Talk:BRCA mutation

Wiki Education Foundation-supported course assignment

This article was the subject of a Wiki Education Foundation-supported course assignment, between 12 May 2020 and 22 June 2020. Further details are available on the course page. Student editor(s): Tavkaur25.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 15:09, 16 January 2022 (UTC)

Article for common effects

I really think an article for both BRCA1 and BRCA2 mutations is needed, because many effects are common, and many articles treat them as one. Mikael Häggström (talk) 14:59, 2 April 2010 (UTC)

Benefits of various prophylactic methods

Please keep an eye on overall mortality when editing this article. Statements like "lowers cancer risk by 90%" are realistic for some methods but very misleading when you look at overall mortality ballance. Richiez (talk) 21:11, 3 February 2011 (UTC)

I agree, but we also want to remember that quantity of life is not always preferable to quality of life. An oophorectomy that leaves the woman with, say, sleep deprivation for 30 years (which apparently happens to some women) is not necessarily an improvement over living 28 years in apparent good health.

Unfortunately, while there are some sources on overall survival, I haven't seen any on the quality-adjusted life years for any given approach. WhatamIdoing (talk) 16:45, 9 June 2011 (UTC)

Notes about cancer risk reporting

There are statistical and logical traps that need to be considered.

• many sources are unclear whether they estimate conditional or absolute risk, unfortunately because of the special nature of the risks the difference is dramatic
• it is not meaningful to state a "lifetime risk" as the life expectancy is substantially influenced by the other cancer and cancer risk appears to be rising cumulatively with age - applies to any other cancer but the effect is much more dramatic with BRCA related cancers
  • for example it is incorrect to say "82% will develop breast cancer" because 56% may develop ovarian cancer and not enough will be left to develop breast cancer
• the risk may be highly heterogenous between families. BRCA2 related breast cancer risk has been reported anywhere between 22-85%

Richiez (talk) 12:40, 3 March 2011 (UTC)

Chromosome image

I'd love to see something like File:BRCA1 location.svg (except in English, and for both chromosomes) in this article. WhatamIdoing (talk) 21:29, 31 May 2011 (UTC)

Getting tested

looks like a good opportunity to make money for people who did not manage to get a degree? Really nothing for an allied health worker.. Richiez (talk) 23:51, 2 June 2011 (UTC)

I think it's not unusual for certified genetic counselors to have a master's degree; "allied health" basically means "anyone except physicians, nurses, and dentists", rather than "low skill". They may not spend as much time in school as psychologists and optometrists (who are also "allied health"), but I think it's rather more than a medical assistant.

It's not really a job that I'd want at any price. Spending all day explaining basic genetics to people who are on the edge emotionally is not my idea of fun. WhatamIdoing (talk) 00:10, 3 June 2011 (UTC)

Advanced age risk

While the example makes sense it is not supported by the data. Most sources and specifically Chen refrain from specifying lifetime risk, for a good reason. The unproven hypothesis that many authors make is that unlike normal breast cancer there is not a fall in risk after say age 75. Some go as far as assuming close to 100% conditional lifetime risk - if that is true it would mean the example would be completely wrong. Currently I do not think we can decide it either way. Richiez (talk) 21:12, 3 June 2011 (UTC)

You're right; I've been sloppy. What I have sloppily described as 'lifetime risk' is the risk by age 70, which is not exactly the same thing, even for women with BRCA1, whose median age at death is not very much older than 70. I will make the basic correction now, and we can talk about more significant improvements later.

Would you take a look at pg 97 in this book (search for "Chen") and tell me what you think about the table at the top? I'd like to present the information in that table (actually, in two tables, because I'd split the BRCA1 and BRCA2 numbers into separate tables). The data is also here in the Chen paper. Do you think that would be a copyright violation? WhatamIdoing (talk) 21:21, 3 June 2011 (UTC)

I think numbers are not subject to copyright although the table should not be reproduced exactly like that, also for practical reasons. Reporting the 10 year risks from the table (looking at original Chen's paper) would seem to be informative? Otoh I would not dwell too much into this numbers, eg the BRCA2 risk group is highly heterogenous - there are families with high ovarian cancer risk and low breast cancer risk and vv. Other aspects of the article might use some love - there is a bunch of possible prophylactic strategies like progestin contraception or tubal ligation to reduce or delay ovarian cancer risk that are not mentioned at all or lack the necessary detail mainly because I did not stumble upon a comprehensive review yet. Richiez (talk) 22:41, 3 June 2011 (UTC)

I'm thinking that the ranges given the Chen table might be more appropriate than trying to reduce it to "the" number. Also, I wonder what you think about expanding the horizontal bar graph to compare not just 'mutation/no treatment' to normal, but also 'mutation/some treatment'. The numbers are currently in the ==External links== section. I don't want it to be too complicated, but it might be appropriate to show at least one treatment scenario.

My expansion list includes expanding the section on prophylactic strategies, adding a section on men with BRCA, and improving the section on non-breast/non-ovarian cancers. Is there anything else on your mind (ideally today or tomorrow, because the library wants this book back ASAP)? WhatamIdoing (talk) 23:08, 3 June 2011 (UTC)

I think these graphics from Kurian et al are hard to beat in any respect: http://jco.ascopubs.org/content/28/2/222/F2.expansion.html and http://jco.ascopubs.org/content/28/2/222/F1.expansion.html . Than again the graphics must be judged with a very sharp mind to get all the fine points - they tend to smooth out the details. So if something like that gets included it should be complemented with some graphics showing age-related risk.. I think I have seen a good one in Chen? Not really sure how many numbers and graphics we want to throw at the reader.

I am undecided on the tables. I consider the screening variant and best known prophylactic strategy as most realistic and useful data (and used in other parts of the article), there could be certainly a third variant before getting too complicated. Richiez (talk) 10:54, 4 June 2011 (UTC)

Maybe we should give up on putting them into the article, and instead link the best tables and graphs in ==External links==. WhatamIdoing (talk) 19:19, 4 June 2011 (UTC)

Half

I'd like to add the fact that BRCA mutations account for slightly less than half of hereditary breast and ovarian cancers. Other single genes account for another 7–8%, and the remaining 45% is unknown (possibly combinations of genes). I'm not sure where to fit this into the page, though. Any ideas? (Source: Morris and Gordon, p. 340). WhatamIdoing (talk) 01:03, 4 June 2011 (UTC)

Not sure but if add that it should also state that all known hereditary factors together amount to some 8-10%. Richiez (talk) 11:25, 4 June 2011 (UTC)

Epidemiology

We seem to be missing a section on the distribution of the gene. Does anyone know if the mutations appear in people without white ancestors? Most of the work I've seen was performed in the US or northern Europe, with obvious racial biases. Do we have a really worldwide prevalence number? WhatamIdoing (talk) 23:18, 5 June 2011 (UTC)

It is known for sure in Japanese: Katagiri T, Emi M, Ito I, et al. Mutations in the BRCA1 gene in Japanese breast cancer patients. Hum Mutat 1996;7:334-9.

The genetics say that it is likely to develop in any genetically isolated polulation. Richiez (talk) 21:58, 6 June 2011 (UTC)

Some more discussion here PMID 19805903 and http://jama.ama-assn.org/content/298/24/2869.full Richiez (talk) 12:53, 7 June 2011 (UTC)

This suggests a prevalence of 1 in 400 to 1 in 800 in "the general population", which from context I believe means "in the US" rather than "in the world". It also deals with questions of ethnicity. Most sources (e.g., from Myriad) give the prevalence with respect to families with known risk factors. This article indicates that BRCA1 mutations are about twice as common as BRCA2 mutations. WhatamIdoing (talk) 19:25, 14 June 2011 (UTC)

LOH

"If the functional copy is harmed, however, then the cell is forced to use alternate DNA repair mechanisms, which are more error-prone. The loss of the functional copy is called loss of heterozygosity (LOH). Any resulting errors in DNA repair may result in cell death or a cancerous transformation of the cell." -- the average reader will interpret this thinking that spontaneous LOH is at the start of carcinogenesis. This is oversimplified - the rate of spontaneous BRCA mutations is so small that it could not explain cancer risk of eg 65%. Richiez (talk) 22:45, 7 June 2011 (UTC)

I agree that it's simplified, but I'm not sure that it's oversimplified. BRCA-related cancer appears in cells that have zero functional copies of the BRCA gene. "Loss of heterozygosity" means, fundamentally, "Loss of the non-mutated, previously functional gene". WhatamIdoing (talk) 23:38, 7 June 2011 (UTC)

Before age 40

About this: I hear that surgically inducing menopause has "substantial adverse effects" even for women who are in their 40s. It might be more accurate to say that these risks affect "pre-menopausal women" rather than "women before age 40". WhatamIdoing (talk) 23:40, 7 June 2011 (UTC)

The risk is substantial even well after menopause - survival impact is statistically significant even for oophorectomies performed at age 60-64, PMID 17476148 . So I fully agree it has substantial adverse effects, on the other hand the balance is different for high risk women. I will amend it if it can be done without getting too unwieldy or feel free to do it. Richiez (talk) 10:02, 8 June 2011 (UTC)

The USPSTF genetic testing recommendation lists the side effects as "at least small". I can't quite decide whether they mean "small, maybe medium-sized, possibly huge" (probable) or "at least the side effects aren't too bad" (my first impression).

