Wikipedia:Featured article candidates/Pancreatic cancer/archive1

Pancreatic cancer

Pancreatic cancer (edit | talk | history | links | watch | logs)

• Featured article candidates/Pancreatic cancer/archive1

Nominator(s): Wiki CRUK John (talk) 15:12, 1 December 2014 (UTC)

This is about one of the most aggressive and hard to treat cancers. Improving the article has been one of the targets of Wikipedia:WikiProject CRUK. I'm very grateful to all the many editors who have contributed. The article has had a thorough peer review, with several editors editing as well as commenting. The article has also been reviewed internally at Cancer Research UK. Wiki CRUK John (talk) 15:12, 1 December 2014 (UTC)

Update January 5, 2015 I think as of now all the points on all the reviews are responded to, and either settled or awaiting a response from the reviewer (mostly the former). I'm sorry some of you have had to wait a while for this. There has been a lot of activity, both on this page and on the article itself, as well as the holidays. It's been great to see so many people getting involved in this. Many thanks to all reviewers and editors. Wiki CRUK John (talk) 20:45, 6 January 2015 (UTC)

Review by SandyGeorgia

Great to see you here, John! I am watchlisting, and will review, but not today, and perhaps not tomorrow. Right off the bat, though, I am seeing several incomplete citations; could you run through and make sure all of your citations are complete and in a consistent format? Your PMID sources seem consistent, but some others are incomplete. Also, could you explain minor deviations from MEDMOS suggested sections? More later. SandyGeorgia (Talk) 15:39, 1 December 2014 (UTC)

Discussion of citation consistency moved to talk. SandyGeorgia (Talk) 17:10, 21 December 2014 (UTC)

Citation consistency and some MOS issues are mostly cleaned up, and I've left some prose and sourcing discussion on article talk.[1] Please ping me for more thorough review when some of that has been processed. Bst, SandyGeorgia (Talk) 17:27, 17 December 2014 (UTC)

Comments from an ip

I don't to make any review comments on sections I've worked on, but I think it's probably ok to help out elsewhere.

• In the "Risk factors" section:
  Some studies have found links with diets high in red meat, processed meat,[35] and sugar-sweetened drinks (soft drinks).[36] The abstract of ref. 36 concludes "a slight correlation between risk of pancreatic cancer and [carbonated soft drinks] consumption has been found." Personally I don't think "a slight correlation" (whatever that may be) would provide strong enough causal evidence to deserve this mention. According to PMID 22194529 (technically, another potential MEDRS), "Although we were only able to examine a modest intake of [carbonated soft drinks], there was a suggestive and slightly positive association for their intakes which reached statistical significance in certain subgroups of participants (e.g., nondiabetics, nondrinkers of alcohol)." [2] Hmm, I'd cut it...
  Fwiw, I'd also perhaps avoid starting the sentence with the phrase "Some studies have..." as this can give an impression of cherry picking of primary studies, whereas the evidence for red and processed meats is actually sourced to a meta-analysis (ref. 35), albeit of observational studies. 86.164.164.29 (talk) 22:37, 1 December 2014 (UTC)

I've cut the soft drinks, although they did actually show up in my survey of advice from NCI, CRUK, ACS et al, which was why they were there. The meta-analysis shows that "some studies have...". Wiki CRUK John (talk) 22:59, 2 December 2014 (UTC)

• • I don't have much to say but it's in regards to the same section as what 86.164.164.29 just commented on so I'll just piggyback off them:
    • "though see the sections on obesity and alcohol" should probably be cut – there is no (longer?) a section on obesity and the section on alcohol is literally a sentence away.

Yes, rewritten, pending including source just below. Wiki CRUK John (talk) 10:33, 18 December 2014 (UTC)

• • • PMID 24403441, a systematic review on nutrition and pancreatic cancer, seems like a useful source for this section. There are a lot of conclusions about particular foods that are drawn from observational studies, so lets skip ahead briefly to the conclusion: "There is an abundance of evidence in the literature on the role of nutrition in pancreatic carcinogenesis. Often the evidence is inconclusive due to confounding factors...The lack of large randomized control trials makes it harder to establish causative associations for various nutrient types. [list of associations follow]."
  • Good luck! NW (Talk) 22:53, 1 December 2014 (UTC)

Included that last ref. Wiki CRUK John (talk) 01:34, 5 January 2015 (UTC)

In risk and prevention in fact. Wiki CRUK John (talk) 19:09, 5 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) 15:13, 8 January 2015 (UTC)

Oops... I see I confused the numbering of PMID 24403441 vs PMID 22240790 (sorry). The former is not actually a true systematic review ^{per [3]} (and is not indexed on PubMed as such), while the latter, which I'd already suggested at PR, is certainly a meta-analysis [4] (and indexed as such in PubMed, though not as a systematic review). Maybe we could ask for a further opinion from NW here? 86.181.67.166 (talk) 18:44, 8 January 2015 (UTC)

The methods section of the paper:

A PubMed search was performed for publications from 1985 through 2013, using the following key words, including both medical subject heading (MeSH) terms and free language words/phrases: “pancreatic cancer”, “nutrition”, “diet”, “dietary factors”, “lifestyle”, “smoking”, “alcohol” and “epidemiology”.

Articles that described and compared the impact of various dietary factors on risk of pancreatic cancer were first screened according to abstracts and titles and the selected articles were assessed for eligibility as full-text articles. No language restriction was applied. Reference lists from studies selected by the electronic search were manually searched to identify further relevant reports. Reference lists from all available review articles, primary studies and proceedings of major meetings were also considered. The quality and strength level of the results were considered.

It's not the best methodology in the world, as it does not meet several components of the PRISMA criteria and fails to follow several other normal protocols (multiple database searches, multiple reviewers, adjudication of disputed data abstraction). The paper describes itself as a systematic review. I don't know if this is something unique about nutritional science, but I would classify it as somewhere between the type of systematic review that you would see in a more quantitative medical subfield and a narrative review. Which isn't to say it's a bad source; I just wouldn't use it to contradict another source. NW (Talk) 22:32, 8 January 2015 (UTC)

Well since MEDRS doesn't permit critical appraisal by editors, I suppose I can only note that it's indexed by PubMed as a "review" but not as a "systematic review". Not a major consideration elsewhere perhaps, but... As I wrote in the edit summary, I felt this source was redundant (except for the somewhat speculative "possibly fat"). I sincerely believe the edit in question is one of several improvements I recently made to the Pancreatic Cancer page. 86.181.67.166 (talk) 23:44, 8 January 2015 (UTC); previously 109.158.8.201, 86.164.164.29, 109.157.83.50

I'll leave the relative merits of the paper to you two, and any others who want to add, but I don't see that "redundant" papers always have to be removed, especially when they undoubtedly add plenty of information. In the sections FeatherPluma has done we follow normal wiki-style and have up to 4 refs for things that aren't at all controversial, eg "... MRI/CT imaging is recommended for those at high risk from inherited genetics.[6][39][50][51]". Having seen something of how large specialist organizations go about things, I'm sure the ACS recommendations (which is extremely similar to its CRUK equivalent) is very carefully weighed using a wide range of sources and expertise, but it is not a published paper, and the methodology, sources used etc etc are not made explicit. The same paper is still used in ?2 other sections btw, so removing the named ref set-up had to be corrected, which a bot quickly did. Wiki CRUK John (talk) 02:10, 9 January 2015 (UTC)

Fair enough. But to put it in WP terms, I felt (and still feel) the previous version [5] gave undue weight to that particular "2014 review":

Specific types of food, as opposed to obesity, have not been clearly shown to increase the risk of pancreatic cancer,^{[NEJM review][Larsson]} although a 2014 review concluded that there was evidence that dietary factors increasing risk were processed red meat and "meat cooked at very high temperatures by methods such as frying, broiling or barbequing", and possibly fat.^{[Pericleous][ACS]}

I also felt that "there was evidence" required some qualification (nuancing) for weight, and that the use of quotation was conferring excessive weight to the subordinate clause. I couldn't see any reason why the ACS factsheet ^{[btw, I think it should be p. 19, rather than 21] done} was being used to source/support the quotation, given that I couldn't see any mention of that "2014 review". I also felt that two sentences would be better than one here, both for readability and to avoid any possible suspicion of wp:synth, and that it would be useful to some readers to provide a link to Diet (nutrition), at least. I therefore revised the point as follows [6]:

Specific types of food (as distinct from obesity) have not been clearly shown to increase the risk of pancreatic cancer.^[NEJM] Dietary factors for which there is some limited evidence of slightly increased risk include processed meat, red meat, and meat cooked at very high temperatures (e.g. by frying, broiling or barbequing).^{[Larsson][ACS]}

On looking at that now I can see that, on stylistic grounds, I rather overdid the hedging to reflect the wording of the ACS document ("may slightly increase"), as quoted – rather appropriately, imo – in the footnote/ref. I also freely apologise to John, NW and everyone else for inadvertently confusing those two pmids. But overall, I think the edit in question provides a significant improvement.

tldr: The changes, which I attempted to outline appropriately in the edit summary, were carefully thought through. 86.181.67.166 (talk) 10:58, 9 January 2015 (UTC)

• The "Classification" section basically provides information on histological types, without really summarizing the clinical classification in terms of grading and staging etc, which as usual in Wikipedia disease articles is located (usefully, imo) under "Diagnosis". Broadly per the spirit of a helpful discussion at Cancer Research UK on the fringes of Wikimania and subsequent proposals at MEDMOS, including this one, I think it would be both more appropriate and more reader-friendly to title the section as "Types". As I feel that the heading "Classification" may needlessly be a source of discouragement to our readers – this is the opening section after the lead - I've boldly implemented the suggestion I made at PR... 86.164.164.29 (talk) 09:21, 3 December 2014 (UTC)

Comments from Curly Turkey

• Way out of my comfort zone (and makes me queasy), but let's see if I can make it through. I'm only looking at the prose, and through the eyes of a (very) non-subject expert.
• I see a mix of serial and non-serial commas

It should all be serial. I've just read through, & I think only 1 needed changing, but I might have missed stuff. Wiki CRUK John (talk) 16:59, 6 January 2015 (UTC)

Lead

• One to two in every hundred cases of pancreatic cancer: there can't be 1 1/2 cases out of a batch of a hundred, so perhaps this would be better as "one to two per cent of"

It's been changed from that for reasons of accessibility; %s are less widely understood than one might think. Wiki CRUK John (talk) 22:31, 2 December 2014 (UTC)

• A smoker's chances of developing the disease are reduced if they cease smoking: I personally think the singular they should be freed from the closet, but I thought I'd point it out just in case this wasn't intended.

No, out and proud. Wiki CRUK John (talk) 22:31, 2 December 2014 (UTC)

• Treatment options are partly based on the cancer stage.: meaning they are decided based on the cancer stage?

Yes, both of those, which seem slightly different but valid statements to me. Wiki CRUK John (talk) 22:31, 2 December 2014 (UTC)

• the top three vary by gender, including breast cancer for women and prostate cancer for men: I wonder if this should be in the article, as it's US-specific

It's in a note (or should be) & that bit is certainly not US-specific. Wiki CRUK John (talk) 09:02, 2 December 2014 (UTC)

• the 5-year survival rate rises about 20%: rises to or by 20%?

