Breast Cancer awareness
October 20, 2015 9:37 AM   Subscribe

The American Cancer Society released new guidelines today recommending that women start getting the tests later, at age 45, and only every other year. posted by roomthreeseventeen (17 comments total) 6 users marked this as a favorite
 
The fact that this study is based on film and not digital imaging means that I would want to know more before I splashed it around the headlines. (But I am old-fashioned like that.)
posted by wenestvedt at 9:53 AM on October 20, 2015


I am certainly biased as I'm currently in treatment for breast cancer at 38 years old, but I don't understand why mammograms get the bad rap when the real problem is overtreatment of low grade DCIS. Mammograms do still find cancer that needs to be treated. They also find DCIS that does not need to be treated. Instead of throwing out the recommendation for mammograms for younger women - who, if they do have invasive cancer, are more likely to have aggressive types that require aggressive treatment - why not focus more on educating both doctors and patients about what DCIS really means?

This part near the end of the article sure made me chuckle:
The National Cancer Institute's assessment tool shows that I'm at average risk for breast cancer, so I'd rather take the small chance that I'm missing an opportunity to avert a cancer death than face the larger risk that my life would be turned upside down by an unnecessary diagnosis.

I doubt she'd feel that way if she could have one glimpse of the terror of an actual high-grade breast cancer diagnosis. I will take those stressful, unnecessary tests over the risk of an early death any fucking day of the week.
posted by something something at 10:24 AM on October 20, 2015 [8 favorites]


I am certainly biased as I'm currently in treatment for breast cancer at 38 years old, but I don't understand why mammograms get the bad rap when the real problem is overtreatment of low grade DCIS. Mammograms do still find cancer that needs to be treated. They also find DCIS that does not need to be treated. Instead of throwing out the recommendation for mammograms for younger women - who, if they do have invasive cancer, are more likely to have aggressive types that require aggressive treatment - why not focus more on educating both doctors and patients about what DCIS really means?


I think the problem is the DCIS is only part of the problem. Fast moving, aggressive cancers that are going to kill someone no matter what are less likely to get caught on mammograms (unless we switch to screening every single woman every six months or so, which is not a good idea, because the harms from false positives will very quickly over take any benefits); slower moving, less aggressive cancers (including, but not limited to DCIS) are more likely to get caught on a mammogram.

I will take those stressful, unnecessary tests over the risk of an early death any fucking day of the week.

The thing is the tests are more than just stressful-- these have real, tangible side effects, that yes, can make somebody die in the end. If there enough false positives in a row-- which will happen, with a large enough sample size (like, say, all women over the age of 40), so this isn't a theoretical problem-- with increase screening, somebody will die during or after surgery for a mastectomy they didn't need, or a different type of cancer from the X-rays, and if the numbers don't work out right, there's a chance that we'll get more deaths from false positives than the number of lives saved through increased screenings (which, for mammograms are currently about 1 in a 1000; see here for more stats and discussion).

I had thyroid cancer, caught by basically me rubbing my neck and feeling the lump. I tell my friends to make sure they feel around their neck every now and then, but I sure as heck do not want to go to the South Korean system, where many, many women get neck ultrasounds to screen for thyroid cancer. The result? An "increased rate" in thyroid cancer there, with women getting thyroidectomies and sometimes, radioactive iodine treatments that they probably really didn't need-- their cancers were either never going to kill them, or weren't going to kill them before they felt the lump and got treatment (or probably were going to kill them no matter when they caught it-- anaplastic thyroid cancer is a fucker).
posted by damayanti at 11:11 AM on October 20, 2015 [6 favorites]


The key point that they keep glossing over is that we don't know which cancers could be safely left alone and which ones will threaten a woman's life. It's lovely to say that some cancers detected by mammogram would go away on their own, or would grow "like turtles" and never spread enough to cause a problem. But the reality is that we can't distinguish those from the "rabbits" that will kill you without prompt treatment. Doing a biopsy gives us a little better picture of the situation, but even then we don't know for sure how the cancer is going to act. Even DCIS, which is getting so much media attention these days with accusations of overtreatment, can sometimes turn deadly if ignored. Until somebody can accurately predict which abnormalities are safe to ignore, I don't want to ignore them.