Unless we've got a source that directly says age 40 is the magic number for side effects, we should not use that number. We can say that it's worse for "younger women" without putting a specific number on it. WhatamIdoing (talk) 15:51, 9 June 2011 (UTC)

I thought that Kurian wrote something to that effect but was wrong, will see how to rewrtie this. Richiez (talk) 23:47, 9 June 2011 (UTC)

Declining lifetime risk

About this:

Do you understand what I'm trying to say? This is not about the annual incidence (which peaks in the later years). This is about lifetime risk, and the fact that a cancer-free baby girl's lifetime risk is bigger than her cancer-free grandmother's risk, because the baby could get breast cancer in her 20s, 30s, and 40s, whereas it is physically impossible for the cancer-free grandmother to travel back in time to develop cancer in those past years. Although the grandmother's short-term risk is higher, the baby girl's lifetime risk is higher. WhatamIdoing (talk) 23:43, 7 June 2011 (UTC)

I do understand what you are trying to say, but (grossly oversimplified) what the example was was saying in effect is that once someone survived cancer free till the age of 60 she is more likely to survive cancer free till 70.. which I think is mostly pointless and misleading because the per year risk is actually much higher at this age.

The original intended meaning is not supported by available data afaics, certainly not by Chen et al. There is no good universally agreed definition what "lifetime risk" means for high risk groups - I have not seen one. For special high risk BRCA2 groups lifetime risk may 100% minus the probability of previously dying of some other cause. Many authors assume that the risk (if the women survives that long) does not drop after age 75 like it does in non-hereditary cancer.

For this reason I think we can not simply subtract accumulated risk from total risk to compute remaining risk and if the arithmetics are so hairy I would prefer not having it there without exact backing by a good source. Richiez (talk) 10:19, 8 June 2011 (UTC)

Ok, let's try this with numbers:

• The cancer-free baby has a 54% chance of developing breast cancer before age 70.
• The cancer-free grandma has a 37% chance of developing breast cancer before age 70.

But you think that the grandma's chance of developing breast cancer before age 70 is higher than the baby's, that is, that 37 is a bigger number than 54?

Note that I'm not simply subtracting accumulated risk, since it doesn't actually work that way. (If you check the cited table, the numbers don't add up that way.) WhatamIdoing (talk) 15:14, 9 June 2011 (UTC)

Looking at this - there was the claim that lifetime risk is smaller for older woman. This can not be supported by the given statistics. Maybe its true or its wrong but not trivially evident from the numbers that are there.

The other part - "For example, a woman born with a deleterious BRCA1 mutation has a 54% chance of developing breast cancer by age 70. If she survives to age 50 without developing breast cancer, her risk of developing breast cancer by age 70 has been reduced to 37%, because it is no longer possible for her to develop breast cancer in her 20s, 30s, or 40s" may be perfectly ok (I have not checked the numbers) but it is as far as I can see pointless. It could be even misleading for a naive reader who could think "ok, I am out of the worst now that I am fifty". Obviously this would be a pretty serious mistake because cancer incidence actually peaks around that age.

Where I wrote "Many authors assume that the risk (if the women survives that long) does not drop after age 75 like it does in non-hereditary cancer" I did mean the annual risk (or incidence). There are few statistics that go beyond 70 but the hypothesis seems reasonable looking at available data.

What do you consider important about the example? Richiez (talk) 13:24, 10 June 2011 (UTC)

First, I consider it important, because it's a little unexpected. The average person doesn't quite 'get' the law of averages. From their perspective, "I didn't get breast cancer last year" directly means "so I'm more likely to get it next year".

In reality, the longer a woman survives without breast cancer, the less likely she is to get breast cancer before she dies of something else.

This matters because a cancer-free 35 year old who is considering prophylactic surgery is going to get a much bigger benefit than an cancer-free 70 year old considering the same surgery. The actual likelihood of a cancer-free elderly woman ever developing breast cancer—while not "small" on an absolute scale—is actually smaller than the young woman's likelihood, no matter how far in the future you want to look. They might both develop breast cancer at age 71, but the cancer-free 70 year old actually has zero risk of developing breast cancer from age 0 to age 70, whereas the cancer-free 35 year old has a real risk of developing breast cancer from age 36 to age 70 (and a zero risk for ages 0 to 35).

Additionally, I think it's important to include, because multiple sources include it. The way the disease incidence and disease prevalence interact has been subjected to more than a little attention by researchers. I suggest that you look up the Morris and Gordon book, and actually read the pages cited. WhatamIdoing (talk) 19:14, 14 June 2011 (UTC)

If you want to have it as a hint not to wait till age 70 with porphylactic surgery then sure but I would put it into the actual prophylaxis section then. I seem to remember that Kurian has the direct comparison PM at 40 or 50 and this example might be more useful because comparing 35 and 70 is not so realistic - nobody would knowingly wait that long.

Otoh if we consider a 70 year old who by accident learns that she has a high risk BRCA mutation the matter is different. If it is a good health 70yr without special comorbidities who was coincidentally oophorectomised sometimes in her 50ies she has (by my opinion) a very good chance to live to an average age of 87 years unless she dies of BC. If we look at table 2 of PMID 14576434 - as it appears to me the risk between 70 and 80 is high enough to be a serious concern. Also 70 is perhaps the highest age where a PM can be performed without spending too many thoughts on the risks of surgery. I am not sure how this example would help this women.

I do not think the numbers in the example are so unexpected - the problem is the definition of lifetime risk as "not dying of something else". The example claimed that "her risk of developing breast cancer by age 70 has been reduced to 37%, because it is no longer possible for her to develop breast cancer in her 20s, 30s, or 40s". I am understanding it so that the risk is only smaller because she is more likely to die of something else before she can develop BC and can not figure out how the preceeding sentence should be helpful to understand that. Or am I hitting a version of the Monty Hall Problem?

The definition of lifetime risk itself is a problem - most people are used to think about lifetime risks of events with a risk of well under 20%. The situation where there is a lifetime risk of say >56% BC and >37% OC escapes the intuitive understanding of most people. Richiez (talk) 16:55, 17 June 2011 (UTC)

I'm not sure if your summary is exactly right. The grandmother isn't that much more likely to die of heart disease than the baby, because breast cancer doesn't normally result in death. These numbers are about "ever get diagnosed with", not "die from" breast cancer. So the baby can get breast cancer at age 65, be treated successfully, and still die of heart disease at age 80, just like the grandmother can live to age 80 without developing breast cancer, and also die of heart disease.

That seems to be the median case, BTW: The median baby with a BRCA1 mutation will develop breast cancer around age 65, be successfully treated, and die from heart disease, while the median grandmother with the same BRCA1 mutation will not ever develop breast cancer, and will die of heart disease.

The grandmother's risk is lower not because she dies from something else (they'll both die of heart disease), but because she is no longer at risk for developing breast cancer at a young age. It's a basic probability question, not really any different from rolling dice. With a six-sided die, your chance of rolling the same number three times in a row is 1/36. If the first two match, then your chance of success is now increased dramatically to 1/6, because it is no longer possible for the first and second numbers to disagree. The numbers are messier with humans, but it's the same principle: Grandma can't get breast cancer in the past, and therefore her chance of living cancer-free is dramatically increased. WhatamIdoing (talk) 16:18, 18 June 2011 (UTC)

I believe saying "For example, a woman born with a deleterious BRCA1 mutation has a 54% chance of developing breast cancer by age 70. If she survives to age 50 without developing breast cancer, her risk of developing breast cancer by age 70 has been reduced to 37%" is fully sufficient and correct assuming it is the raw data from the statistics. This example could be further expanded, for example providing the same the data for ages 30,40. We can not say anything more from this statistics. Saying "because it is no longer possible for her to develop breast cancer in her 20s, 30s, or 40s" seems unnecessary and I feel very uncomfortable with the because.

The statement "Additionally, the longer a woman lives without developing breast or ovarian cancer, the lower her lifetime risk becomes, as she has outlived the risk of developing cancer at a younger age" is a further unnecessary complication of the picture because the 35 year old with a high risk BRCA1 mutation has a different median residual life span than a a women who already survived to 50. I believe in the case of BRCA1 mutations the difference in expected remaining lifespan may be quite substantial - the women who survived to 50 has survived some natural selection process that has dramatically changed the conditions.

Do we have an expert on statistical paradoxes? Richiez (talk) 22:19, 18 June 2011 (UTC)

The "because" claim is one of the major points in that source. Explaining how an older woman's lifetime risk is lower (even thought the annual incidence is higher) is IMO relevant. WhatamIdoing (talk) 01:02, 26 June 2011 (UTC)

Oophorectomy at age 25?