"to", done. Wiki CRUK John (talk) 22:31, 2 December 2014 (UTC)

• signet ring cell carcinomas: I don't think "signet ring" is linking where you intended

No indeed, someone had changed it. Sorted Wiki CRUK John (talk) 09:02, 2 December 2014 (UTC)

Classification

* ("invasive" and "ductal" may be added to this term): are "invasive" or "ductal" carry any meaning?

Yes, the ones you'd expect: invasive because cancer (and a very agressive one) and arising in the ducts. Meaning is not really the issue, it's a question of the variable names used. Wiki CRUK John (talk) 10:23, 2 December 2014 (UTC)

This has now been rephrased somewhat. Wiki CRUK John (talk) 09:17, 18 December 2014 (UTC)

• endocrine cancers (see next section): perhaps "(see the [[#Endocine|Endocrine section]])" in case someone rearranges sections (say, by alphabetizing them)

Anything is possible, but they are more likely to rename them, messing up the link, aren't they? I thought such links were discouraged. Wiki CRUK John (talk) 10:23, 2 December 2014 (UTC)

You can avoid having the links break by using {{anchor}}. Curly Turkey ¡gobble! 23:19, 3 December 2014 (UTC)

See Sandy Georgia on her Talk review complaining about "bouncing the reader around". Wiki CRUK John (talk) 09:17, 18 December 2014 (UTC)

This has now gone. Wiki CRUK John (talk) 03:43, 6 January 2015 (UTC)

Signs and symptoms

• This is one of the main reasons for the generally poor survival rates. Exceptions to this are the functioning PanNETs, whose over-production of hormones is likely to give a range of symptoms according to the type.: is this supposed to share an inline cite with the previous statement?

No, needs another. Wiki CRUK John (talk) 22:38, 2 December 2014 (UTC)

Done Wiki CRUK John (talk) 13:12, 5 January 2015 (UTC)

• In the UK, about half of new cases: is this comparable to other countries?

I don't have comparative figures for anywhere, which may not exist, but it should be fairly comparable for developed countries, perhaps on the high side; probably rather more so than US figures which are endlessly given in these articles without arousing any comment, and are often also at the extremes for the developed world. Wiki CRUK John (talk) 22:38, 2 December 2014 (UTC)

• Up to two-thirds of people have abdominal pain, < !-- Wolgang 341 cites 2007 study that says "nearly 75%" -- >: "nearly 75%" is more than two-thirds---what's the story here?

Refined per the source to "In up to two-thirds of people abdominal pain is the main symptom" which probably accounts for the difference, though actually such statistics do jump around between different studies. Wiki CRUK John (talk) 18:01, 4 December 2014 (UTC)

Risk factors

• CDKN2A is a disambiguation page

Yes, though I doubt that it should be (is anybody familiar enough to confirm?). Linked to p16 Wiki CRUK John (talk) 10:29, 2 December 2014 (UTC)

There was another one, fixed: [7] SandyGeorgia (Talk) 02:46, 6 December 2014 (UTC)

• There a commented-out bit on gingivitis. What's the story?

From the old version, but most main sources don't mention. I should cut. Wiki CRUK John (talk) 10:29, 2 December 2014 (UTC)

This and partial gastrectomy now removed,; noted on talk at Talk:Pancreatic_cancer#Risk_factors_removed. Wiki CRUK John (talk) 14:21, 15 December 2014 (UTC)

• While the association between alcohol abuse and pancreatitis is well established ... the majority of studies have found no association: the association is well established, yet the majority of studies find no association?

2 diseases: "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." Wiki CRUK John (talk) 10:29, 2 December 2014 (UTC)

Changed now anyway Wiki CRUK John (talk) 20:36, 6 January 2015 (UTC)

Diagnosis & Staging

• < !-- Yes, refs needed; I know; Wolfgang I think -- >: so you plan to deal with this? The following paragraph lacks cites, too.

Refed/removed (see article talk also) Wiki CRUK John (talk) 16:18, 5 January 2015 (UTC)

• Pancreatic cancer is usually staged using a CT scan.: also uncited

Done, but also moved down Johnbod (talk) 00:22, 8 December 2014 (UTC)

• (e.g. 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.: where should the closing parenthesis fall? Also, lacking a cite

Changed. Re the ref, I have implicit ones, & I'm pretty sure this is correct, but I don't have a ref now I can use. Does anyone? The point is worth keeping I think. Johnbod (talk) 00:25, 8 December 2014 (UTC)

Removed for now, after rewriting (see article talk also) Wiki CRUK John (talk) 16:18, 5 January 2015 (UTC)

I'm going to take a break---I've got a cold that's making it hard to concentrate on this. If I don't return, ping me. Curly Turkey ¡gobble! 07:39, 2 December 2014 (UTC)

Curly Turkey, feedback from someone who is "out of their comfort zone" is a good thing! Suppose your long-lost Aunt Mabel is diagnosed with Pancratic cancer tomorrow? You, as a layperson, should be able to get from this article info you seek. Layperson review is desirable ... so please keep going! (Since you've inquired many times at WT:FAC about how to manage a review, see my section above as a sample of how to use article talk to shorten very long FACs :) I hope you're feeling better soon. Best, SandyGeorgia (Talk) 14:43, 6 December 2014 (UTC)

Thanks. I was actually over it by Friday but neglected to return. Curly Turkey ¡gobble! 01:28, 8 December 2014 (UTC)

Mechanism

• use of CT scans for other reasons are not all treated: because it's decided they don't need to be?

Yes, the surgery is still usually major, with its own risks, and as mentioned elsewhere, the criteria for choosing to intervene are not fully established. Wiki CRUK John (talk) 18:39, 11 December 2014 (UTC)

This bit rewritten now. Wiki CRUK John (talk) 03:45, 6 January 2015 (UTC)

• three types of precancerous lesion are recognized.: The "first" and "last" are enumerated—you might want to do the same with the second for consistency and ease of skimming

Done. Wiki CRUK John (talk) 18:39, 11 December 2014 (UTC)

• neoplasia. These are: "neoplasia" is plural?

Yes, for neoplasm. There's a neoplasms I need to lose too. Wiki CRUK John (talk) 17:50, 11 December 2014 (UTC)

Neoplasia in general and intraepithelial neoplasia in particular are definitely singular. And there's nothing wrong with neoplasm. NikosGouliaros (talk) 19:52, 11 December 2014 (UTC)

hmmm. Neoplasia just goes to neoplasm, but on further investigation refers to the process of formation of neoplasms, or the condition of having one (or more). And neoplasms is the correct plural it seems. Neoplasm doesn't really explain this very well. Wiki CRUK John (talk) 21:44, 11 December 2014 (UTC)

• (the so-called "Mothers against decapentaplegic homolog 4"): this aside seems more confusing than helpful at this scope

It's a name, & a nice break from all the code ones I thought. Wiki CRUK John (talk) 01:42, 5 January 2015 (UTC)

• < !-- Ryan 25% risk?? -- >: is this going to be dealt with?

Decided to add the point in the end. Wiki CRUK John (talk) 01:42, 5 January 2015 (UTC)

Prevention

• 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%.: I might append this to the previous paragraph

Done. Wiki CRUK John (talk) 18:40, 11 December 2014 (UTC)

Subsequently removed anyway (see article talk) Wiki CRUK John (talk) 03:46, 6 January 2015 (UTC)

Management

• Gemcitabine is administered intravenously on a weekly basis.: uncited

Removed, too close to dosage info which we don't do. Wiki CRUK John (talk) 14:12, 15 December 2014 (UTC)

• Treatment of pancreatic endocrine tumors, including the less common malignant tumors, may include a number of approaches. Watchful waiting of small tumors identified incidentally, for example on a CT scan performed for other purposes, may not ultimately need treatment, but the criteria for watchful waiting are unclear. This is mainly because the risk of surgery far outweighs the overall risk of aggressiveness for small tumours.: uncited

Changed and referenced. Wiki CRUK John (talk) 13:16, 5 January 2015 (UTC)

• If the tumor is not amenable to surgical removal and is causing symptoms by secreting functional hormones, targeted medication for PanNETs can be effective, both in reducing symptoms and slowing the progression of the disease. These are covered in more detail in the relevant section of the main article on neuroendocrine tumors.: uncited

Changed and referenced. Wiki CRUK John (talk) 13:16, 5 January 2015 (UTC)

• Radiation therapy is occasionally used if there is pain due to anatomic extension, such as metastasis to bone. Radiolabeled hormone: some PanNETs absorb a hormone called norepinephrine and these may respond to nuclear medicine medication, radiolabeled MIBG therapy (or, experimentally, other hormones), given intravenously. Radiofrequency ablation (RFA), cryoablation, and hepatic artery embolization may also be used.: uncited

Changed and referenced. Wiki CRUK John (talk) 13:16, 5 January 2015 (UTC)

Now cited, and indeed we have a new article for hepatic artery embolization from User:FeatherPluma. Wiki CRUK John (talk) 18:53, 11 December 2014 (UTC)

• Radiolabeled hormone: some PanNETs absorb a hormone called norepinephrine: I don't understand this---what is the colon for?

Semi-colon by the time I got to it. But the 5 commas after nearly bust some form of record: "... therapy, such as iobenguane (I131-MIBG), or, experimentally, other hormones, given intravenously." so I've rejigged. Wiki CRUK John (talk) 18:47, 11 December 2014 (UTC)

Outcomes

• although the outlook greatly varies according to the type.: uncited

Actually this point was always cited at the end of the lead, but I have added a more detailed ref here. Wiki CRUK John (talk) 02:07, 5 January 2015 (UTC)

• <! -- Sp p27, but Table 12 shows the much better uterine rates have been flat -- >: I don't know what this means, but will it be dealt with?

Removed to talk, for other reasons. Wiki CRUK John (talk) 17:20, 5 January 2015 (UTC)

Distribution

• Deaths from pancreatic cancer have changed little over time.: meaning the proportion of deaths?

Yes I think so - removed as really this needs more elaboration to make a useful point I think. Source not to hand. Wiki CRUK John (talk) 03:51, 6 January 2015 (UTC)

• over the course of their lifetime is very low": how is this punctuated in the original?

?That's a cut n' paste quotation - see the page. There is a missing word "cancer" however", which I'll point out to them. Wiki CRUK John (talk) 09:48, 18 December 2014 (UTC)

Missing word added on the site & the note here. Is there a remaining issue here? Wiki CRUK John (talk) 10:24, 18 December 2014 (UTC)

• risk for African Americans is over 50% greater than for whites: I'd expect "black" to be paired with "white", or "African American" with "caucasian" (or whatever the current accepted terms are).