I am a breast cancer nurse navigator, and my mom is a breast cancer survivor. Every day I debate what I would choose for myself and recommend for the women I love, informed by reading scientific articles and by the stories I see of the hundreds of women who come through my clinic. I would NEVER choose to forgo mammograms like the author of the Mother Jones article. I question the judgment of a science writer who reports that the three main organizations collating the scientific research are fiddling over the question of starting mammograms at age 40, 45, or 50, and then concludes that she should never have a mammogram.
posted by vytae at 11:19 AM on October 20, 2015 [3 favorites]


If there enough false positives in a row-- which will happen, with a large enough sample size (like, say, all women over the age of 40), so this isn't a theoretical problem-- with increase screening, somebody will die during or after surgery for a mastectomy they didn't need.

This is the problem, right here.

It's not so much what happens in any single instance. It's what happens when you have recommendations that millions and millions and millions of people ought to follow - what happens with the inevitable effects of this mass action.

Furthermore, Here is the abstract from the Canadian mammogram study that is important to this issue. For 25 years they followed 89,835 women; some received mammograms, and some did not. In the group of women who received annual mammography, there was not a lower rate of deaths due to breast cancer.
posted by entropone at 11:23 AM on October 20, 2015 [7 favorites]


vytae: The key point that they keep glossing over is that we don't know which cancers could be safely left alone and which ones will threaten a woman's life.

And if I may offer a gloss on this: some women are not comfortable with the slightest chance of cancer, and for them, even DCIS is too frightening to not treat or test for.
posted by wenestvedt at 1:27 PM on October 20, 2015


It's a choice between risks; miss a cancer you might be able to stop, or have your health degraded by unnecessary treatment/over-screening. It's going to stay that way till we get better diagnostic tools. We're at a sort of uphill part in medicine where we know a lot more than we did 50 years ago, but not nearly enough to fight cancer in the most effective ways.

You can lose either way; there is no clear-cut "safer" path for women who don't have any obvious genetic risks.
posted by emjaybee at 1:27 PM on October 20, 2015 [1 favorite]


I have a strong negative bias against mammograms, since they are so effective at diagnosing DCIS but were completely incompetent at finding a palpable lump that I, my doctor, and the technician could all feel perfectly well. Prior to my surgery for IDC, my breast surgeon suggested an MRI to be sure that we weren't all missing something and sure enough, there was another malignant lesion.

I get a mammogram every year because I'm the world's most compliant patient but I have already told my doctor that if I'm ever diagnosed with DCIS, I'm not going to have it treated. I find it horrifying that any woman with "pre-cancer" (if that's even accurate) should have to go through the shit I did, and that's having a relatively positive experience with cancer. I can't even begin to imagine what cancer treatment is like for women without my resources -- great insurance, great doctors, easy transportation, a great prognosis.
posted by janey47 at 1:28 PM on October 20, 2015 [1 favorite]


I'm not clear on the concern over false positives. If the mammogram comes back positive, isn't a biopsy performed prior to a mastectomy? This is what happened when my mom was first diagnosed in 1989 at the age of 40. In her case it was a true positive, but she was able to live another 26 years thanks to catching it relatively early and getting treated (it wasn't caught early when it returned, unfortunately).
posted by Thoughtcrime at 3:58 PM on October 20, 2015


Biopsies have risks and costs same as every other treatment. And biopsies can return false positives same as any other test.

This is how medicine works; find the recommended testing regimen which results in the greatest good for the greatest number of people, taking into account all risks, benefits, and costs. A lot of times we want to ignore the last part. But we can't. Every unnecessary procedure is a necessary procedure that doesn't happen.
posted by Justinian at 4:20 PM on October 20, 2015 [2 favorites]


Consider: We could save some people's lives by recommending full body MRIs every two years from the age of 20. Why don't we? Because that benefit is massively outweighed by the huge costs associated with such a program.
posted by Justinian at 4:21 PM on October 20, 2015


recommended testing regimen which results in the greatest good for the greatest number of people, taking into account all risks, benefits, and costs

This recommendation doesn't do that, though. The studies have only looked at mortality as an endpoint. They don't consider how much money and suffering we save by catching breast cancers early enough to avoid chemotherapy, early enough to do a lumpectomy instead of mastectomy, early enough to limit the number of lymph nodes that need to be removed. Maybe women have the same odds of survival with breast cancer caught early by mammogram vs. a bit later when they notice a lump, but what they have to endure to achieve those equal survival statistics is NOT the same.

