Development and Evaluation of Prediction Models for Breast Cancer Prognosis

描述（由申请人提供）：八分之一的美国女性将在其一生中被诊断出患有乳腺癌，使其成为美国女性癌症死亡的第二大原因。目前，美国约有310万女性患有乳腺癌。尽管乳腺癌的总体预后良好，但在患者的剩余生命中，复发的风险仍然存在，15年的复发率超过40%。因此，准确预测乳腺癌预后对于制定个性化治疗至关重要，这将最大限度地提高治疗效果，避免患者接受不必要的治疗，并识别复发风险高的妇女进行预防性干预。然而，现有的乳腺癌预后预测工具和模型，包括佐剂！online是美国使用最广泛的预测工具，仅包括临床病理预后因素（如年龄、肿瘤分级和大小、淋巴结、激素受体和雌激素受体状态），而没有考虑公认的生活方式预测因素，如肥胖、体重增加和诊断后的身体活动。此外，还没有针对亚洲女性乳腺癌患者的预测工具或模型得到开发和验证。在本研究中，我们将利用上海乳腺癌生存研究的资源，建立结合临床病理和生活方式因素的预测模型。上海乳腺癌生存研究是一项基于人群的前瞻性队列研究，共有5042名乳腺癌幸存者，并提供了详细的生活方式因素信息。单独的预测模型将首先建立5年和10年的总生存，乳腺癌特异性生存，和无病生存基于临床病理因素，包括在辅助治疗！然后扩展到生活方式因素。将验证和评估模型的预测能力，然后与现有的乳腺癌预测工具/模型进行比较。该研究具有成本效益和可行性，将填补乳腺癌预后预测研究的空白。虽然生活方式因素的流行程度在美国和中国之间可能有所不同，但我们在几项大规模研究中对生活方式因素和乳腺癌结果的调查中特别证明了这一点

DESCRIPTION (provided by applicant): One in eight US women will be diagnosed with breast cancer during her lifetime, making it the second leading cause of cancer deaths among US women. Currently, there are about 3.1 million women living with breast cancer in the US. Despite the overall favorable prognosis for breast cancer, the risk of recurrence persists for the remainder of the patient's lifetime, with a 15-year recurrence rate greater than 40%. Accurate prediction of breast cancer outcomes, therefore, is essential for developing personalized treatment, which would maximize treatment efficacy, spare patients unnecessary treatment, and identify women at high risk of recurrence for preventive intervention. However, existing prediction tools and models for breast cancer prognosis, including Adjuvant! online, the most widely used prediction tool in the US, include only clinicopathological prognostic factors (e.g., age, tumor grade and size, and lymph node, hormone receptor , and estrogen receptor status), without considering well-recognized lifestyle predictors such as obesity, weight gain, and post-diagnosis physical activity. In addition, no prediction tool or models have been developed and validated among Asian women with breast cancer. In the proposed study, we will utilize the resources of the Shanghai Breast Cancer Survival Study, a well characterized, population-based prospective cohort study of 5,042 breast cancer survivors with detailed information on lifestyle factors, to build prediction models that incorporate both clinicopathological and lifestye factors. Separate prediction models will be built first for 5-and 10-year overall survival, breast cancer-specific survival, and disease-free survival based on the clinicopathological factors that are included in Adjuvant! online, and then expanded to incorporate lifestyle factors. Models will be validated and evaluated for predictive ability, then compared with existing breast cancer prediction tools/models. The proposed study is cost-efficient, feasible, and will fill gaps in breat cancer outcomes prediction research. While the prevalence of lifestyle factors may differ between the US and China, we have specifically demonstrated this in our investigations of lifestyle factors and breast cancer outcomes in several large-scale studies that physical activity, pre-diagnosis BMI, and soy food intake, are associated with breast cancer outcomes, regardless of country or race/ethnicity. Therefore, the knowledge gained from the proposed study should be generalizable for building prediction models for breast cancer patients of other racial and ethnic groups.