Adding Tomosynthesis to Digital Mammography Decreased Recall Rate, Increased Cancer Detection Rate in Breast Cancer Screening


Get Permission

Sarah M. Friedewald, MD

Addition of tomosynthesis to digital mammography was associated with a decrease in recall rate and an increase in cancer detection rate. Further studies are needed to assess the relationship to clinical outcomes.

—Sarah M. Friedewald, MD, and colleagues

In a large multicenter study reported in JAMA, Sarah M. Friedewald, MD, of the Advocate Lutheran General Hospital, Center for Advanced Care, Park Ridge, Illinois, and colleagues found that use of tomosynthesis plus digital mammography reduced the recall rate and increased the cancer detection rate in breast cancer screening compared with digital mammography alone.1

Tomosynthesis provides image acquisition from an x-ray source that moves over an arc of excursion and permits reconstruction into thin slices to minimize the effect of overlapping breast structures. The device can be used simultaneously with conventional digital mammography. Single-institution studies have indicated that tomosynthesis plus mammography increases cancer detection and reduces false-positive rates vs mammography alone.

Study Details

This retrospective study compared performance of breast cancer screening before and after introduction of tomosynthesis at 13 U.S. institutions over two periods. Period 1 consisted of 1 full year of screening with digital mammography alone that ended on the date of tomosynthesis introduction at each institution. Period 2 consisted of screening with tomosynthesis plus digital mammography until December 31, 2012. The start dates of period 2 ranged from March 2011 to October 2012.  The average duration of period 2 was 17 months (range, 3–22 months).

For tomosynthesis, all 13 sites used the Selenia Dimensions device (Hologic), the only U.S. Food and Drug Administration–approved device at the start of the study. Outcome measures consisted of recall rate for additional imaging, biopsy rate, cancer detection rate, and positive predictive values for recall and biopsy. Outcomes were assessed using mixed models adjusting for site as a random effect.

Totals of 281,187 digital mammograms and 173,663 tomosynthesis/digital mammograms were evaluated by 139 radiologists. The volume of examinations across institutions ranged from 4,801 to 53,181 for digital mammography alone and from 2,613 to 34,119 for tomosynthesis/digital mammography. Mean ages were 57.0 years in the digital mammography group and 56.2 years in the tomosynthesis/digital mammography group.

Recall, Biopsy, and Detection Rates

Among women undergoing digital mammography alone, 29,726 were recalled, 5,056 underwent biopsy, and cancer was diagnosed in 1,207, including invasive cancer in 815 and ductal carcinoma in situ in 392. Among women undergoing tomosynthesis/digital mammography, 15,541 were recalled, 3,285 underwent biopsy, and cancer was diagnosed in 950, including invasive cancer in 707 and ductal carcinoma in situ in 243.

Model-adjusted rates per 1,000 screens for digital mammography vs tomosynthesis/digital mammography were: 107 vs 91 for recalls (difference = −16, 95% confidence interval [CI] = −18 to −14, P < .001); 18.1 vs 19.3 for biopsies (difference = 1.3, 95% CI = 0.4 to 2.1, P = .004), 4.2 vs 5.4 for cancer detection (difference = 1.2, 95% CI = 0.8 to 1.6, P < .001), 2.9 vs 4.1 for invasive cancer detection (difference = 1.2, 95% CI = 0.8 to 1.6, P < .001), and 1.4 vs 1.4 for ductal carcinoma in situ detection (difference = 0.0, 95% CI = −0.2 to 0.2, P = .95).

Changes per 1,000 screens by study site with tomosynthesis/digital mammography vs digital mammography ranged from −42 to 18 for recalls, −6.3 to 10.5 for biopsies, −1.5 to 3.7 for cancers detected, −0.9 to 3.0 for detection of invasive cancers, and −0.7 to 0.7 for detection of ductal carcinoma in situ.

Positive Predictive Values

The positive predictive value for digital mammography vs tomosynthesis/digital mammography was 4.3% vs 6.4% for recalls (difference = 2.1%, 95% CI = 1.7%–2.5%, P < .001) and 24.2% vs 29.2% for biopsies (difference = 5.0%, 95% CI = 3.0%–7.0%, P < .001). Changes in positive predictive value by study site with tomosynthesis/digital mammography vs digital mammography ranged from −3.2% to 4.8% for recalls and from −6.9% to 27.8% for biopsies.

The investigators concluded, “Addition of tomosynthesis to digital mammography was associated with a decrease in recall rate and an increase in cancer detection rate. Further studies are needed to assess the relationship to clinical outcomes.” ■

Disclosure: The study was supported by a grant from the National Cancer Institute and by Hologic. For full disclosures of the study authors, visit jama.jamanetwork.com.

Reference

1. Friedewald SM, Rafferty EA, Rose SL, et al: Breast cancer screening using tomosynthesis in combination with digital mammography. JAMA 311:2499-2507, 2014.

 


Related Articles

Breast Cancer Screening Using Tomosynthesis in Combination With Digital Mammography

In a study reported in JAMA and reviewed in this issue of The ASCO Post, Friedewald and colleagues1 showed that the addition of tomosynthesis to digital mammography2 resulted in a decrease in the screening recall rate3 and an increase in the cancer detection rate.4,5 This retrospective analysis of...


Advertisement

Advertisement



Advertisement