Computer-Aided Breast Cancer Detection in Women Undergoing Screening Mammography

A Prospective Evaluation of Computer Aided Detection (CAD) in the NHS Breast Screening Programme [CADET II]

RATIONALE: A computer-aided detection program may help doctors find breast cancer sooner, when it may be easier to treat, in women undergoing screening mammography. PURPOSE: This randomized clinical trial is studying how well computer-aided breast cancer detection works in women undergoing screening mammography.

OBJECTIVES: Compare, prospectively, the breast cancer detection rate and recall rate of single reading using computer-aided detection vs standard double reading in women undergoing screening mammography. OUTLINE: This is a prospective, randomized, multicenter study. Participants are randomized to 1 of 3 screening arms. Arm I (double reading): Screening mammograms are independently reviewed by two readers (i.e., radiologist or radiographer), each of whom reports on any abnormalities and decides whether a participant is recalled for further assessment or returned home to routine screening. In the event there is disagreement between the readers' findings, a third reader acts as an arbitrator by rendering an interpretation of the mammogram and making the final decision to recall the participant or return to routine screening. Arm II (single reading with computer-aided detection [CAD]): Screening mammograms are reviewed initially by a single reader who reports on any abnormalities. The reader then re-examines the mammogram with the aid of CAD, a system that uses computer prompts to recall suspicious features or abnormalities that may have been overlooked or previously dismissed as being normal. Based on these evaluations, a recommendation is made by the reader to either recall the participant for further assessment or return home to routine screening. If there is a discrepancy between the reader's interpretation and the CAD findings, another reader may be consulted to review the mammogram. Arm III (double reading followed by single reading with CAD): Screening mammograms are reviewed by double reading as in arm I followed by single reading with CAD as in arm II. Peer Reviewed and Funded or Endorsed by Cancer Research UK PROJECTED ACCRUAL: A total of 30,000 participants will be accrued for this study.

DISEASE CHARACTERISTICS: Currently undergoing 2-view mammography for routine breast cancer screening at any of the following participating National Health Services Breast Cancer Screening Program centers: Manchester Coventry Nottingham Hormone receptor status not specified PATIENT CHARACTERISTICS: Female Menopausal status not specified PRIOR CONCURRENT THERAPY: Not specified

James JJ, Gilbert FJ, Wallis MG, Gillan MG, Astley SM, Boggis CR, Agbaje OF, Brentnall AR, Duffy SW. Mammographic features of breast cancers at single reading with computer-aided detection and at double reading in a large multicenter prospective trial of computer-aided detection: CADET II. Radiology. 2010 Aug;256(2):379-86. doi: 10.1148/radiol.10091899.

Gilbert FJ, Astley SM, Gillan MG, Agbaje OF, Wallis MG, James J, Boggis CR, Duffy SW; CADET II Group. Single reading with computer-aided detection for screening mammography. N Engl J Med. 2008 Oct 16;359(16):1675-84. doi: 10.1056/NEJMoa0803545. Epub 2008 Oct 1.