• Alberto S Tagliafico, Giovanna Mariscotti, Francesca Valdora, Manuela Durando, Jacopo Nori, Daniele La Forgia, Ilan Rosenberg, Francesca Caumo, Nicoletta Gandolfo, Maria Pia Sormani, Alessio Signori, Massimo Calabrese, and Nehmat Houssami.
• Department of Health Sciences (DISSAL), Radiology Section, University of Genova, Via Pastore, 1-16132, Genova, Italy; Emergency Radiology, Policlinico San Martino, Largo Rosanna Benzi, 10-16132, Genova, Italy. Electronic address: alberto.tagliafico@unige.it.
• Eur. J. Cancer. 2018 Nov 1; 104: 39-46.

BackgroundIncreased risk of breast cancer (BC) and increased risk of an interval BC at mammography screening are associated with high mammographic density. Adjunct imaging detects additional BCs not detected at mammography screening in women with dense breasts.AimThe aim is to estimate the incremental cancer detection rate (CDR) and false-positive recall for each of tomosynthesis and ultrasound, as adjunct screening modalities in women with mammography-negative dense breasts.MethodsA multicentre prospective comparative trial of adjunct screening with tomosynthesis or ultrasound in women with mammography-negative dense breasts (ASTOUND-2) recruited asymptomatic women attending Italian breast screening services. All participants had independently interpreted tomosynthesis and ultrasound. Outcomes were ascertained from excision histopathology or completed assessment. Paired binary data were compared using McNemar's test.ResultsWe recruited 5300 screening participants with median age of 50 (interquartile range 43-79) years who had negative mammography and dense breasts (April 2015-September 2017). Adjunct screening detected 29 additional BCs (27 invasive, 2 in situ): 12 detected on both tomosynthesis and ultrasound, 3 detected only on tomosynthesis, 14 detected only on ultrasound. Incremental CDR for tomosynthesis (+15 cancers) was 2.83/1000 screens (95% confidence interval [CI]: 1.58-4.67) versus ultrasound (+26 cancers) with an incremental CDR of 4.90/1000 screens (95% CI: 3.21-7.19), P = 0.015. Mean size of these cancers was 14.2 mm (standard deviation: 7.8 mm), and six had nodal metastases. Incremental false-positive recall was 1.22% (95% CI: 0.91%-1.49%) and differed significantly between tomosynthesis (0.30%) and ultrasound (1.0%), P < 0.001.ConclusionsUltrasound detected more BCs but caused more false positives than tomosynthesis, underscoring trade-offs in screening outcomes when adjunct imaging is used for screening dense breasts.Copyright © 2018 Elsevier Ltd. All rights reserved.

30 keywords...

hide…