Department of Gastrointestinal Surgery and Laboratory of Gastric Cancer, State Key Laboratory of Biotherapy, West China Hospital, Sichuan University and Collaborative Innovation Center for Biotherapy, No. 37 Guo Xue Xiang Street, Chengdu, 610041, Sichuan, China.
Division of Digestive Surgery, Xijing Hospital of Digestive Diseases, Fourth Military Medical University, 127 West Changle Road, Xi'an, 710032, Shanxi, China.
Gastric Cancer. 2019 May;22(3):506-517. doi: 10.1007/s10120-018-0890-2. Epub 2018 Nov 2.
The new 8th TNM system attributes AEG Siewert type II to esophageal classification system. However, the gastric and esophageal classification system which was more suitable for type II remains in disputation. This study aimed to illuminate the 8th TNM-EC or TNM-GC system which was more rational for type II, especially for patients underwent transhiatal approaches.
We collected the database of patients with AEG who underwent radical surgical resection from two high-volume institutions in China: West China Hospital (N = 773) and Xi Jing Hospital of Fourth Military University (N = 637). The cases were randomly matched into 705 training cohort and 705 validation cohort. All the cases were reclassified by the 8th edition of TNM-EC and TNM-GC. The distribution of patients in each stage, the hazard ratio of each stage, and the separation of the survival were compared. Multivariate analysis was performed using the Cox proportional hazard model. Comparisons between the different staging systems for the prognostic prediction were performed with the rcorrp.cens package in Hmisc in R (version 3.4.4. http://www.R-project.org/ ). The validity of these two systems was evaluated by Akaike information criterion (AIC) and concordance index (C-index).
By univariate analysis, the HRs from stage IA/IB to stage IV/IVB were monotonously increased according to TNM-GC scheme in both cohorts (training 2.63, 3.91, 5.02, 8.64, 15.51 and 29.64; validation 1.54, 3.55, 4.91, 7.14, 11.67, 18.71 and 48.32) whereas only a fluctuating increased tendency was found when staged by TNM-EC. After the multivariate analysis, TNM-GC (P < 0.001), TNM-EC (P = 0.001) in training cohort and TNM-GC (P < 0.001) TNM-EC (P < 0.001) in the validation cohort were both independent prognostic factors. The C-index value for the TNM-GC scheme was larger than that of TNM-EC system in both training (0.721 vs. 0.690, P < 0.001) and validation (0.721 vs. 0.696, P < 0.001) cohorts. After stratification analysis for Siewert type II, the C-index for TNM-GC scheme was still larger than that of TNM-EC in both training (0.724 vs. 0.694, P = 0.005) and validation (0.723 vs. 0.699, P < 0.001) cohorts.
The 8th TNM-GC scheme is superior to TNM-EC in predicting the prognosis of AEG especially for type II among patients underwent transhiatal approaches.
新的第 8 版 TNM 系统将 Siewert Ⅱ型 AEG 归入食管分类系统。然而，更适合 Siewert Ⅱ型的胃食管分类系统仍存在争议。本研究旨在阐明第 8 版 TNM-EC 或 TNM-GC 系统对 Siewert Ⅱ型更合理，特别是对经食管裂孔入路的患者。
我们收集了来自中国两家高容量机构（华西医院[N = 773]和第四军医大学西京医院[N = 637]）的接受根治性手术的 AEG 患者的数据库。病例被随机分为 705 例训练队列和 705 例验证队列。所有病例均按第 8 版 TNM-EC 和 TNM-GC 重新分类。比较各期患者的分布、各期的危险比以及生存的分离情况。使用 Cox 比例风险模型进行多变量分析。使用 R（版本 3.4.4. http://www.R-project.org/ ）中的 Hmisc 中的 rcorrp.cens 包比较不同分期系统的预后预测。通过赤池信息量准则（AIC）和一致性指数（C-index）评估这两种系统的有效性。
单因素分析显示，在两个队列中，根据 TNM-GC 方案，从 IA/IB 期到 IV/IVB 期的 HR 呈单调递增（训练组 2.63、3.91、5.02、8.64、15.51 和 29.64；验证组 1.54、3.55、4.91、7.14、11.67、18.71 和 48.32），而根据 TNM-EC 方案则呈波动增加趋势。多因素分析后，TNM-GC（P<0.001）、TNM-EC（P=0.001）在训练组和 TNM-GC（P<0.001）、TNM-EC（P<0.001）在验证组均为独立的预后因素。TNM-GC 方案的 C 指数值在训练组（0.721 比 0.690，P<0.001）和验证组（0.721 比 0.696，P<0.001）均大于 TNM-EC 系统。在 Siewert Ⅱ型分层分析后，TNM-GC 方案的 C 指数在训练组（0.724 比 0.694，P=0.005）和验证组（0.723 比 0.699，P<0.001）均大于 TNM-EC 方案。
第 8 版 TNM-GC 方案在预测 AEG 尤其是经食管裂孔入路的 Siewert Ⅱ型患者的预后方面优于 TNM-EC 方案。
Zotero 插件
