Supplemental imaging may not be necessary in all women with dense breasts, study finds

Supplemental imaging may not be necessary for all women with dense breasts on mammography, according to a new study.

Researchers performed a prospective cohort study using data from the Breast Cancer Surveillance Consortium (BCSC) to examine breast cancer risk and Breast Imaging Reporting and Data System (BI-RADS) density categories and determine which combinations are associated with high rates of interval cancer. The study's main outcome measures were BI-RADS breast density, 5-year breast cancer risk according to the BCSC risk model, and an interval cancer rate per 1,000 mammography exams. Interval cancer rate was defined as invasive cancer 12 months or less after a normal mammogram, and higher interval cancer rate was defined as more than 1 case per 1,000 exams. The study results appeared in the May 19 Annals of Internal Medicine.

The study involved 365,426 women 40 to 74 years of age and 831,455 digital screening mammograms. A total of 2,696 women received a diagnosis of invasive breast cancer within 12 months of their screening mammogram. Such a diagnosis was more likely in older women, white women, women with heterogeneously or extremely dense breasts, those with a BCSC 5-year risk of 1.67% or higher, and those with a family history of breast cancer. Denser breasts were more common among younger women. The researchers found that interval cancer rates were high in women with a 5-year breast cancer risk of 1.67% or more and extremely dense breasts and in women with a 5-year risk of 2.50% or more and heterogeneously dense breasts. Women with a 5-year risk of 2.50% or greater and heterogeneously or extremely dense breasts had the highest interval rate of advanced-stage disease (>0.4 case per 1,000 exams). Just over half of women with heterogeneously dense or extremely dense breasts (51.0% and 52.5%, respectively) had a low 5-year risk; their interval cancer rates were 0.58 to 0.63 case per 1,000 exams and 0.72 to 0.89 case per 1,000 exams, respectively.

The researchers noted that their study did not address the potential benefit of supplemental imaging, alternative imaging strategies, or more frequent mammography. However, they concluded that digital mammography has sufficiently high rates of breast cancer detection and reasonably low false-positive rates and that use of supplemental imaging should not be based only on breast density, since interval cancer rates are not high in all women with dense breasts. “Primary care providers can calculate 5-year breast cancer risk using the BCSC risk calculator and use this information in their discussions about supplemental or alternative screening methods in women with dense breasts,” they wrote.

The authors of an accompanying editorial said that the study provides “compelling evidence” that breast density should not be the only factor considered in decision making about supplemental imaging and suggests that federal legislation on screening in women with dense breasts is “premature.” The editorialists agreed with the study authors that risk assessment should also factor into supplemental imaging decisions.

“Given the lack of scientific consensus, resources targeted for breast density legislation would be better devoted toward more accurate identification of women at high risk for interval breast cancer, research on optimal use of imaging methods, reduction of disparities in screening and early detection, and training of front-line primary care providers on breast cancer risk assessment,” the editorialists wrote.

The same issue of Annals includes high-value advice from ACP on screening average-risk adults with no symptoms for 5 common types of cancer—breast, colorectal, ovarian, prostate, and cervical—as well as a companion framework to help clinicians think about the value of different intensities of cancer screening.

Members of ACP's High Value Care Task Force developed the high-value screening advice by reviewing clinical guidelines and evidence synthesis from ACP and other organizations. The Task Force defined high value as “the lowest screening intensity threshold at which organizations agree about screening recommendations” and low value as “agreement about not recommending overly intensive screening strategies.” In addition to providing current high-value screening advice, the paper also looks at future directions to enhance cancer screening value by reducing overly intensive screening.

The value framework, meanwhile, highlights the following 5 concepts:

1. 1. Screening is a cascade of events rather than a single test.
2. 2. Cancer cases are heterogeneous.
3. 3. Patients are heterogeneous.
4. 4. Screening leads to important benefits for some cancer types and some patients but can lead to significant harms for many more.
5. 5. Determining the value of screening strategies is complex but not impossible.

The authors of the framework noted that while there is extensive pressure to use a maximal detection approach for cancer screening, the value approach would reduce harms, reduce costs, and reduce time spent on low-value screening. “Countering the pressures to screen intensively will likely require multilevel interventions over time … and finding what works is a research priority,” the authors wrote. “Ultimately, the interventions will need to lead to adoption by both the medical profession and the public of a new, value-based framework for thinking about screening.”

Both the high-value care advice and the companion framework are available free of charge online.