the one you removed a while ago about the projected place in future cancer "rankings" (essentially another way of saying the same thing), but both need better sources. Wiki CRUK John (talk) 15:22, 8 January 2015 (UTC)

Comments
I meant to clarify a couple of my comments today but the FAC had closed (congrats!) so posting here to avoid posting there after the closing.
 * All I meant about the needle biopsy/surgery sentence is maybe to split into two sentences. Not to split into separate sections or anything so drastic.
 * Margins - this edit seems fine to me.
 * RE the opioids sentence - I only meant to maybe rephrase so as to avoid "significant negative side effects". Maybe simply remove "significant"? But depends on what the sources say.
 * Re preserving the duodenum. Yes, agree with your proposed fix. I thought of suggesting something along those line.
 * Also, I wanted to mention on the FAC page (and would have today), that it's really good work on an important article. Well done. Victoria (tk) 22:12, 14 January 2015 (UTC)

numbers in head/body/tail
RE this edit, etc –

The statement made in the 2014 NEJM review cited by us is somewhat cryptic: Approximately 60 to 70% of pancreatic cancers are located in the head of the pancreas, and 20 to 25% are located in the body and tail of the pancreas. The source cited for this statement in the NEJM review is clearer: "Approximately 60-70% of pancreatic ductal adenocarcinomas are localized in the head of pancreas, being [sic] 20-25% in the body and tail, with the remaining 10-20% of carcinomas involving the whole organ at the time of diagnosis." I started by following the source we actually cite by conserving the word "located" (whereas "localized" is more correct"confined to" might be better given that many are already invasive) and by not specifying "at the time of diagnosis". However, I think it may (arguably) be preferable for our wording should reflect the intended meaning of the source we cite (as in this edit), though I don't have strong feelings about that, as I feel "located" probably conveys the sense clearly enough here. 86.134.203.235 (talk) 12:14, 22 January 2015 (UTC)

"Four types" research
Removed as primary, but should probably be added to research directions when secondary sources appear. Nb a CRUK COI for me: "Recent research has found that pancreatic cancer can be classified into one of four subtypes ref http://medicdaily.co/pancreatic-cancer-comes-in-four-types /ref>. Wiki CRUK John/ Johnbod (talk) 15:36, 1 March 2015 (UTC)
 * Another story of that Johnbod (talk) 17:55, 11 March 2015 (UTC)

Refs
June 2015 reviews: Approach to Patients With Pancreatic Cancer Without Detectable Metastases Heestand et al  Progress in the Treatment of Metastatic Pancreatic Cancer and the Search for Next Opportunities  Ko  FeatherPluma (talk) 14:25, 30 June 2015 (UTC)
 * Thanks, both paywalled unfortunately. What do we think of this free access one? World J Gastroenterol. 2015 Feb 28; 21(8): 2281–2293. Published online 2015 Feb 28. doi: 10.3748/wjg.v21.i8.2281  New challenges in perioperative management of pancreatic cancer.

- not the best journal, as I recall. Johnbod (talk) 14:44, 30 June 2015 (UTC)

Stromal biology
10.1136/gutjnl-2015-309304 from Gut. JFW &#124; T@lk  10:49, 11 August 2015 (UTC)

BRCA2
Jason Hoffman, PharmD, RPh October 07, 2015 BRCA2 Mutations Account for >50% of Mutations in Pancreatic Adenocarcinoma

Salo-Mullen EE, O'Reilly EM, Kelsen DP, et al. Identification of germline genetic mutations in patients with pancreatic cancer [published online ahead of print October 6, 2015]. Cancer. doi: 10.1002/cncr.29664. FeatherPluma (talk) 17:37, 14 October 2015 (UTC)