This article is more than 5 years old. Information may no longer be current.

Computer, physician analysis produce similar breast density measurements

The risk for screen-detected cancer and interval cancer was predicted equally with automated and clinical assessments of breast density, according to research published in Annals of Internal Medicine.

“In 30 states, women who have had screening mammography are informed of their breast density on the basis of Breast Imaging Reporting and Data System (BI-RADS) density categories estimated subjectively by radiologists,” Karla Kerlikowske, MD, from the University of California, San Francisco, and colleagues wrote. “Variation in these clinical categories across and within radiologists has led to discussion about whether automated BI-RADS density should be reported instead.”

Kerlikowske and colleagues performed a case-control study to compare automated and clinical BI-RADS density measures for breast cancer risk and detection. The researchers enrolled 1,609 women with screen-detected cancer, 351 women with interval invasive cancer and 4,409 matched controls.

Results showed that women with extremely dense breasts as categorized by automated BI-RADS more than 6 months to 5 years prior to diagnosis were 5.65 times more likely to have interval cancer and 1.43 times more likely to have screen-detected cancer, compared with those with scattered fibroglandular densities.

Regardless of when density was measured, there were similar associations of interval and screening-detected cancer between automated and clinical BI-RADS density measures. Discriminatory accuracy for interval (0.7 vs. 0.62) and screening-detected cancer (0.72 vs. 0.62) was also similar between automated and clinical BI-RADS density measures.

Additionally, automated and clinical BI-RADS categories showed similar mammography sensitivity for fatty breasts (93% vs. 92%), scattered fibroglandular densities (90% vs. 90%), heterogeneously dense breasts (82% vs. 78%) and extremely dense breasts (63% vs. 64%).

“These findings suggest that automated or clinical BI-RADS measures may be used to inform women of their breast density and predict their risk for interval and screen-detected cancer, even as long as 5 years before cancer diagnosis,” Kerlikowske and colleagues concluded.

“Because automated BI-RADS breast density is more reproducible than clinical density and is being used increasingly in the clinical setting, our results suggest that automated density measures may be used to predict risk and help identify women most in need of supplemental screening,” they added. “Future research should focus on developing prediction models comparing automated with clinical BI-RADS density to determine whether repeated automated or clinical measures more accurately predict the 5-year cumulative risk for interval cancer.”

In an accompanying editorial, Joann G. Elmore, MD, MPH, from David Geffen School of Medicine at the University of California, Los Angeles, and Jill Wruble, DO, from Yale School of Medicine, wrote that although the study by Kerlikowske and colleagues contains several methodological strengths, it also contains limitations regarding the study population.

They noted that the study only compared women with extremely dense breasts with those with scattered fibroglandular densities, though many more women have heterogeneously dense breasts which are still considered dense.

“Like [computer-aided detection], automated density measurement has the potential to improve reproducibility and workflow efficiency,” they wrote. “However, we are in an era of ‘choosing wisely’ and seeking value in health care. Therefore, we must be cautious before implementing and paying for medical technology.” – by Alaina Tedesco

Disclosure: Kerlikowske and Wruble report no relevant financial disclosures. Please see study for all other authors’ relevant financial disclosures.