Answers are grounded in this paper and cite their sources.
Cohort study demonstrates improved 5-year breast cancer risk prediction by adding prior mammograms in screened women, highlighting the potential for personalized screening strategies.
Current image-based long term risk prediction models do not fully utilize existing data.Dynamic risk prediction models that incorporate current and prior screening mammogram images have not been investigated for use in routine care.The dynamic model was trained on a prospective WashU clinic-based cohort of 10,099 cancer-free women at cohort entry (recruited from women attending routine screening between Nov 3, 2008 and February 2012).Follow-up through 2020 identified 478 pathology-confirmed breast cancer cases.The cohort included 27% non-Hispanic Black women.Two external validation cohorts were included in this study.The first is Emory cohort included 18,360 women screened from 2013 and followed through 2020.This included 42% non-Hispanic Black women and 332 pathology-confirmed breast cancer cases.The second is a prospective Cohort of 208,313 women 40-74 drawn from the British Columbia Cancer (BC Cancer) Breast Screening program screened from Jan 2013 to Dec 31, 2019.We identified 3021 pathology-confirmed incident breast cancers diagnosed through June 20, 2023.In the Emory cohort, using 3 years of prior mammogram images available at the current screening mammogram, after excluded cases diagnosed within 6 months of screening, we obtained a 5-year AUC of 0.80 (95% CI 0.78, 0.83) when controlling for age and BI-RADS density.Evaluating model performance in the same women, this represents a significant improvement over only using the current visit mammogram AUC 0.74 (95% CI 0.71, 0.77) (p < 0.01).In the BC Cancer cohort, using up to 4 years of prior mammogram images in addition to the current mammogram, we obtained a 5-year AUC of 0.78 (95% CI 0.76-0.80)based on analysis of images alone, again representing a significant gain comparing to using the current visit mammogram [AUC = 0.71 (95% CI 0.69, 0.73)].When calibrated to SEER 5-year incidence rates, a risk ratio of 21.1 was observed in the Emory Cohort comparing high (> 4%) to very low (< 0.3%) 5-year risk.Similarly, a risk ratio of 24.4 was observed in BC Cancer. Key finding: Adding previous screening mammogram images improves 5-year breast cancer risk prediction beyond static models.Impact on screening strategy: It can identify women at high risk who might benefit from supplemental screening or risk reduction strategies.
No takes yet. Share an insight, caveat, or question.
A 2026 study studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: