In the News: The new mammogram recommendations

Welcome to the first installment of what will hopefully be a regular installment spotlighting stories relating to breast cancer research. While I claim no expert knowledge regarding breast cancer or oncology, I am a trained biomedical scientist with a passion for research advocacy. As such, I wanted to use my unique point of view to highlight some advancements (or set backs) in breast cancer research over the coming months.

To start, I wanted to focus on the newly released guidelines for mammogram screenings. This story has been playing out in the national media over the last few weeks, and while much of the hype around it has quieted down, I think it is critically important that everyone understands the what the study itself  was designed to address and what the actual findings of the study are. 

To learn more about the United States Preventative Services Task Force, the mammogram effectiveness study and my thoughts, please 
About the United States Preventative Services Task Force

From the USPSTF webpage:
The U.S. Preventive Services Task Force (USPSTF), first convened by the U.S. Public Health Service in 1984, and since 1998 sponsored by the Agency for Healthcare Research and Quality (AHRQ), is the leading independent panel of private-sector experts in prevention and primary care. The USPSTF conducts rigorous, impartial assessments of the scientific evidence for the effectiveness of a broad range of clinical preventive services, including screening, counseling, and preventive medications. Its recommendations are considered the "gold standard" for clinical preventive services.  The mission of the USPSTF is to evaluate the benefits of individual services based on age, gender, and risk factors for disease; make recommendations about which preventive services should be incorporated routinely into primary medical care and for which populations; and identify a research agenda for clinical preventive care.
It is also important to recognize that (again from the USPSTF website)
"Recommendations issued by the USPSTF are intended for use in the primary care setting. The USPSTF recommendation statements present health care providers with information about the evidence behind each recommendation, allowing clinicians to make informed decisions about implementation."
That is, the recommendations of the USPSTF are not intended to direct individuals in making decisions about their personal treatments nor are they intended to direct what services should be covered by health insurance plans.  These recommendations are solely to help primary care physicians make more informed decisions when recommending treatments to their patients.

This USPSTF is made up of 17 individuals that are appointed by the head of the Agency of Healthcare Research and Quality (AHRQ) and vetted by Health and Human Services (HHS).  These individuals come from various areas of health care, including doctors, nurses and public health policy.  About half of the members are women and the majority of them come from an academic background.  Of these members, I think it is important to note that there are no oncologists, oncological surgeons, radiologists or Ob/Gyns.  There is one member, Dr. Kimberly Gregory, who is the Director of Maternal-Fetal Medicine and Women's Health Services Research at Cedars Sinai Medical Center, but otherwise, there are no other experts in women's health.  In addition to the members of the panel, there are also a number of partners that "contribute their expertise in the peer review of draft USPSTF documents and help disseminate the work of the USPSTF to their members." (USPSTF website).  While these groups represent a wide swath of the health services industry, it is again worth noting that there are no women's health, oncology or radiology groups represented.  Because of the potential implications of the findings of the USPSTF on health insurance policies, I also wanted to point out that both the Centers for Medicare and Medicaid as well as AHIP, the association of health insurers ARE USPSTF partners.  That is not to say that insurance companies are behind these findings, but I do think it is important to recognize who is actually involved in this process when thinking about the findings of the Task Force.

More about the USPSTF, including a list of all members and partners can be found at http://www.ahrq.gov/clinic/uspstfab.htm.

About the Screening for Breast Cancer Recommendations and Study

This particular study and the resulting Recommendation Statement were designed to update the original recommendations on breast cancer screening for women of "normal risk" that were released by the USPSTF in 2002.  The USPSTF identified a series of specific areas from the 2002 report that they felt warranted more detailed investigation.  Specifically, the USPSTF felt that the 2002 findings regarding film mammography for women between the ages of 50 and 70 were strongly supported, but that questions still remained for women ages 40-49 and 70+.  In addition, the USPSTF wanted to reexamine the evidence surrounding the efficacy of breast self exams (BSEs) and clinical breast exams (CBEs).  Finally, the USPSTF wanted to begin to examine digital mammography and MRI in lieu of film mammography for women of all ages.   For each screening method that was reviewed, the USPSTF was solely interested in examining the effect of that screening method on the breast cancer mortality rate.  This is an important point that I will be returning to later, so I wanted to make sure to emphasize it here.

