This Clinician’s View is written by AdventHealth internist and high-risk breast health specialist Lauren Cornell, MD.
While the medical field has made great strides in calculating and communicating a woman’s risk of developing breast cancer, one limitation remains at the center of our traditional approach: risk is not static.
An individual woman’s risk today or over the next five years may look very different from her risk 10 or 20 years from now. Hormonal changes, family history, genetic discoveries, biopsies and lifestyle factors can all modify risk over time. As that risk evolves, screening strategies must evolve with it. A one-time risk assessment cannot reliably guide decades of clinical decision-making. At the same time, more testing doesn’t necessarily lead to better outcomes.
That is why I believe it is time to refine how we approach breast cancer screening. Rather than relying on long-term, population-based risk estimates and protocols, we should move toward shorter-term, risk-based strategies that incorporate ongoing reassessment and truly individualized care.
For me, this issue is personal.
I carry a BRCA1 gene mutation. My grandmother died of ovarian cancer when I was in elementary school, and when I was a sophomore in college, my mother was diagnosed with an aggressive form of breast cancer. Fortunately, it was detected early, but the effects of her diagnosis and treatment forever changed her life. Learning that I carry the same mutation has allowed me to make informed decisions about my own health and inspired my career in high-risk breast care and survivorship.
My family’s cancer journey continues to shape my philosophy today. Our goal should not just be more testing. It should be the right testing for the right patient at the right time.
The Challenge with Traditional Screening Protocols
Historically, assessment tools like the Tyrer-Cuzick (IBIS) model have been used primarily to calculate an individual’s lifetime risk of developing breast cancer, taking into consideration factors like age, race/ethnicity, reproductive history, family history and breast density. Those who exceed the 20% lifetime risk threshold for developing breast cancer are then placed in a more aggressive screening protocol that incorporates supplemental imaging, including annual breast MRI. But are we truly capturing the right patients when we focus on lifetime risk?
I believe lifetime risk models leave too much room for error. They can overestimate risk in younger women and underestimate risk in older women. We don’t know when a woman is going to go into menopause or if she will require a breast biopsy. We also don’t know if someone else in her family is going to get diagnosed with breast cancer. Her hormonal exposures and lifestyle factors will also change over time. Clinical decisions today should not rely solely on predictions extending to age 85.
If risk is dynamic, then our screening strategies should be as well. Recently, the WISDOM (Women Informed to Screen Depending On Measures of Risk) study tested whether a more personalized, five-year risk assessment approach could safely replace routine population-based screening for breast cancer. The results of this prospective, randomized clinical trial were published online in JAMA (The Journal of the American Medical Association) last December and presented at the San Antonio Breast Cancer Symposium (SABCS), the world’s largest conference dedicated solely to breast cancer research.
The WISDOM Trial: A Shift in How We Screen for Breast Cancer
The WISDOM study was designed to answer a fundamental question: Can a personalized, risk-based approach to breast cancer screening be just as safe as annual mammography while reducing unnecessary interventions and better aligning with patient preferences?
To find out, researchers enrolled more than 46,000 women between 2016 and 2023. Over 28,000 participants agreed to be randomized, with half assigned to annual mammography and the other half placed on a personalized screening schedule based on their individual risk. A key feature of the study was its use of a woman's five-year breast cancer risk to guide screening recommendations.
Women in the personalized arm completed the Breast Cancer Surveillance Consortium (BCSC) risk calculator, which estimates the likelihood of developing breast cancer within five years. Researchers then combined those results with additional information, including genetic testing, polygenic risk scores, breast density, and other biological and lifestyle factors to create a more comprehensive risk assessment.
Based on that evaluation, participants were assigned to one of four risk categories with screening recommendations aligned to their needs:
- Lowest risk: Primarily women under age 50 who were advised to delay screening mammography until age 50, unless their risk profile changed.
- Average risk: Recommended screening mammography every two years beginning at age 50.
- Elevated risk: Recommended annual mammography beginning at age 40.
- Highest risk: Recommended screening every six months, alternating between mammography and MRI.
The WISDOM study revealed four key findings that should shape how we think about breast cancer risk assessment and screening moving forward:
- Personalized, risk-based screening did not lead to more advanced cancer diagnoses (Stage 2B or higher), demonstrating that it can be a safe alternative to routine annual screening.
- Women overwhelmingly embraced the personalized approach. When given a choice, nearly 90% selected risk-based screening, underscoring the growing demand for more individualized care.
- Researchers found that nearly one in 38 women carried a pathogenic breast cancer mutation. About 30% of those women had no family history of the disease and likely would not have been identified under current clinical guidelines. This finding highlights the potential value of broader genetic testing.
- Biopsies and other intensive diagnostic procedures could be more effectively concentrated among women at highest risk, while reducing unnecessary interventions for those at average risk.
We’re already starting to see some momentum build because of these findings. In March 2026, the National Comprehensive Cancer Network (NCCN) updated its Breast Cancer Screening and Diagnosis Guidelines to recommend considering supplemental MRI screening based on a woman’s five-year high-risk estimate ( 1.7%), rather than relying solely on lifetime risk. This represents an important step toward more personalized screening recommendations, and I hope it signals a broader shift across future guidelines.
