Expert Commentary

Should supplemental MRI be used in otherwise average-risk women with extremely dense breasts?

Author and Disclosure Information

Recent data show that supplemental MRI screening in women with extremely dense breasts and negative screening mammograms decreases the rate of interval breast cancers. It remains unclear, however, whether supplemental MRIs will improve other outcomes, such as breast cancer mortality.



While the frequency of dense breasts decreases with age, approximately 10% of women in the United States have extremely dense breasts (Breast Imaging, Reporting, and Data System [BI-RADS] category D), and another 40% have heterogeneously dense breasts (BI-RADS category C).1 Women with dense breasts have both an increased risk for developing breast cancer and reduced mammographic sensitivity for breast cancer detection compared with women who have nondense breasts.2

These 2 observations have led the majority of states to pass legislation requiring that women with dense breasts be informed of their breast density, and most require that providers discuss these results with their patients. Thoughtful clinicians who review the available literature, however, will find sparse evidence on which to counsel patients as to next steps.

Now, a recent trial adds to our knowledge about supplemental magnetic resonance imaging (MRI) breast screening in women with extremely dense breasts.

DENSE trial offers high-quality data

Bakker and colleagues studied women aged 50 to 74 who were participating in a Netherlands population-based biennial mammography screening program.3 They enrolled average-risk women with extremely dense breasts who had a negative screening digital mammogram into the Dense Tissue and Early Breast Neoplasm Screening (DENSE) multicenter trial. The women were randomly assigned to receive either continued biennial digital mammography or supplemental breast MRI.

The primary outcome was the between-group difference in the development of interval breast cancers—that is, breast cancers detected by women or their providers between rounds of screening mammography. Interval breast cancers were chosen as the primary outcome for 2 reasons:

  • interval cancers appear to be more aggressive tumors than those cancers detected by screening mammography
  • interval cancers can be identified over a shorter time interval, making them easier to study than outcomes such as breast cancer mortality, which typically require more than a decade to identify.

The DENSE trial’s secondary outcomes included recall rates from MRI, cancer detection rates on MRI, positive predictive value of MRIs requiring biopsy, and breast cancer characteristics (size, stage) diagnosed in the different groups.

Between-group difference in incidence of interval cancers

A total of 40,373 women with extremely dense breasts were screened; 8,061 of these were randomly assigned to receive breast MRI and 32,312 to continued mammography only (1:4 cluster randomization) across 12 mammography centers in the Netherlands. Among the women assigned to the MRI group, 59% actually underwent MRI (4,783 of the 8,061).

The interval cancer rate in the mammography-only group was 5.0 per 1,000 screenings (95% confidence interval [CI], 4.3–5.8), while the interval cancer rate in the MRI-assigned group was 2.5 per 1,000 screenings (95% CI, 1.6–3.8) (TABLE 1).3

Key secondary outcomes

Of the women who underwent supplemental MRI, 9.49% were recalled for additional imaging, follow-up, or biopsy. Of the 4,783 women who had an MRI, 300 (6.3%) underwent a breast biopsy, and 79 breast cancers (1.65%) were detected. Sixty-four of these cancers were invasive, and 15 were ductal carcinoma in situ (DCIS). Among women who underwent a biopsy for an MRI-detected abnormality, the positive predictive value was 26.3%.

Tumor characteristics. For women who developed breast cancer during the study, both tumor size at diagnosis and tumor stage (early vs late) were described. TABLE 2 shows these results in the women who had their breast cancer detected on MRI, those in the MRI-assigned group who developed interval cancer, and those in the mammography-only group who had interval cancers.3 Overall, tumor size was smaller in the interval group who underwent MRI compared with those who underwent mammography only.

Continue to: Study contributes valuable data, but we need more on long-term outcomes...

Next Article: