BACKGROUND: Each year, millions of women undergo breast biopsies. Of these, 80% are negative for malignancy but some may be at elevated risk of invasive breast cancer (IBC) due to the presence of benign breast disease (BBD). Cellular senescence plays a complex but poorly understood role in breast cancer development and the presence or absence of these cells may have prognostic value. METHODS: We conducted a case-control study, nested within a cohort of 15,395 women biopsied for BBD at Kaiser Permanente Northwest between 1971 and 2006. Cases (n = 512) were women who developed a subsequent invasive breast cancer (IBC) at least one year after the BBD biopsy
controls (n = 491) did not develop IBC during the same follow-up period. Using H&E-stained biopsy images, we predicted senescence based on deep learning models trained on replicative senescence (RS), ionizing radiation (IR), and various drug treatments. Age-adjusted and multivariable odds ratios (ORs) and 95% confidence intervals (CI) were estimated using unconditional logistic regression. RESULTS: The RS- and IR-derived senescence scores for adipose tissue and the RS-derived score for epithelial tissue were positively associated with the risk of IBC (adipose tissue - RS model: OR CONCLUSIONS: This study suggests that nuclear senescence scores predicted by deep learning models in breast epithelial and adipose tissue can predict the risk of breast cancer development among women with BBD.