"Women with BRCA1 are encouraged to have salpingo-oophorectomy as soon as childbearing is complete" - this is patent nonsense as it is formulated now. Perhaps Morris did not read the latest survival analysis of Kurian. Ovarian cancer risk is rather small before age 35 even in BRCA1 risk groups. Kurian explicitly states that oophorectomy before age 40 yields no additional benefits but increases risk for side effects. Much better prophylaxis might be tubal ligation at an early age and oophorectomy around 40. Richiez (talk) 13:40, 8 June 2011 (UTC)

Also note, for BRCA2 risk groups an oophorectomy seems to improve 25-70 year survival by 5% (Kurian). Not a terribly convincing margin although the pure cancer prevention effect is surely worth a lot more. Richiez (talk) 14:24, 8 June 2011 (UTC)

The fact that the recommendation is not based on the best available evidence doesn't change the fact that it's the actual recommendation being made.

In practice, most women in the affected populations (=the developed world) start childbearing between age 25 and 30 and complete it in their mid-30s, so "at age 35" and "after childbearing is complete" are basically the same thing.

We could cite ESMO's official recommendation, which mentions both age and childbearing status if you want to provide the full details rather than a summary. WhatamIdoing (talk) 15:47, 9 June 2011 (UTC)

That seems reasonable though a bit vague - after age 35 could mean a lot of things. Pruthi et al used a similarly vague formulation to the same effect. Is there some good reason not to stick with Kurian et al who say age 40 explicitly? Richiez (talk) 22:01, 9 June 2011 (UTC)

Well, an official practice guideline (like EMSO's) is a much more authoritative source for what is recommended. Kurian is basically one researcher's opinion's about what Kurian (et al) believes should be recommended. WhatamIdoing (talk) 19:17, 14 June 2011 (UTC)

Looked at Kurian again and got the impression they choose 40 perhaps by coincidence and perhaps because a large part of POs occurs around that age - and most likely not really suggesting that it would be better than 35 or any earlier age. So anything like "after age 35 and after childbearing is complete" should be fine. PMID 21123638 and other suggest that the optimal age is still not agreed upon. The ESMO guideline was based on the weak data of PMID 16510331 and in my opinion it was quite daring label it as IIa evidence level. Even PMID 20810374 has still a very short followup and from a decision making viewpoint I consider it very unlucky that they excluded those who had evidence of OC at the time of PO from the analysis. Richiez (talk) 17:08, 17 June 2011 (UTC)

Thinking more about it I would not consider it very good write that PO is recommended "after age 35 and after childbearing is complete" without mentioning that there is no basis to say whether it should be 35 or 40. Kurian data at least provides good basis to estimate the effect of delaying it after 40. So would not it be better to just list the estimated benefits and the effect of delaying it too much? Richiez (talk) 21:23, 17 June 2011 (UTC)

Deciding whether the expert sources have anything behind their recommendations isn't really our job. I personally happen to think it strange for EMSO to give the same answer for BRCA1 and BRCA2 mutations, but the practice guidelines are the definition of what's recommended. Kurian, BTW, cites the NCCN's as one source of "before age 40". It says "ideally between 35 and 40 y, and upon completion of child bearing, or individualized based on the earliest age of onset of ovarian cancer in the family" (IIa). WhatamIdoing (talk) 00:24, 18 June 2011 (UTC)

The "earliest age of onset of *** cancer in the family" aspect should certainly be mentioned. Somehow I got the feeling that it is perhaps especially important for BRCA2. Btw the same age for PO for BRCA1/2 is is perhaps justifiable by the effect on BC risk in the absence of PM. Richiez (talk) 15:57, 18 June 2011 (UTC)

Could be, or could be simply that they've got to start making recommendations somewhere, and they'll refine it if/when further data suggests that's warranted.

Adding the family history aspect seems reasonable. We mention it under ==Screening==, but I don't think we mention it under ==Surgery==.

BTW, the preventive surgery section seems a little long. I'm going to subdivide it; if you don't like it, feel free to revert. WhatamIdoing (talk) 16:26, 18 June 2011 (UTC)

Can certainly use subdivision. Perhaps there should be a short section giving general overview of the various surgeries and separate BRCA1 and BRCA2 strategies subsections? Particularly for BRCA2 there is a bewildering variety of strategies like no PM, PM at 25,30,40 and/or PO at 40 or 50 which certainly would warrant some detail although it seems that for this we would have to rely almost solely on the data provided by Kurian.. I have noticed that the PM description is quite extensive, while there seems to be little knowledge if the various types of PM affect residual risk - so perhaps this could be moved to the separate article? Richiez (talk) 17:39, 18 June 2011 (UTC)

I think I would now make sections "BRCA1 prophylaxis strategies" and "BRCA2 prophylaxis strategies" after the overview in ==Surgery== and before the sections treating the particular surgeries? Richiez (talk) 20:20, 20 June 2011 (UTC)

At the moment, all those sections are about the same size, and I'm not sure that there's so much difference (mostly timing) between the two mutations. Dividing them might make for very short, very redundant sections. Perhaps we should hold off on this for a while. WhatamIdoing (talk) 01:06, 26 June 2011 (UTC)

Mammograms before 30

Are there any details on the risk analysis in BRCA risk groups before age 30? Is mircocalcification relevant at this age? Richiez (talk) 07:46, 9 June 2011 (UTC)

Everything I've seen is too vague to be called details.

Unless a family member developed cancer before age 40, then mammograms are not done on women in their 20s. Vague worries about radiation causing breast cancer in this group are apparently de rigueur for the sources. But I haven't seen any details, and this worry may be nothing more than data-free handwringing. WhatamIdoing (talk) 15:19, 9 June 2011 (UTC)

I certainly recall this: http://jnci.oxfordjournals.org/content/101/3/205.abstract Richiez (talk) 22:06, 9 June 2011 (UTC)

Whether and when

If someone ever decides to challenge the obvious statement that "Whether and when to perform preventive surgeries is a complex personal decision", it can be trivially supported by the Morris and Gold book. That is the main point of the whole book, and one short chapter addresses the issue directly. WhatamIdoing (talk) 16:00, 9 June 2011 (UTC)

I do not insist on "whether" but would feel better if the whole thing was not formulated like there is really no choice to be made. The choice must be made even though the odds are pretty clear. Richiez (talk) 22:13, 9 June 2011 (UTC)

I'm perfectly happy with "whether" and would also be happy to make sure that we're adequately conveying reality, i.e., that BRCA mutations are often discovered in older women, post-cancer diagnosis. Much of what we've been writing is a bit more focused on the people whose family members tested positive when they were young. Also, Morris and Gold would support a claim that a lot of male carriers don't choose to be tested at all (no estimates at proportions given, but clearly some grad student needs to run the study). WhatamIdoing (talk) 00:04, 10 June 2011 (UTC)

Parity

In this edit, you replace a secondary source that discusses multiple conflicting studies with a single, uncontrolled, retrospective primary study with weak results (e.g., the first data given has a one in 16 chance of the results being random, and the CI crosses 1.0, which means that there could be no effect at all).

By contrast, the most review, PMID 20975359, uses words like "complex", "limited", and "imprecise". The recent change conveys none of the uncertainty or limitations. WhatamIdoing (talk) 16:35, 9 June 2011 (UTC)

Might be best to leave out the whole paragraph, I have tried to change it because I considered the conclusion of Morris too speculative. If we stick with whats undisputed there is nothing left.

Out of curiosity, do you have some hypothesis explaining the date of birth effect shown eg here http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1180265/table/TB5/ Richiez (talk) 22:50, 9 June 2011 (UTC)

I'd like to get a copy of the 2010 review. In between now and then, we could shelve that subject, perhaps saying only that the studies are contradictory, but the effect is not expected to be as significant as it is for typical women. (I think everyone more or less agrees on that point?)

I can make up stories about the birth cohort effects. You would not wish to mistake any of them for expert speculation, much less an evidence-supported anything, but perhaps you'll find them entertaining:

• As the 20th century went on, women were less likely to die (young) of other things (like the Spanish flu, which would have killed a lot of the people in the pre-1920 birth cohort), so more of them lived long enough to develop cancer.
• Food insecurity during two world wars and the Great Depression—not to mention the general effects of widespread poverty—produced occasional nutritional issues. Cancer cells, being metabolically deranged, might be disproportionately susceptible to short-term nutritional stress.
• The much more physical nature of housekeeping back before microwaves, washing machines, and vacuums might have provided far more exercise than we gave them credit for.
• The social taboo against women drinking alcohol faded during the last century; increased alcohol consumption almost certainly causes some of the increased risk. (It'd be fun to run those numbers among Mormons in Utah, where alcohol consumption is very clearly associated with birth cohort.)
• Atmospheric nuclear testing and other irresponsible uses of radioactive stuff caused a mid-century spike in radiation, to which BRCA mutation carriers are uniquely susceptible.
• Greater screening in the late 20th century resulted in overdiagnosis of tumors that would probably have receded harmlessly.
• Too much time in front of old CRT televisions (which apparently emit radiation) caused the risk to rise.
• More women are aware that breast and/or ovarian cancer runs in their families, and thus more women get properly diagnosed, rather than just going into a decline or dying of "stomach problems".
• The mutations themselves are somehow getting worse, or combining with other factors to make the result worse. I seem to recall speculation about ovarian cancer appearing earlier in each generation because of telomere loss; what if you had telomere problems (or something similar) plus a BRCA mutation?

