Annual Mammography Not Superior to Standard Care on Breast Cancer Mortality

Review

Miller AB, Wall C, Baines CJ, Sun P, To T, Narod SA. Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. BMJ. 2014;348:g366. http://www.bmj.com/content/348/bmj.g366.

Study Methods

In this follow-up study to a randomized controlled trial (RCT) on mammography screening that used cancer registries and vital statistics databases, participants included female volunteers aged 40-59 years who were independently randomized to either annual mammography or no mammography at each screening center.

The intervention group received 4 annual mammography screens, while the control group received 4 annual follow-up questionnaires. The screening period was the first 5 years from randomization, with follow-up from years 6 to 25.

Included in the study were 89,835 Canadian women who had not undergone mammography in the previous 12 months, had no history of breast cancer, and were not pregnant. Participants were recruited at 15 screening centers in 6 Canadian provinces between 1980 and 1985 and were then followed through 1996. Follow-up of all participants was obtained through medical record linkage of the Canadian Cancer Registry and the Canadian National Mortality Database.

Intervention and Control

Among women aged 40-49 years, the intervention group received annual mammography screening and physical breast exams, while the control group received usual care, or remained under the care of their family physician. For women aged 50-59, the intervention group received annual mammography screening and clinical breast examination, while the control group received only clinical breast examination.

Results and Outcomes

Breast cancer incidence, survival, and mortality during the 5-year screening period and the follow-up period were reported in each study group, in addition to breast cancer size at time of detection, node status, and palpability.

During the screening period, 666 breast cancers were detected in the mammography arm, compared to 524 in the control arm. However, the mean size of cancers diagnosed during the screening period was smaller in the mammography arm than the control arm.

Despite a slightly improved “25-year breast cancer survival” rate, there was no difference in breast cancer mortality over the study period between the mammography and control arms. Most importantly, regarding all-cause mortality, the 25-year cumulative mortality rate between the mammography and control groups was not statistically different.

Additionally, 106 “excess” breast cancers were detected in the mammography arm, compared to the control arm, which accounted for 22% of all screen-detected cancers and amounted to one overdiagnosed breast cancer for every 424 women who received mammography screening in the study.

Conclusion

Among women aged 40-59 years, those who received annual mammograms had an increased incidence of diagnosed breast cancers, yet there was no reduction in breast cancer or all-cause mortality beyond clinical breast exam for the 50-59 year-olds or beyond usual care for the 40-49 year-olds.

Commentary

Breast cancer is the most frequently diagnosed non-skin cancer and ranks second for cancer deaths among US women. According to the National Cancer Institute (NCI) and the US Centers for Disease Control and Prevention (CDC), there were 206,966 new cases of breast cancer and 40,996 related deaths in 2010 alone. Approximately 12.3% of women will be diagnosed with breast cancer in their lifetime, according to 2013 estimates. Breast cancer risk increases with age, with the highest rate of new diagnoses and deaths among women aged 55-64 years.

Mammography has been shown to detect pre-symptomatic breast cancer. Its efficacy is presumed to stem from early detection of smaller, earlier-stage tumors, which are believed to be more responsive to treatment and generally carry a more favorable prognosis. However, the likelihood a tumor identified on a mammogram will progress to invasive cancer remains unknown. As a result, the surgical removal and treatment of tumors that may not become clinically apparent in a patient’s lifetime may represent overdiagnosis, which occurs when the benefit of identifying a problem does not result in improved outcomes.

Mammography screening has conflicting efficacy results in the medical literature, as some clinical studies have shown little or no effect, while others have demonstrated decreased mortality in women aged 40 and older, with the largest effect in those aged 50-59 years. Taken together with the lack of robust RCTs, this conflicting evidence prompted this Canadian follow-up study.

Current screening recommendations vary widely by organization and country. The US Preventive Services Task Force (USPTF) and American Academy of Family Physicians (AAFP) recommend biennial mammography screening for women aged 50-74 years. For women under 50 or over 75, it is recommended that the decision to start or continue regular, biennial screening mammography be an individual one that weighs patient context, values, benefits, and harms.

Data from this RCT suggest there is no benefit in annual mammography screening for women aged 40-59 years when compared to usual care, as the mammography study group showed an increased breast cancer diagnosis, but no improved mortality.

Furthermore, the overdiagnosis of breast cancer in the mammography arm could infer added risk with additional diagnostic tests, treatment, treatment complication, and psychological harm. One in 424 women who received a mammogram in this study were diagnosed with a clinically insignificant cancer. The emotional toll alone should spur all healthcare providers to rethink their testing approach and should also be explained to patients before the study is ordered.

While the strengths of this study involve its RCT methodology, long follow-up period, and complete record retrieval, the limitations include its generalizability to the US, as Canadian study participants differ genetically from the US population, and Canada has a universal healthcare system with its resultant increased access to preventive care services.

Based on these new data, current recommendations for mammography screening should be reassessed. Another difference between the US and Canada is the frequency and risk of lawsuits. Malpractice claims when no malpractice occurs is a burden US providers suffer, and only tort reform will lessen this burden. Until then, providing a patient-centered discussion of the risks and benefits of screening and documenting such discussion will keep all parties safe.

Other implications of this study include:

• Waiting until age 50 to begin mammographic screening is reasonable.
• Repeat a screening mammogram every other year.
• Although it was not directly studied, conduct a clinical breast exam annually starting at age 40.
• There is no data to support a benefit for self breast examination, and it may even cause harm, but it is still considered a recommendation by the American Cancer Society (ACS).

About the Authors

Shaula Woz, MS IV, is a fourth-year medical student at the University of Massachusetts Medical School (UMMS) in Worcester, MA.

She was assisted in writing this article by Frank J. Domino, MD, Professor and Pre-Doctoral Education Director for the Department of Family Medicine and Community Health at UMMS and Editor-in-Chief of the 5-Minute Clinical Consult series (Lippincott Williams & Wilkins).