Remember, mammography screening is, by definition, something we do to healthy women. And when you do something to healthy women, you don’t want to make them worse, so it has to be clear to everyone that you will benefit and not harm them. The benefit of screening should be that finding the thing you are looking for will result in fewer deaths.
The issue of whether mammography screening actually reduces deaths, particularly in younger women, has been debated for decades. There have been at least 7 prospective randomized clinical trials on the issue, and that is the gold standard of evidence. Those trials did not resolve the matter. Now less robust – and less clear – methods of evidence, known as statistical modeling, are being touted as a way to find benefit. These methods are complex, require a number of assumptions, and their reliability is often debated. Yet those in power want to take the outputs from these complex, far from clear, methods, tie them up in a pretty bow, and reduce them to a sound bite for all women aged 40 and over. History tells us that these recommendations will become the equivalent of a mandate and take away women’s right to information and their power to choose. Don’t women deserve better?
What the debate and the various models and the statistics show is that this is a complicated issue with no clear evidence of benefit but known harms. That should mean that women at average breast cancer risk should decide for themselves if, given the uncertainty, they want to be screened.
Background
Screening mammography for breast cancer in average-risk women with no symptoms is an extremely complex and controversial topic. Because it takes place in a healthy population, the National Breast Cancer Coalition (NBCC) has long held that the benefits of screening (e.g., reduction in mortality) MUST significantly outweigh the risks to justify the recommendations to women. That is why NBCC was disappointed to learn about the recent proposed change in the United States Preventive Services Task Force (USPSTF) recommendations for breast cancer screening.
On May 9, 2023, the USPSTF announced updated draft recommendations for breast cancer screening among women at average risk, making a blanket recommendation for biennial (every two years) screening in women ages 40 to 74 years. Under its prior recommendations, which NBCC reluctantly supported, the Task Force recommended biennial screening mammography for women ages 50 to 74 years, with the option for biennial screening among women ages 40 to 49 years following a conversation with their doctor about the risks and benefits of screening in this age group. Women got to choose. The reasons appropriately cited at that time were the known harms of screening and the totality of randomized controlled trial data that has shown limited benefit for all women and especially in this age group.
Why Now?
So what has changed to justify this new recommendation? Not the clinical trial data. No new experimental data has emerged regarding the benefits and harms of screening mammography. NBCC is now attempting to explain as best we can in lay terms, a very complex situation. The explanation is dense, we know. But you have to bear with it if you want to truly understand this debate and why NBCC takes the position it does on screening.
The USPSTF made this new recommendation based on a collaborative modeling analysis using the six CISNET breast cancer models (Cancer Intervention and Surveillance Modeling Network). NBCC has written about CISNET before.
These six statistical models, which rely on a number of assumptions, were used to independently estimate what breast cancer outcomes would look like in a hypothetical group of 1,000 average-risk 40-year-old females (all races and Black females alone) with and without breast cancer screening (using either digital mammography or digital breast tomosynthesis [DBT]). The models were employed to assess a variety of screening strategies (e.g., annual, biennial, or mixed intervals based on age, and different starting and stopping ages). The key benefit outcomes they considered were how many deaths would be averted, the percent reduction in breast cancer mortality, life years gained, and quality-adjusted life years. They also estimated the rate of known harms, such as the number of false positives, overdiagnosis, and the number of benign breast biopsies conducted unnecessarily.
Each of the six modeling approaches found different estimates for each outcome and harm considered. Yes, that’s right, the models did not agree. The median value across all of the models was used to provide the best estimates of benefit and harm in each scenario considered.
What’s Different Now?
Remember, there is no new clinical trial evidence. In 2016, using the same statistical modeling approach and the same six CISNET models, the USPSTF found that lowering the starting age for screening from 50 years to 40 years would result in 1.0 additional breast cancer death averted (going from 7 to 8 deaths averted) for every 1,000 women screened over a lifetime. As a result, screening in this age group (40-49 years) was given a “C” level recommendation, making the decision to start screening during this period, an individual one.