After defining these questions, a study was commissioned to address these particular questions.  This study was performed by a group from the Oregon Health & Science University; Veterans Affairs Medical Center; and the Women and Children’s Health Research Center, Providence Health & Services in Portland, Oregon.  For this study, all of the relevant published literature was reviewed and the data from independent studies that met the inclusion criteria of the researchers were pooled together.  A meta-analysis of this data was then performed to best identify the trends across multiple, independent studies.  The results of this study were published in the November 17th edition of Annals of Internal Medicine (vol 151, issue 10).  

The members of USPSTF reviewed the findings of the study linked to above  in order to make their recommendations.  After weighing the benefits of the various screening methods for the age groups of interest against the identified harms that are associated with those same methods, the final  recommendations of the USPSTF were released in a Recommendation Statement on the HHS website.  Each recommendation was also graded using this classification system.  Their recommendations, copied from the above linked Recommendation Statement, are as follows:
  • The USPSTF recommends biennial screening mammography for women aged 50 to 74 years.
    Grade: B recommendation.
  • The decision to start regular, biennial screening mammography before the age of 50 years should be an individual one and take patient context into account, including the patient's values regarding specific benefits and harms.
    Grade: C recommendation.
  • The USPSTF concludes that the current evidence is insufficient to assess the additional benefits and harms of screening mammography in women 75 years or older.
    Grade: I Statement.
  • The USPSTF recommends against teaching breast self-examination (BSE).
    Grade: D recommendation.
  • The USPSTF concludes that the current evidence is insufficient to assess the additional benefits and harms of clinical breast examination (CBE) beyond screening mammography in women 40 years or older.
    Grade: I Statement.
  • The USPSTF concludes that the current evidence is insufficient to assess the additional benefits and harms of either digital mammography or magnetic resonance imaging (MRI) instead of film mammography as screening modalities for breast cancer.
    Grade: I Statement
A detailed explanation for each recommendation can be found in the Recommendation Statement here:
http://www.ahrq.gov/clinic/uspstf09/breastcancer/brcanrs.htm  I highly recommend that anyone that is interested in this issue (and if you're still reading this, I assume that you are) read the Recommendation Statement.  While the study on which these findings were based is fairly dense and written in "science-ese," I found the Recommendation Statement very straight forward and easy to follow.

Specific Points of Discussion

1. The Cost/Benefit Question
 The type of study on which the USPSTF based their recommendations is what is known as "comparative effectiveness research" or CER.  This type of research is focused on evaluating whether the benefits of a specific type of disease screening or treatment justify the costs of doing that screening or treatment in the first place.  CER has been pretty controversial over the years, in part because the relative values of both the benefits and the harms are highly subjective to the reviewer.  Personally, after reading this study, I found myself in strong disagreement with USPSTF almost entirely because I feel that they undervalued the advantages to breast cancer screening while simultaneously overvaluing the "harms" of screening.

So what are the advantages to breast cancer screening?  According to the USPSTF, the only advantage of breast cancer screening worth considering is a decrease in breast cancer mortality.  That is, saving lives that would otherwise be lost to breast cancer.  And the screening study did find that screening saves lives, stating that
"Our meta-analysis of mammography screening trials indicates breast cancer mortality benefit for all age groups from 39-69 years, with insufficient data for older women."    (my own emphasis added)
It is not under debate that breast cancer screening, and specifically mammography saves lives.  The question then is at what age does number of lives saved outweigh the risks associated with screening.  But before we get into the specifics of that, I wanted to take a minute to talk about the additional advantages to mammography and breast cancer screening in general that the USPSTF failed to consider.  Mammograms have been shown to extend one's life span when battling breast cancer, something that I personally value very highly.  In addition, when cancers are caught at earlier stages, the treatments required are much less invasive, take less time and are more affordable than treating a later stage cancer.  Without considering any of these very tangible advantages to mammography, I think this study falls incredibly short.

