Mammographic calcifications association with risk of advanced breast cancer.
Objective: Mammographic calcifications on mammograms with a negative/benign assessment are associated with increased breast cancer risk. Associations with advanced breast cancer risk are unknown. We evaluated whether calcifications recorded on mammography reports are associated with advanced invasive breast cancer risk.
Methods: We included 3,710,313 screening mammograms with a negative/benign final assessment performed on 991,991 women aged 40-74 in the Breast Cancer Surveillance Consortium associated with 7229 advanced cancers. We calculated cumulative 5-year advanced (prognostic pathologic stage ≥II) breast cancer risk and hazards ratios (HR) adjusted for clinical risk factors according to presence or absence of calcifications by menopausal status, dense (heterogeneously or extremely dense) vs. non-dense (almost entirely fatty or scattered fibroglandular density) breasts, body mass index (BMI) < 25 kg/m2 vs. ≥ 25 kg/m2.
Results: Prevalence of calcifications was 6.1% among women who developed advanced breast cancer vs. 3.6% among others. Overall associations of advanced cancer with calcifications were similar for premenopausal (HR = 1.4; 95% CI 1.1-1.9) and postmenopausal (HR = 1.5; 95% CI 1.2-1.7) women. Compared to postmenopausal women with non-dense breasts and BMI < 25 kg/m2 without calcifications [cumulative 5-year advanced cancer incidence = 1.6 (95% CI 1.3-2.0) per 1000 women], postmenopausal women with dense breasts, BMI ≥ 25 kg/m2, and calcifications had 5.5-fold (95% CI 3.9-7.7) higher advanced cancer risk [cumulative 5-year advanced cancer incidence = 10.2; (95% CI 7.0-13.3) per 1000 women]. Results were similar for premenopausal women.
Conclusions: Mammographic calcifications increase advanced cancer risk beyond having dense breasts and being overweight/obese. Future research should investigate strength of associations by type of calcification and incorporation of calcifications into advanced cancer risk models for improvement in model accuracy.