Advancing breast cancer risk prediction in national cohorts: the role of mammogram-based deep learning

推进国家队列中的乳腺癌风险预测：基于乳房 X 光检查的深度学习的作用

• 批准号: 10734544
• 负责人: Kimberly A Bertrand
• 金额: $ 69.83万
• 依托单位: COLUMBIA UNIVERSITY HEALTH SCIENCES
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-09-21 至 2026-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10734544
• 关键词: Address; Adoption; Artificial Intelligence; Black race; Breast; Breast Cancer Detection; Breast Cancer Prevention; Breast Cancer Risk Assessment Tool; Breast Cancer Risk Factor; Calibration; Classification; Clinical; Clinical Data; Data; Data Set; Discrimination; Disease; Disparity; District of Columbia; Early Diagnosis; Electronic Medical Records and Genomics Network; Ensure; Epidemiology; Equity; Estrogen Receptor Status; Estrogen Receptors; Ethnic Origin; Family; Genetic; Genomics; Goals; Guidelines; High Risk Woman; Knowledge; Laws; Malignant Neoplasms; Mammographic Density; Mammography; Measures; Methods; Modeling; Not Hispanic or Latino; Notification; Outcome; Pathology; Performance; Population; Positioning Attribute; Prognosis; Publishing; Race; Recording of previous events; Reduce health disparities; Research; Resources; Risk; Risk Assessment; Risk Factors; Risk Reduction; Role; Sister; Stage at Diagnosis; Statutes and Laws; Stratification; Testing; Validation; Woman; Women' s Health; advanced breast cancer; algorithmic bias; black women; breast density; cancer subtypes; clinical application; clinical risk; cohort; deep learning; deep learning model; digital; epidemiologic data; evidence base; follow-up; genetic resource; genetic risk assessment; genomic data; genomic tools; health disparity; improved; learning strategy; malignant breast neoplasm; mortality; personalized screening; polygenic risk score; racial disparity; racial diversity; risk prediction; risk prediction model; risk stratification; screening; standard of care; study population; tool; tumor; uptake

To date, significant efforts devoted to developing breast cancer risk prediction models that produce modest discriminatory accuracy in the range of 59-68% and underperform in Black women compared to non-Hispanic white (NHW) women, limiting their clinical impact on equitable precision-based/personalized breast cancer prevention and screening. Emerging data suggest that deep learning (DL) models based on mammographic images outperform traditional breast cancer risk prediction models based on clinical and risk factors and breast density alone. Yet there remain major gaps to resolve before widespread roll out of DL methods in order to enhance outcomes and reduce health disparities in breast cancer. These models have been developed in clinical or screening cohorts and require further validation in “real world” settings. Equally unknown is how performance of MDL risk models may differ between Black and NHW women, as few previous studies included sufficient numbers of Black women for meaningful interpretation. Finally, the combination of polygenic risk scores (PRS) and MDL risk scores could significantly improve clinical utility of these risk stratification tools, yet MDL and PRS clinical rollout are largely being evaluated separately, primarily due to lack of availability of both mammographic and genetic data in the same study population. We propose to leverage existing unparalleled resources from two complementary national cohorts, the Sister Study and the Black Women’s Health Study, with digital screening mammograms, genomic data, and extensive epidemiologic and clinical data to address these evidence gaps (ncases, 971; ncontrols, 8793). In Aim 1, we will validate the performance and clinical risk stratification of Mirai, a published MDL model with the highest accuracy in screening cohorts to date, in our epidemiologic cohorts. In addition to considering breast cancer risk overall, we will evaluate performance of this MDL model for early and advanced stage and for ER+ and ER– breast cancer, outcomes with significant clinical implications for prognosis and treatment. In Aim 2, we will evaluate the addition of PRS to MDL risk scores in improving risk stratification, and determine net reclassification. In Aim 3, we will determine whether MDL models perform equally well and impacts risk stratification within strata of breast density, as defined in widespread implementation of breast density notification legislation (dense vs. non-dense). These results may provide crucial data for guiding follow up and surveillance of women with and without dense breasts. To ensure equitable application of results, we will examine differences in results between Black and NHW women in all aims. MDL models hold great promise as breast cancer risk assessment tools, and offer clinical efficiency and optimization by eliminating the need to collect detailed family history and risk factor data. Validation in “real world” epidemiologic cohorts will advance the field substantially. The research proposed here has great potential to identify women at both high and low risk of breast cancer, inform personalized screening and risk reduction strategies, and even to reduce racial disparities in breast cancer outcomes.

到目前为止，致力于开发乳腺癌风险预测模型的重大努力产生了适度的 与非西班牙裔相比，黑人女性的歧视准确率在59-68%之间，表现不佳 白人(NHW)女性，限制她们对公平精确/个性化乳腺癌的临床影响 预防和筛查。新出现的数据表明，基于乳房X光检查的深度学习(DL)模型 图像优于基于临床和风险因素以及乳腺的传统乳腺癌风险预测模型 仅密度一项。然而，在广泛推广数字学习方法之前，仍有重大差距需要解决，以便 改善结果，减少乳腺癌患者的健康差距。这些型号是在 临床或筛查队列，并需要在“真实世界”环境中进一步验证。同样未知的是它是如何 MDL风险模型在黑人和nhw女性中的表现可能不同，因为以前很少有研究包括在内 有足够数量的黑人女性来进行有意义的解读。最后，多基因风险的组合 评分(PR)和MDL风险评分可以显著提高这些风险分层工具的临床实用性，但 MDL和PRS的临床推广在很大程度上是单独进行评估的，主要是因为两者都缺乏可用性 同一研究人群中的乳房X光检查和遗传数据。我们建议利用现有的无与伦比的 资源来自两个互补的国家队列，姐妹研究和黑人妇女健康研究， 通过数字筛查乳房X光照片、基因组数据以及广泛的流行病学和临床数据来处理 这些证据差距(n病例，971；n对照，8793)。在目标1中，我们将验证性能和临床风险 Mirai的分层，一个发表的MDL模型，到目前为止在队列筛选中具有最高的准确性，在我们的 流行病学队列。除了全面考虑乳腺癌的风险外，我们还将评估 对于早期和晚期以及ER+和ER-乳腺癌，这种MDL模型的结果具有显著意义 对预后和治疗的临床意义。在目标2中，我们将评估将PRS添加到MDL风险 在改进风险分层方面的得分，并确定净重新分类。在目标3中，我们将确定是否 MDL模型执行得同样好，并影响乳房密度层内的风险分层，如 广泛实施乳房密度通知立法(密集与非密集)。这些结果可能 为指导对乳房致密和乳房不致密的妇女的随访和监测提供重要数据。为了确保 公平应用结果，我们将检查黑人和nhw妇女在结果上的所有差异 目标。MDL模型作为乳腺癌风险评估工具具有很大的前景，并提供临床效率和 通过消除收集详细的家族病史和风险因素数据的需要进行优化。“REAL”中的验证 世界“流行病学队列将大大推动这一领域的发展。这里提出的研究有很大的意义 有可能识别乳腺癌高风险和低风险女性，提供个性化筛查和风险信息 减少战略，甚至减少乳腺癌预后的种族差异。