Heckenbach Indra, Peila Rita, Benz Christopher, Weinmann Sheila, Wang Yihong, Powell Mark, Scheibye-Knudsen Morten, Rohan Thomas
Center for Healthy Aging, Department of Cellular and Molecular Medicine, University of Copenhagen, Copenhagen, Denmark.
Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, USA.
Breast Cancer Res. 2025 Mar 11;27(1):37. doi: 10.1186/s13058-025-01993-z.
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.
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.
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=1.69, 95% CI 1.03-2.77, and IR model: OR=1.73, 95%CI 1.06-2.82; epithelial tissue- RS model: OR=1.53, 95% CI 1.05-2.22). The results were stronger among postmenopausal women and women with epithelial hyperplasia with/without atypia, and postmenopausal women also showed a positive association for stromal tissue with the RS model (OR=1.84, 95%CI 1.12-3.04). There was an elevated risk of IBC in those with higher senescence scores in both epithelial and adipose tissue compared with those with low senescence scores in both (IR epithelium-IR fat: OR=2.14, 95% CI 1.30-3.51; and IR epithelium-RS fat: OR= 2.24, 95% CI 1.15-4.35).
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.
每年，数百万女性接受乳房活检。其中，80%的活检结果为恶性阴性，但由于存在良性乳腺疾病（BBD），一些女性可能患浸润性乳腺癌（IBC）的风险较高。细胞衰老在乳腺癌发展中起着复杂但尚不清楚的作用，这些细胞的存在与否可能具有预后价值。
我们进行了一项病例对照研究，该研究嵌套在1971年至2006年间在西北凯撒永久医疗中心因BBD接受活检的15395名女性队列中。病例组（n = 512）为在BBD活检至少一年后发生后续浸润性乳腺癌（IBC）的女性；对照组（n = 491）在同一随访期内未发生IBC。使用苏木精和伊红（H&E）染色的活检图像，我们基于在复制性衰老（RS）、电离辐射（IR）和各种药物治疗上训练的深度学习模型预测衰老。使用无条件逻辑回归估计年龄调整和多变量优势比（OR）以及95%置信区间（CI）。
脂肪组织的RS和IR衍生衰老评分以及上皮组织的RS衍生评分与IBC风险呈正相关（脂肪组织 - RS模型：OR = 1.69，95% CI 1.03 - 2.77，IR模型：OR = 1.73，95% CI 1.06 - 2.82；上皮组织 - RS模型：OR = 1.53，95% CI 1.05 - 2.22）。绝经后女性以及有/无异型性的上皮增生女性的结果更强，绝经后女性的基质组织与RS模型也呈正相关（OR = 1.84，95% CI 1.12 - 3.04）。与上皮和脂肪组织衰老评分均低的女性相比，上皮和脂肪组织衰老评分均高的女性患IBC的风险升高（IR上皮 - IR脂肪：OR = 2.14，95% CI 1.30 - 3.51；IR上皮 - RS脂肪：OR = 2.24，95% CI 1.15 - 4.35）。
本研究表明，深度学习模型预测的乳腺上皮和脂肪组织中的核衰老评分可以预测BBD女性患乳腺癌的风险。
