Stephen W. Duffy, Abstract Co-Author: Nothing to Disclose

Oliver William Edmund Morrish MSc, Abstract Co-Author: Nothing to Disclose

Lorraine Tucker, Abstract Co-Author: Nothing to Disclose

Richard Black MS, Abstract Co-Author: Nothing to Disclose

Fiona Jane Gilbert MD, Presenter: Speaker, Bracco Group Research Grant, GlaxoSmithKline plc Research Grant, General Electric Company

To estimate the extent to which automatic measures of density are predictive of breast cancer risk, and in particular risk of the potentially fatal cancers which are prime targets of early detection.

In a retrospective study of breast tomosynthesis in addition to standard full field digital mammography, we had Volpara automated volumetric density on 7019 subjects (1157 cancers) and Quantra automated volumetric density on 7005 subjects (1156 cancers). Data were analysed using logistic regression.  

These was a significant (p<0.001) 3% (95% CI 1-5%) increased risk of breast cancer per 10 cm3 increase in fibroglandular (dense) tissue measured by Volpara, and a significant (p<0.001) 2% (95% CI 1-3%) increase per 10 cm3 as measured by Quantra. In both cases, the effect was stronger in invasive grade 3 cancers than in grade 2 or grade 1. Risk of grade 3 breast cancer increased by 4% per 10 cm3 increase in dense tissue measured by Volpara (95% CI 1-7%) and by 3% per 10 cm3 as measured by Quantra (95% CI 1-5%). The effect of neither density measure on risk varied substantially by lymph node status. The effect of Volpara density was considerably stronger for risk of invasive tumours of size greater than 20 mm, at 6% increased risk per 10 cm3 increase dense tissue (95% CI 3-9%). The effect of the Quantra measure had a weaker association with tumour size.

Automated volumetric breast density is predictive of breast cancer risk. There is evidence that it has stronger predictive power for potentially fatal large and grade 3 invasive cancers. This suggests that automated volumetric breast density has a potential role in risk stratification and management of breast cancer risk.

Automated calculation of fibroglandular volume appears to indicate risk of developing breast cancer.

Duffy, S, Morrish, O, Tucker, L, Black, R, Gilbert, F, Automated Volumetric Breast Density and Risk of Cancer Stratified by Pathological Attributes.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006134.html