Now, what about those negatives?  In this study, the authors conclude that there are five major "harms" to be considered when evaluating the effectiveness of mammography and other screening methods, which I'm going to address one at a time.

1.  Radiation exposure during screening:  This is a very valid concern for many women and a legitimate potential harm of screening, to my mind.   In this study, the authors found that the radiation exposure during mammography is very low.  They felt that the ultimate harm of exposure was unclear but that some women may be at increased risk of developing breast cancer in response to these levels of radiation. The USPSTF noted that "radiation exposure, although a minor concern, is also a consideration."

2.  Pain during procedures:  A major concern for CER is the level of discomfort and pain caused by the test being evaluated.  If a test is overly painful to the patient, I think that would constitute a legitimate  argument against the widespread use of that test.  The screening study here found that "many women consider mammography painful but don't consider it a deterrent to further screening."  Moreover, the USPSTF only mentioned pain as a deterrent to mammography once and concluded that "it had little effect on mammography use."  As such, for the purposes of this study, the pain of mammograms was not considered a significant harm of mammography.

3.  Anxiety, distress and other psychological responses:  This is the issue where I really begin to disagree with both the authors of the study and especially the USPSTF.  One of the major "harms" that was considered in evaluating mammography effectiveness was that in the event that something was detected on a mammogram that warranted some form of follow-up, the women suffered from anxiety and breast cancer worry.  Personally, I would gladly endure a bought of nerves in order to know for sure what my questionable lump or lesion was.  This is doubly true for BSEs.  If I find something abnormal in my breast, I want to know what it is.  Trying to protect me from being concerned for my own health is a futile exercise.  If I have a bad headache, a strange freckle or a sore leg, I'm worried about what it might be and I highly doubt I'm the only one.  This idea that women are too fragile psychologically to potentially get bad news is ludicrous.  That this was even a concern of the effectiveness study really bothers me.  Interestingly, the actual conclusion of the authors of the study was that "patient adverse experiences like anxiety are common but seem to be transient and do not adversely influence future screening practices." The USPSTF, on the other hand, seems to consider anxiety and distress to be a much more serious concern and reference it repeatedly.  My personal feeling is that this harm is seriously over-valued by the USPSTF.

4.  False positives/negatives:  The main "harm" here is that too many mammograms produce unclear results, leading to excessive follow up imaging and/or unnecessary biopsies to determine the true nature of the breast abnormality.  While mammograms may result in higher than ideal rates of false positives, the lack of a better test makes this "harm" less of a concern, at least to me.  An imperfect test is better than no test at all.  False negatives, while still more common than would be ideal, are less of a concern in regards to breast cancer screening.  Based on all of this, I consider this harm to be over-valued by the USPSTF as well.

5.  Overtreatment:  Overtreatment is defined as the treatment of a disease that would not become a risk to one's life during their life expectancy.  This is a major concern of both the study authors and the USPSTF.  Given what is currently known about breast cancer progression, I disagree with the authors and the task force about the current magnitude of this problem, as I will discuss in detail below.  As a result, I consider this harm to also be over-valued by the USPSTF. 

In conclusion, I feel the incredibly narrow focus on the effects of breast cancer on mortality rate  to the exclusion of other additional benefits coupled with at least three significantly overvalued risks has seriously distorted the findings of the authors of the study and especially the USPSTF.  Many other individuals may disagree with me.  Unfortunately, that is the nature of CER - it is based on subjective values.

2. High False Positives?
As I mentioned in the previous section, one of the significant limitations of breast cancer screening is the high false positive rate.  This is often the case with tests used to routinely screen otherwise healthy people.  This is especially true for tests that are self-administered, like BSEs.  In the specific case of breast cancer, there are a number of lesions or masses that frequently occur in breast tissue, most of which will turn out to be benign and/or non-cancerous.  However, when you are checking yourself or even on a mammogram, it is impossible to know what those abnormal lumps and bumps really are.  I think there is a common misconception that a mammogram is a diagnostic test.  It is not.  Breast cancer is should never be diagnosed from a screening mammogram.  Instead, a doctor will order additional screening using more specific and/or more advanced techniques or a biopsy in order to properly identify the mass.  Only after the actual cells of the mass are examined can a true diagnosis of breast cancer be given.

