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 screening mammography performance was conducted at 13 sites (8 nonacademic and 5 academic).
Hybrid Study Sites
In this study, 454,850 examinations (281,187 digital mammography alone and 173,663 digital mammography combined with tomosynthesis) were evaluated. The introduction of tomosynthesis at the 13 sites was nonuniform. Two sites made a complete conversion for screening from digital mammography to digital mammography combined with tomosynthesis.
The other 11 sites maintained a hybrid environment, with some patients receiving digital mammography alone while the study patients underwent digital mammography combined with tomosynthesis. At the 11 sites with hybrid screening environments, the possibility of selection bias existed.
At the 11 hybrid sites, there were 245,985 concurrent cases with digital mammography alone during the same time period in which 173,663 screening examinations were performed with digital mammography combined with tomosynthesis. During that period, there was a statistically significant improvement in cancer detection of 0.6 cases per 1,000 examinations (P < .001) along with a decrease in the recall rate of 5.4 cases per 1,000 studies (P < .001) for the women who underwent digital mammography combined with tomosynthesis compared to digital mammography alone.
Recall Rate
The recall rate for screening with digital mammography alone was 10.7%, compared to 9.1% for digital mammography combined with tomosynthesis (P < .001). The recall rate decreased for digital mammography combined with tomosynthesis compared to digital mammography alone at all but two sites—the site with the lowest tomosynthesis volume and the site with the fourth lowest tomosynthesis volume. These two low-volume sites also had the most and the second most radiologists interpreting the studies—20 radiologists at one site and 18 radiologists at the other site. These findings suggest that there is a learning curve for interpreting tomosynthesis studies.
Cancer Detection
The cancer detection rate was 4.2 cancers per 1,000 examinations for digital mammography and 5.4 cancers per 1,000 studies for digital mammography combined with tomosynthesis (P < .001). Twelve of the 13 sites showed increased cancer detection rates with the addition of tomosynthesis. Eleven of the 13 sites simultaneously increased cancer detection rates and decreased recall rates when tomosynthesis was added to digital mammography for breast cancer screening.
Histopathology
When the cancers were classified according to histopathology, the invasive cancer detection rate was 2.9 cancers per 1,000 examinations for digital mammography alone and 4.1 cancers per 1,000 examinations for digital mammography combined with tomosynthesis. The invasive ductal carcinoma detection rate increased from 2.46 cases per 1,000 examinations with digital mammography to 3.27 cases per 1,000 when tomosynthesis was combined with digital mammography.
For invasive lobular carcinoma, the detection rate was 0.27 cases per 1,000 examinations with digital mammography alone and 0.55 cases per 1,000 examinations with digital mammography plus tomosynthesis. For invasive ductal carcinoma combined with invasive lobular carcinoma, the detection rate with digital mammography was 0.14 cases per 1,000 examinations. When tomosynthesis was added to digital mammography, the detection rate for invasive ductal carcinoma combined with invasive lobular carcinoma was 0.17 cases per 1,000 studies.
Twelve of 13 sites demonstrated an increased detection rate for invasive cancer with the addition of tomosynthesis. The single site that did not report an increased detection rate for cancer or invasive cancer with tomosynthesis had the highest number of study radiologists (20) and the lowest number of tomosynthesis examinations (2,613).
The detection rate for ductal carcinoma in situ was 1.4 cases per 1,000 examinations for digital mammography combined with tomosynthesis and for digital mammography alone.
‘Win-Win’
With the addition of tomosynthesis, the investigators, in both academic and nonacademic settlings, recalled fewer patients yet found more cancers. These findings are a real “win-win,” indicating that the relative yield for each recall increases with the introduction of tomosynthesis.
The positive predictive value for recall, the likelihood of cancer diagnoses in women recalled for additional imaging, increased from 4.3% to 6.4% with the addition of tomosynthesis. Although the addition of tomosynthesis led to an increase in the biopsy rate (from 18.1 per 1,000 cases to 19.3 per 1,000 cases), it also led to a 5% increase in the positive predictive value for biopsy, the proportion of cancers found in women undergoing biopsies based on screening-detected findings, from 24.2% to 29.2%.
This study showed that adding tomosynthesis to digital mammography leads to the detection of more invasive cancers without decreasing the detection rate for ductal carcinoma in situ. Tomosynthesis combined with digital mammography should allow for the identification of more subtle and/or obscured masses, which may be invasive cancers, without failing to identify suspicious calcifications, which may represent ductal carcinoma in situ.
The increased invasive cancer detection rate with tomosynthesis is likely to be of great value in optimizing screening outcomes, since the mortality reduction associated with mammographic screening is thought to be based on the identification and appropriate treatment of small, asymptomatic invasive cancers before they have had the opportunity to metastasize. ■
Disclosure: Dr. Whitman reported no potential conflicts of interest.
References
1. Friedewald SM, Rafferty EA, Rose SL, et al: Breast cancer screening using tomosynthesis in combination with digital mammography. JAMA 311:2499-2507, 2014.
2. Pisano ED, Gastonis C, Hendrick E, et al: Diagnostic performance of digital versus film mammography for breast-cancer screening. N Engl J Med 353:1773-1783, 2005.
3. Haas BM, Kalra V, Geisel J, et al: Comparison of tomosynthesis plus digital mammography and digital mammography alone for breast cancer screening. Radiology 269:694-700, 2013.
4. Ciatto S, Houssami N, Bernardi D, et al: Integration of 3D digital mammography with tomosynthesis for population breast-cancer screening (STORM): A prospective comparison study. Lancet Oncol 14:583-589, 2013.
5. Caumo F, Bernardi D, Ciatto S, et al: Incremental effect from integrating 3D-mammography (tomosynthesis) with 2D-mammography: increased breast cancer detection evident for screening centres in a population-based trial. Breast 23:76-80, 2014.
Dr. Whitman is Professor in the Department of Diagnostic Radiology, The University of Texas MD Anderson Cancer Center, Houston.