It would be interesting to (someday) split the "1960 onward" group into "1960–1979" and "1980 onward" (and so forth) groups, to see whether the risk really continues rising. Also, the BRCA2 risk rises rapidly, but not at the frightening pace that BRCA1 does, so it would be interesting to know what factors might have a different effect on the two genes. WhatamIdoing (talk) 00:23, 10 June 2011 (UTC)

PMID 14576434O has the breast cancer risk by age 50 for 2 birth cohorts which I consider an interesting data point. The physical exercise and obesity certainly appear significant but parity and OC use will also play a role. I would think that other developments - like use of alcohol, poor diagnosis or healthcare, poverty, early death from other cause may play a role but probably less so in this population than in the general population?

The article is complicated enough without it so I would leave it out for now. Richiez (talk) 14:31, 10 June 2011 (UTC)

Oh, and add "maybe more women got hysterectomies back in the day" to my list of stories. I have the impression that it was a routine surgery for a couple of decades in there. WhatamIdoing (talk) 00:13, 13 June 2011 (UTC)

Yes, also have this impression. Btw in the prophylaxis section I wrote strongly oversimplified that hysterectomy has no direct effect on BRCA related cancer risk which is arguable - just as tubal ligation it should have an albeit smaller effect on OC risk. However as far as I know this has not been really demonstrated for BRCA risk groups and is irrelevant because most of these women would do a PO at the same time. Not quite correct but the article is complicated enough without including speculation like this? Richiez (talk) 17:17, 17 June 2011 (UTC)

Dr Fishman very graciously sent me a pre-print copy of her recent review! The summary runs like this:

• Typical: Pregnancy associated with early-onset breast cancer, but protects significantly against breast cancer after age 40. Pregnancy protects against ovarian cancer. (Not mentioned, but verifiable elsewhere: early age at first pregnancy significantly protective.)
• BRCA1: Age at first birth irrelevant to breast cancer. "Modest protective effect" against breast cancer iff the woman has five or more births. Pregnancy protects against ovarian cancer (any pregnancies are protective, and more pregnancies are more protective). Breastfeeding is protective against breast cancer, but probably not against ovarian cancer.
• BRCA2: "Each additional pregnancy" produces a "statistically significant increase" in breast cancer. (I think that means that the first pregnancy is 'free'.) For ovarian cancer, the studies are "not definitive". One study shows pregnancy increasing ovarian cancer, and another had too few cases to tell anything. (Which makes me wonder: Are there a lot more people with BRCA1 mutations than BRCA2? I've seen several unrelated studies with too few BRCA2 carriers to draw conclusions.) Breastfeeding has no effect on either cancer.

In general, the review sticks to its conclusion that the current state of knowledge is "imprecise" and therefore declines to put a specific number on anything.

In the category of "even more things I didn't know", the introduction says, "The fertility issues are often considered by a young BRCA mutation carrier to be at least as important as the cancer risk, and seem to enhance and complicate the associated emotional load." So we might want to say that this is a particularly important issue to younger mutations carriers.

I'd guess that this is a multifactorial issue: Do I want to have children at all, knowing that each child has a 50-50 chance of getting this mutation from me? Will having children hurt me? Even if it has no harmful effect on me, is it responsible of me to have children if I might die before they're grown? And among people diagnosed at an older age, Did my decision to postpone childbearing or limit my family size contribute to my cancer (answer: probably not with BRCA1, and that choice was probably beneficial for someone with BRCA2)?

This pretty much lines up with the Morris and Gordon book, and since a review is better than picking out any single study, I think we should probably go with this—being very careful to not present it as the definitive, final answer. WhatamIdoing (talk) 00:13, 13 June 2011 (UTC)

Thanks for reporting, the pregnancy related BC risk is a really mind boggling relationship. I have the impression there is not much agreement how pregnancy protects in the non BRCA case so I am not that much surprised.

As of the mutation risk of children, preimplantaion or prenatal diagnostics may become common very soon and we are also speaking about children who will grow into a world that is completely different from what we see today. People must be optimists to have children in any case. Richiez (talk) 17:28, 17 June 2011 (UTC)

I'll write this up sometime. Perhaps I'll put it under ==Fertility issues==, rather than adding yet another factor to the cancer sections. WhatamIdoing (talk) 01:04, 26 June 2011 (UTC)

Ovarian cancer screening

This news story claims a JAMA paper says screening for ovarian cancer in average-risk women is more harm than benefit. Do you think we should include this piece of context, i.e., that these tests are not recommended for average-risk women, but they are recommended for women with BRCA mutations? Or is that too tangential to bother? WhatamIdoing (talk) 23:06, 15 June 2011 (UTC)

For BRCA1 I see at least 2 reasons to think that screening is much more beneficial than in general population but both are not or to a much lesser extent valid for BRCA2. But as long as we write that screening is recommended I think we should also mention the difference to general population. Richiez (talk) 14:42, 18 June 2011 (UTC)

OT, ref group parameter

Noticed that there is a group= parameter now, was there some discussion about using it medical articles? I have no idea what else it does but it looks quite nice to have author name instead of just a number in the note. Richiez (talk) 20:32, 19 June 2011 (UTC)

The group parameter has been around for years, although very few people have ever noticed it. It's used to divide "refs" into separate lists. The most common use is to have one "reflist" that contains citations to actual references, and another that contains explanatory footnotes.

If you want the author's names to be more prominent in the text, then WP:PAREN is the best option. WhatamIdoing (talk) 00:58, 26 June 2011 (UTC)

Odds ratio

Are we being accurate here?

• Tubal ligation is the least invasive of these surgeries and and appears to reduce ovarian cancer risk for BRCA1 carriers (but not BRCA2 carriers) by over 60%. (Pruthi)

Pruthi gets an odds ratio of .37, which I'm not sure is exactly the same as "reduces risk by over 60%". Their actual data is that women with BRCA1 mutations who had ovarian cancer were half as likely as women with the same mutation but no ovarian cancer to have gotten a tubal ligation.

I don't think that an odds ratio is exactly the same thing as a relative risk. Are we being misleading here, or is this probably about accurate? WhatamIdoing (talk) 23:58, 23 July 2011 (UTC)

I have been sloppy here but as far as I know it is widely assumed that there is a real protective effect of that magnitude. Richiez (talk) 21:15, 27 July 2011 (UTC)

Good, then I won't worry about it. WhatamIdoing (talk) 23:25, 27 July 2011 (UTC)

Radiation risk

Some news on that - http://www.bmj.com/content/345/bmj.e5660. Richiez (talk) 21:27, 9 September 2012 (UTC)

Screening changes

The net result of the recent changes under ===Screening==== seems to be mostly this:

Original

Current

Breast imaging studies usually include a mammogram once a year, beginning between ages 20 and 30, depending on the age at which any relatives were diagnosed with breast cancer.(ref name=Morris7 /> Breast MRIs (magnetic resonance imaging) are typically also performed once per year, six months after the mammogram. Breast MRIs use a contrast medium to enhance the image, and are far more expensive than mammograms. Mammograms, which are X-rays of the breast, expose the patient to some ionizing radiation, and BRCA mutations make the person somewhat more likely to develop radiation-induced cancers, so they may be used less frequently in younger carriers. However, the risk-benefit analysis shows that annual mammograms are appropriate. Both tests are commonly conducted because they each have strengths and weaknesses. Mammograms are better at identifying calcium microdeposits that are characteristic of some precancerous tissue changes. MRIs show more details in women with dense breast tissue, but the false positive rate (the tissue is actually healthy, but MRI suggests that cancer may be present) can be as high as 50%. Additionally, clinical breast exams and breast self-exams, which do not save lives in the general, asymptomatic population, are appropriate for women with BRCA mutations.(ref name=Morris7 />

Breast imaging studies usually include a Breast MRI once a year, beginning between ages 20 and 30, depending on the age at which any relatives were diagnosed with breast cancer. Mammograms are typically used only at advanced age as there is reason to believe that BRCA carriers are more susceptible to breast cancer induction by X-ray damage than general population.(ref name=Pijpe_2012>{cite pmid|22956590}}</ref>

Are we happy with this? It seems to me that we've removed a lot of basic information (use of contrast medium in MRIs, the strengths and weaknesses, breast self-exams, etc). WhatamIdoing (talk) 23:31, 20 January 2013 (UTC)

Tubal ligation

This changes "tubal ligation" to "salpingectomy". AFAICT, the (old) cited source doesn't mention salpingectomy at all, and the 60% statement is about tubal ligation. Does the new source claim 60% for salpingectomy? Should we mention tubal ligation anywhere? WhatamIdoing (talk) 23:36, 20 January 2013 (UTC)

Yes I think tubal ligation should be still mentioned with the old source, sorry I got distracted for longer time than expected trying to merge and reorganize mastitis related things. The new source suggest that salpingectomy might have a much bigger potential to reduce ovarian cancer risk than tubal ligation, however evidence is still limited. Salpingectomy for OC prevention is also a bigger surgery than tubal ligation so it may or may not be worth scheduling it inbetween the end of childbearing whenever that is and oophorectomy. Tubal ligation still has a possible use when the tradeoff seems infavorable for Salpingectomy. Richiez (talk) 11:46, 21 January 2013 (UTC)

The articles about women's health are generally in bad shape, so thanks for your work at Mastitis, too.

