Department of Clinical Laboratory, Guangxi Medical University Cancer Hospital, Nanning, China.
Department of Injection Room, The People's Hospital of Yingtan, Yingtan, Jiangxi, China.
Front Endocrinol (Lausanne). 2024 Aug 7;15:1388861. doi: 10.3389/fendo.2024.1388861. eCollection 2024.
We aim to develop a new prognostic model that incorporates inflammation, nutritional parameters and clinical-pathological features to predict overall survival (OS) and disease free survival (DFS) of breast cancer (BC) patients.
The study included clinicopathological and follow-up data from a total of 2857 BC patients between 2013 and 2021. Data were randomly divided into two cohorts: training (n=2001) and validation (n=856) cohorts. A nomogram was established based on the results of a multivariate Cox regression analysis from the training cohorts. The predictive accuracy and discriminative ability of the nomogram were evaluated by the concordance index (C-index) and calibration curve. Furthermore, decision curve analysis (DCA) was performed to assess the clinical value of the nomogram.
A nomogram was developed for BC, incorporating lymphocyte, platelet count, hemoglobin levels, albumin-to-globulin ratio, prealbumin level and other key variables: subtype and TNM staging. In the prediction of OS and DFS, the concordance index (C-index) of the nomogram is statistically greater than the C-index values obtained using TNM staging alone. Moreover, the time-dependent AUC, exceeding the threshold of 0.7, demonstrated the nomogram's satisfactory discriminative performance over different periods. DCA revealed that the nomogram offered a greater overall net benefit than the TNM staging system.
The nomogram incorporating inflammation, nutritional and clinicopathological variables exhibited excellent discrimination. This nomogram is a promising instrument for predicting outcomes and defining personalized treatment strategies for patients with BC.
我们旨在开发一种新的预后模型，该模型结合炎症、营养参数和临床病理特征，以预测乳腺癌（BC）患者的总生存（OS）和无病生存（DFS）。
该研究纳入了 2013 年至 2021 年间共 2857 例 BC 患者的临床病理和随访数据。数据被随机分为两个队列：训练（n=2001）和验证（n=856）队列。基于训练队列的多变量 Cox 回归分析结果，建立了一个列线图。通过一致性指数（C 指数）和校准曲线评估列线图的预测准确性和判别能力。此外，还进行了决策曲线分析（DCA），以评估列线图的临床价值。
我们为 BC 开发了一个列线图，该列线图纳入了淋巴细胞、血小板计数、血红蛋白水平、白蛋白/球蛋白比值、前白蛋白水平和其他关键变量：亚型和 TNM 分期。在 OS 和 DFS 的预测中，列线图的一致性指数（C 指数）显著大于单独使用 TNM 分期获得的 C 指数值。此外，时间依赖性 AUC 值超过 0.7 的阈值，表明列线图在不同时间段具有令人满意的判别性能。DCA 显示，该列线图比 TNM 分期系统提供了更大的总体净效益。
该列线图纳入了炎症、营养和临床病理变量，具有出色的判别能力。该列线图是预测 BC 患者结局和制定个性化治疗策略的一种很有前途的工具。
