TOSYMA: Tomosynthesis Bests Mammography in Detecting Invasive Breast Cancers

Detection of invasive breast cancers in women ages 50-69 years was 48% higher with digital breast tomosynthesis plus s2D mammography compared with digital mammography alone, according to results from the randomized, controlled TOSYMA screening study.

Results are published in The Lancet Oncology.

"Two dimensional (2D) full-field digital mammography is the current standard of breast cancer screening. Digital breast tomosynthesis generates pseudo-three-dimensional datasets of the breast from which synthesized 2D (s2D) mammograms can be reconstructed. This innovative approach reduces the likelihood of overlapping breast tissues that can conceal features of malignancy. We aimed to compare digital breast tomosynthesis plus s2D mammography with digital screening mammography for the detection of invasive breast cancer," wrote investigators for the TOSYMA Screening Trial Study Group, led by Walter Heindel, MD, of the Clinic for Radiology and Reference Center for Mammography Munster, University of Munster and University Hospital Munster, Germany.

For this open-label superiority trial, the study authors randomized women ages 50-69 years taking part in a population-wide, mammography screening program in two states in Germany. From July 2018 through December 2020, 99,689 women were randomized to either digital breast tomosynthesis plus s2D mammography or digital mammography. Most were 50-54 years old, and for most, this screening was a regular subsequent round of breast screening.

Heindel et al found that the detection rate of invasive breast cancer via digital breast tomosynthesis plus s2D was significantly higher compared with digital mammography (7.1 cases/1,000 versus 4.8 cases/1,000 women screened, respectively; odds ratio [OR]: 1.48; 95% CI: 1.25-1.75; P<0.0001).

For digital breast tomosynthesis plus s2D mammography, median image acquisition time was similar to that for digital breast mammography, at 220.0 versus 186.0 seconds, respectively. Median image reading times were higher, however, in the digital breast tomosynthesis plus s2D mammography group, at 109.0 versus 54.0 seconds.

"We hypothesize that these longer reading times, the larger sample size of TOSYMA, and the multicenter setting might have contributed to the findings (i.e., the detection rate for invasive breast cancer in the digital breast tomosynthesis plus s2D mammography group of the TOSYMA study was higher than that of the digital mammography group). Sensitive digital breast tomosynthesis reading generally requires more time than digital mammography reading because thorough reading of slices is presumably a prerequisite to increase invasive cancer detection," noted researchers.

Median thickness of the compressed breast was 59.0 mm in both groups, and median glandular doses were also similar, at 1.86 mGy for the digital breast tomosynthesis plus s2D mammography group, compared with 1.36 mGy for the digital mammography group.

Only six adverse events occurred in each group, and no events were serious or device-related. There were 23 device deficiencies in the digital breast tomosynthesis plus s2D group (0.05%) compared with five in the digital mammography group (0.01%).

"The currently available evidence, including findings from TOSYMA-1, seems to suggest that digital breast tomosynthesis plus s2D mammography is superior to standard digital mammography for the detection of invasive breast cancer. The study design with extended follow-up (TOSYMA-2) will provide further insights about the prognostic implications of increased detection of invasive breast cancer at screening by determination of interval breast cancer incidence at 24 months in both groups," concluded Heindel and fellow TOSYMA researchers.

In her accompanying editorial, Nehmat Houssami, MD, of The Daffodil Centre, University of Sydney and Cancer Council NSW, Sydney, Australia, commended Heindel and colleagues for their efforts, particularly through the pandemic.

"In The Lancet Oncology, Walter Heindel and colleagues report the results of the TOmosynthesis plus SYnthesised MAmmography Study (TOSYMA), a multicenter trial embedded in the German population-wide mammography screening program. TOSYMA is the largest of three randomized controlled trials providing evidence on breast tomosynthesis versus mammography screening. Heindel and colleagues persevered with TOSYMA despite the Covid-19 pandemic stalling recruitment, resulting in a marginally lower sample size than planned in their adaptive design," she wrote.

"Results from TOSYMA showed that, compared with mammography alone, tomosynthesis with synthetic two dimensional (s2D) imaging significantly increased the cancer detection rate in participants aged 50–69 years, specifically increasing invasive cancer detection by an additional 2.3 cancers per 1,000 women screened…" noted Houssami, adding that these results are similar to those from a recent study conducted in Italy.

But how will these findings play out in the real-world? According to Houssami, they are not enough to lead to the widespread adoption of tomosynthesis, especially for population-based programs.

"In the absence of sufficient evidence for an effect on interval cancer rate, and no evidence on long-term outcomes, the concern that tomosynthesis could lead to overdetection or could extend lead time without improving long-term outcomes will deter population screening programs from replacing mammography with tomosynthesis. The increased screen-reading time for tomosynthesis, which was roughly double that of mammography in TOSYMA,4 although not the only barrier to adoption of tomosynthesis, represents a real-world resourcing challenge for screening programs tasked with screening a considerable proportion of the female population," she wrote.

She added that "[a]t present, improved screening performance metrics (ie, cancer detection rates, recall rates) are unlikely to be the sole impetus for a widespread transition to tomosynthesis in programmatic screening. However, these metrics have supported the adoption of tomosynthesis as replacement to digital mammography in many radiology services in developed health systems; the reduced recall rate observed in annual screening has accelerated the shift to tomosynthesis particularly in the USA, and this is likely to happen in other settings."

"Strategically, population-based breast screening programs need to proactively prepare for the possibility of a transition to tomosynthesis through program-embedded trials to accelerate evidence on key screening outcomes, and to identify the most appropriate model to use tomosynthesis, which might entail novel screen-reading approaches and adaption of the double-reading process. Such trials might require multicenter, and possibly international, efforts to establish whether tomosynthesis improves health outcomes for women compared with mammography screening," she concluded.

Study limitations include the analysis of only one screening round and lower patient participation compared to previous studies.

This study was funded by the Deutsche Forschungsgemeinschaft (German Research Foundation).

Heindel has received grants from the German Research Foundation; runs the National Reference Center Mammography Münster, a third-party-funded project at the University of Münster and the associated Reference Screening Unit; and is involved in the educational courses of the Reference Center and receive payments for his lectures.

Houssami reported no disclosures.

Heindel W, et al, for the TOSYMA Screening Trial Study Group "Digital breast tomosynthesis plus synthesized mammography versus digital screening mammography for the detection of invasive breast cancer (TOSYMA): A multicenter, open-label, randomized, controlled, superiority trial" Lancet Oncol 2022; DOI: 10.1016/ S1470-2045(22)00194-2.

Houssami N "Should tomosynthesis replace mammography for breast cancer screening?" Lancet Oncol 2022; DOI: 10.1016/S1470-2045(22)00215-7.