Dr. Susan Harvey, Vice President of Global Medical Affairs for Breast and Skeletal Health, Hologic02.07.22
The year 2021 marked a decade since the U.S. Food and Drug Administration (FDA) approved the first digital breast tomosynthesis (DBT) system—Hologic’s Selenia Dimensions Mammography System1—which revolutionized mammographic breast cancer screening. Since the approval of that first DBT system, patients and clinicians have experienced improved patient outcomes related to mammographic breast imaging. In addition, recent innovations have led to utilization of artificial intelligence (AI) and deep learning software with DBT technology, which has further augmented the skills and accuracy of the radiologist’s interpretation. Looking ahead, DBT technology is expected to further evolve, along with the full breast imaging field, including ultrasound and MRI imaging.
A Decade of DBT
Following FDA approval, DBT was quickly adopted at breast imaging centers across the U.S. due to improved radiologist performance metrics and resulting patient outcomes. In the United States, DBT is widely recognized as standard of care for performing mammography imaging—both screening and diagnostic. With DBT, radiologists benefit from the lack of overlying tissue obscuring lesions due to the slices of a DBT image. Less tissue overlap has improved cancer detection rates as well as false positive rates compared to full field digital mammography (FFDM), which produces images that rely on summation of all the breast tissue from each view.2 As an example, a DBT exam was found to detect 20 to 65 percent more invasive cancers than a 2D mammogram alone.3
The introduction of DBT was especially important for women with dense breast tissue, as breast tissue appears white on FFDM mammograms and can obscure lesions, including breast cancers. Patients with dense breasts have experienced significantly improved cancer detection rates and lower false positive rates with DBT imaging. In fact, some DBT technologies have received FDA approval for better detection of breast cancers across all tissue densities compared to FFDM alone.3,4 Lesions, including breast cancers, that were previously obscured on 2D exams are now evident to radiologists via DBT, leading to a more accurate interpretation of images.
Along with increased cancer detection rates, DBT also decreased patient callback rates—meaning fewer false positive interpretations. Several studies demonstrated DBT reduced recall rates by 15 to 40 percent.5 Both clinicians and patients benefit from fewer recalls, as clinicians can focus on patient communication and clinically important cases. For patients, a reduced recall rate improves their experience by reducing anxieties about unnecessary procedures and limiting the physical and financial toll of tests.
While DBT led to higher detection rates and reduced callbacks, the modality also created an increased number of images for radiologists to review. To combat this new challenge, companies developed AI technology to assist in reviewing the images and to augment radiologists’ performance.
These software tools have become capable of identifying calcifications, architectural distortions, and masses, and they can highlight these findings of interests in real time. The images can be prioritized for interpretation based on the number and severity of regions that may represent cancer. Collectively, these deep learning-enabled tools have not only assisted radiologists with reading, but they have also improved workflow, as they can safely reduce the amount of time spent on interpretation, which can lead to less reader fatigue.
Today’s DBT technology has evolved to prioritize the patient experience and comfort in an effort to improve compliance with recommended screening intervals. This has taken many forms—one of which is the introduction of curved breast paddles. These breast paddle systems are designed to conform to the contours of a woman’s breast, therefore reducing pinching and making compression more even from the back of the breast near the chest wall to the anterior part of the breast near the nipple, which is thinner.
The World Health Organization has noted that early detection, diagnosis, and treatment can be highly effective, achieving survival probabilities of 90 percent or higher for breast cancer.6 Today’s advanced DBT imaging is designed to help radiologists clearly visualize subtle masses, asymmetries, distortions, and calcifications to help pinpoint cancers early. The earlier breast cancer is detected, the greater the number of less invasive and aggressive treatment options may be available to patients.
The Future of DBT and Breast Imaging
Patient comfort will always remain an important consideration, as it may impact some patients’ willingness to receive an annual exam. In fact, in one survey, 49 percent of women reported fear of anxiety and pain as the reason they have never had a mammogram.7 Therefore, it is expected that innovations in DBT gantries will bring additional features that increase comfort.
These changes will have little impact on patient outcomes, however, if we do not increase the availability of DBT for all women in the United States. While DBT gantries are widely available and now accepted as the standard of care in the United States, a study published by the Journal of the American College of Radiology noted that Black women were less likely to be screened via DBT and less likely to be screened multiple times during the five-year period than Caucasian women.8 This is particularly alarming given that Black women are almost 40 percent more likely to die from breast cancer than non-Hispanic white women.9 We must strive to increase access to DBT for underserved populations, so they too may benefit from the improved patient outcomes the technology imparts, which may help to decrease disparities in breast cancer care.
For radiologists, ongoing advances in DBT imaging and deep learning systems will likely continue to improve the speed at which they can review mammogram exams and the accuracy of interpretation. It is anticipated deep learning technologies will enable the integration of multiple different imaging modalities, such as DBT and ultrasound, which allow radiologists to correlate exams in real time between different modalities and determine next steps for a patient quickly.
Conclusion
Today, a decade after the first DBT system transformed breast cancer screening, its capabilities continue to evolve. There is hope that in the coming years, earlier detection of breast cancer and further reduction of false positive interpretations will offer all women in the United States the best experience and highest quality of care possible.
