Fall 2024 Issue

Risky Business
By Rebecca Montz, EdD, MBA, CNMT, PET, RT(N)(CT), NMTCB RS
For The Record
Vol. 36 No. 4 P. 16

Advanced breast density evaluation and risk assessment increase personalized screening.

According to the National Cancer Institute, nearly 50% of women aged 40 and older have dense breasts, presenting difficulties in the diagnosis of breast cancer. Breast density is the proportion of fibrous and glandular tissue, also referred to as fibroglandular tissue, in comparison to fat tissue within the breast. Understanding breast density is crucial, as women with dense breasts face an elevated risk of breast cancer.

The presence of dense breast tissue poses challenges for radiologists in detecting cancer on mammograms. Fatty tissue appears almost black on mammograms. Dense breast tissue appears white on mammograms, and cancers also exhibit a white appearance, making it more challenging to distinguish a tumor in dense tissue. People with dense breasts may require additional imaging tests beyond a mammogram to help find cancers.

Scott Pohlman, the director of outcomes research at Hologic, says patients now have a heightened awareness of breast density, partly due to recent legislative mandates that enhance reporting standards and FDA auditing capabilities. In addition, a new federal rule by the FDA mandates that by September 10, 2024, all mammogram reports sent to patients in the United States must incorporate an assessment of breast density, providing patients with the following statement, “Breast tissue can be either dense or not dense. Dense tissue makes it harder to find breast cancer on a mammogram and raises the risk of developing breast cancer. Your breast tissue is dense. In some people with dense tissue, other imaging tests in addition to a mammogram may help find cancers.”

Decoding Breast Density
Pohlman says a key difficulty in assessing breast density is that analyzing mammography images and assigning a Breast Imaging-Reporting and Data System (BI-RADS) breast density category is a subjective process without the support of a software tool. Because many medical guidelines recommend that women with dense breast tissue consider supplemental screening in addition to routine mammography, the decision to classify breast tissue as “dense” or “nondense” has downstream implications that could affect patient care.

Subjective assessments of breast density classification may vary between radiologists and may confuse patients, Pohlman says. Inconsistency can create uncertainty if, for example, a radiologist categorizes breast tissue as “nondense” one year and another radiologist classifies it as “dense” the next year. One way to address these challenges is by incorporating software tools specifically designed for objective, quantitative breast density assessment.

“These software solutions offer detailed insight into the breast composition, contributing to a more comprehensive strategy for assessing breast cancer risk in cases of dense breast tissue,” Pohlman says.

Reducing Variability
Volpara’s volumetric breast density assessment software uses artificial intelligence (AI) to generate an objective TruDensity algorithm in Volpara Scorecard, which has been proven to reduce reader variability. TruDensity automatically assesses the volumetric breast density percentage (VBD%) of each mammogram and differentiates each woman on a continuum of density—whether her density is a “high B” or a “low C.”

Volpara volumetric breast density assessment is the only automated, continuous measure validated for use with the clinically popular Tyrer-Cuzick v8 (TC8) Cancer Risk Evaluation Tool. The TC8 model is the first to include data on breast density in the patient calculation and is used to identify patients at elevated lifetime risk appropriate for MRI.

“The key improvement that Volpara brings to density reporting is an objective, reproducible density value that can be used in risk assessment models. These models are increasingly used to determine whether a woman qualifies for MRI-based screening and to decide if the risk is high enough to warrant preventive therapy to reduce risk,” says Jack Cuzick, CBE, PhD, FRS, FMed-Sci, the John Snow Professor of Epidemiology at Queen Mary University of London in the United Kingdom, who developed the TC8 model.

Texture and Pattern Analysis
Hologic has introduced Quantra 2.2 breast density assessment software, which uses a proprietary machine learning algorithm to analyze mammographic images for the distribution and texture of parenchymal tissue. The algorithm automatically creates a breast density score that can be used in breast cancer risk assessment models that utilize breast density as one of the model inputs.

Quantra 2.2 is consistent with the American College of Radiology (ACR) BI-RADS Atlas, 5th Edition. The software uses breast texture and pattern analysis to categorize breasts into four breast composition categories based on ACR breast composition guidelines, creating standardized and reliable assessments, Pohlman says. The software also focuses on pattern and texture analysis. There is a growing body of evidence that the pattern and texture of fibroglandular tissue play an equally significant role in mammographic cancer risk prediction compared with breast tissue volume alone. This feature allows Quantra 2.2 to provide a more comprehensive assessment of breast density.

More consistent and reliable scoring can help elevate the standard of care and standardize reporting across radiology practice, Pohlman says. Improved scoring allows for more personalized care plans and potentially earlier detection of breast cancer, especially in patients with dense breasts, and will provide a more stable year-over-year variance in scoring, he adds.

Integrating Risk Assessment
Breast density is just one factor that can impact a patient’s risk of developing breast cancer over their lifetime. Additional factors, including family history, noncancerous breast conditions such as atypical ductal hyperplasia or lobular carcinoma in situ, and reproductive/hormonal experiences, also contribute to determining the risk of breast cancer. There is strong medical consensus from the ACR, the Society for Breast Imaging, and the American College of Obstetricians and Gynecologists, who agree that all women should have their risk assessed to determine whether they are at elevated risk and would benefit from early or more intensive screening and prevention. This means different data sets, risk models, and care pathways for each patient must be managed in a cancer risk assessment program.

Bruce Lin, risk pathways product manager at Volpara, says, “Ideally, every mammography patient should have a comprehensive risk assessment as part of their patient record. The biggest danger with high-risk breast cancer patients is that they do not know they are high risk. Software is needed to bring standardization to the assessment to help ensure that patients do not fall through the cracks and that each patient receives the agreed upon risk protocol.”