In the USPSTF-commissioned study, any time a mass was detected either in BSE or in routine mammography, the screening test was classified as "positive".  However, as I described above, the masses detected in a large number of "positive mammograms" are found to be benign, non-cancerous  lesions with follow-up imaging or biopsy.  In this study, all of those tests, then, fall under the category of "false positives".  So you can see now why mammograms and especially BSEs have such a high false positive rate.  In the opinion of the USPSTF, any follow-up test that is administered that doesn't detect cancer is considered a "wasted" test.  That would be like considering an x-ray that reveals a sprain instead of a fracture a waste, as Nancy Brinker wrote in this wonderful USA Today column.  Why then, does the USPSTF consider any follow up test that doesn't detect cancer as wasteful??   
    In an ideal world, the screening test that we use not just for breast cancer, but for any disease, would also be a diagnostic tool, telling us exactly which lumps and bumps are cancer and which aren't.  But right now, that just isn't the case.  To my mind, the only time a recommendation against a certain screening test based on a high false positive rate is warranted is if there's a better test out there.  Right now, that isn't the case for breast cancer.  Rather than recommending that women forgo the flawed but effective test that we currently have, we should be advocating for better screening methods.  But until we get those tests, having to perform too many "wasteful" follow-up procedures is a small price to pay for saving someone's life.

    3. Overdiagnosis and overtreatment?
    Another area of concern for the USPSTF and the authors of the study on which the Task Force's recommendations are based is overdiagnosis and overtreatment.  Overdiagnosis is defined in the Task Force's Recommendation Statement as "detection of cancer that would never have become clinically apparent... and it is usually followed by overtreatment."  In other words, overtreatment is unnecessarily treating a cancer that would not have killed the woman during the normal span of her life.  This is particularly an issue for women ages 75 and up because of their relatively shortened remaining life span.

    To better understand why this is an issue, we must think for a moment about the actual biology of cancer.  Oftentimes, the words tumor and cancer are used interchangeably to refer to an abnormal mass of cells that can leave the site of the mass and spread to other parts of the body.  However, they are not exactly the same thing.  A tumor is a mass of cells that may or may not be malignant (multiplying and/or migrating to other parts of the body).  Cancer, on the other hand, refers only to malignant tumors, not benign masses.  While most people believe that all tumors will become cancerous if given enough time, that isn't usually the case.  In actuality, the nature of a tumor (benign or malignant) is often determined when the cells first become abnormal.  Not all cells undergoing what we commonly think of as "cancerous growth" have the ability to migrate to other parts of the body and establish new tumors.  Many "cancerous cells" are dividing very slowly and are limited to the immediate surrounding tissue.  Now, if one of these slow growing, non-migrating tumors happens to be in or on a critically important organ, they can also wreak havoc on your health without ever becoming "malignant cancer".  But more often than not, if the cells stay where they are and reproduce relatively slowly, then you could live with that tumor for a very long time without any impact on your day to day life.

    The most commonly diagnosed form of breast cancer is what is known as ductal carcinoma in situ or DCIS.  This type of breast tumor is confined to the milk duct within the breast.  By it's very nature, DCIS is not considered malignant because it is still confined to it's site of origin.  A diagnosis of DCIS is often at a very early stage and is rarely a precursor for more advanced, invasive breast cancer.  Because of this, some oncologists and breast cancer specialists consider DCIS to be "pre-cancer", not breast cancer.  Because DCIS is often has a long asymptomatic phase, it is almost always initially detected on a routine screening mammogram.  As a result, with the increase in routine screening, there has been a concomitant increase in the incidence of DCIS without a clear understanding of the long term nature of this type of tumor.  The other primary type of breast cancer is lobular carcinoma in situ or LCIS.  Unlike DCIS, LCIS often leads to an increased risk of developing later, invasive breast cancer.  Whether those tumors themselves become malignant or whether they just signal that other, more aggressive cancer cells may be present in the breast is still up for debate.

