Changing Paradigms in Breast Cancer Screening

New imaging approaches such as tomosynthesis and abbreviated MRI are suitable to overcome the limitations of mammography. Experts believe that these possibilities should ultimately be integrated into screening strategies tailored to personal risk.

While mammography screening has been shown to be capable of reducing breast cancer mortality by an estimated 25 to 40%, breast cancer remains the leading cause of cancer death among women worldwide.[1] Even in the best existing screening programs, some cancers go undetected, particularly due to the limited sensitivity of digital mammography, the current standard screening method. On the other hand, screening can also lead to overtreatment, that is, therapies that would probably not have been necessary because the detected tumor grows too slowly to be clinically relevant. Many experts agree that these shortcomings make more advanced approaches to breast care necessary.
In fact, the possibilities for better breast cancer detection have grown considerably, as highlighted at this year’s European Congress of Radiology (ECR) in Vienna. In particular, more sensitive imaging methods as well as personalized screening strategies could improve detection performance while being patient friendly and efficient. At the same time, it seems obvious that no single solution can be implemented for all countries or settings, and that it is necessary to continue to develop breast cancer screening.
 

An important incremental step towards this goal is the use of tomosynthesis, also known as 3D mammography. According to a recent trial in a Swedish screening population, the method clearly allows more cancerous lesions to be detected than traditional mammography – even if the breast is only X-rayed in one projection (as opposed to the traditional two). This approach (one-view tomosynthesis) seems to achieve an equally good detection performance as more extensive tomosynthesis examinations such as the combination of two-view mammography with two-view tomosynthesis, while limiting the necessary radiation dose.[2]

The procedure also requires less compression of the breast compared to mammography, as Sophia Zackrisson, lead author of the study, underlined in her presentation at the ECR. This makes the examination more comfortable and could increase participation in screening.

While tomosynthesis provides higher accuracy than 2D mammography, it also yields considerably more images, and reading time increases. AI-supported reading might therefore prove to be important for the procedure to become widely feasible in everyday practice. In addition, it is not yet clear whether tomosynthesis can find not only a larger total number of tumors compared to mammography, but also more fast-growing lesions known as interval carcinomas. "It is also difficult with tomosynthesis to find the aggressive cancers," conceded Zackrisson.

Multifocal intraductal Carcinoma (IDC) and Papillomatosis
A 41-year-old female patient with a palpable lump in the left breast. TiCEM shows very good correlation with MRI and was even better than DBT in the preoperative assessment of breast cancer.
Courtesy of Prof. Luis Pina, MD, University of Navarra, Pamplona/Spain.

Exactly this should be one of the main strengths of MRI. As radiologist Christiane Kuhl from the University of Aachen, Germany, emphasized at the ECR, contrast-enhanced MR imaging of the breast has not only a significantly higher sensitivity than mammography in a wide range of age and risk groups, but also finds a large portion of interval carcinomas. These are particularly fast-growing lesions that remain undetected in current screening programs and are instead diagnosed between two regular appointments.
According to Kuhl, contrast-enhanced MRI, unlike traditional mammography or tomosynthesis, maps the angiogenic and growth potential of a cancer. “MRI is probably the most powerful breast cancer detection method we have right now,” Kuhl states.

So far, high costs, long examination times, and limited availability have prevented the broad use of breast MRI.[3] However, this could change with shortened imaging protocols. In a pilot study, a team led by Kuhl was able to show that an abbreviated protocol reduces the MRI scan time from 17 to three minutes and also considerably shortens the reading time, while maintaining diagnostic accuracy.[4]
Such time- and cost-saving “keep it short and simple” approaches are increasingly seen as a new opportunity for screening.[5,6] “Adaptation of an abbreviated MRI protocol significantly increases patient throughput and may allow MRI to serve as a stand-alone breast cancer screening tool,” write the authors of a recent review on the topic.[7]

Ultimately, these advanced imaging approaches should be integrated into personalized screening strategies, experts believe. The basic idea is that women will be screened depending on their individual risk, which could make it more reliable to detect cancers in women with an increased disease probability while saving many healthy women unnecessary examinations.
“Of course, one-size-fits-all approaches are easier to implement, but there is a growing intent to do better,” comments Mireille Broeders, screening specialist at Radboud University Medical Center, the Netherlands.

In particular, recent risk prediction models that take into account not only classic risk factors such as age and family history but also breast density and various genetic markers, could form the basis for graded screening in the future.[8,9] The idea is currently being tested in a multinational study in Europe in which women with a low-risk score receive a mammogram only every four years, while medium-risk and high-risk women undergo mammography and possibly additional ultrasound every one to two years, and very high-risk women are screened annually by mammography and MRI.[10]

Assuming the success of such risk-stratified approaches, Broeders predicts that over the next five to 10 years they could find their way into the everyday practice of established screening programs. However, due to the diverse risk constellations and imaging possibilities, it is to be expected that different countries and health systems will develop somewhat varying frameworks, and that model calculations of screening success will also be used to find the optimal strategy.

Meanwhile, it seems clear that risk communication and counselling is critical for any personalized screening program.[11] “Women do understand the concept and they want to know how it works, but they also want to know what to do about their individual risk,” says Broeders. For example, the risk of breast cancer can be influenced by lifestyle, and hormonally effective drugs such as estrogen receptor modulators or aromatase inhibitors can also play an important role, especially at high risk. Ultimately, risk-based screening should therefore be integrated into a more comprehensive health program, combining optimized breast cancer prevention and detection.

Martin Lindner is a science writer in Berlin, Germany.