The updated NCCN guidelines also introduced a notable advancement: incorporation of artificial intelligence (AI) to estimate a woman’s five-year breast cancer risk directly from mammograms. It’s another example of how precision medicine is moving from concept to clinical reality in breast cancer screening.
The Role of AI in Enhancing Breast Cancer Risk Prediction
Most women diagnosed with breast cancer have no identifiable genetic mutation or strong family history of the disease. Similarly, while breast density is a common risk factor, it is a relatively broad measure that provides limited ability to accurately stratify risk at the individual level. These limitations underscore a fundamental gap in current risk assessment approaches. AI-based analysis of mammograms offers a promising new direction, with the potential to improve risk prediction and advance more precise, individualized breast cancer risk assessment.
In May 2025, Clairity Breast became the first AI mammography-based risk assessment tool to receive De Novo authorization from the FDA. The platform analyzes a standard, routine screening mammogram at the pixel level to identify subtle tissue patterns that human eyes cannot see. In a cohort study published in JAMA Network Open, Clairity’s deep-learning risk model Mirai outperformed breast density in estimating risk of future breast cancer and stratified false-negative screening results across risk groups. The researchers concluded by recommending a transition from density-based policy triggers toward more precise image-derived risk models to guide access to supplemental imaging like MRI.
More recently, an abstract presented at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting delved into the clinical implications of incorporating this new AI technology and how it may impact clinical care, comparing it to traditional clinical risk models like Tyrer-Cuzick. Their findings indicated modest overall agreement between assigned risk assessment categories. The AI model demonstrated the potential to reduce the intensity of short-term breast cancer screening for most patients in a high-risk clinic by providing a more accurate, shorter-term risk estimate that did not depend on historical data.
Additional AI-based platforms are in development, including one by Lunit. A large validation study for that model, also published in JAMA Network Open, explored its accuracy and applicability across diverse groups of women and in multiple settings for estimating future risk of breast cancer. The researchers found that the method performed consistently across racial and ethnic groups and provided better predictive performance than a single screening mammogram time point.
In all, these AI-based platforms appear to outperform our standard clinical risk models.
At AdventHealth, I am working with Wes Walker, MD, AdventHealth's Vice President of Genomics and Precision Health , to continually build from what we have already learned through our Genomics Risk Assessment for Cancer and Early Detection (GRACE) program, with the goal of informing short-term risk and making more personalized breast screening decisions.
All the studies to date suggest that AI may help us move beyond broad population-based estimates and toward more precise, individualized risk prediction. As these tools continue to mature, they have the potential to become an important component of precision breast cancer screening, helping clinicians identify which women need intensified surveillance and which can avoid unnecessary testing.
But better risk prediction alone is not enough.
Moving Beyond Imaging: Building Stronger Education and Comprehensive High-Risk Care
Identifying risk is only the first step. The challenge is helping patients understand what that risk means and translating the information they have received into meaningful prevention strategies.
However, many providers feel unequipped to counsel patients on breast density implications, genetic testing and proper MRI usage as well as behavior modification and prevention medications. We know lifestyle factors can contribute significantly to breast cancer risk. Exercise, nutrition, alcohol use and obesity all remain under-addressed, and they need to be part of the conversation.
The demand for additional support is there. In the WISDOM trial, uptake for preventive interventions was 96% for the breast health decision tool or a breast health specialist consult, and most patients who were polled afterward made changes -- whether medication or lifestyle. That is huge.
Unfortunately, there are very few clinicians formally trained in high-risk breast medicine, and no standard specialty pathway exists. As a result, high-risk care and survivorship are often fragmented and under-resourced. My own path to this specialty was unique. I was fortunate to have the opportunity to complete a Mayo Clinic Scholar Year where I rotated through both Dana Farber Cancer Institute and Yale Cancer Center before leading the high-risk and survivorship clinic at Mayo Clinic Florida and now pioneering another at AdventHealth. I hope we will see access to more formal and even accredited training programs in both high-risk and survivorship care for physicians in the years to come.
Ushering in a New Era in Personalized Breast Cancer Prevention
We are entering a pivotal moment in breast cancer prevention. For decades, screening recommendations have been driven largely by age and lifetime risk estimates. Today, advances in genetics, risk modeling and artificial intelligence are giving us the ability to evaluate risk with far greater precision and to adapt screening strategies as that risk evolves.
Yet technology alone is not the answer. As clinicians, our responsibility extends beyond identifying risk. We must help patients understand it, contextualize it and act on it. That means discussing genetic testing when appropriate, addressing modifiable lifestyle factors, considering preventive therapies and ensuring patients receive the level of screening that matches their individual needs.
The future of breast cancer prevention is about screening smarter. Some women will benefit from earlier imaging, supplemental MRI or genetic testing. Others can be spared the unnecessary procedures and anxiety. Both outcomes represent success.
Risk changes. Our patients change. The way we assess and manage breast cancer risk must evolve as well.
By embracing dynamic risk assessment, investing in high-risk breast expertise and continuing to advance personalized prevention strategies, we have an opportunity to detect cancers earlier, reduce unnecessary interventions and ultimately improve whole-person outcomes for the women we serve.
