Risk prediction of breast cancer treatment-related cardiotoxicity to guide clinical decision making

项目总结 在美国300万乳腺癌幸存者中，近20%患有心血管疾病。这个 国家癌症研究所，NHLBI，以及专业肿瘤学和心脏病学会都支持 通过早期识别和干预减轻乳腺癌幸存者心血管负担的重要性。 尽管被诊断为I至III期乳腺癌的妇女预后良好，但其5年亲属 存活率：90%，据报道，特定的辅助治疗会导致心血管(CV)事件，损害 与健康相关的生活质量和/或导致心血管疾病过早死亡。心血管事件包括急性心肌梗死 据报道，脑梗塞、中风和静脉血栓栓塞症与佐剂有关。 化疗、生物制剂、放射治疗和/或激素治疗。这些与治疗相关的简历 这些事件构成了一个重大的公共卫生问题，因为它们将影响到越来越多的乳腺癌患者 幸存者在较长预期寿命内与健康相关的生活质量。目前，还不存在标准风险模型来 预测与多种辅助乳腺癌治疗相关的心血管事件的风险 已确定的心血管危险因素(如高血压、高脂血症、吸烟)，为实践提供参考 指导方针，促进共享的临床决策。这样的模特可以在治疗前告知女性 关于心血管疾病的潜在风险的替代治疗策略，同时保持最佳的机会 癌症治疗。这些模型还可以帮助识别治疗后心血管疾病风险最高的女性 通过常规成像和/或预防性应用，可能受益于更早和更密集的心血管监测 减少心血管事件风险的药物治疗。为了解决这一差距，我们的研究将通过以下方式创建风险预测模型 对新诊断的成年女性(N=40,500)进行大规模、人口统计学上不同的队列分析 在真实世界的卫生保健环境中，I期到III期浸润性乳腺癌。我们将研究被诊断为 2008-2020年，并使用最大健康电子记录之一的综合电子记录进行了长达15年的跟踪调查 Kaiser Permanente在美国的计划。在目标1中，我们将评估发生的心血管事件(急性心肌梗死、 中风、心力衰竭)乳腺癌辅助治疗后，根据肿瘤特征和心血管风险进行调整 年龄、种族/民族、既往心血管疾病、心血管疾病药物(他汀类药物、抗高血压药物、抗高血压药物)等因素 糖尿病患者)、高血压、糖尿病、体重指数和吸烟。然后，我们将评估心血管事件的风险是否 与年龄、种族匹配的非癌症队列相比，乳腺癌队列中的比例更高。在目标2中，我们将创建 并验证早期(1年)和晚期(最长15年)CV事件的风险预测模型。我们的项目将是 首次评估多种已确定的心血管危险因素与乳腺癌辅助治疗风险的关系 在真实世界、种族和社会经济多样化的社区队列中发生与治疗相关的简历事件。 我们的风险预测模型将提供新的信息来指导循证临床决策。 关于乳腺癌的辅助治疗以及同期和治疗后的心脏肿瘤护理。

PROJECT SUMMARY Nearly 20% of the 3 million breast cancer survivors in the U.S. have cardiovascular disease (CVD). The National Cancer Institute, NHLBI, and professional oncology and cardiology societies have all endorsed the importance of reducing CVD burden in breast cancer survivors through earlier recognition and intervention. Although women diagnosed with stages I to III breast cancer have an excellent prognosis with 5-year relative survival >90%, specific adjuvant therapies have been reported to lead to cardiovascular (CV) events that impair health-related quality of life and/or lead to premature CVD death. CV events including acute myocardial infarction, stroke, and venous thromboembolism have been reported to be associated with adjuvant chemotherapy, biological agents, radiation therapy, and/or hormonal therapies. These treatment-related CV events pose a significant public health problem because they will affect the increasing number of breast cancer survivors’ health-related quality of life over a long-life expectancy. Currently, no standard risk model exists to predict the risk of CV events associated with multiple adjuvant breast cancer therapies in the presence of established CV risk factors (such as hypertension, hyperlipidemia, smoking) to inform practice guidelines and promote shared clinical decision-making. Such models can inform women before treatment about the potential risks of CVD from alternative treatment strategies while maintaining the best chances for cancer cure. These models can also help to identify women at highest risk of CVD after therapy who would potentially benefit from earlier and more intensive CV monitoring via routine imaging and/or use of preventive medications to mitigate risk of CV events. To address this gap, our study will create risk prediction models by analyzing a large, demographically heterogeneous cohort of adult women (N=40,500) with newly diagnosed stages I to III invasive breast cancer in real-world health care settings. We will study women diagnosed from 2008-2020 and followed up to 15 years using the comprehensive electronic records of one of the largest health plans in the U.S., Kaiser Permanente. In Aim 1, we will assess incident CV events (acute myocardial infarction, stroke, heart failure) following adjuvant breast cancer therapies, adjusting for tumor characteristics and CVD risk factors such as age, race/ethnicity, pre-existing CVD, CVD medications (statins, anti-hypertensives, anti- diabetics), hypertension, diabetes, BMI, and smoking. We will then estimate whether the risk of CV events is greater in the breast cancer cohort versus an age, race- matched cancer-free cohort. In Aim 2, we will create and validate risk prediction models for early (<1 year) and late (up to 15 years) CV events. Our project will be the first to estimate the association of multiple established CVD risk factors with the risk of breast cancer adjuvant treatment-related CV events in a real-world, ethnically and socioeconomically diverse community-based cohort. Our risk prediction models will provide new information to guide evidence-based clinical decision-making concerning adjuvant therapy use for breast cancer and concurrent and post-treatment cardio-oncology care.