Comparison of Full-Field Digital Mammography With Digital Breast Tomography for Screening Call-Back Rates

Comparison of Full-Field Digital Mammography With Digital Breast Tomosynthesis Image Acquisition in Relation to Screening Call-Back Rate

This multicenter trial using Hologic digital mammography units will evaluate the specificity of 2-D full field digital mammography (FFDM) versus a combination of 2-D and 3-D tomosynthesis imaging in breast cancer screening. Specificity, in this study, will be measured by the participant call-back rate by each modality. Varying combinations of 2-D mammography and tomosynthesis projections will be evaluated to optimize the screening paradigm and limit radiation exposure when tomosynthesis is incorporated. Both prospective and retrospective imaging data will be assessed.

Hypothesis: Digital breast tomography (DBT) will improve the specificity of breast cancer screening as measured by a reduction in the call-back rate while maintaining the sensitivity of cancer detection. This improved accuracy will be achieved by the optimization of the imaging sequence and number of views obtained at a capped radiation dose in the combined DBT and 2-D screening sequence.

Study Overview

• Status: Completed
• Conditions: Breast Neoplasms
• Intervention / Treatment: Device: Screening Tomosynthesis; Device: Diagnostic Tomosynthesis

Detailed Description

Asymptomatic women 25 years and older with no history of breast cancer will be recruited from a prospective population of patients scheduled for screening mammography (Group A). A similar population of women called back from screening for 2-D FFDM-detected abnormalities will also be recruited to provide an enriched population of true-positive and false-positive 2-D FFDM and tomosynthesis cases (Group B). Pregnant women, women unable to tolerate compression of the breast associated with mammography, women with implants, and women with breasts too large to accommodate adequate positioning of the breast for DBT are excluded from trial participation.

A total of 550 participants will be recruited--500 women will enroll for collection of prospective imaging data in this trial (Group A); 50 additional participants, recalled for diagnostic assessment after positive screening findings, will be recruited for DBT imaging data collection and retrospective image analysis (Group B). Participating institutions for this trial will be clinical research institutions in Pennsylvania with Hologic tomosynthesis units.

• Study Type: Interventional
• Enrollment (Actual): 558
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • Pennsylvania
    • Philadelphia, Pennsylvania, United States, 19118
      • Hospital of University of Pennsylvania
    • Philadelphia, Pennsylvania, United States, 19141
      • Albery Einstein Medical Center

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 25 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

• Women 25 years of age or older;
• No history of breast cancer;
• Group A only: Asymptomatic and scheduled for screening mammography;
• Group B only: Asymptomatic and recalled for diagnostic testing due to positive findings on recent screening using FFDM, completed within 30 days prior to registration (BI-RADS 0: additional imaging needed);
• Willing to provide a written informed consent.

Exclusion Criteria:

• Pregnancy or intent to become pregnant;
• Unable or unwilling to tolerate compression associated with mammography;
• Breast implants;
• Breasts too large to allow for adequate positioning for the DBT examination;
• Group B only: Patients with FFDM taken at screening who are unwilling or unable to submit images to ACRIN;
• Group B only: Unwilling to undergo tomosynthesis on both breasts as well as potentially additional diagnostic imaging based on tomosynthesis findings;
• Unable or unwilling to complete screening and (as necessary) diagnostic imaging at same facility;
• Tomosynthesis or mammography within 11 months prior to registration (excluding the screening mammography required for Group B).

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Screening
• Allocation: Non-Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Group A: Screening; Group A comprises 500 asymptomatic women with no history of breast cancer who are scheduled for routine screening of the breasts with FFDM.	Device: Screening Tomosynthesis; Tomosynthesis imaging sets (limited tomosynthesis set and then a sequential read with the low-dose CC view added for the tomosynthesis plus set).; Other Names: Hologic; Dimension
Experimental: Group B: Diagnostic Enriched Population; Approximately 50 asymptomatic women with no history of breast cancer who have been informed of positive (abnormal) findings from a recent (within 30 days) FFDM screening will be recruited to Group B prior to their diagnostic imaging (e.g., diagnostic FFDM and/or ultrasound and/or other).	Device: Diagnostic Tomosynthesis; Tomosynthesis imaging sets (limited tomosynthesis set and then a sequential read with the low-dose CC view added for the tomosynthesis plus set).; Other Names: Hologic; Dimension

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Recall Rate in Group A Participants	Each participant will undergo both FFDM and DBT examinations. Each of the two exams will be interpreted separately by readers at the participating sites and any positive finding for diagnostic followup will be considered a recall.	at the first imaging session

