A large study of older women with screen-detected breast cancer suggests that 3D breast imaging does not deliver the same advantage at every age. Digital breast tomosynthesis, commonly known as DBT, was associated with lower-grade and node-negative tumors in women both younger and older than 75, but its apparent ability to identify smaller tumors and invasive lobular cancers disappeared in the oldest group. The findings raise a question at the heart of breast-screening policy: as women age, does the extra information provided by 3D imaging continue to translate into meaningfully earlier or more biologically favorable cancer detection?
The analysis, published in Breast Cancer Research and Treatment, examined 17,201 women aged 67 and older who were diagnosed with screen-detected, estrogen-receptor-positive and HER2-negative breast cancer between 2015 and 2021. The researchers linked cancer-registry information from the Surveillance, Epidemiology, and End Results program with Medicare claims, creating a large real-world cohort rather than a tightly controlled clinical-trial population. Overall, 60.4 percent of the cancers were detected after DBT screening, while 43.6 percent of the women were at least 75 years old when diagnosed. The investigators compared DBT with conventional digital mammography and used multivariable logistic regression to account for measured differences between patients and screening groups.
Digital mammography produces two-dimensional X-ray images in which normal tissue can overlap, sometimes obscuring a tumor. DBT addresses that problem by acquiring multiple low-dose projections as the X-ray tube moves across the compressed breast. Computer reconstruction converts those projections into a stack of thin virtual slices, allowing radiologists to inspect breast structures layer by layer. The technique can reduce the visual clutter caused by overlapping tissue and has shown benefits in studies of women aged 50 to 74, including increased detection of invasive cancers and tumors diagnosed at smaller sizes. But the breast changes with age: glandular tissue generally becomes less dense and more fatty, potentially making conventional mammography easier to interpret and reducing the incremental value of tomosynthesis.
Among women aged 67 to 74, DBT was associated with several signs of more favorable disease compared with digital mammography. The adjusted odds of a smaller tumor were 18 percent higher with DBT, corresponding to an adjusted odds ratio of 1.18 and a 95 percent confidence interval of 1.08 to 1.29. DBT was also associated with a greater likelihood of grade I disease rather than grade II or III disease, with an adjusted odds ratio of 1.14. The odds of node-negative cancer were 15 percent higher, although the confidence interval reached the boundary of statistical significance. In addition, DBT was associated with a 29 percent higher likelihood of invasive lobular carcinoma, or ILC, being detected rather than other histologic types.
The pattern changed sharply among women aged 75 and older. DBT remained associated with lower-grade tumors, with an adjusted odds ratio of 1.21 for grade I disease, and with node-negative disease, with an adjusted odds ratio of 1.18. However, the technique was not significantly associated with smaller tumors: the adjusted odds ratio was 0.99, with a 95 percent confidence interval from 0.90 to 1.10. Nor was DBT significantly associated with invasive lobular histology in this age group, where the adjusted odds ratio was 1.13 and the confidence interval, 0.98 to 1.32, included the possibility of no difference. In practical terms, the study indicates that tomosynthesis may still identify cancers with less aggressive pathological features in older women, but it may not consistently expose those cancers at an earlier, smaller stage than standard mammography.
Tumor grade and lymph-node status provide different kinds of biological information. Grade describes how abnormal cancer cells look under a microscope and how actively they appear to be dividing; grade I tumors generally resemble normal tissue more closely and tend to grow more slowly than higher-grade tumors. Node-negative disease means that cancer was not found in the sampled regional lymph nodes, a finding often associated with a lower risk of regional or distant spread. Tumor size is a direct measure of how much a cancer has grown before detection. These variables are clinically related but not interchangeable, so the study’s results do not mean that DBT failed to detect early cancers in women over 75. Rather, its association with favorable grade and lymph-node status persisted even when its association with smaller size did not.
The result involving invasive lobular carcinoma is especially notable because ILC can be difficult to see on standard breast imaging. Unlike invasive ductal carcinoma, which often forms a more compact mass, lobular cancer cells typically lose adhesion to one another and infiltrate breast tissue in a dispersed, single-file pattern. This can produce subtle architectural distortion rather than a sharply defined lump. DBT’s reconstructed slices might, in principle, make such distortions easier to recognize by reducing tissue overlap. The association observed in women aged 67 to 74 is consistent with that possibility. Yet the absence of a statistically significant association in women over 75 suggests that age-related tissue composition, tumor biology, screening patterns, or other clinical factors may influence how much additional information DBT contributes.
The researchers caution that their findings should not be interpreted as proof that tomosynthesis itself caused the more favorable tumor characteristics. This was a retrospective observational study based on administrative and registry data, not a randomized trial in which women were assigned to one imaging method or the other. Women receiving DBT may differ from those receiving digital mammography in ways that are difficult to measure, including access to advanced imaging centers, healthcare engagement, geographic location, socioeconomic conditions, overall health, and the frequency or continuity of prior screening. The analysis focused only on screen-detected estrogen-receptor-positive, HER2-negative cancers, so the results may not apply to triple-negative tumors, HER2-positive cancers, symptomatic disease, or cancers diagnosed outside screening.
Age also complicates the balance between screening benefits and harms. A screening program can find cancers earlier, but it can also detect tumors that would never have caused symptoms during a person’s lifetime. This phenomenon, known as overdiagnosis, becomes increasingly important when competing health conditions and life expectancy vary widely among older adults. Finding a small tumor does not automatically mean that screening has improved survival if the tumor is biologically indolent or if treatment would never have become necessary. Conversely, a lack of an observed size advantage with DBT does not establish that the technology is useless after age 75, particularly because the study found associations with lower grade and node-negative disease. Decisions about continuing screening should therefore consider individual health, functional status, personal preferences, prior screening history, and the likely benefits and burdens of follow-up and treatment rather than age alone.
The study’s central message is that the apparent advantage of 3D mammography may be age-dependent. For women in their late 60s and early 70s, DBT was linked to a broader set of favorable findings, including smaller tumors and detection of invasive lobular cancers. For women aged 75 and older, its associations narrowed to tumor grade and lymph-node status, making DBT and digital mammography appear more comparable for some outcomes. The findings do not dictate which test any individual should receive, nor do they resolve when screening should stop. Instead, they point to a more precise future for screening research—one that evaluates not only how many cancers an imaging technology detects, but also which cancers it finds, how dangerous those cancers are, and whether detection changes outcomes that matter to older women.
Subject of Research: Tumor characteristics associated with digital breast tomosynthesis versus digital mammography in older women with screen-detected breast cancer
Article Title: Tumor characteristics associated with tomosynthesis versus digital mammography in older women with screen-detected breast cancer
Article References: Huang S, Westvold SJ, Soulos PR, et al. “Tumor characteristics associated with tomosynthesis versus digital mammography in older women with screen-detected breast cancer.” Breast Cancer Research and Treatment 218, article 4 (2026). Original research article
Image Credits: AI Generated
DOI: 10.1007/s10549-026-08013-4
Keywords: breast cancer, digital breast tomosynthesis, screening mammography, older women, tumor grade, lymph-node status, invasive lobular carcinoma, tumor size, overdiagnosis
Tags: age-related differences in mammogram effectivenessbreast cancer detection policies for elderlyBreast cancer screening in older womendigital breast tomosynthesis vs. traditional mammographyestrogen-receptor-positive and HER2-negative breast cancersimpact of imaging modality on tumor size and gradeimplicationsinvasive lobular carcinoma detectionreal-world impact of 3D mammographyscreening outcomes in women over 75tumor biology and imaging technologytumor characteristics by ageuse of Medicare data in breast cancer research
