Breast MRI Finds More Invasive Cancers than Digital Breast Tomosynthesis in Women with Dense Breasts Undergoing Screening

Take-Away Points

• ■ Major Focus: Comparing screening performance of abbreviated breast MRI to digital breast tomosynthesis (DBT) in women with dense breasts, a population that experiences increased breast cancer risk and increased risk of failed early diagnosis with mammography.

• ■ Key Result: Abbreviated breast MRI detected more breast cancers (11.8 per 1000 women) than digital breast tomosynthesis (4.8 per 1000 women, P = .002).

• ■ Impact: Breast MRI screening, in comparison to mammography and DBT, will allow for a significantly higher detection rate of invasive cancers in a clinical setting, particularly in women with dense breast tissue. Early and accurate detection of invasive cancers may improve prognosis and treatment outcomes.

Mammography is the current clinical standard for breast cancer screening. However, in women with dense breast tissue, mammography has reduced sensitivity due to masking of breast cancers by dense tissue. Approximately half of screened women have dense breast tissue which increases a woman’s risk for an advanced cancer or interval cancer diagnosis. Many women seek supplemental screening with US achieving moderate increases in sensitivity but at the expense of high false-positive rates. Determining which test is best for supplemental breast cancer screening in women with dense breasts is unclear.

This study compares abbreviated breast MRI and DBT screening in women with dense breasts using a cross-sectional study design at multiple institutions with longitudinal follow-up. A total of 1444 asymptomatic women from 40 to 75 years who had dense breasts (77% heterogeneous and 15% extremely dense) underwent both abbreviated breast MRI and DBT, which were interpreted independently. Women who had a screening breast US in the last 12 months, ever had a breast MRI study, or contrast-enhanced mammogram were excluded. All individuals underwent imaging with both DBT and abbreviated breast MRI at baseline, 1 year, and 3 additional years, with central randomization determining order of imaging examinations. The primary endpoint of cancer detection yielded a statistically significant (P = .002) rate of 11.8 per 1000 for breast MRI versus 4.8 per 1000 for DBT. MRI had greater sensitivity (95.7% MRI vs 39.1% DBT, P = .001), detecting all invasive cancers and five of six cases of ductal carcinoma in situ. Overall results suggest that contribution of DBT is nominal with concurrent abbreviated breast MRI screening; no invasive cancers were detected with DBT alone. However, abbreviated breast MRI had more false-positive (benign) biopsies and 6-month follow-up recommendations. Although specificity of abbreviated breast MRI was lower than DBT (86.7% vs 97.4%, P < .001), positive predictive values were not different (positive predictive value 19.6% MRI vs 31.0% DBT, P = .15).

In conclusion, among women with dense breasts undergoing screening, abbreviated breast MRI was associated with a significantly higher rate of invasive breast cancer detection compared with digital breast tomosynthesis. Early detection from improved MRI screening methods may be an important means to reduce breast cancer mortality in individuals with dense breast tissue.

Highlighted Article

• Comstock CE, Gatsonis C, Newstead GM, et al. Comparison of abbreviated breast MRI vs digital breast tomosynthesis for breast cancer detection among women with dense breasts undergoing screening. JAMA 2020;323(8):746–756. doi:10.1001/jama.2020.0572