I don't like saying blacks, which even with the context of a US para risks suggesting this is also true for Africans, which it isn't. Wiki CRUK John (talk) 23:55, 8 December 2014 (UTC)

Hmmm ... there's a White American article, and it distinguishes "whites" from "caucasians". Is this a distinction that plays a role here? Curly Turkey ¡gobble! 00:23, 9 December 2014 (UTC)

Surely we should take no account of WP articles on racial groups...? I certainly wouldn't expect to see "Caucasian" paired with "African American" myself; I don't think it's a respectable term these days, used for whites. SEER, who produce the US figures, report on "whites, blacks, Asian/Pacific Islanders, American Indian/Alaska Natives (footnote here), and Hispanics (who they note are "not mutually exclusive from" the other groups. Race and ethnicity in the United States Census has more on this, though the census categories jump around like anything between censuses, and are not directly comparable, which I expect SEER tries to minimize. But the main thing is to avoid at all costs giving the impression that African Americans (generally with up to 40% white genetics I believe) will have a similar risk profile to global "black" figures. Typically African people have lower cancer risks because they have very different lifestyle factors and (yet) live less long. I'll link to White American, but am against changing African American, to which "Black American" unsurprisingly redirects. Wiki CRUK John (talk) 18:08, 11 December 2014 (UTC)

Officially, in the US: "“Black or African American” refers to a person having origins in any of the Black racial groups of Africa. It includes people who indicated their race(s) as “Black, African Am., or Negro” or reported entries such as African American, Kenyan, Nigerian, or Haitian." (in the 2010 US census) Page 3 here. Wiki CRUK John (talk) 16:20, 15 December 2014 (UTC)

• Pancreatic cancer is the tenth most common cancer in the UK (around 8,800 people were diagnosed with the disease in 2011), and it is the fifth most common cause of cancer death (around 8,700 people died in 2012).: seems to jump out of nowhere---perhaps merge with the other UK rates immediately above?

The one UK figure above is risk, in a bit on risk. I'll see if the whole section can be better arranged, but the flow is fairly logical. No-one ever complains about US figures seeming "to jump out of nowhere", though they often do. Wiki CRUK John (talk) 18:15, 11 December 2014 (UTC)

History

• but the person only survived: or "patient"? "the person" could be the surgeon (I'd've just changed it to "patient", but I wnat to make sure that's the right word)

We don't like using "patient" so no doubt someone will change it back before I get to it. Wiki CRUK John (talk) 23:52, 8 December 2014 (UTC)

Well, then it needs to be otherwise reworded to avoid ambiguity. Curly Turkey ¡gobble! 23:54, 8 December 2014 (UTC)

How is it ambiguous? Wiki CRUK John (talk) 23:57, 8 December 2014 (UTC)

"the person" could be Codivilla. Curly Turkey ¡gobble! 00:05, 9 December 2014 (UTC)

He couldn't be really. Maybe it's ok to use "patient" when referring to a single specific (and long-dead) individual? I think so - what do others think? Wiki CRUK John (talk) 18:29, 11 December 2014 (UTC)

See below, in JDW section; I'm rooting for "patient" in the history section only. Wiki CRUK John (talk) 17:21, 17 December 2014 (UTC)

• In the 1900s, after some: does this mean the early 1900s, or the whole century?

1900-1909, always, no? Wiki CRUK John (talk) 23:52, 8 December 2014 (UTC)

Not where I come from—"1900s" is usually a synonym for "the 20th century". How about "the opening decade of the 20th century"? Curly Turkey ¡gobble! 23:56, 8 December 2014 (UTC)

No thanks! Precise range used. Wiki CRUK John (talk) 18:29, 11 December 2014 (UTC)

• total removal of the duodenum is compatible with life: "compatible with life" sounds somehow odd to me

Any alternative suggestion? Is "compatible with survival" better? Wiki CRUK John (talk) 18:29, 11 December 2014 (UTC)

Now "mistaken beliefs that it was essential for life to preserve the duodenum" Wiki CRUK John (talk) 09:31, 18 December 2014 (UTC)

• until recent decades: recentism; should be more specific

It is certainly not WP:RECENTISM! I think the next paragraph conveys very adequately which "recent decades" are meant. Wiki CRUK John (talk) 18:29, 11 December 2014 (UTC)

• [[Vitamin K deficiency|bleeding]]: that's quite the Easter egg!

it would be if we weren't already talking about Vitamin K I think. Wiki CRUK John (talk) 23:52, 8 December 2014 (UTC)

If I wasn't reading the source in Emacs I never would have guessed that "bleeding" was pointing to Vitamin K deficiency (and I'd be far less likely to click through—after all, I "know" about "bleeding"). Curly Turkey ¡gobble! 00:03, 9 December 2014 (UTC)

Fair enough - link now piped from "bleeding with jaundice", which I hope resolves. Wiki CRUK John (talk) 09:34, 18 December 2014 (UTC)

• "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.": who is being quoted, and can this not be paraphrased?

The reference. What's wrong with quoting? I don't understand why we are so scared of it. Any encyclopedic paraphrase would be far more bland. Wiki CRUK John (talk) 23:52, 8 December 2014 (UTC)

There's nothing wrong with quoting per se, but it often integrates poorly into the surrounding text. The important thing is to be clear what's being quoted—the vast majority of readers with ignore the inline cites, and even if they don't, is the quotation from the author of the reference, or quoted by the reference? Another problem is with the way this particular quote is introduced—very abrupt. If read out loud, there would be no clue at all that this was a quote. Curly Turkey ¡gobble! 00:00, 9 December 2014 (UTC)

Another point here is that the fiddly past ranges of years are best handled by quoting. Wiki CRUK John (talk) 18:11, 5 January 2015 (UTC)

Research

• are being intensely looked at: sounds like a stare-down

Changed to "investigated", though I should make it clear I don't actually think that's an improvement. Wiki CRUK John (talk) 16:26, 15 December 2014 (UTC)

That brings me to the end of the article. I'll be back again to take a look at your responses to my first batch of comments. Curly Turkey ¡gobble! 01:28, 8 December 2014 (UTC)

Comments from NikosGouliaros

I've previously reviewed the article, making minor contributions; I'll try not to repeat myself. I feel the need to congratulate Wiki CRUK John and everyone else who has worked for this excellent result.

Classification

• "("invasive" and "ductal" may be added to this term)": I still feel that the meaning - i.e., that some sources refer to the same disease as "invasive pancreatic adenocarcinoma" and "ductal pancreatic adenocarcinoma" - isn't clear; to this attests the comment by Curly Turkey. Some minor rephrasing might be warrantied.

I see the problem, but I certainly don't want to list of the variants (I make it 7 or more, with "of the pancreas" etc). Gone with "dominated by pancreatic adenocarcinoma (variations of this name may add "invasive" and "ductal")". Wiki CRUK John (talk) 16:31, 15 December 2014 (UTC)

• "This cancer arises in the tiny ducts that carry certain hormones and enzymes away from the pancreas": I'm not sure what the sources exactly say; however, the epithelial cells adenocarcinomas arise from aren't only found in the "tiny ducts", but also in the main pancreatic duct. Moreover, strictly speaking, these tiny ducts carry pancreatic products outside the organ not directly but through the pancreatic duct. Some rephrasing might be warrantied.

"tiny" removed; don't think I put that in. Wiki CRUK John (talk) 22:51, 2 December 2014 (UTC)

• I'm a bit baffled by the link to islet cell carcinomas. Is the linked article article about the PanNETs (and therefore an article with a mistaken/obsolete name), or to another pathologic entity? In the latter case, the link might confuse the reader.

Jumping in here... Hum, fair point... The problem is that Wikipedia is a bit behind the curve here, and that page should perhaps (imo, at least) be renamed as Pancreatic neuroendocrine tumor and updated accordingly, based on current MEDRS. Thoughts? 86.164.164.29 (talk) 18:07, 2 December 2014 (UTC)

The main, far longer, article is neuroendocrine tumor, covering a wider range of sites, which I think is reasonable. Pancreatic neuroendocrine tumor (you had a typo, 86) redirects to Pancreatic cancer; I'm not sure it should. islet cell carcinoma should be merged to one of these (NET or here) I think. Mind you the codes are still there in the ICD/Mesh databases - would they get removed? It doesn't feature in the most recent literature, as the text says. I'd welcome other views. Wiki CRUK John (talk) 18:13, 2 December 2014 (UTC)

Yup, those are all questions I'd been asking myself too (but wasn't really in a hurry to get involved in... help!) At present, the information at Neuroendocrine tumor#Pancreatic neuroendocrine tumors is largely genetic, and I feel "Pancreatic cancer" probably is the more appropriate redirect, for the moment at least. 86.164.164.29 (talk) 18:33, 2 December 2014 (UTC)

OK, I see exactly what's going on, and can "unbaffle" it. It will take about 2 hours. I can aim to get to this in about 10 days as I have a surge in non-wikipedia activities. FeatherPluma (talk) 03:10, 13 December 2014 (UTC)

What was "nesidioblastoma" until October 16, and was correctly that topic (despite wrongly having general statistics for "islet cell tumors" - removed) is moved back to nesidioblastoma. Keeping all neuroendocrine tumor content within that main article has proven unwieldy. Pulmonary neuroendocrine tumors already have a specific article, and in usage carcinoid marches forward through time (despite WHO). The traditional pancreatic terms, "islet cell tumor" and "islet cell carcinoma", are now redirected to pancreatic neuroendocrine tumor. PanNET is now linked to pancreatic cancer but no longer redirects there. I reorganized (and will further edit) the contents. I think these changes address these suggestions for this bullet point item. FeatherPluma (talk) 21:32, 24 December 2014 (UTC)

Agree - these wide-ranging improvements address the concerns I raised above (as 86.164.164.29). 109.158.8.201 (talk) 16:03, 26 December 2014 (UTC)

Yes, I'm happy with this, and thanks. Wiki CRUK John (talk) 00:01, 6 January 2015 (UTC)

Signs and symptoms

• Image caption: I propose "Man with jaundice etc" instead of "Painless jaundice" (pain isn't shown!)

But he's smiling! [joke]. I agree, done. Wiki CRUK John (talk) 18:24, 2 December 2014 (UTC)

• "Painless jaundice": The problem here is that, in the line just above, we say that jaundice with pain is much more common than painless one. We could just delete the adjective painless here. On the other hand, cancer (pancreatic, bile duct, ampulla of Vater, metastatic liver) is a typical cause of painless jaundice. This is not a contradiction: a disease can be the most common cause of a given symptom, and this symptom can still be uncommon in this disease.

Yes. Am I right in thinking that the jaundice itself is never the cause of the pain as such? Wiki CRUK John (talk) 22:22, 2 December 2014 (UTC)

Yes you are. In painful jaundice jaundice and pain only have a common cause. NikosGouliaros (talk)

Rewritten to:"Jaundice, a yellow tint to the whites of the eyes or skin, with or without pain, and possibly in combination with darkened urine. This results when a cancer of the head of the pancreas obstructs the common bile duct as it runs through the pancreas."

• It might be better if every paragraph in the bullet list begins with a mention of the actual manifestation; e.g. one can rephrase: "Nausea and a feeling of fullness, caused by compression of neighboring organs by the tumor, which disrupts digestive processes and makes it difficult for the stomach to empty".

But this point, which has been rewritten by about 6 people including you, also includes the floating stools. I think it's ok & I'm reluctant to reopen this one. We are going round and round in circles on several of these points. Wiki CRUK John (talk) 22:48, 2 December 2014 (UTC)

Point taken. NikosGouliaros (talk)

• I still disagree with the subheading "other symptoms" in this context. It just isn't clear what separates symptoms below the subheading from symptoms above. This must be solved somehow. Just deleting the subheading might be enough.

Doc James has changed to "Other findings", which I think works well. As I said in the peer review, I want to keep the main list of symptoms very simple and accessible. This is how it was in early May, by the way. Wiki CRUK John (talk) 17:06, 17 December 2014 (UTC)

It does work better. NikosGouliaros (talk) 22:25, 18 December 2014 (UTC)

• This bullet point could be added: "Other non-specific manifestations of the disease include: weakness and fatigability; dry mouth; sleep problems; constipation; dyspepsia; and a palpable abdominal mass."