We could save some people's lives by recommending full body MRIs every two years from the age of 20. Why don't we? Because that benefit is massively outweighed by the huge costs associated with such a program.

The other reason we don't do this is that we believe we would actually harm more people than we'd help. Incidental findings on full-body MRI would lead to unnecessary biopsies and surgeries, which carry their own risks of harming people. That's the same argument that's currently being made about breast screening with mammography for women in their 40s. I buy this argument when it's used against full-body scans, but not when it's used against mammography. Although no biopsy experience is pleasant, it's a heck of a lot easier and safer to biopsy a breast than a lung, a liver, a bone, or almost any other organ that might show an abnormality on a full-body scan.
posted by vytae at 6:59 PM on October 20, 2015 [4 favorites]


They don't consider how much money and suffering we save by catching breast cancers early enough to avoid chemotherapy

You also save money and suffering by avoiding unnecessary procedures. The mortality statistics are one piece of the puzzle and an important piece. I'm not implying that we shouldn't keep looking at the data and take into account money and suffering. But that cuts both ways; some people suffer unnecessarily from both early screening and late screening.
posted by Justinian at 8:16 PM on October 20, 2015


Harvard radiologist still recommends women get mammograms at age 40:
Q: What about the issues surrounding overdiagnosis and overtreatment of breast cancer — are they largely exaggerated or should women be worried?

The arguments suggesting massive overdiagnosis — that is, when a woman is diagnosed with breast cancer that would not become a threat during her lifetime — have been completely manufactured. There is no evidence that invasive breast cancers found by mammography, if left alone, would not go on to become lethal. One of the authors of a major article published in the New England Journal of Medicine that has promulgated this misinformation has stated that his analysis was based on his best guess. Subsequent analyses have shown that the study's conclusions were completely false.
posted by purpleclover at 4:38 PM on October 22, 2015 [2 favorites]


So, I have many thoughts about this, mostly about the linked Mother Jones story.

tl;dr: While every sentence is not false, in aggregate, I think it provides poor medical advice.

First and foremost, why must every story be debunking something? Have we really fallen so far down the Freakanomics hole that we can't read a news story or analysis that doesn't promise to show us that "EVERYTHING YOU KNOW IS WRONG!!!>!>!>1"? (Don't answer that. I fear the answer. But I would suggest that probably not everything you know is wrong.)

There are a bunch of these stories floating around right now, including, notably, the very similar Time cover story, Why Doctors Are Rethinking Breast-Cancer Treatment. (That one gets very low marks from me for the misleading cover line, "What if I do nothing?" The person in the story, did not do nothing. The person in the story gets alternating MRIs and mammograms every 6 months, and takes tamoxifen, which is most certainly not nothing.)

But back to our linked Mother Jones story:
Despite what her doctor said ("It's indicative of cancer"), the fact was that the abnormality on Taylor's mammogram—ductal carcinoma in situ, or DCIS— is not considered a cancer by many experts, and it had only a small chance of ever progressing into an invasive cancer.
A few things about DCIS first:
- DCIS is not a single and complete diagnosis. As something something said earlier, all breast cancer (at least the kind I have personally had) is also assigned a grade, 1 through 3. Grade 1 means that the cells look only somewhat abnormal and are probably growing slowly. Grade 3 means that it looks extremely different from normal cells. Getting a grade 1 diagnosis and a grade 3 diagnosis are quite different.
- We do not know how, when, or why DCIS breaks out of the milk duct and becomes invasive ductal carcinoma (IDC), the most common kind of breast cancer. That's because from the time when we could first find it on a mammogram, we cut it out. Surgery has been the traditional first-line treatment for breast cancer. I don't know that surgery will always be the first-line treatment for breast cancer, but just eschewing surgery seems to me like a peculiar, reactionary stance.
- It is true that there have been autopsy results that suggest that some percentage of women die with DCIS. That's been taken to mean that DCIS can exist in the body for a long time without progressing into IDC. One of the first lay descriptions of this was in Peggy Orenstein's 2013 NYTimes magazine story Our Feel-Good War on Breast Cancer (Previously). Orenstein goes on to say, with no source, that 50 to 80 percent of DCIS cases will not progress to IDC. I have no corroboration of her percentages, but everyone, even her, admits that no one knows which 20 to 50 percent will become IDC.
The probability that it would kill her was even slimmer, about 3 percent. The thing in her breast was not a ticking time bomb, and were it not for the mammogram, she probably never would have known it was there.
This is a big NOPE. The 3 percent figure, which comes from the study Breast Cancer Mortality After a Diagnosis of Ductal Carcinoma In Situ is for women who ARE TREATED FOR DCIS. They had a lumpectomy, lumpectomy with radiation, or a mastectomy. This Mother Jones story makes it sound like without treatment, this woman would have only had a 3 percent chance of being killed by breast cancer. That's flatly untrue.
... a 2012 study published in the New England Journal of Medicine calculated that over the last 30 years, mammograms have overdiagnosed 1.3 million women in the United States. Millions more women have experienced the anxiety and emotional turmoil of a second battery of tests to investigate what turned out to be a false alarm. Most of the 1.3 million women who were overdiagnosed received some kind of treatment—surgical procedures ranging from lumpectomies to double mastectomies, often with radiation and chemotherapy or hormonal therapy, too—for cancers never destined to bother them.
I feel like you can't mention this study without adding some context: H. Gilbert Welch, the second author of that study, is (in my opinion) a notorious crank, extremely unloved by oncologists who actually treat women with breast cancer, and not a fan of early detection of cancer. He's built his career around arguing that all kinds of cancer is overdiagnosed.