    As I hope you are starting realize, the sticky point to all of this is that in the end, we do not know what causes a breast tumor or cancerous breast tissue cells to become invasive and malignant.  Let me say that again - we do not know yet what causes a breast tumor or cancerous breast tissue cells to become malignant.  We don't know whether cells are destined to be malignant from the moment they become abnormal or if a benign tumor can evolve into a malignant tumor with time.  This is an important point for me because it represents a serious diversion between myself and the USPSTF.  As the USPSTF said in it's Recommendation Statement:
    "Because the likelihood that DCIS will progress to invasive cancer is unknown, surgical removal—with or without adjuvant treatment—may represent overdiagnosis or overtreatment."
    In essence, what the USPSTF is saying is that because we don't know for sure that DCIS will become a malignant cancer that could kill you, we shouldn't worry about treating it because we don't want to expose women to unnecessary surgery or treatments.  While I recognize that only treating the tumors that we know for sure will become life threatening is the ideal, the fact of the matter is that we have no idea of know which tumors those are.  I know that for me, if I had a tumor that even had the slightest chance of becoming life threatening, I would want to treat it as soon as possible, when only a lumpectomy would be required rather than wait and see and ultimately end up having to go through a much more radical mastectomy and intense rounds of chemotherapy and/or radiation.  I suspect that the most women would agree with me, putting us at odds with USPSTF.  Again, while their recommendations are sound based on their values, they are out of line with the values of regular women.

    4. So how did they decide to start recommending screening at age 50?
    I mentioned at the beginning of this article that one of the primary questions that the USPSTF wanted to address was the effectiveness of breast cancer screening for women ages 40-49.  They felt confident in their previous recommendations for women ages 50-69, but felt that the evidence thus far for women in their 40s was less clear.  To address this question, the authors of the study commissioned by the USPSTF stratified the results of the studies they reviewed in order to compare the effectiveness of breast cancer screening for each group.  The primary results that they reported in the Annals of Internal Medicine were:

    That is, for women aged 39-49 years old, mammography screening resulted in a 15% reduction in breast cancer mortality, with screening of 1904 women needed in order to save one life.  Interestingly, for women aged 50-59, mammography screening resulted in a comparable 14% reduction in breast cancer mortality compared to a control group.  Because death from breast cancer is more frequent in this age group, that means that 1339 women need to be screened in order to save one life.  Finally, for women aged 60-69, mammography screening resulted in a 32% reduction in breast cancer mortality and only 377 women needed to be screened in order to save one life.

    Based on these results, it is clear that a large jump in the effectiveness of regular mammograms in reducing breast cancer mortality actually occurs at age 60, not at age 50.  In fact, mammograms are just as effective for women in their 40s as for women in their 50s at reducing breast cancer mortality, even though breast cancer in those women occurs less frequently.  Herein is what I consider to be one of the major design flaws of this study.  Just looking at this data, one might logically conclude that the most important age to start mammograms is actually at age 60, not age 40 or 50.  Without a lower age group to compare these results to, I don't think you can properly assess the effectiveness of mammography for women in their 40s.  Let's say that for women ages 30-39, regular mammography also results in a 15% reduction in breast cancer mortality and that the number of women that need to be screened to save one life is about 2500.  Those findings probably wouldn't change your conclusions much: there is a moderate benefit that increases with each decade of life and that the big jump in effectiveness occurs around age 60.  But what if the results for women in their 30s were a little different?  What if for women in their 30s, regular mammography only results in a 5% reduction in breast cancer mortality and that a whopping 5000 women needed to be screened to save one life?  If that were the case, it would be incredibly clear that there is a boost in the effectiveness of mammography at reducing breast cancer mortality around age 40 and again at age 60.  I doubt anyone would look at that data and conclude that it was okay to wait to start screening when a significant effect at age 40 is so obvious.