Is the "palpable mass" the same as in Courvoisier's law (which I know you want to move up here)? The rest are really non-specific, and don't feature in the main sources. Wiki CRUK John (talk) 17:13, 17 December 2014 (UTC)

It is not the same; a palpable mass can be anywhere, and it's the actual tumor, not the gallbladder. It's rare though, so I insist no more. NikosGouliaros (talk) 22:25, 18 December 2014 (UTC)

Ok, I've added this, but in the diagnosis section. This was you not a quote from the source, yes? Wiki CRUK John (talk) 16:01, 5 January 2015 (UTC)

• A more general issue: Symptoms and signs are different types of disease manifestations. Are we supposed to make a distinction clear and list them separately? (No distinction is made in the FA lung cancer.)

Imo, no. Symptoms can also be signs (though not all signs are symptoms). 86.164.164.29 (talk) 18:11, 2 December 2014 (UTC)

• I was and I am in favor of a subsection dedicated to Metastasis; however, one might expect that, if this will be in the "Signs and symptoms" section, it must include some symptoms of metastatic disease.

Too various, allowing for the major possibilities, aren't they? Also they are not given in any of the main sources I've used on pancreatic cancer, which makes me think we have gone far enough down this path. Wiki CRUK John (talk) 22:42, 2 December 2014 (UTC)

Point taken. NikosGouliaros (talk)

We now have a metastasis section though (see elsewhere this section) Wiki CRUK John (talk) 17:06, 17 December 2014 (UTC)

• I'm just reminding that the faulty picture must go.

Gone. Wiki CRUK John (talk) 15:37, 4 December 2014 (UTC)

Risk factors

Everything looks great, aside of the paragraph on rare hereditary syndromes, which looks wedged in the bullet list. Why don't we just connect it to the previous paragraph?

Indented, which works I think. It's all too long for one para. Wiki CRUK John (talk) 03:03, 3 December 2014 (UTC)

Diagnosis

• I'm still not sure that repeating the disease symptoms in this section is justified. Could other reviewers say what they think?

I don't think it's appropriate either. (Commented here, under "Anything else outstanding?). 86.164.164.29 (talk) 18:23, 2 December 2014 (UTC)

And I must add that this repetition is not present in the Diagnosis sections of any of the 4 FA's on maligancies (acute myeloid leukemia, cholangiocarcinoma, endometrial cancer, and lung cancer). Unfortunately the medical MOS is not clear on what should be included in this section. NikosGouliaros (talk) 23:35, 2 December 2014 (UTC)

Trimmed at the moment, though I would like to get the relationship between pain and stage back in - currently on the talk page as a MEDRS ref is needed. Wiki CRUK John (talk) 23:59, 5 January 2015 (UTC)

• Similarly, I think the place for Courvoisier sign is the Symptoms and signs section.

Well, that's one that actually is diagnostically relevant. Since MEDMOS doesn't really follow the usual practice for journal review articles anyway, I think a reasonable case could be made, from a WP perspective, for including this sort of diagnostic red flag here. 2c, 86.164.164.29 (talk) 18:23, 2 December 2014 (UTC)

I do not agree; the same case could be made for painless jaundice. Courvoisier sign is a sign and is probably warrantied to be in the S&S section. NikosGouliaros (talk) 23:26, 2 December 2014 (UTC)

• Only now do I notice that the CT image has the arms of a disconnected cross pointing at the tumor. I was thinking of making this a bit more obvious, e.g. by coloring the arms - this could be requested at the graphics lab.

Caption changed, to "... Cross lines towards top left surround a macrocystic adenocarcinoma of the pancreatic head.", but yes, added colour is an achievable improvement. Wiki CRUK John (talk) 02:43, 3 December 2014 (UTC)

Requested at the graphics lab. NikosGouliaros (talk) 08:53, 3 December 2014 (UTC)

And I'm grateful to Centpacrr for doing it. I think it looks better. NikosGouliaros (talk) 15:06, 3 December 2014 (UTC)

Indeed. Thanked, and added color noted in caption. Wiki CRUK John (talk) 17:32, 3 December 2014 (UTC)

My pleasure. Glad this worked out well for you. Centpacrr (talk) 18:17, 3 December 2014 (UTC)

Staging

• The mention of Stage III seems a bit out of the blue; the average reader does not know how many the AJCC-UICC stages are. I propose the following rewriting of the first paragraph (my additions in italics):

The cancer staging system used internationally for pancreatic cancer is that of the American Joint Committee on Cancer and Union for International Cancer Control, referred to as AJCC-UICC; it designates 4 stages, which correspond to specific TNM stages. Stage I and II tumors are resectable, and so are the subset of Stage III tumors that do not involve the celiac axis or superior mesenteric artery and are classified as "borderline resectable"; the remaining Stage III and Stage IV (metastatic) tumors are unresectable. The TNM staging system builds up an overall stage by assessing the Tumor size, spread to lymph Nodes and Metastasis. Stage T1 is localized tumors less than 2 cm in size, T2 tumors over that size but still wholly in the pancreas, and the other T stages are defined by the degree of spread beyond. A simpler practical classification groups the tumors as "resectable", "borderline resectable", and "unresectable" because of locally advanced or metastatic disease.

Rewritten along these lines, but with shorter sentences: "...referred to as AJCC-UICC. This designates 4 stages, which correspond to specific TNM stages. Stage I and II tumors are candidates for successful surgical removal or resection. The staging system makes an important distinction within Stage III between tumors that are classed as "borderline resectable" because they do not involve the celiac axis or superior mesenteric artery, and "unresectable". Surgery is likely to be possible for the former, but is not usual for the latter." Wiki CRUK John (talk) 17:44, 17 December 2014 (UTC)

I've tried to clarify the text here [8] though please ignore some of the less helpful edit summaries (maybe a simple table would help?) 86.181.67.166 (talk) 14:50, 1 January 2015 (UTC)

...and further streamlined (I hope) here. 86.181.67.166 (talk) 01:24, 2 January 2015 (UTC)86.181.67.166 (talk) 00:16, 2 January 2015 (UTC)

• Is it acceptable that the images have more details on the TNM staging than the text mentions? (It's a question, not an opinion.)

I think so. There's not much more. But I can add if desired. Wiki CRUK John (talk) 17:29, 3 December 2014 (UTC)

Agree that it's ok - and it also helps give an to the lay reader what TNM evaluation is in practice. (? by isp 86 I think)

• The TNM image gallery isn't centered but its heading is. Could it be formatted a bit better? Maybe center the gallery too?

For me the gallery is left-aligned and the heading right-aligned, the latter looking a bit odd. At Help:Gallery tag I'm not seeing options for these at all (on IE). Maybe there are different templates or tags to use. Wiki CRUK John (talk)

By "Heading" I mean the words "Pancreatic cancer staging". And what I mean is this. NikosGouliaros (talk)

Sorted, thanks Wiki CRUK John (talk) 15:37, 4 December 2014 (UTC)

Mechanisms

• Intraepithelial neoplasia merits a wikilink. The article on it is little more than a disambiguation page; imo it should become a redirection page to carcinoma in situ, where our wikilink should also be directed to (despite the fact that this page too is in heavy need of improvement). (Or perhaps Carcinoma in situ should be renamed intraepithelial neoplasia; this is a tricky subject, expert opinion is warrantied, and not very relevant to pancreatic cancer).

linked; won't tangle with the rest now. Wiki CRUK John (talk) 23:57, 5 January 2015 (UTC)

• I'm not entirely sure that all or most exocrine pancreatic cancers evolve from cysts; is it supported by the sources? (I've no access to the NEJM article). In my mind, intraepithelial neoplasia is the universal preliminary lesion of adenocarcinoma. NB: The existence of several types of precancerous cysts that often evolve to frank cancer does not mean that these lesions are the major precursor lesion of adenocarcinomas.

I'm not an expert on the subject; but after rereading this section (in more depth than I did while peer reviewing the article) I wonder if it might not be totally accurate. I admit not having the time to dig into the matter in detail (not being an expert on it).

Having checked (Wolfgang mostly) I think you're right - rewrite to come. Wiki CRUK John (talk) 17:25, 17 December 2014 (UTC)

Thank you for your answer, and your rewrite. It looks excellent! (And note that I was mistaken that intraepithelial neoplasia [IN] is the universal preliminary lesion). One could add that IN is by far the most common of the three. However, the first paragraph seems to need adjusting now that intraepithelial neoplasia has been added.

A question though Besides pancreatic serous cystadenomas (SCNs), which are almost always benign, three types of precancerous lesion are recognized. Does it mean that SCNs are precancerous too? NikosGouliaros (talk) 23:09, 18 December 2014 (UTC)

No, but changed to the clearer "Apart from..." Wiki CRUK John (talk) 13:19, 5 January 2015 (UTC)

I hope these points are sorted now; do you agree? Wiki CRUK John (talk) 23:56, 5 January 2015 (UTC)

• The ominous image of liver metastases might be better off in the Metastasis section of Signs and Symptoms.

Moved Wiki CRUK John (talk) 15:37, 4 December 2014 (UTC)

I'm wondering... is this illustration really so pertinent? (Though I suppose it could be argued that it's illustrating where symptoms are coming from.) 109.158.8.201 (talk) 10:07, 28 December 2014 (UTC)

• I'm not sure why the parenthesis "(cystic lesions)" is necessary to accompany "cysts".

Indeed - leaving for rewrite of this bit, per point 2 up. Wiki CRUK John (talk) 17:25, 17 December 2014 (UTC)

Changed. Wiki CRUK John (talk) 23:51, 5 January 2015 (UTC)

Management

• "Encasement": maybe, for the sake of clarity, make it "encasement by the tumor"?

Now rewritten to: "...but not "encasement" by the tumor, defined as the tumor touching more than 180° round the vessel." Wiki CRUK John (talk) 17:29, 17 December 2014 (UTC)

• One could link median survival to Survival rate#median survival.

It needed rewriting, and old vandalism reverting, but now done at first occurence, in the chemo section. Wiki CRUK John (talk) 14:09, 15 December 2014 (UTC)

• "Τhe changes of the last few years have only increased average survival times by a few months": And let's always pay attention never to refer to "average" when it's about "median".

Time to read the lead section of average, I think. I'd be amazed if it was not about equally true of both, but will check the sources. Wiki CRUK John (talk) 23:46, 3 December 2014 (UTC)

"average" removed; not needed I think. Wiki CRUK John (talk) 14:09, 15 December 2014 (UTC)

• "Watchful waiting of small tumors identified incidentally, for example on a computed tomography (CT) scan performed for other purposes, may not ultimately need treatment, but the criteria for watchful waiting are unclear.": Something sounds wrong here.

Yes, needed a "which" I think, but now rewritten to: "Small tumors that are identified incidentally, for example on a CT scan performed for other purposes, may be followed by watchful waiting. Ultimately, these small tumors may not need treatment, but the criteria for watchful waiting are unclear. " Clear, if a tad repetitive. Wiki CRUK John (talk) 17:36, 17 December 2014 (UTC)

• "Radiolabeled hormone: some PanNETs absorb a hormone called norepinephrine and these may respond to nuclear medicine medication, radiolabeled MIBG therapy (or, experimentally, other hormones), given intravenously.": May I suggest rephrasing?

