Interval and Consecutive Round Breast Cancer after Digital Breast Tomosynthesis and Synthetic 2D Mammography versus Standard 2D Digital Mammography in BreastScreen Norway

Research output: Contribution to journalArticle

Abstract

Background: Screening that includes digital breast tomosynthesis (DBT) with two-dimensional (2D) synthetic mammography (SM) or standard 2D digital mammography (DM) results in detection of more breast cancers than does screening with DM alone. A decrease in interval breast cancer rates is anticipated but is not reported. Purpose: To compare rates and characteristics of (a) interval breast cancer in women screened with DBT and SM versus those screened with DM alone and (b) screen-detected breast cancer at consecutive screenings with DM. Materials and Methods: This prospective cohort study from BreastScreen Norway included women screened with DBT and SM (study group) or DM alone (control group) between February 2014 and December 2015 (baseline). All women, except nonattendees, women with breast cancer, and those who exceeded the upper age limit, were consecutively screened with DM after 2 years. Interval breast cancer, sensitivity, and specificity were estimated for women screened at baseline. Recall, screen-detected breast cancer, and positive predictive value were analyzed for consecutively screened women. A χ2 test, t test (P < .001 after Bonferroni correction indicated a significant difference), and binomial regression model were used to analyze differences across groups. Results: A total of 92 404 women who underwent baseline screening (mean age, 59 years ± 6 [standard deviation]) were evaluated; 34 641 women in the study group (mean age, 59 years ± 6) were screened with DBT and SM and 57 763 women in the control group (mean age, 59 years ± 6) were screened with DM. A total of 26 474 women in the study group (mean age, 60 years ± 5) and 45 543 women in the control group (mean age, 60 years ± 5) were consecutively screened with DM. Rates of interval breast cancer were 2.0 per 1000 screened women in the study group and 1.5 per 1000 screened women in the control group (P = .12). No differences in histopathologic characteristics of interval breast cancer were observed. In the consecutive screening round, rates of screen-detected breast cancer were 3.9 per 1000 screened women (study group) and 5.6 per 1000 screened women (control group) (P = .001). Rates of histologic grade 1 invasive cancer were 0.5 per 1000 screened women (study group) and 1.3 per 1000 screened women (control group) (P = .001). Conclusion: No differences in interval breast cancer rates or tumor characteristics were observed in women screened with DBT and SM compared with women screened with DM. Higher rates of low-grade screen-detected tumors were observed in the control group at consecutive screening.

Details

Authors
  • Tone Hovda
  • Åsne S. Holen
  • Kristina Lång
  • Judy Lynn Albertsen
  • Hilde Bjørndal
  • Siri H. B. Brandal
  • Kristine Kleivi Sahlberg
  • Per Skaane
  • Pål Suhrke
  • Solveig Hofvind
Organisations
External organisations
  • University of Oslo
  • Cancer Registry of Norway Institute of Population-Based Cancer Research
  • Vestfold Hospital
  • Vestre Viken Hospital Trust
  • Oslo university hospital
  • Oslo Metropolitan University
  • ETH Zürich
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Radiology, Nuclear Medicine and Medical Imaging
  • Cancer and Oncology
Original languageEnglish
Pages (from-to)256–264
Number of pages9
JournalRadiology
Volume294
Issue number2
Early online date2019 Dec 10
Publication statusPublished - 2020 Feb
Publication categoryResearch
Peer-reviewedYes