Now, in 2023, the collaborative model analysis found that for every 1,000 women screened over a lifetime, decreasing the starting age from 50 years to 40 years for biennial mammography would result in 1.3 more breast cancer deaths averted (all women; going from about 7 to 8.3). Note: this is 0.3 more breast cancer deaths averted compared to the 2016 recommendation. Looking specifically at Black women, the collaborative modeling analysis also found that 1.8 more breast cancer deaths would be averted for every 1,000 Black women screened over a lifetime. In terms of harms, model estimates include:
Based on the above, the USPSTF gave biennial screening of women 40 to 49 years of age, a grade “B” recommendation, meaning a practice that health care professionals should provide.
Harms of Screening
The harms associated with breast cancer screening are common and should not be underestimated. False positives are associated with psychological effects, such as anxiety and distress, that can last years. False positives will also lead to unnecessary biopsies. Although biopsies are relatively simple surgeries, they are still surgeries and can cause distress, scarring, disfigurement, and/or infectious complications. Another important harm of screening is overdiagnosis—and consequently, overtreatment which presents its own risks and harms. Detecting, removing, and treating breast cancers that would otherwise never have harmed a woman does not save any lives. But it does subject them to toxic treatments they never needed that could lead to significant life-long health issues, including other cancers. This is a serious concern that is often underestimated. Until researchers are able to determine which breast cancers will eventually spread, reduce quality of life, or result in death, overdiagnosis will continue to be a consequence of the current screening technologies.
It Isn’t a Perfect World
A KEY limitation of the modeling approach that underpins the proposed new recommendations is that all of the models assumed 100% screening adherence, prompt evaluation of abnormal screening results, and appropriate and timely access to treatment. They assume everyone gets screened, everyone gets timely and up-to-date treatment. Unfortunately, this is not how it works in the real world, so the above-cited modest potential benefits are a best-case, unlikely scenario and do likely underestimate the magnitude of harm.
In its draft statement and supporting documents, the USPSTF cited epidemiological data that shows the incidence rate (i.e., the number of new cases) of invasive breast cancer in women ages 40-49 has increased by 2.0% annually between 2015-2019. But one should know this increase is likely to be due, at least in part, to the fact that extensive screening already takes place among women in this age group. According to the Centers for Disease Control and Prevention (CDC), more than 60% of all women between the ages of 40 and 49 years in the United States, report use of screening mammography within the past 2 years. That would of course increase the number diagnosed.
Addressing Disparities
It is clear from the draft USPSTF report that a key impetus for changing the recommendation is to address the mortality gap between White and Black women. While the ultimate goal, of course, is not to have Black women die at the same rate as White women but to eliminate breast cancer mortality for all, we must figure out and work on this gap. We have long known that despite having a comparable incidence of breast cancer, death from breast cancer is 40% higher in Black women. Moreover, Black women are also twice as likely to develop the most aggressive form of breast cancer, Triple-Negative Breast Cancer. However, NBCC is not clear on why the USPSTF believes that lowering the age of starting breast cancer screening to 40 years of age will have any effect on the mortality gap, particularly given that, as noted above, ~60% of all women (both Black and White) in this age range are already being screened. The gap exists even though Black and White women are being screened at the same rate. Screening is not the reason for the mortality gap.
Screening with mammography will not eliminate the long-standing disparities that exist in breast cancer outcomes regardless of the start and stop age. They are the result of structural racism, and the health care policies that create inequitable access to appropriate, timely, and quality care.
Saving Lives
Diverting billions more dollars each year to an ineffective or at best, weakly effective intervention like screening that includes so much harm diverts those resources from addressing the tough questions, like how to prevent breast cancer or stop it from metastasizing (becoming lethal) or how to create an equitable health care system.
Mammography screening is not the answer to ending breast cancer and saving lives, and continued emphasis on it as a dominant strategy is misplaced.
