Validation of a new fully automated software for 2D digital mammographic breast density evaluation in predicting breast cancer risk

Paolo Giorgi Rossi,Emilien Gauthier,Donella Puliti,Suzette Delaloge,Camille Baron,Fiona J Gilbert,Elaine F Harkness,Pierpaolo Pattacini,Corinne Balleyguier,André Grivegnée,Valerie Hélin,Paola Mantellini,Olivera Djuric,Stefan Michiels,Andrea Nitrosi,Susan Astley

doi:10.1038/s41598-021-99433-3

Abstract

We compared accuracy for breast cancer (BC) risk stratification of a new fully automated system (DenSeeMammo—DSM) for breast density (BD) assessment to a non-inferiority threshold based on radiologists’ visual assessment. Pooled analysis was performed on 14,267 2D mammograms collected from women aged 48–55 years who underwent BC screening within three studies: RETomo, Florence study and PROCAS. BD was expressed through clinical Breast Imaging Reporting and Data System (BI-RADS) density classification. Women in BI-RADS D category had a 2.6 (95% CI 1.5–4.4) and a 3.6 (95% CI 1.4–9.3) times higher risk of incident and interval cancer, respectively, than women in the two lowest BD categories. The ability of DSM to predict risk of incident cancer was non-inferior to radiologists’ visual assessment as both point estimate and lower bound of 95% CI (AUC 0.589; 95% CI 0.580–0.597) were above the predefined visual assessment threshold (AUC 0.571). AUC for interval (AUC 0.631; 95% CI 0.623–0.639) cancers was even higher. BD assessed with new fully automated method is positively associated with BC risk and is not inferior to radiologists’ visual assessment. It is an even stronger marker of interval cancer, confirming an appreciable masking effect of BD that reduces mammography sensitivity.

Highlights

We compared accuracy for breast cancer (BC) risk stratification of a new fully automated system (DenSeeMammo—DSM) for breast density (BD) assessment to a non-inferiority threshold based on radiologists’ visual assessment
Breast density is being used to stratify women based on their breast cancer risk to decide if they need further imaging assessment in clinical trials, such as ultrasound or M RI5,6, or if they would benefit from a shorter mammography screening interval
5359 2D full-field digital mammography images were available for breast density analysis since some images were performed with mammography equipment (GE 2000D) which is not supported by DenSeeMammo software

Summary

Introduction

We compared accuracy for breast cancer (BC) risk stratification of a new fully automated system (DenSeeMammo—DSM) for breast density (BD) assessment to a non-inferiority threshold based on radiologists’ visual assessment. BD assessed with new fully automated method is positively associated with BC risk and is not inferior to radiologists’ visual assessment It is an even stronger marker of interval cancer, confirming an appreciable masking effect of BD that reduces mammography sensitivity. Breast density is being used to stratify women based on their breast cancer risk to decide if they need further imaging assessment in clinical trials, such as ultrasound or M RI5,6, or if they would benefit from a shorter mammography screening interval. BD evaluation plays a central role in the MyPeBS trial as mammographic density contributes to the risk assessment in the intervention arm and it determines whether women should receive supplemental ultrasound

Objectives

Methods

Results

Conclusion