Tag: breast cancer risk

AI Risk Prediction’s Long-Term Value

Posted on by Brian Casey

AI-based calculations of breast cancer risk derived from screening mammograms can track cancer risk as it evolves over time, giving clinicians a longitudinal tool for following patients who might need additional care. A new study in Radiology adds to the growing body of knowledge on AI-based risk analysis. 

Cancer risk prediction has emerged as a promising new application for AI, as exemplified by a study earlier this month in which three commercial AI models for screening mammograms were also able to predict risk as much as six years before diagnosis. 

  • At least one AI model – Clairity Breast from Clairity – has received FDA clearance for image-based risk prediction, with others under review at the agency. 

But most studies of AI-powered breast cancer risk prediction calculate risk at a single point in time. 

  • While that’s useful, a woman’s breast cancer risk can evolve with factors such as breast tissue density, which is known to change over time – thus changing their risk profile. 

So authors of the current study tracked breast cancer risk longitudinally using the Mirai algorithm, an open-source model that’s been validated in previous studies as more accurate than clinical risk prediction models like Tyrer-Cusick and BCRAT.  

  • They retrospectively applied Mirai to 54k women who got mammograms from 2009 to 2019, and compared changes in risk scores between women who developed cancer and those who didn’t. 

Researchers found… 

  • Median risk scores six years before diagnosis changed from 2.1 to 6.6 in women eventually diagnosed with cancer.
  • Risk scores were essentially stable in women who were cancer-free (1.8 to 2.2).
  • Risk scores rose at a higher annual rate longitudinally in those with cancer versus those without (1.13 vs. 0.09 per year).
  • Women in the group who developed cancer tended to be older and had dense breast tissue or a personal or family history of breast cancer. 

Exactly what is the AI detecting if cancer isn’t visible to radiologists reading the mammograms?

  • Most likely, AI is detecting changes in patterns of breast parenchymal tissue that “may precede radiographic detection.” These changes are basically biomarkers that can be used to develop personalized screening intervals, supplemental modalities, and other preventive strategies. 

The Takeaway

The new study on AI-based breast cancer risk prediction advances our understanding of how risk can be calculated far in advance of a cancer diagnosis. It’s easy to see this knowledge put to use with earlier intervention strategies that exemplify the rise of personalized medicine. 

AI for Breast Cancer Risk

Posted on by Jake Fishman

Artificial intelligence may be capable of identifying subtle mammographic signs of breast cancer years before conventional diagnosis, according to a new study published in Radiology. Researchers from Sweden found that three commercially available AI algorithms for mammography screening generated elevated cancer scores as early as 10 years before diagnosis, with detection signals strengthening as diagnosis approached.

Predicting breast cancer risk offers the prospect not only of detecting cancer earlier, but also of tailoring mammography screening to women most likely to benefit from it.

  • Clinical risk calculators like Tyrer-Cuzick and breast density analysis are available, but AI-based algorithms are showing promise by predicting risk from screening mammograms.

In the new study, researchers analyzed 89k mammograms from 31.4k women collected over a 10-year period, drawn from Sweden’s national screening program, where women aged 40-74 undergo biennial mammography interpreted by two radiologists.  

  • During the study period, 12.1k women (39%) were ultimately diagnosed with breast cancer. Three commercially available AI algorithms were used to generate risk scores (Vara AI from Vara, Lunit Insight MMG from Lunit, and MammoScreen from Therapixel). (It’s worth noting all three were originally designed for cancer detection rather than risk prediction.) 

AI scores increased progressively over time in women who later developed cancer, while remaining relatively stable among cancer-free participants…

  • At 90% specificity, AI systems flagged 19%-20% of future breast cancer cases six years before diagnosis.
  • Detection increased to 23%-25% at four years before diagnosis.
  • Performance rose further to 35%-39% at two years before diagnosis.
  • Even 10 years before diagnosis, the systems identified 13%-17% of future cancers.
  • Across all pre-diagnostic examinations, AI achieved AUC values of 0.63-0.67, outperforming mammographic density alone (AUC = 0.57).

The findings suggest that AI tools developed for cancer detection may also have value as early-alert systems for identifying women who could benefit from closer surveillance or supplemental imaging.

  • While prospective validation is still needed, sequential AI scoring may ultimately help identify women who would benefit from supplemental imaging, closer surveillance, or earlier intervention.

