Lester Litchfield, Head of Data Science at Volpara Health, talks about the future of artificial intelligence in breast health
Early breast cancer detection leads to the best breast health outcomes for patients — with five-year relative survival rates in the United States falling from 99% for “local” stage cancer to 29% for “distant” stage cancer that has metastasized to other areas of the body Patients.(1) As artificial intelligence (AI) finds its place in the field of medical imaging, its role in early cancer detection is becoming increasingly important. But it can also be used to predict cancer before it occurs and put patients on the path to prevention.
In breast cancer imaging, AI is often used as a decision aid in reading mammograms and characterizing suspicious tissues.(2) However, AI can also help clinicians predict who may be likely to develop cancer through more intensive screening programs and others offers preventive breast health assessment techniques.
AI to automatically, objectively, and accurately assess breast density is commercially available, and researchers are investigating its use as a tool to aid in prediction and prevention.(3)
Assessment of breast density with AI
Breast density refers to the relative amounts of different types of fibrous, glandular, and adipose tissue(4) and is both a strong, independent risk factor for breast cancer(5,6) and a factor affecting mammography sensitivity(7 , 8th)
It is believed that the sensitivity of mammography is influenced by the fact that dense tissue can serve to hide potentially cancerous lesions on an X-ray.(7,8) Nearly 50% of screening-age patients in the United States have dense breasts , a population that could benefit from additional screening.(9)
Traditionally, radiologists visually and subjectively assess a patient’s breast density to place her into one of four categories, as defined by the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS®).(10)
Alternatively, AI can be used to automatically assess the volume of dense tissue present in a patient’s breast on a continuous scale to ensure more accurate identification and consistent application of screening protocols. Research has shown that radiologists using only visual assessment inconsistently assign density categories to patients, and the variability between how different radiologists categorize density is quite high.(11,12)
The Volpara Breast Volumetric Density (VBD) measurement from the Volpara® ScorecardTM has demonstrated consistency and reliability in assessing density in several independent peer-reviewed journal articles (13,14,15,16) and was validated in an independent study from the Year 2021 review which showed it had the strongest scientific evidence of effectiveness in this sector.(17)
Breast cancer risk prediction based on breast density
Not only has the use of Volpara VBD demonstrated its usefulness in the objective assessment of breast density, but it has also been shown to be a good mechanism for personalizing breast healthcare and triaging patients for complementary screening with alternative screening modalities such as MRI.(18)
Changing medical treatment toward personalized screening regimens gives patients with dense breasts a higher chance of successfully detecting breast cancer earlier and preventing late-stage diagnosis, and this begins with a risk assessment.
The Tyrer-Cuzick (TC) risk model assesses a patient’s 10-year and lifetime risk of developing breast cancer based on personal and family history of cancer and other personal risk factors, including her breast density in version 8 of the model. Density inputs to the risk model include the results of visual scoring (BI-RADS 4th edition or visual analog scale) or Volpara VBD (the vendor’s only density scoring technology built into the model).(19)
According to a study in a screening setting in the United States, the use of Volpara VBD in the TC risk model resulted in more women being identified as high-risk than screening alone, making them eligible for additional reimbursed screening .(19)
Breast density and tumor biology
Aside from general risk assessment, breast density measures have been correlated with tumor characteristics and used to unequivocally predict interval and node-positive breast cancer.
In a study at Elizabeth Wende Breast Care (EWBC) in New York, Volpara VBD had the strongest association with tumor size among the 406 types of breast cancer studied. Women with category D dense breasts had a 3.8 times greater risk of being diagnosed with a tumor size greater than 2 cm compared to women with category A or B dense breasts combined.(20)
A second study by EWBC examined 318 breast cancers detected at screening and 100 interval cancers and found that the Volpara® Density GradeTM measure (VDG®), calculated from Volpara VBD, had a stronger association with breast tumor characteristics than BI-RADS density scores. In all cancers, VDG was associated with HER2 status and a tumor size of at least 2 cm, while BI-RADS showed an association with tumor size only in screened cancers. Finally, Volpara VBD had stronger associations with tumor characteristics associated with poorer outcomes than either VDG or BI-RADS.(21)
A recent 2021 study showed that Volpara VBD can strongly predict interval and node-positive breast cancer, with women in the highest dense-volume quartile having a 4.5- to 5-fold increased risk of these forms of breast cancer.(22)
Monitoring of preventive measures based on breast density
In addition, breast density has been studied in the context of various chemopreventive agents (23) along with some other personal and lifestyle factors, some of which may signal new opportunities for cancer prevention. Three recent studies have shown relationships, or lack thereof, between breast density and caffeine, cigarettes, and commonly used medications. No significant associations between smoking or the substances studied provided any evidence of a reduction in breast density (24,25,26).
The role of AI is constantly evolving. For example, many researchers are beginning to develop methods of predicting risk and breast health analyzes from images (27,28) and examining a person’s mammogram images over successive rounds of screening to identify changes over time and determine whether those changes should indicate the need for a specialized pathway of care.(29,30,31) The meaning of these changes is not yet clear, such as distinguishing between early signs of cancer and patterns that indicate risk(32) – but AI is becoming a possibility his Pursue them and help define them.
From reaching patients earlier to raising the standard of care, AI has a promising role to play in the future of breast cancer prediction and prevention. Prevention will only be possible if efforts are made to assess breast cancer risk at an early age so that appropriate screening regimens and other interventions can be offered and used in high-risk patients.
References available upon request