Magnetic Resonance Imaging and Mammography Outcomes by Advanced Breast Cancer Risk

Importance Few studies have examined advanced breast cancer rates in women undergoing routine mammography and/or breast magnetic resonance imaging (MRI). Objective To compare screening outcomes of breast MRI vs mammography by advanced breast cancer risk and screening round. Design, Setting, and Participants This cohort study included women aged 40 to 74 years who had a screening MRI performed with or without a screening film, digital, or digital breast tomosynthesis (DBT) mammogram within 12 months (MRI group) or a screening digital or DBT mammogram (mammography-only group) performed at breast imaging facilities participating in the Breast Cancer Surveillance Consortium (BCSC) between January 1, 2005, and December 31, 2021, and breast cancers diagnosed between January 1, 2005, and December 31, 2022. Data analysis was performed from April 2025 to August 2026. Exposures MRI and mammography; advanced breast cancer risk score calculated using the BCSC advanced cancer risk model and categorized as low (0% to <0.172%), average (0.172% to <0.3797%), intermediate (0.3797% to <0.6582%), or higher (≥0.6582%). Main Outcomes and Measures Advanced (prognostic stage II or higher) and screen-detected early-stage (prognostic stage I) invasive cancer and false-positive biopsy recommendation rates per 1000 examinations adjusting for registry, examination year, age, race and ethnicity, family history, prior breast biopsy, and breast density. Results A total of 172 865 examinations performed in 105 532 women (median [IQR] age, 52 [46-59] years) were studied. After mammography, the intermediate to high vs low to average risk group had significantly higher advanced cancer (0.6 vs 0.3; difference, 0.4; 95% CI, 0.1-0.6), screen-detected early-stage cancer (2.8 vs 1.8; difference, 1.1; 95% CI, 0.5-1.7), and false-positive biopsy recommendation (18.1 vs 14.7; difference, 3.4; 95% CI, 1.7-5.2) rates per 1000 examinations. In the MRI group, rates per 1000 examinations of advanced cancer (0.8 vs 1.1; difference, −0.3; 95% CI, −1.3 to 0.7), screen-detected early-stage disease (10.3 vs 11.4; difference, −1.1; 95% CI, −4.9 to 2.7), and false-positive biopsy recommendation (94.0 vs 93.3; difference, 0.7; 95% CI, −9.9 to 11.3) were not significantly different between intermediate to high and low to average advanced cancer risk groups. Advanced cancer rates were not significantly different for MRI vs mammography between risk groups (difference, 0.6; 95% CI, −0.4 to 1.6). Screen-detected early-stage rates were significantly higher for MRI vs mammography for low to average risk (11.4 vs 1.8; difference, 9.6; 95% CI, 7.3-12.0) and intermediate to high risk (10.3 vs 2.8; difference, 7.5; 95% CI, 4.9-10.1) groups as were false-positive biopsy recommendation rates for low to average risk (93.3 vs 14.7; difference, 78.6; 95% CI, 72.5-84.8) and intermediate to high risk (94.0 vs 18.1; difference, 75.9; 95% CI, 68.3-83.5) groups. Women who underwent subsequent vs first MRIs had significantly lower advanced cancer rates (0.4 vs 2.0; difference, 1.6; 95% CI, 0.3-2.8) and higher advanced cancer risk (median [IQR] BCSC advanced cancer risk score, 0.348 [0.238 to 0.490] vs 0.301 [0.213 to 0.429]; P < .001). Conclusions and Relevance In this cohort study, intermediate to high advanced breast cancer risk was associated with higher advanced cancer rates with screening mammography. MRI screening was associated with higher screen-detected early-stage cancer rates compared with mammography but not lower advanced cancer rates, except among women who underwent repeat MRIs. These results suggest that women with elevated advanced cancer risk should discuss MRI screening with their physician.