Extensive changes in esophageal and esophagogastric junction cancer staging published in the 7th edition of the American Joint Committee on Cancer’s (AJCC) Cancer Staging Manual have been deemed an improvement over the previous staging system, based on a recent, retrospective comparison of nearly 400 patients staged using both the new and old systems.

Definitive new staging criteria for esophageal cancer and amended criteria for staging other types of cancer were based on the results of numerous worldwide studies. However, only recently has research been conducted to compare the predictive ability and performance of the 6th and 7th editions of the AJCC staging systems in esophageal cancer, according to a report in the Annals of Thoracic Surgery.

“The new staging recommendations for cancer of the esophagus and esophagogastric junction are data driven and harmonized with stomach cancer. This required changes in the tumor, node, and metastasis (TNM) definitions and additions of nonanatomic cancer characteristics,” according to Dr. Thomas W. Rice of the Cleveland Clinic and his colleagues who commented on the evolution of the new system earlier this year (J. Thorac. Cardiovasc. Surg. 2010 March;139:527-9).

Stages in both editions are based on severity in the degree of TNM involvement (number of tumors, number of nodes involved, and presence of metastasis). For example, stage I is defined as T=1; N=0; M=0, whereas stage IV is defined as T=Any; N=Any; M=1.

The changes they list in the TNM anatomic classifications in the 7th edition, versus the 6th edition, are as follows:

T Classification

(Tis is redefined and T4 is subclassified.)

– Tis – high-grade dysplasia.

– T4a – resectable cancer invades adjacent structures (for example, pleura, pericardium, diaphragm).

– T4b – unresectable cancer invades adjacent structures (for example, aorta, vertebral body, trachea).

N Classification

(Regional lymph node is redefined.)

– N0 – no regional lymph node metastasis.

– N1 – 1 to 2 positive regional lymph nodes.

– N2 – 3 to 6 positive regional lymph nodes.

– N3 – greater than or equal to 7 positive regional lymph nodes.

M Classification

(M is redefined.)

– M0 – no distant metastases.

– M1 – distant metastases.

Nonanatomic cancer characteristics added to the 7th edition system are as follows: histologic cell type (either adenocarcinoma or squamous cell carcinoma); histologic grade – from G1 (well differentiated) to G4 (undifferentiated); and cancer location – upper, middle, or lower thoracic or esophagogastric junction.

To compare the new and old staging systems, Dr. Po-Kuei Hsu and colleagues evaluated 392 esophageal squamous cell carcinoma patients who received primary surgical resection through the tri-incisional approach at the Taipei (Taiwan) Veterans General Hospital, during 1995-2006. The investigators described tumor specimens according to appearance, invasion depth, and differentiation; recorded the number of lymph nodes involved; and examined lymph nodes histologically for each station.

The mean patient age was 63.8 years, and 18/392 were women (4.6%); 81/392 patients received adjuvant chemoradiation (20.7%). The mean follow-up was 32.8 months (Ann. Thorac. Surg. 2010 April;89:1024-31).

Patients were retrospectively staged based on this information using both the 6th and 7th edition staging systems. Survival analysis was performed with a Cox regression model and plotted using the Kaplan-Meier method; the homogeneity, discriminatory ability, and monotonicity of the gradients of the systems were also compared using several statistical methods.

The overall 5-year survival rate was 27.1%. Only sex (with women having better survival) and the TNM classifications according to the 7th edition staging system were significant predictors of survival, the investigators found.

Age, tumor length, histologic grade, adjuvant treatment, and cancer location were not significant predictors of survival in a multivariate analysis (although tumor length was a significant factor in a univariate analysis). In addition, the subclassification of seven or more positive lymph nodes as N3 in the 7th edition was seen as unnecessary because survival was not significantly different from that of N2 patients (5-year survival rates were 2.9% vs. 0.0%, respectively).

The patient Kaplan-Meier plot applied to the 6th edition staging system showed overlapped survival curves among stages IIB, II, and IV. Even classification using only I, II, III, and IV as the major stages without subclassification found a similarity between the survival curves for stages III and IV. In the 7th edition classification system, using all eight substages, survival curves were similar between stages IIA and IIB and between stages IIIB and IIIC. But when classification of the four major stages was used, the Kaplan-Meier plot showed good discriminatory ability among stages I through IV for the 7th edition system.

