And Why Women Need to Be Included in the Screening Discussion

The never-ending controversy regarding mammographic screening for breast cancer is once again in the news. The study “Swedish Two-County Trial: Impact of Mammographic Screening on Breast Cancer Mortality During 3 Decades” published last month in the journal Radiology, is the longest trial of mammographic screening follow-up. It showed convincing evidence that the longer the screened population is evaluated the greater the decrease in number of deaths due to breast cancer. This resultant long-term impact on patient survival was much larger than had been initially estimated. The study, which followed patients for 29 years demonstrated overall reduction in cancer death rate by 31 percent. However since only 70 percent of eligible women who were in the screening portion of the study actually were screened, the actual reduction in deaths would have been 40 percent if all women presented for screening. More important, to those who question the number of women needed to be screened per “life saved” the Swedish study lowers that number to between 414 to 519 women vs. the 1,000 to 1,500 that is projected or presumed by prior studies.

Also intriguing and powerful was the unexpected result that while women were screened for only seven years the beneficial effect of finding their tumors earlier persisted for many years later as far fewer women died of breast cancer even after 29 years than those who were not screened at all. In fact most of the deaths averted would occur more than 10 years after the screening period and therefore the long-term benefit of screening is more than double the benefit reported in shorter studies. Countering this argument are those who would say that recent advances in chemotherapy, during the study period, has leveled the playing field such that those with larger cancers (found in women who are not screened) would do as well as those with smaller (screening detected). However this is  not substantiated by evidence-based science and there is little doubt that over all women with smaller, lower stage tumors do better than those with larger higher stage tumors. Further women with smaller cancers have better treatment options, often lumpectomy and less toxic chemotherapy than those with larger tumors who more likely have mastectomy and more aggressive treatment.

With screening the potential harm of over-diagnosis and false positive mammograms is real but its impact has not been quantified and its use as an argument against screening is specious to those women who choose screening. There is little doubt that if asked women would choose to chance these negative aspects associated with screening if they were given the opportunity to find their tumors earlier with routine mammograms. Not that women have been asked.

There is no denying that regular and consistent mammographic screening saves lives. Even the controversial U.S. Preventive Services Task Force (USPSTF) recommendation statement is preceded by its evidence update article, which analyzed screening mammography trials and agreed that there is a 15 percent reduction in breast cancer deaths in young  women ages 39 to 49 regularly screened by mammography, i.e. 15 percent of breast cancer deaths would be avoided if their cancers are detected by screening mammography even in this young age group. In older women, 50 to 69 years of age the death rate decreases to a much greater degree, to about 30 percent. Yet, using mathematical models weighed benefits vs. potential harms and concluded that there was no need to screen younger women at all or any need to screen older women annually because doing so was “inefficient.” How they were able to recommend that women ages 40 to 49 abandon, or make up their own minds, as to whether to participate in annual screening mammography, which would significantly decrease all screen-detected breast cancer deaths is at best difficult to understand. While denying that economic considerations were utilized the USPSTF drew a subjective line in the sand, a non-Solomonic assessment of what life was worth. Not that women were asked.

More interesting, analyzing the same USPSTF modeling data that was used to recommend essentially abandoning screening among women 40 to 49 and offering bi-annual, instead of annual screening to women 50 to 69,  the authors of a recent scientific analysis titled “United States Preventive Services Task Force Screening Mammography Recommendations: Science Ignored” published in the American Journal of Radiology, February 2011, found very different results.  They reported that the same data indicated that “screening mammography shows greatest benefit — a 39.6 percent mortality reduction — from annual screening of women 40 to 84 years old. This screening regimen saves 71 percent more lives than the USPSTF-recommended regimen of biennial screening of women 50 to 74 years old, which had a 23.2 percent mortality reduction. For U.S. women currently 30 to 39 years old, annual screening mammography from ages 40 to 84 years would save 99,829 more lives than USPSTF recommendations if all women comply, and 64,889 more lives with the current 65 percent compliance rate. The potential harms of a screening examination in women 40 to 49 years old, on average, consist of the risk of a recall for diagnostic workup every 12 years, a negative biopsy every 149 years, a missed breast cancer every 1,000 years, and a fatal radiation-induced breast cancer every 76,000 to 97,000 years. Evidence made available to the USPSTF strongly supports the mortality benefit of annual screening mammography beginning at age 40 years, whereas potential harms of screening with this regimen are minor.”

The report concludes:

The irony of screening mammography is that it is one of the most studied of medical interventions. Over three-quarters of a million women have participated in randomized controlled trials (RCT) of screening mammography, over half of them in women 39 to 49 years old. Millions more women have been followed longitudinally through databases such as the Breast Cancer Surveillance Consortium, Swedish, and Canadian screening programs. Mammography is one of the few medical screening interventions that has been shown to have statistically significant mortality benefits (saving lives), even when broken down into age subgroups never intended by the original RCT investigators. Rather than following the established criterion for evaluating medical screening interventions (i.e., the presence of a statistically significant mortality benefit), the USPSTF chose to ignore the science available to them and overemphasized the potential harms of screening mammography, to the serious detriment of U.S. women who follow their flawed recommendations.”

