Study: Routine breast cancer screening leads to overdiagnosis

IN DEPTH REVIEW OF RESEARCH

Study background:

Breast cancer screening is a controversial topic—health care providers and researchers are not sure when women at average risk should start screening or how often they should be screened. One reason for the controversy is that health care providers and researchers are still trying to determine the benefit of screening: does it prevent more advanced-stage breast tumors, and will doing that lead to fewer deaths due to breast cancer? Or does screening catch smaller, nonadvanced tumors that may never lead to advanced disease or death?

Karsten Jørgensen and colleagues from The Nordic Cochrane Centre and other institutions published work in the Annals of Internal Medicine in March 2017, using data collected through two Danish cancer registries to determine whether overdiagnosis of breast tumors was occurring.  

Researchers of this study wanted to know:

What types of breast tumors are found through routine breast cancer screening and what rate of overdiagnosed breast cancer occurred after screening was implemented in Denmark?     

Population(s) looked at in the study:

The researchers collected data from women in two registries: the Danish Breast Cancer Group (DBCG) and the Danish Cancer Registry (DCR). These women were between the ages of 35 and 84 and had tumors that were either nonadvanced (20 mm or smaller) or advanced (greater than 20 mm).  Diagnoses of both invasive cancer and ductal carcinoma in situ (DCIS) were considered in this study. Women in the study received screening every other year.

Study findings: 

1. When routine screening programs were introduced in Denmark, more nonadvanced tumors were identified.
2. The number of overdiagnosed tumors (including both invasive tumors and DCIS) were calculated as the difference between the number of tumors before and after screening programs were implemented in Denmark.  The rate of overdiagnosis ranged from about 16% to about 48%, depending on the specifics of the women included in the estimate (age range, where they lived, etc.) and the type of tumor (invasive cancer or DCIS).

Limitations:

This research study looked at the number of nonadvanced and advanced breast tumors, but it did not include data on breast cancer mortality. This means that the study authors can claim incidence of either type of tumors increasing or decreasing, but it cannot connect that data to mortality, because they do not know how many of the nonadvanced tumors would have advanced and caused death. Additionally, this study was done in Denmark, and may not directly translate to the United States. Also, this was a retrospective study; because the authors did not collect their own data, some of it  may have been incomplete, possibly excluding or including factors that may have affected the findings and control for them. Finally, this study included women with or without a family/personal history of breast cancer, and women with or without genetic mutations that put them at higher risk for developing breast cancer. Because of this, it is not known whether the study findings would have been different if the study included only women at higher risk of breast cancer.

Also of importance is the fact that the published study results rely on the authors’ definition of “overdiagnosis” and does not consider the benefits to women whose cancers were caught by routine screening before they had symptoms. This study also excluded women whose tumors were tested with prognostic tests that are now widely used to help guide treatment decisions. 

Conclusions:

This study suggests that overdiagnosis occurs after breast cancer screening. However, more work needs to be done to determine which women benefit or don’t benefit from screening, and when and how often to screen. In an accompanying editorial, Dr. Otis Brawley of the American Cancer Society reviewed other data showing overdiagnosis of breast cancer, and argues that “a significant minority” of breast cancers detected through routine screening do not threaten the health or life of the women.  Identifying which of these tumors do not require aggressive treatment is an area of active research, and as Dr. Brawley writes, “…we must carefully examine screening, realize its limitations, maximize its effectiveness, and try to improve it.”

Not all women have the same risk for breast cancer, and a “one-size-fits-all” screening guideline for women does not exist. Current guidelines already call for distinct screening for women who are known to be at high risk for breast cancer due to mutations in BRCA or other genes associated with increased cancer risk, a strong family history of breast cancer, and/or history of radiation treatment to the chest.  In some cases, guidelines recommend combining magnetic resonance imaging (MRI) with mammography.

Current breast cancer screening guidelines for women at average risk of breast cancer are controversial and differ among reputable organizations. The United States Preventative Services Task Force (USPSTF) recommends a screening mammogram every other year for women ages 50-74. Guidelines of numerous other organizations, including the American Cancer Society (ACS), the National Comprehensive Cancer Network (NCCN), the American Medical Association (AMA), the American College of Radiology (ACR), and the American Congress of Obstetricians and Gynecologists (ACOG) recommend annual screening mammograms beginning at younger ages—the ACS recommends age 45, while the other organizations recommend age 40. Recently, the NCCN added that women and their doctors consider using 3D mammography (tomosynthesis), which is not addressed in this study. Patients should work with their health care providers to determine when and how often they are screened, and to be sure to discuss all of their concerns.

Posted 4/4/17

Share your thoughts on this XRAYS article by taking our brief survey.