I've made a small change to restore tubals and still mention salpingectomy. WhatamIdoing (talk) 05:14, 27 January 2013 (UTC)

Five or ten

This:

No additional benefit is seen for more than six years' use, and hormonal contraception appears to carry a modest increased risk of causing breast cancer in BRCA mutation carriers.

was removed. Is this no longer accurate? I don't follow the literature here, but I thought that we'd concluded that this had some good data behind it last time we talked about it. WhatamIdoing (talk) 03:08, 21 January 2013 (UTC)

Last time we have looked at it the concern about increased breast cancer risk was the limiting factor, it is becoming less of a concern now. The benefit for ovarian cancer protection seems clear. No final data yet but slightly shifted tradeoffs. Also may depend on other factors - like if and when preventive mastectomy is planned. Richiez (talk) 12:00, 21 January 2013 (UTC)

So it sounds like this statement might be a bit UNDUE, given the most recent things you've read. I don't mind leaving it out. WhatamIdoing (talk) 05:16, 27 January 2013 (UTC)

Teens

I'm not sure why you're reverting these. This is not WP:SYNTHESIS; only one source is involved (SYNTH requires two or more) and the source actually says this is an issue for teens.

This is based on a reliable source and the information in it is not contradicted by your source. The meta-analysis is concerned with the lifetime likelihood of getting breast cancer, not the likelihood of getting breast cancer at the age of 18 or 20. I didn't see anything in it about teenagers. (Did you?) So there's no contradiction here: the sources are talking about separate issues. WhatamIdoing (talk) 00:37, 27 May 2013 (UTC)

I removed the following sentence and called it WP:SYNTHESIS:

Because taking birth control pills before age 20 is associated with very early development of breast cancer in BRCA mutation carriers, teenage girls may need to be tested before deciding to take hormonal contraceptives for any reason.^{[Morris & Gordon (Feb 23, 2010). Positive Results, pp. 275–302]}

because I could not find the cited source saying anything like: "teenage girls may need to be tested before deciding to take hormonal contraceptives for any reason."
The book discusses testing children on page 82 and discusses using oral contraceptives for ovarian cancer protection on pages 290–291, but does not combine these topics.

I removed the following sentences because they were out-of-date and gave medical advice (which is against Wikipedia policy):

However, the use of hormonal contraception during teenage years may cause unusually early development of breast cancer in women with BRCA mutations and is discouraged before age 20.^{[Morris & Gordon (Feb 23, 2010). Positive Results, pp. 275–302]}
The recommendation therefore is for women to take birth control pills for at least five years in their late 20s or early 30s, as the possible side effect of promoting breast cancer is believed to be lowest during these years.^{[Morris & Gordon (Feb 23, 2010). Positive Results, pp. 275–302]}

which cited:

• Morris & Gordon (Feb 23, 2010). Positive Results: Making the Best Decisions When You're at High Risk for Breast or Ovarian Cancer, pp. 290–291:

Ovarian Cancer Prevention
Oral Contraceptives
Studies show at most a modest increase in breast cancer risk in BRCA carriers with oral contraceptive use.^{[Narod SA (Sep 2006) PMID 16998497] which cites: [Narod SA, et al. (Dec 4, 2002) PMID 12464649]}
One study has shown that oral contraceptive use in the teenage years is linked to increased risk of early-onset breast cancer in BRCA-mutation carriers.^{[Jernström H, et al. (Oct 2005) PMID 16118051]} [a dubious claim, see below what study actually says]
Oral contraceptives may have less impact in women who are older, and the benefits in terms of decreased ovarian cancer risk are sufficiently significant that most doctors do not discourage oral contraceptive use in BRCA-positive women over the age of twenty; however, some recommend that women consider deferring oral contraceptive use until around age twenty-five to thirty when the risk of ovarian cancer begins to rise and when the drugs are less likely to increase breast cancer risk.
Dr. Gordon Recommends:
• Women considering oral contraception for birth control are encouraged to do so for a total of five years (use does not need to be continuous) because this will decrease ovarian cancer by 50 percent.
• Increased breast cancer risk with oral contraceptives use is modest at best and should not be a reason to avoid oral contraceptives.
• The ideal timing for conceptive use is still under study, so a discussion about risks and benefits should always be sought in teens and older women who are BRCA-mutation carriers.

Instead of Morris & Gordon's out-of-date Feb 23, 2010 book, I used the Aug 2010 meta-analysis of 18 studies of oral contraceptive use in 2,855 breast cancer cases and 1,503 ovarian cancer cases in women with BRCA1 or BRCA2 mutations:

• Iodice S, et al. (Aug 2010). "Oral contraceptive use and breast or ovarian cancer risk in BRCA1/2 carriers: a meta-analysis" Eur J Cancer 46 (12): 2275–2284. PMID 20537530

because it was the main reference cited by both the current National Cancer Institute PDQ article and the current UpToDate article:

• PDQ Cancer Genetics Editorial Board (Mar 4, 2013). "PDQ Genetics of breast and ovarian cancer (Health professional version)". National Cancer Institute.

• Isaacs C, et al. ( Apr 10, 2013). "Management of hereditary breast and ovarian cancer syndrome and patients with BRCA mutations". UpToDate.

The Narod, et al. 2002 study indirectly cited by Morris & Gordon's Feb 23, 2010 book is included in the Iodice Aug 2010 PMID 20537530 meta-analysis on breast cancer risk in BRCA1/2 carriers and contributed 1,311 (45.92%) of the 2,855 breast cancer cases in women with BRCA1/BRCA2 mutations. It was also 1 of 3 studies included in the Cibula (Aug 2011) PMID 21916573 case-control studies meta-analysis.

The Jernström, et al. 2005 study cited by Morris & Gordon's Feb 23, 2010 book is also only included in the Iodice Aug 2010 PMID 20537530 meta-analysis on the association between of OC use and BRCA1/2 mutation status in cancer patients but contributed only 19 (7.9%) of the 241 breast cancer cases with BRCA1/BRCA2 mutations, and its individual study results were not included in Figure 3. It was excluded from the Cibula (Aug 2011) PMID 21916573 meta-analysis because "no effect on breast cancer after COC use for carrier women or women with familial risk is published."

The Jernström, et al. 2005 study cited by Morris & Gordon's Feb 23, 2010 book is about 221 breast cancer cases diagnosed at age 40 or earlier in southern Sweden between 1990–1995, only 19 of whom had BRCA1/BRCA2 mutations—all of whom had used OCs, so they could not do a case-control analysis, but instead did a case-case analysis:

• 100% of breast cancer cases with BRCA1/BRCA2 mutations had used OCs vs. 84% of breast cancer cases without BRCA1/BRCA2 mutations
• a non-significant 81% (13 of 16) breast cancer cases with BRCA1/BRCA2 mutations had used OCs before age 20 vs. 58% of breast cancer cases without BRCA1/BRCA2 mutations
• 9% of breast cancer cases were BRCA1/BRCA2 mutation carriers; 12% of breast cancer cases who started OCs before age 20 were BRCA1/BRCA2 mutation carriers vs. 4% of breast cancer cases who started OCs after age 20; but
• there was no significant difference between breast cancer cases with or without BRCA1/BRCA2 mutations in:
  • total duration of OC use
  • duration of OC use prior to age 20
  • duration of OC use at age 20 or older
    

The Aug 2010 meta-analysis found:

• breast cancer relative risk was not significantly increased in BRCA1/BRCA2 mutation carriers who used OCs
• breast cancer relative risk was not significantly increased in BRCA1 mutation carriers who used OCs
• breast cancer relative risk was not significantly increased in BRCA2 mutation carriers who used OCs
• breast cancer relative risk was not significantly increased in BRCA1/BRCA2 mutation carriers who began using OCs before age 20
• breast cancer relative risk was not significantly increased in BRCA1/BRCA2 mutation carriers who began using OCs after age 20
• breast cancer relative risk was not significantly increased in BRCA1/BRCA2 mutation carriers who began using OCs after 1975
• breast cancer relative risk was significantly increased (1.17) in BRCA1/BRCA2 mutation carriers who began using OCs before 1975

All studies measured the relative risk of breast cancer or ovarian cancer in women who used oral contraceptives.