That said, hey, maybe they're right, and that in many women diagnosed with breast cancer would have died of something else first. (Median age of diagnosis is 61, after all, and the two biggest risk factors are: being a woman and getting older.) But this is very, very much not settled science. And the women whose screening is most a matter of debate (those age 40 to 50) are the ones who have the most years of life to lose.
In an Elle magazine story in June, Duke University breast surgeon Shelley Hwang described the "terrible feeling" that overcomes her every time she's asked to perform an elective double mastectomy on a woman with DCIS who "almost certainly" would have lived a long life without the procedure.
So, again, context. Shelley Hwang is one of the few breast surgeons in the country (the other one being her former colleague, Laura Esserman from UCSF) who are into watching and waiting on DCIS. This is also wildly taken out of context. In the Elle story, she was talking about how PATIENTS want bilateral mastectomies even when there's no medical indication for them. This is not something the breast cancer industrial complex is pushing on women.
... according to an analysis published in JAMA last December of just that many women, most will receive a false positive—6,130 women will get called back for more testing for something a doctor ultimately deems not to be cancer.
This is not awesome. No one likes the part where you get called back for another mammogram, and then for an ultrasound, and then, if the radiologist still doesn't like what she sees, a biopsy. However, all of this shit has happened to me personally (btw, it was all ultimately not as bad as actually having cancer), and for me, it all happened in one day. Much of the anxiety and stress of these "false positives" could be avoided by just staffing women's imaging centers better. More techs. More doctors to read the images. Stretching it out over a matter of days or weeks makes the callbacks worse. Doing it all at once? Not that bad. Abandoning screening mammography altogether? Nonsensical.
An analysis published in the New England Journal of Medicine in November 2012 calculated that most of the decline in breast cancer deaths over the last 30 years was due to the development of newer treatments such as tamoxifen and targeted chemotherapy and radiation. When they work, these treatments can even stop cancers that have become big enough to feel or are causing symptoms like pain, says Welch, one of the study's authors (which can explain those 173 women out of 10,000 who survive cancer regardless of whether they are screened). All of this suggests that early detection on a mammogram makes little difference in the outcome.
As vytae mentioned upthread, this assumes that surviving is the only important factor. I can assure you, surviving is great, but it's not the only thing. A lumpectomy is better than a mastectomy. Losing no lymph nodes is better than losing 5 or 10 or all of your lymph nodes, which can cause lifelong pain and swelling in your arm. Doing an "easy" course of chemo is way better than doing a terrible course of chemo. Just because you don't die doesn't mean you don't suffer.
Aggressive treatments like surgery have become alarmingly common for women diagnosed with DCIS—from 1998 to 2005, their rate of elective double mastectomy nearly tripled, and just more than 30 percent of women diagnosed with DCIS in 2012 opted for a mastectomy, even though the vast majority of them wouldn't ever have developed breast cancer.
First, I'm confused by the first part of this paragraph. Surgery is the only treatment for DCIS. (Esserman is doing neoadjuvant therapy with an antihormonal, and I'm interested in the results of her study, but an experimental treatment is not standard of care.)