References
Susan C. Harvey, MD, has more than three decades of experience as a breast imager working in academic settings. She currently serves as the vice president of global medical affairs in the breast and skeletal health division at Hologic Inc.
A Decade of DBT
Following FDA approval, DBT was quickly adopted at breast imaging centers across the U.S. due to improved radiologist performance metrics and resulting patient outcomes. In the United States, DBT is widely recognized as standard of care for performing mammography imaging—both screening and diagnostic. With DBT, radiologists benefit from the lack of overlying tissue obscuring lesions due to the slices of a DBT image. Less tissue overlap has improved cancer detection rates as well as false positive rates compared to full field digital mammography (FFDM), which produces images that rely on summation of all the breast tissue from each view.2 As an example, a DBT exam was found to detect 20 to 65 percent more invasive cancers than a 2D mammogram alone.3
The introduction of DBT was especially important for women with dense breast tissue, as breast tissue appears white on FFDM mammograms and can obscure lesions, including breast cancers. Patients with dense breasts have experienced significantly improved cancer detection rates and lower false positive rates with DBT imaging. In fact, some DBT technologies have received FDA approval for better detection of breast cancers across all tissue densities compared to FFDM alone.3,4 Lesions, including breast cancers, that were previously obscured on 2D exams are now evident to radiologists via DBT, leading to a more accurate interpretation of images.
Along with increased cancer detection rates, DBT also decreased patient callback rates—meaning fewer false positive interpretations. Several studies demonstrated DBT reduced recall rates by 15 to 40 percent.5 Both clinicians and patients benefit from fewer recalls, as clinicians can focus on patient communication and clinically important cases. For patients, a reduced recall rate improves their experience by reducing anxieties about unnecessary procedures and limiting the physical and financial toll of tests.
While DBT led to higher detection rates and reduced callbacks, the modality also created an increased number of images for radiologists to review. To combat this new challenge, companies developed AI technology to assist in reviewing the images and to augment radiologists’ performance.
These software tools have become capable of identifying calcifications, architectural distortions, and masses, and they can highlight these findings of interests in real time. The images can be prioritized for interpretation based on the number and severity of regions that may represent cancer. Collectively, these deep learning-enabled tools have not only assisted radiologists with reading, but they have also improved workflow, as they can safely reduce the amount of time spent on interpretation, which can lead to less reader fatigue.
Today’s DBT technology has evolved to prioritize the patient experience and comfort in an effort to improve compliance with recommended screening intervals. This has taken many forms—one of which is the introduction of curved breast paddles. These breast paddle systems are designed to conform to the contours of a woman’s breast, therefore reducing pinching and making compression more even from the back of the breast near the chest wall to the anterior part of the breast near the nipple, which is thinner.
The World Health Organization has noted that early detection, diagnosis, and treatment can be highly effective, achieving survival probabilities of 90 percent or higher for breast cancer.6 Today’s advanced DBT imaging is designed to help radiologists clearly visualize subtle masses, asymmetries, distortions, and calcifications to help pinpoint cancers early. The earlier breast cancer is detected, the greater the number of less invasive and aggressive treatment options may be available to patients.
The Future of DBT and Breast Imaging
Patient comfort will always remain an important consideration, as it may impact some patients’ willingness to receive an annual exam. In fact, in one survey, 49 percent of women reported fear of anxiety and pain as the reason they have never had a mammogram.7 Therefore, it is expected that innovations in DBT gantries will bring additional features that increase comfort.
These changes will have little impact on patient outcomes, however, if we do not increase the availability of DBT for all women in the United States. While DBT gantries are widely available and now accepted as the standard of care in the United States, a study published by the Journal of the American College of Radiology noted that Black women were less likely to be screened via DBT and less likely to be screened multiple times during the five-year period than Caucasian women.8 This is particularly alarming given that Black women are almost 40 percent more likely to die from breast cancer than non-Hispanic white women.9 We must strive to increase access to DBT for underserved populations, so they too may benefit from the improved patient outcomes the technology imparts, which may help to decrease disparities in breast cancer care.
For radiologists, ongoing advances in DBT imaging and deep learning systems will likely continue to improve the speed at which they can review mammogram exams and the accuracy of interpretation. It is anticipated deep learning technologies will enable the integration of multiple different imaging modalities, such as DBT and ultrasound, which allow radiologists to correlate exams in real time between different modalities and determine next steps for a patient quickly.
Conclusion
Today, a decade after the first DBT system transformed breast cancer screening, its capabilities continue to evolve. There is hope that in the coming years, earlier detection of breast cancer and further reduction of false positive interpretations will offer all women in the United States the best experience and highest quality of care possible.
References
- bit.ly/mpo220121
- bit.ly/mpo220122
- bit.ly/mpo220123
- FDA submissions P080003, P080003/S001, P080003/S004, P080003/S005
- bit.ly/mpo220125
- bit.ly/mpo220126
- Kadence International, Ten Thousand Quantitative Findings Research Study (5107), April 2017.
- bit.ly/mpo220128
- bit.ly/mpo220129
Susan C. Harvey, MD, has more than three decades of experience as a breast imager working in academic settings. She currently serves as the vice president of global medical affairs in the breast and skeletal health division at Hologic Inc.