Now: "Some PanNETs absorb specific peptides or hormones, and these PanNETs may respond to nuclear medicine therapy with radiolabeled peptides or hormones such as iobenguane (iodine-131-MIBG).[55][56][57][58] Radiofrequency ablation (RFA), cryoablation, and hepatic artery embolization may also be used.[59]". Wiki CRUK John (talk) 17:36, 17 December 2014 (UTC)

Outcomes

• "increased average survival rates": As above: is it truly average or actually median? If unsure, it might be better to just mention "increased survival rates".

As above; the rates are calculated on the medians, but I avoid that term for accessibility reasons (and even more "measures of central tendency"), so your dichotomy is strictly incorrect (a median is an average). But the whole issue can be skipped by omitting any measure I think, and the source often does so in discussing them. Wiki CRUK John (talk) 15:20, 4 December 2014 (UTC)

I have no strong view on it (I would if it wrote "mean" instead of "average"); why don't you just leave it as it is. NikosGouliaros (talk)

Distribution

• The fact that I do not support "Distribution" as a substitute for "Epidemiology" as a section heading still applies (see the discussion in MEDMOS), but it wouldn't oppose a FA candidacy for it.
• "In recent clinical guidelines": I'm not sure this offers anything

Otherwise people will tag/query "These are variously estimated ..." I think. Mention of divergence between sources tends to produce such a reaction in my experience. Wiki CRUK John (talk) 17:46, 17 December 2014 (UTC)

History

• Maybe add a couple of words on how vitamin K helped with postoperative survival?

Will do, though I now see from this that my source appears to be mistaken in calling this a discovery of 1940, since the key papers came out in 1938 (notes 2 & 3)! Irritating & I'd better say something. Probably a mistaken confusion with the line following. Wiki CRUK John (talk) 15:30, 4 December 2014 (UTC)

Done as: "The discovery in the late 1930s that vitamin K prevented bleeding with jaundice,..." Wiki CRUK John (talk) 23:34, 22 December 2014 (UTC)

Research

• I'm not sure that "the tissue surrounding the pancreatic tumor" is the most accurate definition of "stroma" we could come up with.

Ane the reason is that it gives the impression of stroma as surrounding the tumor like a capsule; stroma is an actual part of the tumor mass. How about "the tissue surrounding the cancer tissue"/? --NikosGouliaros (talk) 08:29, 28 December 2014 (UTC)

• I've tried to address the issue here (per cited sources). 109.158.8.201 (talk) 10:38, 29 December 2014 (UTC)

That's helpful, a bit, but doesn't explain stroma. I'm not sure whether the link should be to Stroma (animal tissue) or Stromal cell - or why we have the 2 different articles! Wouldn't the explanation be better much higher up, when desmoplasia is discussed? At the moment this doesn't mention stroma, though one of the sources (Ryan) uses the term plenty (though the other, Wolfgang, seems not to). I'll try to add up there. Wiki CRUK John (talk) 14:52, 6 January 2015 (UTC)

Done that. Useful I think - there might even be more on the implications. Wiki CRUK John (talk) 15:21, 6 January 2015 (UTC)

References

• Are citations supposed to be totally uniform? E.g., have all the same links (linkable title, DOI, PMID), uniformly mention either full journal titles or initialsms, etc? I don't particularly mind, but I can start working on it

Yes, very much so, see the top of the page (Sandy Georgia comment). But there has been a lot of re-editing by various hands, especially in the peer review. I think we may need a section on the talk page to confirm what the style for the article actually is - I don't have very strong views, since there appears to be (very wrongly imo) a local consensus that page refs should NEVER be used for articles, except that it should as close as possible to the auto-template style in the standard editing window menu. But various people have adjusted the refs to suit their own preferences and we need to be consistent. Any help on this greatly appreciated. Wiki CRUK John (talk) 15:27, 4 December 2014 (UTC)

I've copy edited the citations from the lead, adding some doi's and url's, and making the use of commas and semicolons more uniform. It's an example of what I have in mind. I haven't touched the journal names though; I see some articles in PMC without uniform style for them. NikosGouliaros (talk)

Lead

• Am I exggerating if I propose to replace the lead image, because of its inaccurate depiction of the pancreatic duct? We could go for this image, which is the lead image in pancreas; or the CT image that shows the tumor, which will be similar (though perhaps a bit inferior) to the lead image in cancer and lung cancer.

I wasn't aware that had a mistake too; what is it? Both of those images are already used, and I'd rather have one that shows the pancreas in its surroundings in that position. I'll look for another.

There's this, thias, or this (COI), Wiki CRUK John (talk) 15:46, 4 December 2014 (UTC)

In the current one, the pancreatic duct should be much longer and continuing further down to the tail.

I think the "COI" one is the best of the three :) NikosGouliaros (talk)

Right, yes, I saw that. Will change to the CRUK one for now. There may be better ones but the categorization on Commons is unreliable so they are hard to find. Wiki CRUK John (talk) 16:19, 4 December 2014 (UTC)

(To be cont'd) --NikosGouliaros (talk) 17:47, 2 December 2014 (UTC) Mostly done. --NikosGouliaros (talk) 15:14, 4 December 2014 (UTC) [Update: I have failed to mention a conflict of interest, as I am the translator of this source to Greek. NikosGouliaros (talk) 10:27, 19 December 2014 (UTC)]

Comments from NikosGouliaros: List of references
References

Comments from WS

Overall a great article which saw a lot of improvement over the last months. I am a bit confused over whether it is supposed to completely cover neuroendocrine tumors or not. Depending on the answer to that question it either contains much too little or too much information about them. Considering the big differences, wouldn't it be more clear to have a short general pancreatic cancer and then separate articles for adenocarcinoma and endocrine tumors? (no doubt this has been discussed before, I haven't looked that up).

The information about resectability is very short and simplistic and should be expanded. Vascular involvement does not necessarily preclude surgery, e.g. splenic artery and vein involvement in more distal tumors is usually not a problem as they can be resected en-bloc (with or without the spleen). There is a lot to be said about borderline resectable tumors, where vascular reconstructions (especially for venous involvement) and neo-adjuvant therapy (to increase the chance of free resection margins) are slowly becoming more commonly used. Also the info about resectability is now spread out across at least three sections, it would be helpful to consolidate that. --WS (talk) 10:14, 7 December 2014 (UTC)

Thanks for these, and the earlier comments. The scope issue has been discussed before, and does present a dilemma. At the moment the article is supposed to cover all, and arguably gives too much weight to PanNETs, while still too little information about them - they are a diverse group its hard to generalize about. We have the same issue at Brain tumor, but there the balance is different, and we are planning to consistently adopt the approach you suggest. Here I think the present solution is the best, given the predominance of PAC and other factors. One issue is that pancreatic cancer is the common name and the natural search term.

I'll look at the resectability question - I'd been meaning to add some bits, like looking for the fat-plane. So far I have avoided specifying blood vessels and similar detail, as once you start you may have to do a fair bit, and Pancreaticoduodenectomy and Pancreatectomy (hmmm) have their own articles, though neither touch on this at present. Wiki CRUK John (talk) 11:56, 7 December 2014 (UTC)

Liver cancer is a good example with a similar situation (HCC vs cholangiocarcinoma), I think it works very well there to have three articles. I think it is almost impossible to comprehensively cover both in one article and at the least PanNet should be split out, leaving only the necessary bits here. Regarding respectability, that should be covered here as it is specific to pancreatic cancer and one of the most important aspects of the diagnosis. It does not need to be very difficult, as there is no definite consensus, but broadly speaking coeliac trunk/superior mesenteric artery/hepatic artery involvement is generally considered unresectable, SMV/portal vein involvement often unresectable but sometimes done with vascular reconstruction, and splenic artery/vein resectable. Assessing resectability with CT is hard to completely cover as various criteria exist, would keep that very general. --WS (talk) 12:51, 7 December 2014 (UTC)

I haven't changed the situation as I found it, which is that we have "Pancreatic cancer", to which PAC redirects, and neuroendocrine tumor covering all sites. There's been discussion but no great consensus to (effectively) move this to PAC and split out the PanNETs. Wiki CRUK John (talk) 13:20, 7 December 2014 (UTC)

Sure it can be done this way, but as you have nominated this as a featured article, in my opinion it completely fails the comprehensiveness criterion with regard to PanNETs. To get it featured, I think it should be decided to keep this in (hard, probably confusing) or out and keep only the basic information here or in an umbrella article. --WS (talk) 13:38, 9 December 2014 (UTC)

The subject of the article is "Pancreatic cancer", of which the many varied types of PanNETs, none predominating in the group, form 1-2% of cases. The comprehensiveness requirement reads "it neglects no major facts or details and places the subject in context". Which major facts or details do you think are neglected, within the scope of an article on the wider topic? It certainly is more comprehensive than the many review articles, several used as sources here, that are titled using "Pancreatic cancer" but then say (usually at the end of para 1) that they are in fact entirely, or almost entirely, about pancreatic adenocarcinoma. Examples, by first-named author, are: Wolfgang, Vincent, Ryan Syl de La Crz. Most sources directed at the general public have made the same choice as this article. Leading examples are: ACS, NCI (professional, also patient version), NHS Choices, WebMD, and so on. Wiki CRUK John (talk) 22:08, 11 December 2014 (UTC)

Well on one hand, there is indeed extensive information in the classification and management sections, but on the other hand, any specific information on diagnosis and prevention/screening is completely lacking. Signs and symptoms are very shortly mentioned but only within the classification section, and it is not clear that the information in the symptoms section does mostly not apply to them. --WS (talk) 02:13, 12 December 2014 (UTC)

It is clear in that the list is introduced as "common symptoms of pancreatic adenocarcinoma". I think the symptoms of PanNETs are too various to describe usefully in an article of this type, but I'm willing to be pursuaded otherwise. The two main journal sources (Burns & Edil and Oberg et al) conspicuously duck giving comprehensive lists of symptoms, but I could add that (from B&E) the ~90% of non-functioning cases "typically present with symptoms related to local mass effect or metastatic disease" which is all they have to say on that. Wiki CRUK John (talk) 10:55, 18 December 2014 (UTC)

1) WS correctly points out thematically that resectibility criteria are examined only briefly herein. However, this is a busily debated / potentially moving target over the past 6 to 7 years e.g. - PMIDs: 16703621, 24787115, 22209537, 22064622, 25519932, 25516657, 25339810, 25071332, 24578248 - and many others. The issue is tangentially touched on within "Research." I would think that a brief encyclopedic synthesis of this corpus is possible, taking WP:NPOV care. I will address the issue tomorrow if I have time. 2) As far as "too much, or too little" on PanNETs, I don't have a strong opinion, although I might opt myself to reduce that content modestly. Such a trim would be procedurally simple, and I think might be a broad modest trim, with a particular focus on "Treatment. I'd be willing to do this - give me a nudge on my Talk page. I might even be tempted to do it boldly tomorrow. FeatherPluma (talk) 00:31, 25 December 2014 (UTC)