Second, I had a mastectomy for DCIS, which is a decision I stand by. My lesion was too large for a lumpectomy without significant disfigurement of the breast. Also, having a mastectomy lets you avoid having adjuvant radiation, which I wanted, because my cancerous breast was on the left side, and I'm super young, so I need my heart to work for a long time in the future.

Third, we're back to the double mastectomy issue, which is, I repeat, a decision by individual patients, not anything recommended by modern surgeons. (An exception: If you have the BRCA mutations, your chance of developing breast cancer can be up to 80 percent. In that case, many surgeons will recommend prophylactic mastectomies. It's a pretty specialized situation.)

Here are some reasons women opt for bilateral mx:
- They don't want to ever have any more screening, and having been diagnosed with DCIS (or IDC) puts them in a high-risk category for developing a second primary cancer. Getting diagnosed with cancer (or pre-cancer) once is quite enough for them, thanks.
- They want to have reasonably symmetrical breasts; an implant wouldn't look anything like their remaining breast, and they're not candidates for autologous tissue transfer (for example: they're too thin.)
- They had highly estrogen- or progesterone-receptive DCIS, and they don't want to take tamoxifen because they're concerned about their fertility window.

I don't find these examples super-compelling for me, personally; I didn't have a bilateral mastectomy. But I'm not really willing to tell women who wanted one that they're wrong.
When celebrity chef Sandra Lee was diagnosed with DCIS via a screening mammogram earlier this year, she shared a video diary of the experience with People magazine. Lee does not have the BRCA genes that indicate a high risk for developing cancer, but she told Good Morning America's Robin Roberts that both her radiologist and her doctor told her, "You're a ticking time bomb." Lee opted for a double mastectomy. But a study published in JAMA Oncology in August of more than 100,000 women with DCIS found that their risk of dying of breast cancer was virtually identical to that of the average woman without DCIS or any signs of breast cancer.
Their odds of dying were the same AFTER THEY'D BEEN TREATED!!! Also, can we please not have Sandra Lee be our spokesperson? Thanks.
The National Cancer Institute's assessment tool shows that I'm at average risk for breast cancer, so I'd rather take the small chance that I'm missing an opportunity to avert a cancer death than face the larger risk that my life would be turned upside down by an unnecessary diagnosis. I have financial considerations, too. If a screening mammogram were to find something, it could send me down a path of more tests and possibly treatments that I don't want and would have trouble paying for.
We live in a weird world, where if the SCIENCE! arguments aren't as slam-dunk as people want, they go to the "well, I can't afford health care anyway" place. That is a problem. That is not a good reason to not find out if you have breast cancer.

I don't know whether screening mammography is a good idea for women 40–50 years, but the arguments against it (anxiety?) don't seem that great to me. Obviously, I have feelings about this, as a patient myself. My cancer was not discovered via a screening mammogram (I had a palpable thickening), and I'm quite young to have breast cancer (36), so I'm a relatively unusual case. Nobody would have watched or waited on my DCIS, because I'm under 40, the lesion was large, it was grade 3, and I had significant comedonecrosis (which means the cells were growing and dying so fast they left behind dead areas of cells). Everybody would have done surgery, and everybody would have been right, because in my case, the DCIS had, in fact, broken out of the milk duct and become HER2+, hormone-negative invasive ductal carcinoma (IDC). Cue up the 12 weeks of chemo I just finished.

It's not that unusual to find IDC with DCIS. The estimates I've heard are that they coexist 20–25 percent of the time with high-grade DCIS, and typically that's only discovered in pathology, after the tissue has been excised. So I'm not that hot on this currently popular storyline that DCIS is harmless and overdiagnosed (especially because the often-cited evidence for it being harmless is a study of women who received treatment.) And, again, there are, like, two surgeons in the US who feel that way: Laura Esserman and Shelley Hwang. Everybody else is still doing surgery.

In summary, this is a complicated issue, DCIS is complicated, and the journalism around it seems most interested in generating a buzzy headline that will turn! our! expectations! upside-down!
posted by purpleclover at 2:28 PM on October 26, 2015 [3 favorites]


Fantastic explanations, purpleclover. I agree with every last word, and I'm glad you took the time to write it out so clearly.
posted by vytae at 6:27 PM on October 27, 2015 [1 favorite]




« Older CreepyPosta   |   Most people have an inaccurate assessment of who... Newer »


This thread has been archived and is closed to new comments