Quantitative Analysis
ScreenPoint Medical provides quantitative breast image analysis and computer-aided detection. The company’s Transpara AI software identifies cancers at earlier stages. It is FDA-cleared for utilization on all major 2D and 3D mammography systems. In November 2023, ScreenPoint Medical announced that its Transpara breast AI had examined over 5 million mammograms, encompassing more than 1 million tomosynthesis exams.

The AI software functions as a “secondary set” of eyes during mammography readings, helping detect cancers early and decrease recall rates. Designed to collaborate with radiologists, Transpara provides decision support to improve clinical accuracy, reduce workload, and optimize workflow. Sanmitra Iwanski, vice president of marketing at ScreenPoint Medical, cites a 2023 publication by Koch et al in European Radiology that reported Transpara can alleviate the impact of dense breast tissue on radiologist sensitivity during mammogram readings. Within the current workflow, this allows radiologists to screen women with high confidence in an effective and consistent manner, she adds.

The prevailing standard for risk assessment is both subjective and labor intensive. Presently, many facilities lack the resources to dedicate the necessary time for manual risk identification. Alejandro Rodriguez Ruiz, vice president of clinical strategy at ScreenPoint Medical, says image-based breast cancer risk software tools have demonstrated higher accuracy compared with the existing standard of care. Rodriguez Ruiz cites two peer-reviewed publications from 2023 in which Transpara, along with other image-based inputs, could be used for risk assessment in the four- to five-year timeframe. In a 2023 publication by Vachon et al in the Journal of Clinical Oncology, researchers from Mayo Clinic and University of California, San Francisco discussed the use of Transpara and a density risk category classification as a useful method to determine invasive breast cancer risk over the 2.5- to five-year horizon. In a 2023 Radiology publication by Lauritzen et al, the approach introduced by Vachon was extended, utilizing Transpara along with a BI-RADS-like density classification that incorporated volumetric breast density. The researchers employed a proprietary texture analysis on the processed image, resulting in a significant enhancement in risk prediction.

In the past, image analysis software typically provided a point-in-time risk, representing the interval between consecutive screenings. Rodriguez Ruiz says Transpara excludes risk by identifying women at low risk and identifying areas of concern during screening. According to Rodriguez Ruiz, when Transpara is combined with a density measurement, it offers a useful long-term risk prognosis. The integration of Transpara with both density and texture measurements surpasses today’s standard of care.

Lifetime Risk
Volpara’s Risk Pathways software provides a single workflow for assessing hereditary and lifetime breast cancer risk of every mammography patient. The software allows providers to send patients an assessment survey before their screening and diagnostic appointment. Once the patient’s personal risk factor data has been collected, the radiology department has access to a set of breast cancer risk models—including Tyrer-Cuzick, Gail, BRCAPRO, Claus, and the National Comprehensive Cancer Network—to identify high-risk patients before they enter the facility. Technologists can confirm and update risk data after dialog with the patient and recalculate once the mammogram is done and breast density is assessed. Risk Pathways can also be integrated with Volpara Scorecard breast density assessment for TC8 calculation.

“Volpara Scorecard and Risk Pathways users have reported that using VBD% in TC8 changes medical management for 4.5% of their screening population,” Lin says.

Risk Pathways also offers decision support for follow-up care and monitoring using clinical guidelines to recommend additional imaging, genetic counseling, or other interventions and preventative measures. This information can be automated and prepopulated into mammography patient letters and referrer summaries. It can also be flagged on a patient’s medical record to support better adherence to recommendations. The software interfaces with major EHRs and mammography reporting systems. Risk Pathways has been used by more than 1,000 providers in the United States to conduct more than 3 million risk assessments annually.

At the Radiological Society of North America 2023 Annual Meeting, Stamatia Destounis, MD, FACR, the managing partner of Elizabeth Wende Breast Center in Rochester, New York, and a member of Radiology Today’s editorial advisory board, presented findings reviewing a five-year period of high-risk screening involving supplemental MRI and genetic testing. The results indicated that the utilization of Volpara’s Risk Pathways and density assessment software in a large, community-based program can enhance outcomes by incorporating effective risk assessment. “Our research found that for the women that got genetic testing, risk assessment evaluation, and had the MRI … those women actually ended up having breast cancers that were smaller and the burden of tumor was a lot less,” Destounis says.

Tomorrow’s Insights
The outlook for better breast density assessment and overall breast cancer risk assessment is bright, thanks to promising AI and other software developments. These innovative software solutions are pivotal to standardizing reporting, improving screenings, and elevating the standard of care throughout the entire medical practice. With increasing attention on breast density, patients are becoming increasingly aware of its implications. Looking ahead, health care providers must be ready to articulate the functionalities of these technologies and elucidate their role in both breast density assessment and comprehensive patient care.

Implementing a cancer risk assessment program can lead to positive outcomes for clinicians, health care institutions, and patients. By using comprehensive risk assessment to identify individuals at higher risk of developing cancer, patients may benefit from closer monitoring, genetic counseling, or preventive interventions. “It has the potential to improve treatment outcomes and reduce mortality rates,” Lin says. “It is also important to remember that it opens the door for preventive measures that may reduce the incidence of cancer in high-risk populations.”

In overtasked health systems, efficient resource utilization is essential. Risk assessment provides the opportunity to identify and target high-risk populations, which allows health care providers to optimize the allocation of health care resources. “Rather than applying a one-size-fits-all approach to screening and interventions, resources can be directed toward those who would benefit most, improving efficiency in health care delivery,” Lin says.

— Rebecca Montz, EdD, MBA, CNMT, PET, RT(N)(CT), NMTCB RS, has worked at the Mayo Clinic Jacksonville and University of Texas MD Anderson Cancer Center in Houston as a nuclear medicine and positron emission tomography technologist.