None of the studies were concerned with absolute risk or lifetime absolute risk.

None of the studies were about teenagers diagnosed with breast cancer between ages 18–20.

Any sentence beginning with the phrase: "The recommendation for women is therefore to ..." is giving medical advice and does not belong in a Wikipedia encyclopedia article.

BC07 (talk) 05:56, 27 May 2013 (UTC)

That's a lot of stuff to deal with, so let's start with the simplest: "The recommendation is (whatever)" is a statement of fact. It is not medical advice. "You should (whatever)" is medical advice. Consider Cancer#Recommendations and Influenza vaccine#Vaccination recommendations: we report the fact that these recommendations are made, without telling any individual person what to do. WhatamIdoing (talk) 22:08, 27 May 2013 (UTC)

Cancer#Recommendations lists old, outdated official cancer screening recommendations by the United States Preventive Services Task Force
  (e.g. old Jan 22, 2003 USPSTF recommendations for cervical cancer screening, which have been superseded by the current Mar 15, 2012 USPSTF recommendations for cervical cancer screening).

The following were WhatamIdoing's recommendations based on their interpretation of a tertiary source
  (a dated Feb 23, 2010 lay medical guidebook written by an attorney with a BRCA2 mutation and her medical geneticist):

However, the use of hormonal contraception during teenage years may cause unusually early development of breast cancer in women with BRCA mutations and is discouraged before age 20.
The recommendation therefore is for women to take birth control pills for at least five years in their late 20s or early 30s, as the possible side effect of promoting breast cancer is believed to be lowest during these years.

WhatamIdoing's recommendations were not the recommendations of any nationally or internationally recognized expert professional or governmental body,
nor were they the recommendations of the medical geneticist coauthor of the cited lay medical guidebook.

BC07 (talk) 01:22, 31 May 2013 (UTC)

Now that we've agreed that Wikipedia can report a recommendation, let's talk about the facts. You have asserted that hormonal contraception does not change the lifetime incidence of breast cancer. I'm happy to accept that, although PMID 20632911 (a review of exsctly the same date as yours) suggests otherwise.

I (based on these sources) am saying that hormonal contraception changes the age at diagnosis. That is, the point is to talk about the findings of PMID 16118051 which concludes "Each year of OC use prior to age 20 years conferred a significantly increased risk for early-onset breast cancer, while there was no risk associated with use after age 20 years."

Do you know of any source that specifically addresses the question of early-onset breast cancer (not overall risk)? WhatamIdoing (talk) 22:52, 7 June 2013 (UTC)

As mentioned above, I cited the Aug 2010 meta-analysis:

  • Iodice S, et al. (Aug 2010). "Oral contraceptive use and breast or ovarian cancer risk in BRCA1/2 carriers: a meta-analysis." Eur J Cancer 46 (12): 2275–2284. PMID 20537530

because it was the main reference cited by both the current NCI PDQ article and the current UpToDate article:

  • Cancer Genetics Editorial Board (Jun 7, 2013). "PDQ Genetics of breast and ovarian cancer (Health professional version)." National Cancer Institute.

  • Isaacs C, et al. (Apr 10, 2013). "Management of hereditary breast and ovarian cancer syndrome and patients with BRCA mutations." UpToDate.

Regarding the Aug 2010 review article you referred to:

  • Gadducci A, et al. (Aug 2010). "Gynaecologic challenging issues in the management of BRCA mutation carriers: oral contraceptives, prophylactic salpingo-oophorectomy and hormone replacement therapy." Gynecol Endocrinol. 6 (8): 568–577. PMID 20632911

Of the ten references cited in the "Oral contraceptive use and breast cancer risk — BRCA mutation carriers" section of the Aug 2010 review article:
◦ five were studies of OCs and breast cancer in BRCA mutation carriers (PMID 16118051, PMID 9288771, PMID 12464649, PMID 17021353, PMID 17635951) which were included in the Aug 2010 meta-analysis;
◦ two were studies of OCs and breast cancer in BRCA mutation carriers that found no increased risk of breast cancer in OC users (PMID 15734957, PMID 18990759) which were not included in the Aug 2010 meta-analysis;
◦ five were studies of OCs and breast cancer in BRCA mutation carriers included in one of the Iodice Aug 2010 PMID 20537530 breast cancer meta-analyses:
  ◦◦ Narod 2002 PMID 12464649 was 1 of 5 studies included in the Iodice Aug 2010 meta-analysis on breast cancer risk in BRCA1/2 carriers, and 1 of 3 studies included in the Cibula Aug 2011 PMID 21916573 case-control studies meta-analysis
  ◦◦ Haile 2006 PMID 17021353 was 1 of 5 studies included in the Iodice Aug 2010 meta-analysis on breast cancer risk in BRCA1/2 carriers, and 1 of 3 studies included in the Cibula Aug 2011 PMID 21916573 case-control studies meta-analysis
  ◦◦ Brohet 2007 PMID 17635851 was 1 of 5 studies included in the Iodice Aug 2010 meta-analysis on breast cancer risk in BRCA1/2 carriers. and 1 of 2 studies included in the Cibula Aug 2011 PMID 21916573 retrospective cohort studies meta-analysis
  ◦◦ Ursin 1997 PMID 9288771 was 1 of 5 studies included in the Iodice Aug 2010 meta-analysis on the association between of OC use and BRCA1/2 mutation status in breast cancer patients and its individual study results were included in Figure 3, and was 1 of 5 studies included in the Cibula Aug 2011 PMID 21916573 case-case studies meta-analysis
  ◦◦ Jernström 2005 PMID 16118051 was 1 of 5 studies included in the Iodice Aug 2010 meta-analysis on the association between of OC use and BRCA1/2 mutation status in breast cancer patients but its individual study results were not included in Figure 3, and the study was excluded from the Cibula Aug 2011 PMID 21916573 meta-analyses
◦ two were studies of OCs and breast cancer in BRCA mutation carriers that found no increased risk of breast cancer in OC users which were not included in the Iodice Aug 2010 PMID 20537530 meta-analyses, but were 2 of 5 studies included in the Cibula Aug 2011 PMID 21916573 case-case studies meta-analysis:
  ◦◦ Milne 2005 PMID 15734957
  ◦◦ Lee 2008 PMID 18990759
◦ one (Gaffield 2009 PMID 19751860) was a review that found no increased risk of breast cancer in OC users among women with a family history of breast cancer, but was not specifically about BRCA mutation carriers;
◦ one (Vessey 2006 PMID 16819539) was a study that found no increased risk of breast cancer in OC users in the general population, but was not about women with a family history of breast cancer nor about BRCA mutation carriers;
◦ one (Pasanisi 2009 PMID 19549808) was a case-case study comparing OC use in 382 breast cancer cases with a predicted ≥45% probability of being a BRCA mutation carrier to 1,333 breast cancer cases with a predicted <5% probability of being a BRCA mutation carrier, and was excluded from the Cibula Aug 2011 PMID 21916573 meta-analysis because "BRCA status was not analyzed."
The results of the Iodice Aug 2010 meta-analysis were not included in the Aug 2010 review article which was published the same month as the meta-analysis.

The last sentence of the abstract of:

  • Jernström H, et al. (Oct 2005). "Impact of teenage oral contraceptive use in a population-based series of early-onset breast cancer cases who have undergone BRCA mutation testing." Eur J Cancer 41 (15): 2312–2320. PMID 16118051

Each year of OC use prior to age 20 years conferred a significantly increased risk for early-onset breast cancer, while there was no risk associated with use after age 20 years.

refers to:

the association of OC use with breast cancer risk in 140 breast cancer cases diagnosed before age 41 between 1990 and 1995 in the South Swedish Health Care Region (Skåne, Kronoberg, and Blekinge counties) vs. 372 controls,
◦ per year of OC use prior to age 20 years, OR: 1.17, 95% CI: 1.03–1.33, P = 0.01)
◦ per year of OC use age 20 years or older, OR: 1.02, 95% CI: 0.98–1.07, P = 0.28)

not to women with BRCA mutations (the topic of this Wikipedia article).
As noted above, all 19 breast cancer cases in the study with BRCA mutations had used OCs, so a case-control analysis could not be done for the subset of breast cancer cases with BRCA mutations.