I'd rather it was discussed first. The subject of "Pancreatic cancer" certainly includes them, and I'm more sympathetic to exapanding rather than reducing the coverage of them, where succinct generalization is possible. Expanding resection criteria somewhat is fine, though anything more than a few lines should go to surgical articles, perhaps even a new one. Wiki CRUK John (talk) 01:25, 25 December 2014 (UTC)

Very good, I'll follow your preference as to PanNET content. Thus, in alignment with input here / other commentators: 1) I've added several WP:MEDRS in response to the comment that there was inadequate sourcing for the section. In so doing, the text mildly consolidates. 2) As far as resectibility considerations for the "adenocarcinoma family" I will add a one or two sentence synthesis with selected reference later today - but this topic is a modestly warm potato that should not to allowed to become disproportionate in an article for general readers. For example, the NEJM review dodges over-engagement by conveying the essence as a sliding scale "continuum of resectibility" i.e. contentious, evolving, and dependent on "risk acceptance". 3) In terms of PanNET symptoms and diagnosis, I think briefly expanding on these as recommended by WS would be reasonable, perhaps along the lines of referencing Burns & Edil and Oberg et al as you suggest. However, substantial enlargement is not needed.FeatherPluma (talk) 14:47, 26 December 2014 (UTC)

I've added content, perhaps too much (and over ref) to fully address resectibility - and in doing so added neoadjuvant Rx. I do not hv time to get to PanNET symptoms and diagnosis today. FeatherPluma (talk) 00:25, 27 December 2014 (UTC)

Great, thanks! That sounds a good approach to me, and I hope others. Wiki CRUK John /Johnbod (talk) 00:43, 27 December 2014 (UTC)

I too definitely support FeatherPluma's improvements. (I've also started a subsection on staging of PanNETs) 109.158.8.201 (talk) 14:31, 27 December 2014 (UTC)

Support promotion to FA. All of my concerns have been addressed. One additional suggestion would be to devote one or two sentences of the diagnosis section specifically to the diagnosis of Pannets. --WS (talk) 16:26, 11 January 2015 (UTC)

Comments from Dwaipayan

• "In people not suitable for curative surgery, palliative chemotherapy may be used to improve quality of life and extend life." There seems to be a minor technical problem in the language. The chemotherapy, whether purely palliative versus curative versus both palliative and curative, depends on the intent. Palliative intent does not necessarily mean that the intent is against lengthening life; however, usually palliative chemotherapy means chemotherapy not particularly intended to lengthen life. So, in this instance, is Gemcitabine (or other regimen) really palliative, if the intent is to increase life? Again, I admit that the same agent can be used with palliative versus curative intent. What I propose is to omit the word "palliative" preceding chemotherapy, and use "chemotherapy may be used to improve quality of life (palliative intent) and extend life". Other comments are welcome. --Dwaipayan (talk) 23:26, 7 December 2014 (UTC)

I see the point, but I think "...chemotherapy may be used to extend life or improve its quality" is neater. Johnbod (talk) 00:10, 8 December 2014 (UTC)

Sorry, I've been doing some edits, here and to the article, as Johnbod, my other a/c, which I try not to do to avoid confusing people. Anyway, we're both going to bed now. Wiki CRUK John (talk) 01:04, 8 December 2014 (UTC)

I've been, and am, travelling, so won't be editing much, but I have some refs, rewrites etc stored up for after I get back. Wiki CRUK John (talk) 00:00, 9 December 2014 (UTC)

This minor point was resolved easily.--Dwaipayan (talk) 23:25, 11 January 2015 (UTC)

Comments from Cas Liber

Reads well - some queries below: Cas Liber (talk · contribs) 03:45, 17 December 2014 (UTC)

In the United States, pancreatic cancer is the fourth most common cause of deaths due to cancer - surely "death" here? (collective noun?)

I don't think so - they count the deaths one by one (in theory anyway). Or you can use both. I don't like "cancer death", which is collective. Wiki CRUK John (talk) 18:08, 17 December 2014 (UTC)

Pancreatic adenocarcinoma typically has a poor prognosis - I'd word it stronger - "dismal prognosis" (but not a deal-breaker)

Happy to do that - it is just the word that a large number of sources use. I anticipate flak from some though. Wiki CRUK John (talk) 18:08, 17 December 2014 (UTC)

While agreeing that "poor prognosis" may be considered an understatement, I think it would be preferable to avoid here the cliché of "dismal prognosis", for the sake of some of our lay readers (personally, I'd prefer "very poor"). 109.158.8.201 (talk) 15:32, 22 December 2014 (UTC)

As I said! Wiki CRUK John (talk) 23:29, 22 December 2014 (UTC)

I've gone "very poor" for now, though I'd prefer dismal. Wiki CRUK John (talk) 02:15, 5 January 2015 (UTC)

Fwiw, I don't have a strong opinion on this - I just felt that "dismal prognosis" was the sort of pro-speak we were trying to avoid for the benefit of our wider readership. 86.181.67.166 (talk) 15:50, 5 January 2015 (UTC)

well "dismal outlook" then. There's nothing technical about "dismal". Wiki CRUK John (talk) 19:05, 5 January 2015 (UTC)

Pancreatic adenocarcinoma has a number of symptoms, but none that are individually distinctive to it or appear in the early stages of disease. - first clause redundant (all conditions have "a number of symptoms") - I would reword to "The symptoms of pancreatic adenocarcinoma are not distinctive, and do not generally appear until the disease is well advanced"

Don't really like that - it may be true that "all conditions have "a number of symptoms"" (though, really, there are no monosymptomatic diseases? Odd if so - oh, 67,300 google results for "monosymptomatic), but pancreatic cancer surely has an unusually large crop. "the disease is well advanced" is a bit less lay-friendly than "the early stages of disease", isn't it? Wiki CRUK John (talk) 19:05, 5 January 2015 (UTC)

The sentence remains problematic. It appears to be unsourced (not covered by the NEJM review) and somewhat equivocal (isn't painless jaundice fairly distinctive?). IMO, this is part of a broader issue regarding the way we're presenting Symptoms/==Diagnosis==, and I've made some suggestions for improvement on the article talk page (permalink). Despite the lack of response there I've so far hesitated to make changes based on silent consensus, given the FAC process (I too am keen to be able to support promotion of this article which I've been working on). 86.181.67.166 (talk) 10:48, 11 January 2015 (UTC)

Changed to "The symptoms of pancreatic adenocarcinoma do not usually appear in the early stages of disease, and are individually not distinctive to the disease." with other refs. No, I don't believe painless jaundice is distinctive; our long and rather technical article on it doesn't seem to mention "pain" once, though I expect it should. Wiki CRUK John (talk) 13:34, 11 January 2015 (UTC)

Thanks. Can we move the following to somewhere near the start of the Symptoms section, where imo this sort of content belongs (and would also be also rather closer to the figure illustrating the head, body and tail of the pancreas)?

The symptoms at diagnosis vary according to the location of the cancer on the pancreas, which anatomists divide (from left to right on most diagrams) into the thick head, the neck, and the tapering body, ending in the tail. About 60–70% of adenocarcinomas are in the head of the pancreas, and 20–25% in the body or tail.<ref name="NEJM14" /> The most common symptom for all locations is unexplained weight loss, which may be considerable. Tumors in the head of the gland typically also cause jaundice, pain, [[Anorexia (symptom)|loss of appetite]], dark urine, and light-colored stools. Tumors in the body and tail typically also cause pain. For all locations, nausea, vomiting and a feeling of weakness are present in a large minority (between 35% and 47%) of people at diagnosis.<ref name="Cruz" />

Fwiw, I'm *very* supportive of making disease articles such as this one more accessible to lay readers, but I don't think that simply listing common presentation symptoms under ==Diagnosis== is actually helpful. Any response to my article talk-page suggestions (permalink)? 86.181.67.166 (talk) 14:12, 11 January 2015 (UTC)

We disagree on this. As I've explained before, I'm very keen to keep the "symptoms" section as clear and patient-friendly as possible, and to keep the relationship between symptoms and precise location on the pancreas out of there. So I think that belongs in diagnosis. Johnbod (talk) 14:17, 11 January 2015 (UTC)

The source of disagreement does not regard accessibility, since these things can be written more simply: eg "Most (60–70%) pancreatic cancers are in the head of the pancreas" - period... thereby saving the reader from wondering why the quoted figures don't add up to 100%. (Though wouldn't this information really be better under ==Types== - with the illustration helpfully alongside?) Please see my talk-page post of 2 December (and subsequent suggestions). 86.181.67.166 (talk) 14:40, 11 January 2015 (UTC)

I'd like to see some quantification/elaboration of defining "family history" or genetic loading. This interests me personally as I've had an uncle and great aunt die of Ca pancreas - so giving the reader some idea, even as a footnote, would be good.

Quantification should all be in terms of increased risk, I think, and these are what the sources give (this sort of thing). Actual incidence is probably too variable in human populations to use. Some of the increases are rather low, others alarmingly high. Wiki CRUK John (talk) 18:08, 17 December 2014 (UTC)

I've added a bit on both aspects. Wiki CRUK John (talk) 02:34, 5 January 2015 (UTC)

Otherwise article looking pretty good -doing a nice job of balancing accessibility and exactness (i.e. minimising jargon and maximising plain English) Cas Liber (talk · contribs) 04:29, 17 December 2014 (UTC)

• Many thanks, and for this review. I think all points are now sorted, or await a response. Wiki CRUK John (talk) 19:06, 5 January 2015 (UTC)

• Alright, I support this being promoted - one minor issue for me is see also section - normally I'd recommend putting these in body of article, so a small section on (say) advocacy and support groups...but the article is so long and all it would be is sentences stating the obvious ("X is a group promoting awareness of..yada yada yada") so I can see a strong case for them being listed at the bottom Cas Liber (talk · contribs) 21:59, 10 January 2015 (UTC)

Thanks! I haven't actually touched the see alsos. In my position as Wikipedian in residence at Cancer Research UK I have a COI here, but I'd argue that probably both in the UK & US as much or more work on pancreatic cancer is done by the big "all cancer" charities like CRUK than the inevitably smaller "one cancer" ones, so if you start a text section, you have to start listing them too, plus groups in other countries ... Johnbod (talk) 23:21, 10 January 2015 (UTC)

Comments by Jfdwolff

Great collaborative effort, and testimony to the hard work by a number of dedicated editors.

• The word "patient" should generally be avoided and replaced with a neutral term.

I have gone through & replaced with "person" or "case", except in the "history" section. See User:Curly Turkey's comments above - he thought that "The first reported partial pancreaticoduodenectomy was performed by the Italian surgeon Alessandro Codivilla in 1898, but the person only survived 18 days before succumbing to complications" was ambiguous, as "person" might refer to the surgeon. I find this "ambiguity" implausible, but "patient" certainly reads more naturally. CT changed it to patient, Doc James changed it back, but I've now changed it to patient again. I understand the general avoidance of "patient", but I don't see that it applies when referring to specific but un-named historical individuals who died decades ago or more (and from their disease). There's also the one just after about one of Whipple's cases in the 1930s: "Only one of the patients had the duodenum totally removed, but he survived for two years before dying of metastasis to the liver". If anyone doesn't agree with these, please supply arguments relating to such historical examples, and suggestions on alternative phrasings - "case" and "person" don't really work. Wiki CRUK John (talk) 12:25, 17 December 2014 (UTC)

• The opening line sounds a little bit clunky. While cancer is thought to arise from a single cell, the current phrasing suggests that it is diffuse/hyperplastic growth rather than a mass lesion. I suggest this is rephrased.