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Accuracy Based on Pathology Results	Malignant pathologic results will be used to determine Accuracy (True Positive[Sensitivity], True Negative [Specificity], Positive and Negative Predictive values[PPV,NPV]) of FFDM and limited DBT set (digital breast two-view tomosynthesis with low-dose MLO) [Groups A and B]. A Positive screening read had an overall site recommendation of additional Workup A Positive diagnostic read is defined as either Breast having a BIRADs score of 4 or 5 on the diagnostic read (where high BIRADS indicates higher confidence of malignancy)	at the first imaging session, at the Diagnostic workup (within 30 Days after Screening), and 1 year (pathologic outcome)
Accuracy, Based on Pathology Results, by Lesion Classification	Malignancy determined by pathology and follow-up will be used to determine Accuracy (True Positive[Sensitivity], True Negative [Specificity], Positive and Negative Predictive values[PPV,NPV]) of FFDM and limited DBT set (digital breast two-view tomosynthesis with low-dose MLO) with results reported by lesion-type characterization (calcification-only lesions versus soft-tissue lesions,)[Groups A and B]. A Positive screening read had an overall site recommendation of additional screening A Positive diagnostic read is defined as either Breast having a BIRADs score of 4 or 5 on the diagnostic read.	at the first imaging session, at the Diagnostic workup (within 30 Days after Screening), and 1 year (pathologic outcome)
Callback Rate of Two-view Limited Tomosynthesis Set (With Low-dose MLO View Alone) With the Tomosynthesis Plus Set (Low-dose MLO View Plus Addition of Low-dose CC View)	Sequential interpretation results for call- back from two-view limited tomosynthesis set (with low-dose MLO view alone) and tomosynthesis plus set (low-dose MLO view plus addition of low-dose CC view)	At the Diagnostic workup (within 30 Days After Screening) and up to 18-months post-screening (Pathology Measure)
Radiation Dose	Radiation dose by modality (FFDM & DBT2D/3D) Dose obtained from the data contained in the acquisition DICOM headers.	at the first imaging session
Regression Model Parameters for Factors Effecting Radiation Dose	a mixed Regression Model to identify the determinants of participant radiation dose, including covariates (X) such as: kVp, mAs, target/filter combination, and breast thickness and compression. Dose = B1*X1 + B2*X2 + B3*X3 + ... The outcome measures is are the estimated coefficients (B) of the covariate (X) in the mixed regression model. The coefficient (B) of the term (X) represents the change in the mean response (dose) for one unit of change in that term. If the coefficient is negative, as the term increases, the mean value of the response decreases. If the coefficient is positive, as the term increases, the mean value of the response increases. Note: the TUNGSTEN/SILVER categorical target/filter combination is used as the reference category and will have a parameter value of 0.	at the first imaging session
Identification and Characterization of Lesion(s) Using Screening Tomosynthesis (Group A)	Group A, screening tomosynthesis, cases will be evaluated using 2 methodologies. Readers will read each screening exams using either (1) the limited tomosynthesis set (with low-dose MLO view alone) or (2) the tomosynthesis plus set (low-dose MLO view plus addition of low-dose CC view) and classify the lesions they are able to visualize per SOC. SOC Lesions characterizations include: Calcification Only lesion , Mass, Focal Asymmetry, Architectural Distortion, Global Asymmetry, Calcification) The characterizations will be tabulated by whether or not each lesion was classified in the given category.	At the Diagnostic workup (within 30 Days After Screening)

Collaborators and Investigators

• Sponsor: American College of Radiology
• Collaborators: Pennsylvania Department of Health
• Investigators: Study Chair: Emily F. Conant, MD, Hospital of University of Pennsylvania

Publications and helpful links

General Publications

• Ray S, Chen L, Keller BM, Chen J, Conant EF, Kontos D. Association between Breast Parenchymal Complexity and False-Positive Recall From Digital Mammography Versus Breast Tomosynthesis: Preliminary Investigation in the ACRIN PA 4006 Trial. Acad Radiol. 2016 Aug;23(8):977-86. doi: 10.1016/j.acra.2016.02.019. Epub 2016 May 25.
• Chen L, Ray S, Keller BM, Pertuz S, McDonald ES, Conant EF, Kontos D. The Impact of Acquisition Dose on Quantitative Breast Density Estimation with Digital Mammography: Results from ACRIN PA 4006. Radiology. 2016 Sep;280(3):693-700. doi: 10.1148/radiol.2016151749. Epub 2016 Mar 22.

Helpful Links

• For ACRIN PA 4006 protocol-specific information, visit us online at ACRIN.ORG.

Study record dates

Study Major Dates

• Study Start (Actual): January 4, 2011
• Primary Completion (Actual): June 1, 2012
• Study Completion (Actual): December 31, 2013

Study Registration Dates

• First Submitted: November 5, 2010
• First Submitted That Met QC Criteria: November 8, 2010
• First Posted (Estimated): November 9, 2010

Study Record Updates

• Last Update Posted (Actual): March 25, 2025
• Last Update Submitted That Met QC Criteria: February 12, 2025
• Last Verified: February 1, 2025

More Information

Terms related to this study

• Keywords: breast cancer; diagnosis; screening; breast neoplasms; comparison; diagnostics; mammography; tomosynthesis; radiation dose; high-risk; Philadelphia; dense breasts; digital mammography; Hologic; Pennsylvania; breast study; mammo; tomo; mammary
• Additional Relevant MeSH Terms: Neoplasms by Site; Neoplasms; Skin Diseases; Breast Diseases; Breast Neoplasms
• Other Study ID Numbers: ACRIN PA 4006