Alfonso Frigerio, a radiologist from Turin, acknowledged that breast cancer screening is among the most complex topics in medicine, but nevertheless believed the overwhelming evidence from the literature was both sound and conspicuous: screening had significantly reduced mortality from breast cancer. He relied on two key sources – the Marmot Report, published in the British Journal of Cancer in 2013, and the EUROSCREEN working group, published in the Journal of Medical Screening in 2012 – and insisted the Canadian trial was an outlier. He concluded, “An extraordinary solution might come in the near future with the introduction of newer technologies such as digital breast tomosynthesis (DBT) and automated breast ultrasound.”
Arguing against mammographic screening was Tony Miller, a physician from Toronto and primary author of the controversial article about the Canadian trial published in the British Medical Journal in 2014. In his opinion, the Swedish two-county trial was the outlier and much of his presentation focused on the “flaws” in this study. He was adamant that the findings of the Canadian trial were beyond reproach, concluding, “the benefit from breast cancer screening derives from the earlier detection of more advanced disease, not the early detection of impalpable cancers, providing modern therapy is used.”
A question-and-answer session, led by Francesco Sardanelli of Milan and Nehmat Houssami of Sydney, explored issues such as: Do we really have reliable estimates for overdiagnosis? Are there technical improvements to be implemented in screening programmes? What are the societal and ethical implications of stopping mammography screening?
Finally, Sardanelli asked the audience – who were mostly radiologists – to vote on various multiple-choice questions. Over 85% said that screening mammography saved lives, 70% felt that a recall rate of 8% was acceptable, while 50% acknowledged that overdiagnosis from screening was a significant problem but was counterbalanced by mortality reduction. When asked to predict which screening modalities would be used in the future, 42% said DBT would dominate in 2025 but almost 80% thought other technologies would be available by 2035.