Many of the studies of OC use and breast cancer risk in women with BRCA mutations are of early-onset breast cancer, e.g.:

  • Milne RL, et al. (Feb 2005). "Oral contraceptive use and risk of early-onset breast cancer in carriers and noncarriers of BRCA1 and BRCA2 mutations." Cancer Epidemiol Biomarkers Prev. 14 (2): 350–356. PMID 15734957

This population-based study found that the use of high-dose formulations of oral contraceptives used before the mid 1970s was associated with increased risk of early-onset breast cancer. There was no evidence, however, that use of current low-dose oral contraceptive formulations was associated with increased risk of breast cancer either for carriers or noncarriers of mutations in BRCA1 or BRCA2.
We are aware of only two published studies on the relationship between oral contraceptive use and breast cancer risk for women known to carry a germ line mutation in BRCA1 or BRCA2. A study of 1,311 case-control pairs of living mutation carriers identified from multiple-case families (PMID 12464649) found a small increased breast cancer risk (OR, 1.2; 95% CI, 1.0–1.4; P = 0.03) for BRCA1 mutation carriers who had ever used oral contraceptives. Interestingly, this effect was most evident for cases diagnosed before 1980 (OR, 2.0; 95% CI, 1.2–3.4; P = 0.01), and for those who first used oral contraceptives before 1975 (OR, 1.4; 95% CI, 1.2–2.8; P < 0.001). More than 50% of the cases who used oral contraceptives had first used them before 1975, so although the effect for those who only used oral contraceptives since 1975 was not reported, it must have been close to unity... No effect was observed for BRCA2 mutation carriers.
In summary, this is the first, large, population-based study specifically addressing breast cancer risk in BRCA1 and BRCA2 mutation carriers and noncarriers. We found no evidence that use of current low-dose oral contraceptive formulations increases risk of early-onset breast cancer, regardless of BRCA1 or BRCA2 mutation status... Given that current formulations of oral contraceptives may reduce, or at least not exacerbate, risk of ovarian cancer for mutation carriers, our data suggest they should not be contraindicated for women with germ line mutations in BRCA1 or BRCA2.

BC07 (talk) 04:51, 11 June 2013 (UTC)

The review that I asked about says:

Each year of OC use prior to age 20 years conferred a significantly increased risk for early-onset breast cancer, while there was no risk associated with use after age 20 years.

Is it your opinion that this source believes this fact to be relevant for women with BRCA mutations? Notice that I'm asking what the source claims is relevant, not whether or not any Wikipedia editor happens to believe that the source is justified in making the claim. WhatamIdoing (talk) 02:16, 12 June 2013 (UTC)

The source you are quoting:

  • Jernström H, et al. (Oct 2005). "Impact of teenage oral contraceptive use in a population-based series of early-onset breast cancer cases who have undergone BRCA mutation testing." Eur J Cancer 41 (15): 2312–2320. PMID 16118051

is not a review, it is a population-based case-control study, i.e. a primary source.

I do not care what the authors of a single 8-year-old primary source believed to be relevant for women with BRCA mutations.

Per WP:MEDREV: Individual primary sources should not be cited or juxtaposed so as to "debunk" or contradict the conclusions of reliable secondary sources.

BC07 (talk) 03:57, 12 June 2013 (UTC)

I'm still trying to find a secondary source that contradicts the specific conclusion. You've provided a source that says women who are taking "current low-dose oral contraceptive formulations"—i.e., taking contraceptives at the time of their diagnosis, which (because "early onset is described as before age 40) is mostly in their 30s, has no effect. It's my impression that this is widely agreed: taking hormonal contraceptives in their 30s has little or no effect on breast cancer rates.

Where's the secondary source that says taking hormones twenty years before that, when they are 15 years old has no effect on the timing of breast cancer? I haven't seen any secondary source in your wall of text that even mentions taking hormones as teenagers. WhatamIdoing (talk) 05:34, 12 June 2013 (UTC)

You seem to be confused.

women who started use of OCs in the last 38 years after 1975 when current low-dose OCs replaced old high-dose OCs

is not the same as:

women who were currently using OCs at the time of their breast cancer diagnosis

"Table 3 – Summary risk estimates of the association between OC use and cancer risk in mutation carriers." on page 2280 of the "Results" section of the secondary source meta-analysis:

  • Iodice S, et al. (Aug 2010). "Oral contraceptive use and breast or ovarian cancer risk in BRCA1/2 carriers: a meta-analysis." Eur J Cancer 46 (12): 2275–2284. PMID 20537530

reports on breast cancer summary relative risk by age at start of OC use from 2,229 breast cancer cases with BRCA mutations from 3 studies:

• breast cancer summary relative risk was not significantly increased in BRCA1/BRCA2 mutation carriers who began using OCs before age 20 (SRR: 1.28, 95% CI: 0.91–1.79)
• breast cancer summary relative risk was not significantly increased in BRCA1/BRCA2 mutation carriers who began using OCs after age 20 (SRR: 1.24, 95% CI: 0.80–1.90)
• breast cancer summary relative risk was not significantly different in BRCA1/BRCA2 mutation carriers who began using OCs before age 20 versus after age 20 (P = 0.53, test homogeneity between strata)

These results were not further discussed and were not included in the abstract of the meta-analysis which ends with:

Conclusions: OC users carrying an ascertained BRCA1/2 mutation have a reduced risk of ovarian cancer, proportional to the duration of use. There is no evidence that recent OC formulations increase breast cancer risk in carriers.

So I'm fine with this BRCA mutation Wikipedia article omitting the age at start of OC use relative risk of breast cancer in BRCA mutation carrier results from Table 3 of the cited meta-analysis, as I did in my original edit:

A 2010 meta-analysis found that oral contraceptive use was associated with a 50% reduced relative risk of ovarian cancer (and a 36% relative risk reduction for each additional 10 years of OC use) in women with BRCA mutations, and that use of oral contraceptives available after 1975 was not associated with a significant increase in the relative risk of breast cancer in women who began use of OCs before or after age 20.

BC07 (talk) 23:55, 13 June 2013 (UTC)

Thank you for that explanation.

We have a source that says, in essence, if you take (modern) OCs from age 14 to 19, you are at somewhat higher risk of getting breast cancer sooner than you would have anyway, e.g., in your 20s. That is, the lifetime risk may be the same, but the risk in your 20s is elevated compared to a person who did not take OCs from age 14 to 19.

We have another source (yours) that says, in essence, if you take (modern) OCs from age 14 to 19, then your summary relative risk is approximately the same. That is, it agrees that your lifetime risk may be the same, and it says nothing at all about whether your breast cancer will appear at age 25 or at age 55.

Do you understand the difference between what these two sources say? Do you understand that these sources do not actually contradict each other? Do you understand that the second source does not actually say that the first source is at all wrong? WhatamIdoing (talk) 19:28, 17 June 2013 (UTC)

I understand that in the single primary source case-control study you cited:

  • Jernström H, et al. (Oct 2005). "Impact of teenage oral contraceptive use in a population-based series of early-onset breast cancer cases who have undergone BRCA mutation testing." Eur J Cancer 41 (15): 2312–2320. PMID 16118051

• all of the odds ratios (ORs) reported are for the general population—not the subject of this "BRCA mutation" Wikipedia article
• none of the odds ratios (ORs) reported are for BRCA mutation carriers—the subject of this "BRCA mutation" Wikipedia article

which is why the study was excluded from analysis in the two meta-analyses of breast cancer risk and ovarian cancer risk with oral contraceptive use in BRCA mutation carriers:

  • Iodice S, et al. (Aug 2010). "Oral contraceptive use and breast or ovarian cancer risk in BRCA1/2 carriers: a meta-analysis." Eur J Cancer 46 (12): 2275–2284. PMID 20537530

  • Cibula D, et al. (Aug 2011). "Oral contraceptives and risk of ovarian and breast cancers in BRCA mutation carriers: a meta-analysis." Expert Rev Anticancer Ther. 11 (8): 1197–1207. PMID 21916573

I understand that in the secondary source meta-analyses I cited:

  • Iodice S, et al. (Aug 2010). "Oral contraceptive use and breast or ovarian cancer risk in BRCA1/2 carriers: a meta-analysis." Eur J Cancer 46 (12): 2275–2284. PMID 20537530

• all of the relative risks (RRs) reported are for BRCA mutation carriers—the subject of this "BRCA mutation" Wikipedia article

  • Cibula D, et al. (Aug 2011). "Oral contraceptives and risk of ovarian and breast cancers in BRCA mutation carriers: a meta-analysis." Expert Rev Anticancer Ther. 11 (8): 1197–1207. PMID 21916573

• all of the effects (odds ratios (ORs), hazard ratios (HRs), interaction relative risks (RRs)) reported are for BRCA mutation carriers—the subject of this "BRCA mutation" Wikipedia article

I understand that there is not consensus for inclusion in this "BRCA mutation" Wikipedia article of results—that are not about BRCA mutation carriers—from a single primary source case-control study.

BC07 (talk) 21:47, 18 June 2013 (UTC)

I revised two of my earlier edits to clarify that:

  • Jernström H, et al. (Oct 2005). "Impact of teenage oral contraceptive use in a population-based series of early-onset breast cancer cases who have undergone BRCA mutation testing." Eur J Cancer 41 (15): 2312–2320. PMID 16118051

• was not one of the 5 studies with 2,855 cases included in the Iodice Aug 2010 PMID 20537530 meta-analysis on breast cancer risk in BRCA1/2 carriers (Section 4.1. Results. OC-associated breast cancer risk: Table 1, Table 3, Figure 2).
• was one of the 5 other studies with 241 cases included in the Iodice Aug 2010 PMID 20537530 meta-analysis on the association between of OC use and BRCA1/2 mutation status in breast cancer patients (Section 6. Association between OC use and mutation status in cancer patients: Table 2, Figure 3) and contributed 19 (7.9%) of 241 cases to the non-significant SRR 1.15 (95% CI: 0.69–1.93) association between BRCA1 and BRCA2 combined carrier status and oral contraceptive (OC) use in breast cancer patients, but was the only one of the 5 other studies whose individual study results were not included in Figure 3.
• was excluded from the Cibula Aug 2011 PMID 21916573 meta-analysis (Table 3. Papers excluded from review) because "no effect on breast cancer after COC use for carrier women or women with familial risk is published."

