乳腺Paget病(PD)是一种少见的基础恶性肿瘤。大约50%至60%的乳腺PD患者同时诊断为浸润性导管癌(PD-IDC),与没有PD的IDC相比,预后较差。到目前为止,一直缺乏准确有效的PD-IDC预后模型,影响这些患者化疗和放疗有效性的因素仍然未知。在这项研究中,我们基于监测流行病学和最终结果(SEER)数据库的数据,开发了基于网络的列线图.我们对模型进行了一系列验证方法,包括曲线下面积(AUC)值,接收器工作特性曲线(ROC)分析,校正曲线,和决策曲线分析(DCA)。我们的结果表明我们的模型表现出很高的区分度,准确度,和临床适用性在预测PD-IDC患者的总生存期(OS)(测试集:三年和五年的AUC,分别为0.831和0.841)。为了进一步验证我们的列线图,我们使用了来自我们机构和姐妹医院的外部数据(外部数据:三年和五年的AUC,分别为0.892和0.914)。多变量Cox回归分析确定了PD-IDC患者OS的几个独立不良预后因素,包括年龄的增长,高品位,寡妇身份,较高的T阶段,和骨转移的存在。此外,进行了倾向得分匹配(PSM)调整分析,揭示了化疗并没有显着提高PD-IDC患者的生存率的分子亚型,除了那些在三级/四级组,它提高了OS和乳腺癌特异性生存率(BCSS)。此外,我们的研究结果表明,只有T4和N3期的PD-IDC患者受益于放疗,导致操作系统和BCSS的改进。总之,我们综合分析了PD-IDC患者的临床特点和预后,最终开发了一个用户友好的基于网络的列线图来预测他们的生存。我们的预测模型不仅精度高,而且提供了简单性,使医疗保健提供者和患者可以使用它。此外,我们的分层分析强调了病理分级,而不是分子亚型,在确定化疗对改善PD-IDC患者预后的疗效方面起着关键作用,而放疗赋予PD-IDC患者T4和N3期的生存益处。
Paget\'s disease (PD) of the breast is a rare underlying malignant tumor. Approximately 50% to 60% of patients with mammary PD are concurrently diagnosed with invasive ductal carcinoma (PD-IDC), a condition associated with a worse prognosis than IDC without PD. Thus far, there has been a lack of an accurate and efficient prognostic model for PD-IDC, and the factors influencing the effectiveness of chemotherapy and radiotherapy for these patients remain unknown. In this study, we developed a web-based nomogram based on the data from the Surveillance Epidemiology and End Results (SEER) database. We subjected the model to a series of validation methods, including area under the curve (AUC) values, receiver operating characteristic curve (ROC) analysis, calibration curves, and decision curve analysis (DCA). Our results demonstrated that our model exhibited high discrimination, accuracy, and clinical applicability in predicting the overall survival (OS) of patients with PD-IDC (testing set: three- and five-year AUCs, 0.831 and 0.841, respectively). To further validate our nomogram, we used external data from both our institution and sister hospitals (external data: three- and five-year AUCs, 0.892 and 0.914, respectively). Multivariable Cox regression analysis identified several independent unfavorable prognostic factors for the OS of patients with PD-IDC, including increasing age, high grade, widowed status, higher T stages, and the presence of bone metastases. Furthermore, propensity score matching (PSM)-adjusted analysis was conducted, revealing that chemotherapy did not significantly improve the survival of patients with PD-IDC across molecular subtypes, except for those in the grade III/IV group, where it improved both OS and breast cancer-specific survival (BCSS). Additionally, our findings indicated that only patients with PD-IDC with T4 and N3 stages benefited from radiotherapy, leading to improvements in both OS and BCSS. In conclusion, we have comprehensively analyzed the clinical characteristics and prognosis of patients with PD-IDC, culminating in the development of a user-friendly web-based nomogram for predicting their survival. Our predictive model is not only highly accurate but also offers simplicity, making it accessible for healthcare providers and patients. Furthermore, our stratified analysis highlights that the pathological grade, rather than the molecular subtype, plays a pivotal role in determining the efficacy of chemotherapy in improving the prognosis for patients with PD-IDC, while radiotherapy confers survival benefits to patients with PD-IDC in T4 and N3 stages.