Department of General Surgery, The Second Affiliated Hospital of Nanjing Medical University, Nanjing, 210011, China.
Department of General Surgery, Sir Run Run Hospital Nanjing Medical University, Nanjing, China.
Updates Surg. 2024 Aug;76(4):1311-1320. doi: 10.1007/s13304-024-01770-1. Epub 2024 Mar 26.
This study attempted to compare the prognostic performance of lymph node ratio (LNR) staging system with different cutoff values relative to American Joint Committee on Cancer (AJCC) pN staging system in stage III colorectal cancer (CRC). Overall, 45,069 patients from the SEER dataset and 69 patients from the Second Affiliated Hospital of Nanjing Medical University (the External set) who underwent surgical resection of the primary tumor and were diagnosed with stage III CRC by postoperative pathology were included. Patients were divided into three subgroups based on the LNR cutoff used in previous studies, Kaplan-Meier curves were plotted, and log-rank test was used to compare the differences among groups in terms of cancer-specific survival (CSS). Cox regression model was applied for survival analysis. To evaluate the discriminatory power of different lymph node staging systems, Harrell's C statistic(C-index) and Akaike's Information Criterion (AIC) were applied. A set of optimal cutoff values (0.11; 0.36; 0.66) of LNR staging system with the most considerable discriminatory power to the prognosis in patients with stage III CRC (SEER set: C-index = 0.714; AIC = 58,942.46, External set: C-index = 0.809; AIC = 164.36) were obtained, and both were superior to the AJCC pN staging system (SEER set: C-index = 0.708; AIC = 59,071.20, External set: C-index = 0.788; AIC = 167.06). For evaluating the prognostic efficacy of patients with stage III colorectal cancer, the cutoff value (0.11; 0.36; 0.66) of LNR staging system had the best discrimination and prognostic ability, which was superior to LNR staging system under other cutoff values and AJCC pN staging system.
本研究旨在比较不同截断值的淋巴结比率（LNR）分期系统与美国癌症联合委员会（AJCC）pN 分期系统在 III 期结直肠癌（CRC）中的预后表现。总体上，纳入了来自 SEER 数据集的 45069 例患者和来自南京医科大学第二附属医院的 69 例患者（外部集），这些患者均接受了原发肿瘤的手术切除，且术后病理诊断为 III 期 CRC。根据之前研究中使用的 LNR 截断值，将患者分为三组，绘制 Kaplan-Meier 曲线，并使用对数秩检验比较各组间癌症特异性生存（CSS）的差异。应用 Cox 回归模型进行生存分析。为评估不同淋巴结分期系统的判别能力，应用 Harrell 的 C 统计量（C 指数）和 Akaike 信息准则（AIC）。确定了 LNR 分期系统具有最佳预后预测能力的最佳截断值（0.11；0.36；0.66）（SEER 集：C 指数=0.714；AIC=58942.46，外部集：C 指数=0.809；AIC=164.36），均优于 AJCC pN 分期系统（SEER 集：C 指数=0.708；AIC=59071.20，外部集：C 指数=0.788；AIC=167.06）。用于评估 III 期结直肠癌患者的预后疗效，LNR 分期系统的截断值（0.11；0.36；0.66）具有最佳的判别和预后能力，优于其他截断值的 LNR 分期系统和 AJCC pN 分期系统。
