Study: MRI or mammograms for detecting breast cancer in families with unknown genetic mutations?

IN-DEPTH REVIEW OF RESEARCH

Study background:

Fifteen percent of all women with breast cancer have a family history of the disease but have no known mutation associated with increased breast cancer risk. Several trials that looked at adding MRI to mammography as a screening method showed that breast cancer cases were detected earlier than with mammography alone. However, these were not randomized trials and MRI was done at the same time as mammography. Additionally, these trials did not determine how breast density affected MRI results.

Current NCCN guidelines in the United States recommend annual mammography and clinical breast exam with MRI for women 30 or older who have a lifetime risk of breast cancer of 20% or more. However, in other countries these recommendations vary. In the Netherlands and U.K., MRI is only recommended for women with BRCA1 or BRCA2 mutations.  For this study, researchers want to know how MRI compares to mammography at detecting breast cancer for women with familial breast cancer.

Researchers of this study wanted to know:

whether MRI or  mammography are better at detecting cancer in women with a family history of breast cancer but without a known genetic mutation.

Populations looked at in this study:

This trial enrolled 1,355 women from 12 outpatient breast cancer clinics or family cancer clinics from 7 academic medical centers in the Netherlands between Jan 1, 2011 and Dec 31 2017. Participants were 30-55 years of age with a lifetime risk of breast cancer of 20% or higher based on family history. Women with a known inherited mutation in BRCA1, BRCA2 or TP53 were excluded because MRI surveillance is already recommended for them. Women with previously diagnosed invasive breast cancer were also excluded. One woman in the MRI group was diagnosed with an invasive breast cancer after randomization and before initial MRI screening and was excluded. After randomization, the total number of participants was 1,354: 674 in the MRI group and 680 in the mammography group.

Women who did not consent to randomization of screening methods were registered, screened by their chosen method and their results were included in analysis of screening method among groups with different breast densities. There were 13 women in the MRI registration group and 218 in the mammography registration group. Combining the randomized and registry groups, there were 206 women with category A breast density (fatty, not dense), 549 women with category B breast density (scattered density), 562 women with category C breast density (heterogeneously dense), and 239 women with category D breast density (extremely dense).

Study design:

Participants were randomized to one of two groups:

• The MRI group received annual MRIs and clinical breast exams and had mammography screening every other year.
• The mammography group received annual mammography testing and clinical breast exams.

The average number of screening rounds was 4.3 screenings per person. Participants were followed for an average of 5.2 years after joining the study.

Study findings: 

A total of 55 breast cancers were detected (32 invasive cancer cases and 23 ductal carcinoma in situ (DCIS) cases). None of the breast cancers were bilateral and none had metastasized.

• More breast cancers were detected in the MRI group than the mammography group.
   
  • In the MRI group, 24 invasive cancers (4%) and 16 DCIS (2%) were detected.
    • Overall, in the MRI group, 14.2 breast cancers were detected for every 1,000 rounds of screening.
  • In the mammography group, 8 invasive cancers (1%) and 7 cases of DCIS (1%) were detected.
    • Overall, in the mammography group, 4.9 breast cancers were detected for every 1,000 rounds of screening.
       
• The breast cancers detected in the MRI group were smaller, of earlier stage, and less often involved lymph nodes than breast cancers detected among the mammography group. Researchers concluded that MRI detected breast cancers at earlier stages than mammography.
   
  • In the MRI group, the median size of the invasive breast cancers was 9 mm.
  • In the mammography group, the median size of the invasive breast cancers was 17 mm.
     
  • In the MRI group, the majority of invasive breast cancer were stage T1a (larger than 1 mm and up to 5 mm) or T1b (larger than 5 mm and up to 10 mm).
    • 14 of 24 invasive breast cancers (58%) were stage T1a or T1b.
    • 10 of 24 invasive breast cancer (42%) were stage T1C (larger than 10 mm and up to 22 mm), T2 (larger than 20 mm and up to 50 mm) or T3 (larger than 50 mm).
       
  • In the mammography group, the majority of invasive breast cancers were more advanced than stage T1b:
    • 1 of 8 invasive breast cancers (13%) were stage T1a or T1b.
    • 7 of 8 invasive breast cancer (87%) were stage T1c or T2.
       
