3-D breast imaging technique increased cancer detection, decreased recalls

The addition of the 3-D breast imaging technique tomosynthesis to digital mammography increased the rate of cancer detection and decreased the rate of recalls, according to results of a retrospective analysis.

Sarah M. Friedewald, MD, of Advocate Lutheran General Hospital in Park Ridge, Ill., and colleagues compared mammography data from 13 centers before and after the initiation of tomosynthesis screenings. The analysis included 454,850 examinations — 281,187 involved digital mammography alone and 173,663 involved digital mammography plus tomosynthesis.

Researchers reported 1,207 patients (0.43%) who underwent digital mammography alone were diagnosed with breast cancer, and 950 patients (0.55%) who underwent digital mammography plus tomosynthesis were diagnosed with cancer.

The model-adjusted recall rate per 1,000 screens improved with the addition of tomosynthesis (107 vs. 91; difference, –16; 95% CI, –18 to –14).

The combination of digital mammography and tomosynthesis was associated with an increased model-adjusted rates of biopsy (19.3 vs. 18.1; difference, 1.3; 95% CI, 0.4-2.1), cancer detection (5.4 vs. 4.2; difference, 1.2; 95% CI, 0.8-1.6) and invasive cancer detection (4.1 vs. 2.9; difference, 1.2; 95% CI, 0.8-1.6).

The model-adjusted rate of in situ cancer detection was 1.4 (95% CI, 1.2-1.6) for both methods.

Overall, the positive predictive value for recall increased from 4.3% to 6.4% with tomosynthesis (difference, 2.1%; 95% CI, 1.7-2.5). The biopsy positive predictive value increased from 24.2% to 29.2% with the 3-D technique (difference, 5.0%; 95% CI, 3.0-7.0).

The study was limited by its retrospective nature, the potential for selection bias given that participating sites converted incrementally to tomosynthesis, and the fact population-level data made it impossible to evaluate the number of repeat examinations, Friedewald and colleagues wrote.

Etta D. Pisano

“While implementation of tomosynthesis in our study was associated with a reduction in recall rate from screening, follow-up data were not available that would allow evaluation of false-negative result rates,” the researchers wrote. “The study did not assess clinical outcomes, so whether the increase in cancer detection rates is of benefit is not known.”

Although the associations with the addition of tomosynthesis are positive, it remains unclear whether the technique should replace digital mammography, Etta D. Pisano, MD, of Medical University of South Carolina, and Martin J. Yaffe, PhD, of University of Toronto’s Ontario Institute for Cancer Research, wrote in an accompanying editorial.

“The continuing controversy surrounding the most effective strategy for deploying the various available technologies continues unabated, and clear consensus is lacking on when to screen, how often, and with what tools, or even which screen-detected cancers could be managed more conservatively,” Pisano and Yaffe wrote. “Only an appropriately powered multisite clinical trial of modern technology can answer the remaining questions definitively. The time is now for the NIH to fund such a much-needed trial to address many of the remaining issues about breast cancer screening.”

For more information:

Friedewald SM. JAMA. 2014;doi:10.1001/jama.2014.6095.
Pisano ED. JAMA. 2014;311:2488-2489.

Disclosure: The researchers report consultant roles and scientific advisory board membership with Hologic; and board membership with Wellmark Blue Cross/Blue Shield for Iowa and South Dakota. See the commentary for a full list of Pisano’s and Yaffe’s relevant financial disclosures.