MRI remains strongly superior to mammography over the long term in screening women who are at increased risk of developing breast cancer, according to a study published online in the Journal of Clinical Oncology.
The advantage in sensitivity was highly significant for BRCA1 mutation carriers, but not for those who carried BRCA2 mutations and were more likely to present with ductal carcinoma in situ (DCIS).
Previous research showed that in the short term, MRI was approximately twice as sensitive as mammography in detecting breast cancer among women susceptible to the disease, and most guidelines now recommend MRI screening in those who carry BRCA1 or BRCA2 mutations.
However, there is no consensus on the screening protocol for other risk groups, few studies have assessed BRCA1 carriers separately from BRCA2 carriers, and until now no studies have evaluated longer-term screening results, said Dr. Adriana J. Rijnsburger of Erasmus University Medical Center, Rotterdam, the Netherlands, and her associates.
To address these issues, the investigators enlarged and extended the Dutch MRI Screening Study (MRISC) and report their findings after following 2,157 women at six cancer or academic centers for 5 years.
The study subjects, aged 25-75 years at enrollment, had never had breast cancer but were at increased risk because they carried either the BRCA1 or BRCA2 mutation (raising their cumulative lifetime risk of developing breast cancer to 50%-85%), had a high-risk family history (raising their cumulative lifetime risk of developing breast cancer to 30%-50%), or had a moderate-risk family history (raising their cumulative lifetime risk of developing breast cancer to 15%-30%).
They underwent biannual clinical breast examination and annual mammography and MRI.
During 5 years of follow-up, 97 breast cancers developed in 94 women, including 78 (80%) invasive tumors and 19 (20%) cases of DCIS.
Sensitivity at detecting breast cancer was 71% with MRI, significantly greater than the 41% sensitivity of mammography. When only invasive breast cancers were considered, MRI sensitivity increased to 78%, while mammography's sensitivity decreased to 36%.
When the analysis was restricted only to women who carried genetic mutations, the sensitivity of MRI (67%) was “strikingly” higher than that of mammography (25%) for BRCA1 carriers. In contrast, MRI sensitivity (69%) was only slightly higher than mammography's sensitivity (62%) in BRCA2 carriers.
This difference can be explained, at least in part, by the higher proportion of DCIS in BRCA2 than in BRCA1 carriers; mammography was much more sensitive in detecting DCIS (69%) than in detecting invasive tumors (36%).
The specificity of the two screening methods was not significantly different.
Overall, 43% of breast cancers were detected by MRI only. This included 46% of the cancers in BRCA1 carriers, 31% in BRCA2 carriers, 41% in women with a high-risk family history, and 47% in the women with a moderate-risk family history, Dr. Rijnsburger and her colleagues said (J. Clin. Oncol. 2010 Nov. 16; doi:10.1200/JCO.2009.27.2294).
These findings “support the recommendation of the American Cancer Society to use annual MRI screening not only for BRCA1/2 mutation carriers, but for all women with an approximately 20%-25% or greater cumulative lifetime risk of breast cancer due to a familial predisposition,” they noted, with the caveat that cost-effectiveness should be evaluated separately in all risk groups.
This also was the first prospective study of screening in this at-risk patient population to report mortality data, the researchers added.
Five women, all BRCA1/2 mutation carriers, developed distant metastases, and four of them died during follow-up. Two of the women who died had had a favorable tumor stage at diagnosis.
This finding underscores the need for clinicians to avoid guaranteeing that all breast cancer deaths can be prevented by early detection via screening, Dr. Rijnsburger and her associates said.
Survival was 84% in the women with BRCA1 mutations and invasive cancer, and 93% in those with BRCA2 mutations and invasive cancer. Survival was 100% in the other at-risk groups.
BRCA1-associated tumors “behaved completely differently” from BRCA2-associated tumors. They developed at a younger patient age, were not detected as well on mammography, were more likely to develop during the interval between screenings, were more likely to be invasive, and were larger at diagnosis, the investigators said.
This indicates that the current screening schedule for BRCA1 carriers may need to be modified, perhaps by increasing MRI screening to twice rather than once yearly, Dr. Rijnsburger and her colleagues said.
This study was supported by the Dutch government and the Cancer Genomics Center in the Netherlands. The investigators reported having no financial conflicts of interest.