Chen Junhong, Zhou Hao, Jin Hengwei, Liu Kai
Department of Hepatobiliary and Pancreatic Surgery II, General Surgery Center, The First Hospital of Jilin University, Changchun, China.
Front Oncol. 2023 May 19;13:1152931. doi: 10.3389/fonc.2023.1152931. eCollection 2023.
Colonic adenocarcinoma, representing the predominant histological subtype of neoplasms in the colon, is commonly denoted as colon cancer. This study endeavors to develop and validate a nomogram model designed for predicting overall survival (OS) in patients with colon cancer, specifically those presenting with perineural invasion (PNI).
The Surveillance, Epidemiology, and End Results (SEER) database supplied pertinent data spanning from 2010 to 2015, which facilitated the randomization of patients into distinct training and validation cohorts at a 7:3 ratio. Both univariate and multivariate analyses were employed to construct a prognostic nomogram based on the training cohort. Subsequently, the nomogram's accuracy and efficacy were rigorously evaluated through the application of a concordance index (C-index), calibration plots, decision curve analysis (DCA), and receiver operating characteristic (ROC) curves.
In the training cohorts, multivariable analysis identified age, grade, T-stage, N-stage, M-stage, chemotherapy, tumor size, carcinoembryonic antigen (CEA), marital status, and insurance as independent risk factors for OS, all with -values less than 0.05. Subsequently, a new nomogram was constructed. The C-index of this nomogram was 0.765 (95% CI: 0.755-0.775), outperforming the American Joint Committee on Cancer (AJCC) TNM staging system's C-index of 0.686 (95% CI: 0.674-0.698). Calibration plots for 3- and 5-year OS demonstrated good consistency, while DCA for 3- and 5-year OS revealed excellent clinical utility in the training cohorts. Comparable outcomes were observed in the validation cohorts. Furthermore, we developed a risk stratification system, which facilitated better differentiation among three risk groups (low, intermediate, and high) in terms of OS for all patients.
In this study, we have devised a robust nomogram and risk stratification system to accurately predict OS in colon cancer patients exhibiting PNI. This innovative tool offers valuable guidance for informed clinical decision-making, thereby enhancing patient care and management in oncology practice.
结肠腺癌是结肠肿瘤最主要的组织学亚型，通常被称为结肠癌。本研究致力于开发并验证一种列线图模型，用于预测结肠癌患者，特别是伴有神经周围侵犯（PNI）患者的总生存期（OS）。
监测、流行病学和最终结果（SEER）数据库提供了2010年至2015年的相关数据，便于将患者按7:3的比例随机分为不同的训练组和验证组。采用单因素和多因素分析，基于训练组构建预后列线图。随后，通过一致性指数（C指数）、校准图、决策曲线分析（DCA）和受试者工作特征（ROC）曲线，严格评估列线图的准确性和有效性。
在训练组中，多因素分析确定年龄、分级、T分期、N分期、M分期、化疗、肿瘤大小、癌胚抗原（CEA）、婚姻状况和保险为OS的独立危险因素，所有P值均小于0.05。随后，构建了一个新的列线图。该列线图的C指数为0.765（95%CI：0.755 - 0.775），优于美国癌症联合委员会（AJCC）TNM分期系统的C指数0.686（95%CI：0.674 - 0.698）。3年和5年OS的校准图显示出良好的一致性，而3年和5年OS的DCA在训练组中显示出优异的临床实用性。在验证组中观察到了类似的结果。此外，我们开发了一种风险分层系统，有助于更好地在所有患者的OS方面区分三个风险组（低、中、高）。
在本研究中，我们设计了一个强大的列线图和风险分层系统，以准确预测表现出PNI的结肠癌患者的OS。这种创新工具为明智的临床决策提供了有价值的指导，从而在肿瘤学实践中加强患者护理和管理。