BC07 (talk) 16:22, 19 June 2013 (UTC)

Teen break

So are we agreed, then, that you have found no published source that directly says anything like, "Taking OCs from age 15 to 19 has absolutely no effect on the age at which a woman with a BRCA mutation will get breast cancer, e.g., whether she will get breast cancer in her 20s instead of in her 50s", right? Notice that a one-word answer of "yes" or "no" would be appropriate; I'm trying to get to the bottom line here rather than wade through another round of your opinion about why this should or shouldn't be so. WhatamIdoing (talk) 12:16, 20 June 2013 (UTC)

It is unnecessary for me to find a published source that says directly:

Taking OCs from 15 to 19 has absolutely no effect on the age at which a woman with a BRCA mutation will get breast cancer, e.g. whether she will get breast cancer in her 20s instead of in her 50s.

because you have not provided a current, reliable secondary source that says directly:

Taking OCs from 15 to 19 has an effect on the age at which a woman with a BRCA mutation will get breast cancer, e.g. whether she will get breast cancer in her 20s instead of her in her 50s.

therefore, the bottom line is: we will not include this statement in this "BRCA mutation" Wikipedia article. BC07 (talk) 14:11, 20 June 2013 (UTC)

The cited source says, "One study has shown that oral contraceptive use in the teenage years is linked to increased risk of early-onset breast cancer in BRCA-mutation carriers...most doctors do not discourage oral contraceptive use in BRCA-positive women over the age of twenty; however, some recommend that women consider deferring oral contraceptive use until around age twenty-five to thirty when the risk of ovarian cancer begins to rise and when the drugs are less likely to increase breast cancer risk."

The cited source is a WP:SECONDARY source and meets all of the requirements for a WP:Reliable source (see WP:NOTGOODSOURCE for the summary of requirements). It is three years old, which certainly counts as "current". Therefore we do have a "current, reliable, secondary source" that says this. WhatamIdoing (talk) 23:03, 23 June 2013 (UTC)

As explained above in detail, an outdated, non-peer-reviewed "secondary source"—a Feb 23, 2010 lay medical guidebook by a lawyer with a BRCA2 mutation and her medical geneticist (misstating the conclusion of a single 8-year-old case-study primary source)—is not an acceptable "secondary source" for the statements about oral contraceptives and breast cancer in BRCA carriers you wish to add to this BRCA mutation Wikipedia article, and there is no consensus to include them. BC07 (talk) 00:05, 24 June 2013 (UTC)

It isn't outdated: Wikipedia defines anything in the last five years, or for rare conditions (e.g., this one), the last ten years to be reasonable time period. In common sense terms, it's silly to claim that it's outdated because a paper that was submitted to the publishing journal (and how many others before it?) on 26 March 2010 is dramatically more current than a book published just 31 (thirty-one) days before. Furthermore, the most relevant guideline explicitly says not to fall into the trap of WP:RECENTISM by using only the single most recent source and ignoring all the ones immediately before it.

Finally, please note that we're not really talking about the evidence in this article; we're talking about what doctors actually do recommend. The doctors might be wrong, but they are actually making this recommendation, and nothing you've added here even mentions what real doctors are telling real patients. This is the difference between "should do" and "are doing". Doctors do not always follow the evidence. They still do knee surgery to shave cartilage despite excellent evidence that it doesn't work at all; they still tell surgery patients that not eating after surgery until bowel sounds return will make them recover faster, despite excellent evidence that the opposite happens; they still tell people that drinking a sip of water four hours before surgery will cause them to vomit under anesthesia, despite excellent evidence that the gut transit time of water in an empty stomach is about ten minutes.

Do look at the sentences we're talking about: "However, the use of hormonal contraception during teenage years... is discouraged before age 20. The recommendation therefore is for women to take birth control pills for at least five years" Think about the difference between "doctors actually recommend this" and "the evidence says this is warranted". We're only making the first claim, not the second. And the premier lay-oriented book with an actual doctor making these actual recommendations is a perfectly reliable source for supporting these statements about what actual doctors actually recommend. WhatamIdoing (talk) 09:30, 26 June 2013 (UTC)

Positive selection of BRCA mutations

I'm going to add something about this study:

http://www.ncbi.nlm.nih.gov/pubmed/10932184

Counteraction (talk) 12:58, 6 June 2013 (UTC)

Recommendations

We say:

A typical recommendation includes frequent breast cancer screening as well as tests to detect ovarian cancer.

I'm comfortable with this being true and verifiable, but what I'd like to know is how many formal recommendations in which an official body has issued a practice guideline or other formal statement on this point are out there. That is, rather than "A typical recommendation includes", could we construct statements like "The UK recommends annual MRIs for BRCA women age 30 to 49" based on this guideline, and could we do it for more countries than just the UK? WhatamIdoing (talk) 12:30, 20 June 2013 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on BRCA mutation. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20050820090304/http://www.jca.gr.jp:80/cs/95/11/866.pdf to http://www.jca.gr.jp/cs/95/11/866.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 12:21, 23 October 2016 (UTC)

Reqdiagram

I removed the reqdiagram as it seem to predate the current gene diagram being placed. Please replace it if there is more needed. Egmason (talk) 00:27, 4 December 2016 (UTC)

Link to brcaexchange.org

Please see Wikipedia_talk:WikiProject_Molecular_and_Cell_Biology#brcaexchange.org Jytdog (talk) 19:37, 11 July 2017 (UTC)

More profound

The lede contains the following sentence:

"Only 5-10% of breast cancer cases in women are attributed to BRCA1 and BRCA2 mutations (with BRCA1 mutations being slightly more common than BRCA2 mutations), but the impact on women with the gene mutation is more profound."

What does that mean? Does it mean that the impact of breast cancer is more profound if one has a damaged gene? The cited source doesn't say (and I'm not particularly impressed by a source that is a mainstream news report about a celebrity; mainstream news does a bad job of science reporting). MrDemeanour (talk) 14:31, 7 November 2018 (UTC)

Short answer: Yes, because the effect of BRCA mutations go far beyond breast cancer. It's the breast-and-ovarian-and-melanoma-and-colon-cancer-and-pancreatic-cancer mutation, not to mention years of medical tests and other costs, even if you never develop breast cancer. WhatamIdoing (talk) 05:24, 24 March 2020 (UTC)

Male breast cancer

It says:

"Unlike other men, men with a BRCA mutation, especially a BRCA2 mutation, may benefit from professional and self breast exams."

That seems to be saying that men without a mutation cannot benefit from breast exams.

Can someone with access to the source please verify that that is what it says? 50% of men who develop breast cancer do not have any mutation; surely early detection would be beneficial, even for them? MrDemeanour (talk) 14:35, 7 November 2018 (UTC)

Given that the costs of self-breast exams exceed the benefits in non-BRCA women, it seems extremely unlikely to me that they would be worthwhile for non-BRCA men. Also, self-exams aren't the only way to achieve early detection. Merely noticing that you have developed a weird lump, maybe the size of a grain of rice, on your chest, and then actually scheduling an exam with your doctor to talk about it (rather than shrugging your shoulders and ignoring it on the assumption that it will eventually go away) is often sufficient. WhatamIdoing (talk) 05:31, 24 March 2020 (UTC)

What is "the costs of a self-breast exam"? Self-examining takes no more than a minute. You don't need to do it more than once a month. I thought all grown women were encouraged to perform regular self-exams? And over a certain age, all women in this country are supposed to be called-in for a mammogram every year. Were you referring to the cost of a professional exam? But even that is pretty low-cost - it's the same as a self-exam, except done by a pro.

Men without a BRCA mutation get breast cancer. Take me, for example. I have no BRCA mutation. I had a single mastectomy, and I would be a fool not to regularly examine my remaining breast, given that it costs nothing.

Note that the breast tissue is covered by a layer of fat; a lump "the size of a grain of rice" would not be detectable in a self-exam. My early-stage cancer was actually detected in an X-ray that was done for another reason; when I was told about the shadow on the X-ray image, I checked, and found a lump a about the size of a hen's egg, along with an inverted nipple. I would not have been able to feel it if it had been the size of a grain of rice. I did not know the significance of the inverted nipple. I would not have noticed it had it not shown up on the X-ray.

Incidentally, I will also be called in for an annual mammogram; it's difficult for the mammogram nurse, because there's not enough breast on a man to be easily squeezed into the machine. MrDemeanour (talk) 11:22, 24 March 2020 (UTC)