Another of the places where the grass has been churned up. See Talk:Pancreatic_cancer#Opening_line. I agree it could be improved, but I'm not sure how at present. Perhaps discussion should continue at that talk page section. Wiki CRUK John (talk) 12:25, 17 December 2014 (UTC)

• It is my personal view that "Classification" should not precede "Signs and symptoms", because as a section it usually requires a lot of context from other sections. It is probably best covered as a subsection of "Diagnosis".

The current position follows WP:MEDMOS. Personally I think (in this case anyway) it is much easier and effective to get this clear early on. It gives even more context than it takes to/from "Signs and symptoms" and "diagnosis", doesn't it? I've noticed that other web pages that don't have it early on get into difficulties, and risk being misleading. Wiki CRUK John (talk) 12:25, 17 December 2014 (UTC)

Imo, without assuming some background knowledge (from the lead?) appropriate contextualization is always going to be an editorial challenge (eg describing a spectrum of symptoms depends on background knowledge of tumor types). However, I do share some misgivings about the choice of heading. Apart from the strictly editorial question of whether it's appropriate to expect our readers to know that we're only referring to pathological (rather than actual clinical) classification, the fact is we're not even really getting close to outlining that – and I wouldn't suggest we attempt to do so (though a ref might be helpful[?]). Imo, 'Types' would be a far more appropriate heading, which is both unpretentious and reader-friendly. And fwiw, I don't think a highly restrictive interpretation of MEDMOS should be a reason for veto (per WP:MOSvT??). 2c, 109.158.8.201 (talk) 13:29, 20 December 2014 (UTC)

• It is not quite clear why "Alcohol" is a subsection of "Risk factors" despite only consisting of three sentences. I would suggest breaking out the genetical causes into their own subsection, seeing that genetics is thought to contribute substantially to many cases.

It's in a sub-section to avoid it being taken as part of the list, because it isn't endorsed as a risk factor, but I thought that there has been so much investigation as to whether it is that it shouldn't just be omitted. I'm thinking what to do with inherited genetic factors in the light of Cas Liber's comments above too. Wiki CRUK John (talk) 12:25, 17 December 2014 (UTC)

• In "Diagnosis", the paragraph beginning with "About 80% to 85%" doesn't have reference despite making a specific numerical claim. Similarly, the next paragraph ("Patients sometimes") duplicates content from earlier in the article, but has no reference. Further on in this section, "Pancreatic cancer has an immunohistochemical profile [...]" has no reference despite being a rather specific claim.

These cut (now at talk section) Wiki CRUK John (talk) 17:58, 5 January 2015 (UTC)

• In "Mechanism", I'm wondering whether it might be an idea to move the highly technical content about mutations into a table. I am concerned that the reader will change channels after being put off by the terminology.

I've thought about this, but they come quite far down the article, by which time the reader has probably learned to sink or skim, as it were, and are over fairly quickly. Not all the info is standard, so either there would be information loss, or a pretty messy table. I'm more concerned about the hereditary factors up at risk factors, which I think are a bigger fence for the reader to jump, but some of the scary information on risk %s is significantly variable between sources, & possibly too alarming to put out in a context like this. Wiki CRUK John (talk) 15:34, 6 January 2015 (UTC)

• • (piggybacking again) I'm also having trouble understanding one of the sentences about the genetics of PanNETs, which currently seems focused on Multiple endocrine neoplasia type 1: Instead, common mutations affect MEN1, as in the inherited Wermer's syndrome, DAXX, mTOR and ATRX.^{[Burns 2012]} I think I know what the original writer probably meant, but without access to the full text of the cited source, it's difficult for me to do a straight fix the wording with confidence. (Another source that might conceivably come in handy here is PMID 22586144 [9].) 109.158.8.201 (talk) 14:03, 28 December 2014 (UTC)

Clearer as "Instead, common mutations affect proteins including MEN1, as in the inherited Wermer's syndrome, and also DAXX, mTOR and ATRX." ? Wiki CRUK John (talk) 18:08, 5 January 2015 (UTC)

I've now expanded this a bit [10] (per a good source, imo). A small, vaguely related point: It occurs to me that perhaps names of genes should really go in italics? (done) 86.181.67.166 (talk) 23:02, 5 January 2015 (UTC), ex-109.158.8.201

• "Prevention" is a short section, and I am unsure why it needs a tiny subsection called "screening". How about calling the whole section "prevention and screening"?

Yes, done that. Wiki CRUK John (talk) 12:50, 17 December 2014 (UTC)

• The "Management" section is very good, although I am wondering if all discussion of chemotherapy could be conducted in a single section. After all, the adjuvant and palliative regimens are effectively identical.

I rather think it was like that until the FAC began .... Wiki CRUK John (talk) 12:50, 17 December 2014 (UTC)

Piggybacking here just to say that I think the subsection on PanNETs still needs some attention (commented here). Also I agree it would be preferable to avoid having a separate "Palliative care" subsection. 109.158.8.201 (talk) 11:15, 21 December 2014 (UTC)

I've re-united all chemo, also radiotherapy. The PanNETS section has been re-written & commented on elsewhere. I don't think JFW was suggesting "to avoid having a separate "Palliative care" subsection", which I don't think a good idea. Wiki CRUK John (talk) 18:50, 5 January 2015 (UTC)

• "History" - is there any source that discussed how the condition was first recognised? Who was the Courvoisier's sign of the eponymous sign? I am a bit unsure about the direct quote at the end of the section. Any reason it can't be paraphrased?

I didn't find any, though I'm sure it's somewhere. Ludwig Georg Courvoisier (1843–1918), a surgeon from Basel, published Casuistisch-statistische Beiträge zur Pathologie und Chirurgie der Gallenwege in 1890, describing the sign as eliminating the possibility of gallstones, but not actually mentioning cancer (according to Wikipedia), which is why I left him out. But he could be mentioned. I also looked for the early radio & chemo history, but couldn't find it easily - I imagine the early chemo results were so poor it hasn't been written up much. Personally I like the quote, and on a general point think Wikipedia medicine doesn't use direct quotation often enough (I think mainly put off by gross overuse by some poor editors). The "virtual explosion" is graphic phrasing I'd like to keep, and would obviously need to quote, and by that time you might as well keep the whole line, imo, rather than a paraphrase that would probably take more words to convey the meaning in a blander fashion. See also Curly Turkey above, who doesn't like this either (but also CasLiber above, rooting for "dismal" prognosis). Wiki CRUK John (talk) 12:50, 17 December 2014 (UTC)

comment moved from Talk:Pancreatic cancer (as requested): 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 [PMID 24206780?] alongside?) 109.158.8.201 (talk) 20:48, 19 December 2014 (UTC)

Request response on this point before FAC process closes. Please note that the concern about the use (and tone) of primary sources (eg case series articles) is not just a formal one. Thanks, 86.181.67.166 (talk) 11:27, 8 January 2015 (UTC), previously 109.158.8.201

I've commented on this under the Curly Turkey section above. I don't understand why "Without consulting the reference, this juxtaposition makes no sense". I'm afraid I can't follow some editors in the drive for maximum blandness in prose at every point. That isn't my conception of encyclopedic prose, at least not all the time. We are an encyclopedia, but even medical textbooks allow themselves the odd moments of "explosive tone", and are generally the better for it imo. The fiddly nature of the point being made, about different periods in the past, also makes a direct quotation the most efficient way of conveying this. On the primary studies, I've added a secondary ref for the "1,000 consecutive". I think this and the other one were referred to in DeVita also, but I'm not sure I have the relevant passage to hand now (at home). Even with a secondary ref, the "1,000" is a useful read, and when the text is talking about a specific historial series in a history section, I think it is good to refer to the original primary source as well as a secondary one, as one would in any normal history section or article. Wiki CRUK John (talk) 15:10, 8 January 2015 (UTC)

Yes, I too support placing notable primary sources alongside the secondary ones in medical history content – as long as the secondary ones are in place. To me, that would typically mean each statement being clearly supported by an appropriate secondary source.

Regarding the use of quotation: I don't see consensus for this particular usage. Also, the WP:QUOTE essay suggests that "Overuse happens when... a quotation is visually on the page, but its relevance is not explained anywhere" – as seems to me to be the case here. 86.181.67.166 (talk) 11:21, 9 January 2015 (UTC)

RE Courvoisier's sign: Based on this review/historical article, which I've used to source/rephrase a statement in the Diagnosis section, I agree with John that no mention of Courvoisier himself is needed here, given that he seems not actually to have mentioned pancreatic cancer in his original publication of 1890 (unsurprisingly perhaps, given the incomplete recognition of the disease at the time he was writing). 109.158.8.201 (talk) 15:25, 22 December 2014 (UTC)

Thanks for expanding this nicely, 109. I'm neutral about Courvoisier, & happy to put him in if wanted. Wiki CRUK John (talk) 18:50, 5 January 2015 (UTC)

Cheers, John. I think that history of Courvoisier's sign here might be wp:undue. 86.181.67.166 (talk) 22:51, 5 January 2015 (UTC), ex-109

• "Research" - good balanced coverage of the current state of research.

Happy to offer unqualified support once the above is addressed. JFW | T@lk 10:24, 17 December 2014 (UTC)

Many thanks for this careful review. I think all points are now sorted, or await a response. Wiki CRUK John (talk) 18:52, 5 January 2015 (UTC)

Support promotion to FA. The history section is now very good. I would still recommend splitting the "genetics" part of the "Risk factors" section into a separate subsection, but that's no biggie. JFW | T@lk 20:49, 10 January 2015 (UTC)

Comments by Maky

A very dear little lemur I knew just died of pancreatic (and liver) cancer... so it's only appropriate that I use this opportunity to learn more and help get this article promoted. Thank you for working on it.

• I think the lead should note that the recent onset of diabetes (and the other signs/symptoms) can also be an indicator. Speaking from experience, I feel this is important to note up front.

Hmm, it is the complicated nature of this, as both a symptom and possible cause, with the difference between new and long-standing diabetes, plus age factors that I think makes it too difficult to fit into a list in the lead, without misleading. De La Cruz, p. 627, says "Also, a sudden onset of atypical type 2 diabetes mellitus that is difficult to control in a thin patient 50 years or older suggests pancreatic cancer." - perhaps I should quote that in a note, but so many qualifications make it hard to include in the lead I think. What do others think? Wiki CRUK John (talk) 18:20, 4 January 2015 (UTC)

That is now quoted in a note. Wiki CRUK John (talk) 18:13, 5 January 2015 (UTC)

I must be missing the note. Regardless, you make a good point. It's a complicated issue. – Maky ^« talk » 21:28, 6 January 2015 (UTC)

Currently #36, in the diagnosis section. Johnbod (talk) 21:34, 6 January 2015 (UTC)

• Under "Signs and symptoms", I suggest putting jaundice before pain in the upper abdomen since the latter also mentions the former (before it's discussed).

But jaundice is (rather narrowly) more common, so I don't like to switch. Wiki CRUK John (talk) 18:34, 4 January 2015 (UTC)

• The "Pancreatic cancer staging" gallery looks very informative, but might be best to split it into two rows with larger images so that the text might be read without clicking on each file.

