This article is more than 5 years old. Information may no longer be current.
Ultrasound improved cancer detection in women with dense breasts
Click Here to Manage Email Alerts
SAN ANTONIO — Women with dense breasts who underwent ultrasound had cancers detected that were missed during screening mammography, according to results of a retrospective analysis presented at the San Antonio Breast Cancer Symposium.
In 2009, Connecticut became the first state to mandate that patients be informed during routine screening mammograms if they have dense breast tissue, Jean M. Weigert, MD, FACR, director of breast imaging at the Hospital of Central Connecticut, said during a presentation.
Jean M. Weigert
“This started by a grassroots effort on part of a group of women who felt they received a delay in diagnosis because they had such dense breast tissue,” Weigert said. “[Since then,] we were able to look at the number of screening breast ultrasounds that were performed to see if they were useful in the detection of breast cancer.”
Weigart reviewed chart information of patients evaluated at five sites during a 4-year period. Information included number of screening mammograms and dense breast ultrasounds, Breast Imaging Reporting Data System (BIRAD) code results, biopsy-proven malignancies and high-risk lesions, and patient demographics.
The analysis included data from women with breast density ˃50% — which represents 40% to 50% of women. All of these women had received negative findings on their screening mammography and were offered screening ultrasound
Approximately 30,000 screening mammograms and 3,000 screening ultrasounds occurred each year.
The number of patients classified as having BIRAD 4 or 5 decreased from 151 the first year to 53 the fourth year, suggesting a learning curve exists in interpreting ultrasound results, Weigert said.
“The first 3 years, we were still doing a significant number of biopsies on patients with findings that we didn’t know were positive or negative,” Weigert said. “In the fourth year there was a significant reduction in naming these as BIRADs 4s or 5s, but we were still finding the same number of cancers. Clearly, we were doing more follow-up in these patients, but I think this is a telling point.”
The positive predictive value of the ultrasound increased from 7.1% the first year to 17.2% in the fourth year; however, the number of cancers detected remained relatively stable from 4 cancers per 1,000 women in the first year to 3.2 cancers per 1,000 women in the fourth year.
The percentage of eligible women with dense breasts who received the screening, despite state-mandated insurance coverage, also remained stable from 27.5% in the first year to 28.1% in the fourth year.
A few cancers that were detected in the first year were larger than 2 sonometers, but in general the cancers detected were small, Weigert said. They were predominately grade 2 and grade 3 tumors, and the majority were ER-positive and PR-positive. Few patients had sentinel node positivity. By year 4, patients were returning for routine ultrasound screening, and most of the cancers detected were smaller than 1 sonometer.
“These are the same types of cancer that we would want to find in screening mammography, but we couldn’t see them on the screening mammogram,” Weigert said. “It took the ultrasound for us to find these cancers. Screening breast ultrasound in women with mammographically dense tissue does find occult cancers.”
For more information:
Weigert JM. Abstract #S5-01. Presented at: San Antonio Breast Cancer Symposium; Dec. 9-13, 2014; San Antonio.
Disclosure: Weigert reports a consultant role with and stock ownership in Tractus Inc.