Tumour size at detection according to different measures of mammographic breast density.

OBJECTIVES: Breast cancer prognosis is better for smaller tumours. Women with high breast density are at higher risk of breast cancer and have larger screen-detected and interval cancers in mammographic screening programmes. We assess which continuous measures of breast density are the strongest predictors of breast tumour size at detection and therefore the best measures to identify women who might benefit from more intensive mammographic screening or alternative screening strategies. SETTING AND METHODS: We compared the association between breast density and tumour size for 1007 screen-detected and 341 interval cancers diagnosed in an Australian mammographic screening programme between 1994 and 1996, for three semi-automated continuous measures of breast density: per cent density, dense area and dense area adjusted for non-dense area.
RESULTS: After adjustment for age, hormone therapy use, family history of breast cancer and mode of detection (screen-detected or interval cancers), all measures of breast density shared a similar positive and significant association with tumour size. For example, tumours increased in size with dense area from an estimated mean 2.2 mm larger in the second quintile (beta = 2.2; 95% CI 0.4-3.9, P < 0.001) to mean 6.6 mm larger in the highest decile of dense area (beta = 6.6; 95% CI 4.4-8.9, P < 0.001), when compared with first quintile of breast density.
CONCLUSIONS: Of the breast density measures assessed, either dense area or per cent density are suitable measures for identifying women who might benefit from more intensive mammographic screening or alternative screening strategies.