In an effort to further reduce the number of breast cancer deaths, what is now emerging is a more focused look at who mammography benefits most and whether “individualized” screening can be beneficial.

To better understand the accuracy of screening mammograms, in 2002 our group published in Radiology  “Comparison of the Performance of Screening Mammography, Physical Examination, and Breast U.S. and Evaluation of Factors That Influence Them: An Analysis of 27,825 Patient Evaluations.” This study  remains the largest to contemporaneously compare mammograms with physical examination and ultrasound determining how often mammograms found and missed small early-stage cancers in women who were sub-grouped by age, breast density and hormonal status. We found that breast density was the most important factor determining how often a screening mammogram failed to detect breast cancer. That is to say, the more dense the breast the more often mammograms missed finding a cancer, i.e. defined as a false negative mammogram. More important than a woman’s age or her menopausal status, density independently determines how effective a women’s mammogram is thus secondarily affecting by what degree lives are saved. Since most premenopausal women are denser than older women (66 percent vs. 25 percent) it is not surprising that mammograms “save more lives” among women 50 and above than those 40 to 49 years of age. In fact we found that in women with the densest breasts mammography alone correctly found only 48 percent of breast cancer. It is most likely a combination of this failure to detect the cancer as well as the fact that cancers in younger premenopausal women are more aggressive, which produces a lower mortality benefit in younger women (fewer lives saved). This makes earlier cancer detection in younger women even more important. Our study also showed that screening breast ultrasound found the vast majority of all small early-stage cancers missed by mammograms. Thus screening ultrasound would be expected to mimic the life saving benefits of screening mammography. (This was not directly tested because lives saved was not used as an end point in this study). Our study, awarded by the American Medical Association scientific paper of the year 2002, ushered in the era of what we termed “individualized” breast cancer screening, i.e. determining when mammographic accuracy suffers (dense breasts) and adding an additional screening test (ultrasound) to achieve the best possible result of finding breast cancer at its smallest size.

As breast density and cancer risk differ widely among women of even the same age, the question is who should be screened and how best to do it. Perhaps the current use of age-based criteria has only secondary importance.

Evaluating the costs and benefits of individualized screening, is the study “Personalizing Mammography by Breast Density and Other Risk Factors for Breast Cancer: Analysis of Health Benefits and Cost Effectiveness” published this week in the Annals of Internal Medicine. In this article the authors argue that “Mammography is expensive if the disutility of false positive mammography results and the cost of detecting non-progressive and non-lethal invasive cancers are considered.” Again, many women would disagree with the subjective characterization of “expensive” when lives are certainly being saved. Especially since physicians cannot know which cancers will be non-progressive or non-lethal in advance of their screening detection and their subsequent surgical removal. While the 15 percent increase in lives saved in the younger age group, is in spite of mammograms not being as accurate in premenopausal women with dense breasts compared with postmenopausal women with more typically non-dense breasts, we are again left with the basic question: Are women ages 40 to 49 willing to dismiss the 15 percent chance that if they have cancer detected by mammographic screening their life would be saved? If they were to be asked, which they have not been, the answer assuredly would be no.

There has been a very real and tangible movement among American women to better understand and begin addressing what women should be told about factors that uniquely affect the effectiveness of their being screened for breast cancer. In 2009, Connecticut became the first state to mandate that women be informed of the degree of their breast density. This information, previously transmitted by the radiologist only to the patient’s physician, allows the patient to more completely evaluate their mammogram report, i.e. the probability of a false negative mammogram. It also allows women the option of understanding whether an additional screening method, such as ultrasound or MRI, to reduce this false negative rate, would be wise. This movement to educate women and legislate changes is spearheaded by the organization areyoudense.org. Recently Texas became the second state with a Breast Density/Inform law on the books. Known as Henda’s Law, it becomes effective in September 2011. Bills in New York and California have passed the state Senates and will be heard in their respective Assemblies in the coming months. There is also a federal initiative for a Mandatory Breast Density Disclosure law. Oddly, while withholding this information from women sounds unethical, physician groups in different states have opposed the bill citing concerns of misuse of this information and the potential effect of scaring women.

As we attempt to further reduce the number of deaths due to breast cancer it is important not to lose sight of the fact that even with the limitations described, screening mammography is the only scientifically proven test that positively impacts survival and it has a significantly large effect. This applies to women 40 to 49 years of age as well as those 50 and older. In younger women, a 15 percent reduction in cancer deaths cannot be trivialized. In these cost-conscious times both younger and older women’s lives should not be put at risk by policymakers who strain to find science that support reduced funding for breast cancer screening programs at all ages.

In conclusion, even normal risk women would be well-advised to have annual mammography from the age of 40 and consider additional screening if they are high risk or have dense breasts. The medical community would be well advised to better educate and include women in the discussion of risk vs. benefit. Not doing so is unethical and potentially life threatening.

Start the conversation

This site uses Akismet to reduce spam. Learn how your comment data is processed.