The evolving landscape of risk assessment
The future of breast cancer risk assessment is evolving. Today, statistical models like TC8 and Gail, along with hereditary risk guidelines such as NCCN, help us assess lifetime risk. Recent advancements have included breast density as an independent factor, marking progress in understanding breast cancer risk.
Despite these advancements, current models have limitations. While they are well-calibrated at a population level, they often lack precision for individuals, with AUC values around 0.6. Most models were developed using data from predominantly white patients, resulting in poor performance for Black and Hispanic populations. These models have yet to be embraced by OB-GYN or primary care specialties before screening age, which has placed data collection and calculation burden on breast imaging providers. Together these highlight the need for a more proactive, personalized and precise approach.
A more complete patient view: integrating image-based risk
We believe the future of breast cancer risk assessment lies in combining, AI powered cancer-detection and image-based risk with traditional models for a more complete view of a patient’s risk:
Immediate: AI will support radiologists in detecting abnormalities and the likelihood of malignancy in the mammogram they are currently interpreting to improve cancer detection rates.
Short-term: If cancer is not detected in that mammogram, then new image-based risk solutions will provide a short-term view of risk by what factors are presenting in the image from factors closer to the biology of cancer development.
Long-term: Whether this is calculated in imaging or primary care, understanding long-term and hereditary risk—from traditional models like TC8 and NCCN—will continue to be valuable in radiology to ensure patients are on the correct care paths.
This combined approach allows radiologists to identify cancer today, identify patients likely to develop cancer soon and work with high-risk specialists to refer those with elevated lifetime or hereditary risk for additional care.
Volpara x Lunit: leading the way forward
Volpara and Lunit are at the forefront of transforming breast cancer risk assessment. By combining our expertise in risk assessment and mammography AI, we are targeting innovative solutions such as image-based risk that will address disparities in care and serve diverse populations. With a massive database of images and patient data, we are uniquely positioned to create a powerful ecosystem to deliver more precise and equitable risk assessments.
Imagine a tool that uses data from baseline mammograms to automatically drive annual risk calculations, while image-based insights from each screening are seamlessly analyzed. This automated process would give clinicians a comprehensive view of patient risk—from screening decisions to genetic testing recommendations—leading to more personalized care without extra administrative work.
Please reach out to share your views on innovation in image-based risk and its place in a comprehensive patient view of risk. Together, we can create a future where early detection is proactive, personalized and precise for every patient.