    Now, I don't know what the numbers look like for women in 30s.  I wish that I did.  But more than that, I wish that the observers knew those numbers.  In order to properly assess the effectiveness of routine mammography for women in their 40s, you can't just look at the next older group and determine that screening is slightly less effective.  You have to compare to the younger group as well.  As a scientist, I think this is a serious flaw in this study.

    We know from the recommendation statement that the USPSTF looked at these results and (I feel, somewhat arbitrarily) decided that the benefits of routine mammography began to outweigh the risks starting at age 50 for most healthy women.  But is that really the right answer?  Personally, I disagree.  In fact, I think the authors of the commissioned study summed up their results the best:
    "Our meta-analysis of mammography screening trials indicates breast cancer mortality benefit for all age groups from 39 to 69 years, with insufficient data for older women. ...  Mammography screening at any age is a trade off of a continuum of benefits and harms.  The ages at which the trade off becomes acceptable to individuals and society are not clearly resolved by the available evidence."
    To conclude, I think that this was an important study to perform and I hope we continue to assess the effectiveness of our screening techniques well into the future, especially as newer, hopefully better techniques emerge.  But I found that my personal values were out of line with the USPSTF and as such, I do not agree with their ultimate conclusions.  I think the one thing highlighted by this study that we can all agree on is that we need better and more accurate tests, less invasive treatments and a better basic understanding of the nature of breast cancer.  If we can use this report as a call to action to increase breast cancer research, I think we will all be better off.

    The Counter Argument

    What I have written here is solely my opinions that I formed after reading extensively about breast cancer screening, the USPSTF and the study that was published in the Annals of Internal Medicine.  There are certainly people out there who disagree with me and in the interest of fairness, I wanted to provide links to two articles discussing the argument in support of the findings of the USPSTF.  The first article that I recommend was written by Devra Davis, PhD and can be found here (click for article), at the Huffington Post. Dr. Davis writes very passionately about her own experience with breast cancer screening in the context of the history of mammography.  It is a very interesting article, even if I don't necessarily agree with her.  The second article that I would recommend was written by Kevin Sack and was published here (click for article), in the Health section of the New York Times.  Mr. Sack focuses on the emerging idea that early detection isn't always the best and the harsh opposition that that point of view has faced.

    Useful Links and Resources

    In writing this piece, I used a number of resources and I tried to directly credit those sources where I could.  I repeatedly referenced both the USPSTF Recommendation Statement and the Annals of Internal Medicine study.  The links for those documents are:

    The USPSTF Recommendation Statement on Breast Cancer Screening
    Screening for Breast Cancer: An Update for the USPSTF, published in the Annals of Internal Medicine

    I gathered much of my information on the USPSTF from their personal webpage (click for link) and from this highly informative article in the New York Times.

    Following the release of the USPSTF recommendations, almost all of the major cancer groups released formal statements, which I have collected here (click the names for the links):

    American Cancer Society
    Komen for the Cure
    Nancy Brinker, Founder and CEO of Komen for the Cure
    National Breast Cancer Foundation
    Breast Cancer Network of Strength (formerly Y-Me)
    National Cancer Institute
    National Breast Cancer Coalition
    Stand Up 2 Cancer Coalition

    I also want to call attention to the official statement of the Secretary of Health and Human Services, Kathleen Sebelius, which can be found in this article (click for link) on CNN.

    While I was preparing this article, Nancy Brinker published an Op Ed in USA Today that I found very interesting.  It can be found here (click for article) if you'd like to read it.

    Finally, in learning about breast cancer, I have found Dr. Susan Love's Breast Book to be a great resource.  I highly recommend it to anyone who wants to learn more about the science of breast cancer and the history of breast cancer screening and treatments.  Dr. Love updates her book every five years, and the fifth edition is due in 2010.

    Thank you to anyone who actually made it to the end of this piece! It was really important to me to use this platform to share my thoughts and insights with all of you.  I encourage everyone and anyone to leave me comments letting me know your thoughts as well!

    1 comment:

    Linda Velott said...

    You continue to amaze me Kristen! Thank you, I think you make great points!