The Takeaway

The study adds to growing evidence that mammography AI can extend beyond cancer detection to long-term risk stratification. By identifying subtle imaging patterns years before diagnosis, AI-derived detection scores could provide an additional layer of longitudinal risk monitoring and help guide more personalized screening strategies.

Risk-Based Mammography Screening Returns

Posted on by Brian Casey

The idea of risk-based mammography screening is back with the publication of a new study in JAMA Network Open claiming that some risk-based strategies averted more breast cancer deaths with fewer false positives than age-based criteria. But like a previous paper on risk-based screening, the new findings raise concerns.

The idea behind risk-based screening is to focus healthcare resources on the people who need them most while sparing low-risk individuals from unnecessary medical interventions.

  • But risk-based breast cancer screening needs more clinical validation before it can be adopted broadly. This was tried with the WISDOM study, but researchers found no statistically significant difference in biopsy rates and only a modest reduction in mammograms performed.

A slightly different tack was taken with the new study, which compared conventional age-based biennial screening to a package of risk-based approaches based on a patient’s five-year breast cancer risk as calculated by widely accepted techniques like the Gail model and BCSC calculator.

  • Out of 50 risk-based strategies, nine averted more deaths than biennial age-based screening for women aged 40-74 (both were compared to no screening), and resulted in fewer false-positive recalls.

One such strategy highlighted by the authors used no screening for younger low-risk women, biennial screening for average-risk women, and annual screening for intermediate- and high-risk women, with the following results…

  • 6% more breast cancer deaths averted per 1k women versus conventional screening (7.2 vs. 6.8).
  • 8% fewer false-positive recalls (1,257 vs. 1,365).
  • While other risk-based strategies saw death reductions as high as 7.5 deaths per 1k women and false-positive reductions of 8-23%.

One key thing to note with the new study is its use of biennial screening as the control group, in line with current USPSTF recommendations for women aged 40-74. 

  • But many clinical organizations like ACR, ACOG, SBI, and NCCN recommend annual screening, and the new study’s findings may have been very different if compared to an annual model.

The Takeaway

This week’s findings are generally more supportive of risk-based screening than those of last year’s WISDOM study. But the new paper’s marginal improvement in cancer deaths averted might disappear when compared with annual age-based mammography. And like WISDOM, its use of clinical models for risk prediction may soon be obsolete given rapid developments in AI-based risk assessment. 

Lunit Acquires Prognosia Breast Cancer Risk AI

Posted on by Brian Casey

AI developer Lunit is ramping up its position in breast cancer risk prediction by acquiring Prognosia, the developer of a risk prediction algorithm spun out from Washington University School of Medicine in St. Louis. The move will complement Lunit and Volpara’s existing AI models for 2D and 3D mammography analysis. 

Risk prediction has been touted as a better way to determine which women will develop breast cancer in coming years, and high-risk women can be managed more aggressively with more frequent screening intervals or the use of additional imaging modalities.

  • Risk prediction traditionally has relied on models like Tyrer-Cuzick, which is based on clinical factors like patient age, weight, breast density, and family history.

But AI advancements have been leveraged in recent years to develop algorithms that could be more accurate than traditional models.

  • One of these is Prognosia, founded in 2024 based on work conducted by Graham Colditz, MD, DrPH, and Shu (Joy) Jiang, PhD, at Washington University.

Their Prognosia Breast algorithm analyzes subtle differences and changes in 2D and 3D mammograms over time, such as texture, calcification, and breast asymmetry, to generate a score that predicts the risk of developing a new tumor.

Prognosia built on that momentum by submitting a regulatory submission to the FDA, and the application received Breakthrough Device Designation.

  • In conversations with The Imaging Wire, Colditz and Jiang believe AI-based estimates like those of Prognosia Breast will eventually replace the one-size-fits-all model of breast screening, with low-risk women screened less often and high-risk women getting more attention.

Colditz and Jiang are working with the FDA on marketing authorization, and once authorized Prognosia’s algorithm will enter a segment that’s drawing increased attention from AI developers.

  • The two will continue to work with Lunit as it moves Prognosia Breast into the commercialization phase and integrates the product with Lunit’s own offerings like the RiskPathways application in its Lunit Breast Suite and technologies it accessed through its acquisition of Volpara in 2024. 

The Takeaway

Lunit’s acquisition of Prognosia portends exciting times ahead for breast cancer risk prediction. Armed with tools like Prognosia Breast, clinicians will soon be able to offer mammography screening protocols that are far more tailored to women’s risk profiles than what’s been available in the past. 

Get every issue of The Imaging Wire, delivered right to your inbox.