  • In the MRI group, the majority of invasive breast cancers did not involve lymph nodes:
    • 4 of 24 invasive breast cancers (17%) were node positive.
  • In the mammography group, the majority of invasive breast cancers involved lymph nodes:
    • 5 of 8 invasive breast cancers (63%) were node positive.
       
• As expected, MRI produced more false positives (suspected cancer that was disproven by biopsy) than mammography:
   
  • 449 false positives results occurred in the MRI group.
    • 275 of 449 (61%) false positives were detected by MRI only.
    • 19 of 449 ( 4%) false positives were detected by both MRI and mammography.
    • 98 of 449 (22%) false positives were detected by mammography (from mammography every other year).
    • 57 of 449 (13%) false positives were detected by clinical exam.
       
  • 276 false positive results occurred in the mammography group.
    • 157 of 276 (57%) false positives were detected by mammography only.
    • 9 of 276 (3%) false positives were detected by MRI.
    • 10 of 276 (40%) false positives were detected by clinical exam.
       
• MRI detected more breast cancer in women among all breast density groups except the very dense breast tissue group. Among women with very dense breast tissue, similar numbers of breast cancer were detected in both the MRI and mammography groups (5 and 6 cases respectively).
   
  • Among women with category A (fatty, least dense)
    • In the MRI group, breast cancer was detected in 5 of 86 (6%) women.
    • In the mammography group, no breast cancer was detected among the 120 (0%) women.
       
  • Among women with category B (scattered density)
    • In the MRI group, 15 of 249 (6%) women had a breast cancer detected.
    • In the mammography group, 7 of 300 (2%) women had a breast cancer detected.
       
  • Among women with category C (heterogeneously dense)
    • In the MRI group, 17 of 238 (7%) women had breast cancers detected
    • In the mammography group, 10 of 324 (3%) women had a breast cancer detected.
       
  • Among women with category D (extremely dense)
    • In the MRI group, 5 of 105 (5%) women had breast cancers detected.
    • In the mammography group, 6 of 134 (4%) women had a breast cancer detected.
       
• The rate of false positives increased with breast tissue density with both MRI and mammography. Overall, MRI produced more false positives than mammography among all breast densities.
   
  • Among women with category A (fatty, least dense)
    • In the MRI group, 38 of 86 (44%) had false positive results.
    • In the mammography group, 37 of 120 (31%) women had false positive results.
       
  • Among women with category B (scattered density)
    • In the MRI group, 150 of 249 (60%) women had false positive results.
    • In the mammography group, 96 of 300 (32%) women had false positive results.
       
  • Among women with category C (heterogeneously dense)
    • In the MRI group, 164 of 238 (69%) women had false positive results.
    • In the mammography group, 157 of 324 (48%) women had false positive results.
       
  • Among women with category D (extremely dense)
    • In the MRI group, 100 of 105 (95 %) women had false positive results.
    • In the mammography group, 79 of 134 (59%) women had false positive results.

Limitations:

A clear limitation of MRI as a screening method is the frequency of false positive results. These results require further testing and biopsies to rule out an invasive cancer diagnosis. On the other hand, the rate of detecting earlier, node negative cancers suggest that MRI screening may lead to improved outcomes in terms of mortality. The researchers note that they intend to follow up with mortality data in the future.

Another potential problem is that MRI uses a gadolinium tracer, which carries a small chance of very small amounts being retained in brain tissue. This is not known to cause harmful effects.

MRI was better at detecting cancers in most breast tissue (groups A-C) than mammography. Contrary to the researchers’ expectation, MRI was not significantly different from mammography at detecting breast cancers in the group of women with the most dense breast tissue. Similar numbers of cancers were detected by both methods, suggesting that neither method is better or worse than the other. The numbers of breast cancers in each breast density group were small, which may have limited the ability to observe small differences. Additionally, compared to mammography, MRI produced far more false positives among women with extremely dense breasts. This may limit its usefulness among this population.

Conclusions:

The researchers concluded that MRI detected breast cancers at earlier stages than mammography and at all but the most extreme breast densities. One drawback is that MRI screening led to more false positives than mammography. Overall, MRI may improve detection for high-risk women who do not have known genetic mutations.

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

Posted 8/15/19