I have, I hope achieved this by using packed mode, and removing one to a normal picture. On my screen all the 5 pancreas images now fit on one row, though no doubt this isn't true on all screens. I think this is better, certainly. Wiki CRUK John (talk) 17:57, 4 January 2015 (UTC)

This looks a thousand times better! There were some issues with the font, where letters were being cut off, so I fixed them in the SVGs... so hopefully I didn't mess anything up. – Maky ^« talk » 21:28, 6 January 2015 (UTC)

• The two Whipple's operation images might be best put together using Template:Multiple image. This is just a suggestion.

I'm not so keen on that, which tends to produce either enormously wide or tall images, or people then make them too small. Wiki CRUK John (talk) 17:57, 4 January 2015 (UTC)

• 2 items: "Management of pancreatic cancer should be in the hands of a multidisciplinary team including specialists in several aspects of oncology, and is therefore best conducted in larger centers." & "treatment should be undertaken in a specialized center" – These almost strike me as a case of WP:NOTADVICE... because of the way it's worded.

Added "Specialists advise that the..." which they all do. Wiki CRUK John (talk) 17:57, 4 January 2015 (UTC)

Good, but there were two: "...treatment should be undertaken in a specialized center..." – Maky ^« talk » 21:28, 6 January 2015 (UTC)

Now "guidelines emphasize that treatment ...", with 2 guideline refs. Johnbod (talk) 21:48, 6 January 2015 (UTC)

• Resectability is explained all the way down under Management, although it is used long before then.

Good catch - thank you, Maky. I've made some tweaks in response. I hope this addresses your point. 86.181.67.166 (talk) 12:46, 1 January 2015 (UTC) previously 86.164.164.29, 109.158.8.201, etc

• "However, the changes of the last few years..." – I'm worried that this could become dated and misleading in a few years.

Sadly, there is little sign of this! If it does change it should certainly get written up as it would be a massive & transformational breakthrough. The preceding sentence begins "By the end of 2013...", which I think is dating enough. Wiki CRUK John (talk) 18:20, 4 January 2015 (UTC)

• "ablation technique" – had to look that one up

Linked Ablation#Medicine Wiki CRUK John (talk) 18:20, 4 January 2015 (UTC)

• Any discussion on the topic of the recent achievement of growing a pancreas from stem cells? I know that any treatments would be a long way off, but has the possibility been mentioned in the literature?

No doubt, like many other things, but it doesn't come up in the sources, nor was it mentioned by the outside researcher-reviewer. I don't see it would help if the disease has metastasized at diagnosis, and replacing the functions of the pancreas is by no means the biggest problem in treatment. Wiki CRUK John (talk) 18:20, 4 January 2015 (UTC)

Also, I know medical articles typically only discuss human disease, but if photos from the lemur necropsy would be helpful, I might be able to get them released. I believe the photos I have showed cancerous growths on the pancreas and liver. If so, I'm not sure what could be said since I didn't see any material on pancreatic cancer in other species at a quick glance. Either way, just let me know. – Maky ^« talk » 08:57, 31 December 2014 (UTC)

I've not seen anything on animal tumours, not that I've looked very hard. Veterinary oncology seems normally only concerned with dogs and cats, & rather sketchy on them. Thanks for these comments, and I'm sorry to hear about the lemur. Wiki CRUK John (talk) 18:34, 4 January 2015 (UTC)

I wish more veterinary articles existed so that articles as wonderful as yours could be linked where appropriate. It just leave Wiki feeling very human-centric, particularly for anyone coming online to learn more about the illness in animals. But that's nothing that can be helped here. Again, thank you for doing such a wonderful job with this topic on the human side of things! – Maky ^« talk » 21:28, 6 January 2015 (UTC)

When I've asked about this, the general consensus was that, apart from the economics, the treatment of animals would typically be unfairly distressing for them (chemo, radio etc) so even the (I imagine rare) early stage diagnoses are normally put to sleep at once. Cancers seem to vary rather a lot between species, so any development of therapy would need to be species-by-species. There is some research interest in what animal cancers can tell us about human cancers. Johnbod (talk) 21:48, 6 January 2015 (UTC)

Update: One of my points above was only partly addressed. Once that last statement is fixed, I will gladly give my support. – Maky ^« talk » 21:28, 6 January 2015 (UTC)

Now done. Thanks very much! Wiki CRUK John/Johnbod (talk) 21:48, 6 January 2015 (UTC)

Support: All of my concerns have been addressed and appropriate changes have been made. I am happy to give my support. And I'm sure Janga would have given you a nose touch and a little friendly face licking if she were still here and able to appreciate what you've done. – Maky ^« talk » 22:14, 6 January 2015 (UTC)

Coord note -- I'll need longer to walk through all commentary but in the meantime I didn't see an image review above, you can post a request on WT:FAC as necessary. Cheers, Ian Rose (talk) 04:53, 11 January 2015 (UTC)

No, there isn't one. Will do. Wiki CRUK John (talk) 13:24, 11 January 2015 (UTC)

Image check - all OK (GermanJoe)

• All images are "own work" or CC (verified via website link or OTRS ticket) - OK.
• All images have sufficient source and author info - OK.
• Maps and graphs include source data information or are taken from reliable sources - OK.
• File:Jaundice08.jpg: personality rights affected, but all relevant templates and information are included (AGF on permission) - OK. GermanJoe (talk) 15:07, 11 January 2015 (UTC)

More Turkey

• I've given the article another read-through and made some minor MoS edits. I find it readable, and I'm going to give it my support. I do have a couple of comments:
• CDKN2A mutations/deletions (about 95%), TP53 inactivations (75%), and SMAD4 deletions/mutations: I don't know the meaning of "mutations/deletions", but I imagine if it's "mutations" first the first time, it should be the second time as well. Also, I don't know whether MOS:SLASH applies here.

I was going to just switch to "mutations", but since we have Deletion (genetics), I've linked that. I'm well out of my depth here, but is Gene silencing the right link for "inactivations"? It appears in the sources. And for CDKN2A and SMAD4, are all the mutations relevant here deletions, so I can drop "mutation", or only some or most? Or do we even know? "Lancet" & "Wolfgang" have the most on this of the commonly used sources. To pursue. Wiki CRUK John (talk) 10:43, 13 January 2015 (UTC)

Since we haven't set out here to provide any real detail about the mechanisms, I feel John's simple fix of just referring to '[gene] mutations' (in general) could be editorially appropriate. Otherwise, I think a bit more explanation is needed for our general readership. 2c, 86.181.67.166 (talk) 16:06, 13 January 2015 (UTC)

• In "Distribution", you have "11th" and "12th", but "tenth" and "fifth" in the next paragraph. I'd homogenize them.

Done, to 10th etc. Wiki CRUK John (talk) 10:43, 13 January 2015 (UTC)

• Curly Turkey ¡gobble! 07:50, 13 January 2015 (UTC)

Thanks for this very comprehensive & thorough review! Wiki CRUK John (talk) 10:43, 13 January 2015 (UTC)

Victoriaearle comments

Support with comments: I read through last week, made notes, carefully watched the edits since. Much that I noted is now taken care of. A few comments though (from a layperson):

• "The exocrine group is dominated by pancreatic adenocarcinoma (variations of this name may add "invasive" and "ductal"), which is by far the most common type, representing about 85% of all pancreatic cancers,[2] although the pancreatic ductal epithelium from which it arises represents less than 10% of the pancreas by cell volume.[18]" >> long sentence at the beginning of a section. Had a hard time getting through it.
• "A biopsy by fine needle aspiration, often guided by endoscopic ultrasound, may be used where there is uncertainty over the diagnosis, but a histologic diagnosis is not usually required for removal of the tumor by surgery to go ahead.[3]" >> I understand what's being said is that needle aspiration isn't required for surgery; but to me it seems these are two separate procedures and so should be split.
• Resection margin - margins are referred to as "edges of the tissue removed". At least in the US, in my limited non-medical experience, margins are always referred to as margins and not edges. Not a big deal but thought I'd mention in case it causes confusion (it confused me).

Isn't a margin always an edge at a rather higher scale? "“Negative Margins” is a term that surgical oncologists and pathologists employ; it means that there are no tumor cells identified at any of the edges of the resected specimen" - US surgeon in the EL at Resection margin- "margin" seems inherently rather unclear to me, so I'm keen to keep the clearer "edge". I had to rewrite Resection margin a while back btw, as it was gibberish, so anyone with ideas as how to make the point clearer here or there is welcome to chip in. We are talking about surfaces here; the margin is supposed to be between the tumour & the cut surface, but on examination it may turn out that the tumour extends further than thought, and there is no such clean margin, and the edge includes cancer. The article says "Even when the operation appears to have been successful, cancerous cells are often found at the edges of the tissue removed when examined microscopically by a pathologist (this will always be done), indicating the cancer has not been entirely removed,[2]". This seems fine to me, I must say. Wiki CRUK John (talk) 02:56, 14 January 2015 (UTC)

Yes, I know what a margin is in regards to cancers and biopsies and so on. The point I think I'm trying to make is that I've always only heard the term "margin" and never "edge". I'm not really fussed about it, but it's something that I noticed and I did click out of the article to look at the link. Maybe put "margins" in parenthesis? I do think the term margins should show up somewhere. It could well be that the terminology is different in the US and UK. Victoria (tk) 03:11, 14 January 2015 (UTC)

I don't think it is (as the surgeon's quote above demonstrates), and I'm quite sure that most people don't know quite what a margin is in this context. I don't like 86's removal of "edge" at all, & will think of a better way later. Outside of page margins, "margin" seems to me to be a word that is inhererently unclear, however common. Wiki CRUK John (talk) 09:28, 14 January 2015 (UTC)

(edit conflict) The usual term is certainly "margin" – a word which corresponds in common English to "edge or border"[11]. My feeling is [12] that "margin" is a sufficiently common word for anyone able to attempt this section to take in without much trouble (especially with the help of the link). I also tried [13] glossing "margins" with "edges", but to my eyes at least, my attempt looked like an unnecessary thicket of words. Anyway, see what you think... 86.181.67.166 (talk) 09:44, 14 January 2015 (UTC)

• Link to cancer pain?
• Palliative care: "CPB is a safe and effective way to reduce the pain, which generally reduces the need to use opioid painkillers, which have significant negative side effects" > One source says "Opioids are the mainstay of pharmacologic therapy for pancreatic cancer pain. Initial therapy may consist of a short-acting agent such as morphine or oxycodone." Another sources says " fewer adverse effects than opioids is important for patients". Should probably be clear that CBP is an option >> when, particularly at end-stage, palliative care is all that can be done. But defer to the medical experts on this.
• "Early operations were compromised partly because of mistaken beliefs that it was essential for life to preserve the duodenum" > I had a hard time parsing this. Maybe delete "for life"?

No, then people will ask "necessary how?"! I will redo along "people would die if ..." lines. Wiki CRUK John (talk) 02:56, 14 January 2015 (UTC)

That's all. Oh agree with Cas re the "See also", and also agree to leave the gene mutations as is. I think it's too tricky to explain that the genes are a.) mutated, and b.) are deletrious. Victoria (tk) 02:16, 14 January 2015 (UTC)

• Thanks for these. I'll address the simpler ones when it's not so late. Wiki CRUK John (talk) 02:56, 14 January 2015 (UTC)