Overall, the 7th edition system had better homogeneity, discriminatory ability, and monotonicity of gradients, and a smaller Akaike Information Criterion (AIC, indicative of a better model for predictive outcome), compared with the 6th edition system. However, the histologic grade did not prove predictive in this study, compared with the results analysis reported in developing the 7th edition system.

“Although our sample size was relatively small compared with the worldwide esophageal collaboration database, we represented single-institution experience. The surgical procedures, pathologic examinations, and patient follow-up were uniform throughout the whole study period. In contrast, the previously published worldwide esophageal cancer collaboration was assembled from 13 centers and the era spanned nearly 30 years; bias may be inevitable,” said the researchers in explaining the discrepancy.

Furthermore, they pointed out that the squamous cell carcinoma histologic type constituted less than half of the worldwide database as compared with all of the patients in their present study, making further external validation necessary.

None of the authors of either journal article reported disclosures relevant to their commentary or analysis.

Copyright (c) 2009 Elsevier Global Medical News. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.

发表在美国癌症联合委员会(AJCC)癌症分期手册第七版中的有关食管和食管胃结合部癌分期的内容较前有大量更新，其被视为对前版分期系统的一些改进，该更新是在近期一项对近400例患者应用新版和前版分期系统进行回顾性分期比较分析的基础上作出的。

 

在全球多项研究结果的基础上，ACJJ最后确定了食管癌新的分期标准并修订的其他癌症的分期标准。而据发表在《胸外科年鉴》(the Annals of Thoracic Surgery)上的一项报告，直到最近才有研究比较了6版和7版AJCC食管癌分期标准的预测能力和表现。

 

来自Cleveland Clinic的Thomas W. Rice博士及其同事在今年早些时候对新版AJCC食管癌分期系统的更新进行了评论，他们说：“研究数据促成了食管癌和食管胃结合部癌新的分期建议，并使其与胃癌相一致。这需要更新肿瘤大小、淋巴结状况和转移情况(TNM)及癌症的其他非解剖学特征等方面的内容。”

 

两个版本的分期标准均基于TNM分期肿瘤发展的严重程度(肿瘤个数、受累淋巴结个数和远处转移情况)。如Ⅰ期定义为T=1、N=0、M=0；而Ⅳ期定义为任意T、任意N、M=1。

 

与6版分期标准相比，他们列出的7版中TNM解剖分类的变化如下：

 

T分级

 

 (重新定义了Tis，亦对T4进行了细分。)

 

—Tis为高度不典型增生。

 

—T4a为侵及邻近组织的可切除癌(如胸膜、心包、膈)。

 

—T4b为侵及邻近组织的不可切除癌(如主动脉、椎体、气管)。

 

N分级

 

 (重新定义了区域淋巴结转移。)

 

—N0为无区域淋巴结转移。

 

—N1为1~2个区域淋巴结阳性。

 

—N2为3~6个区域淋巴结阳性。

 

—N3为≥7个区域淋巴结阳性。

 

M分级

 

 (重新定义了M分级)

 

—M0为无远处转移。

 

—M1为有远处转移。

 

7版分期标准中增加的非解剖学癌症特征为：癌细胞组织学类型(腺癌或鳞状细胞癌)、癌症组织学分级[从G1(分化良好)到G4(未分化)]、癌症部位(上、中或下胸或食管胃结合部)。

 

为了比较新旧分期系统，Po-Kuei Hsu博士及其同事对1995~2006年在台北荣民总医院接受三切口一期外科切除治疗的392例食管鳞状细胞癌患者进行了评价。研究者根据临床表现、浸润深度、分化程度、获取阳性淋巴结个数、行组织学检测的各部位淋巴结个数等对肿瘤样本进行了描述。

 

患者的平均年龄为63.8岁，其中18例(4.6%)为女性，81例(20.7%)接受了辅助放化疗。平均随访时间为32.8个月(Ann. Thorac. Surg. 2010 April;89:1024-31)。

 

根据这些资料，依照6版和7版分期标准，研究者对患者分别进行了回顾性分期。应用Cox回归模型对患者进行了生存分析，并用Kaplan-Meier方法绘制了生存率曲线。同时应用数种统计方法对两版系统分级的同质性、区分能力和梯度单调性进行了比较研究。

 

5年总体生存率为27.1%。研究者发现，依据7版分级系统，仅性别(女性生存率较佳)和TNM分级为患者预后的显著预测因子。

 

多因素分析发现，年龄、肿瘤大小、组织学分级、辅助治疗和肿瘤部位并非生存的显著预测因子(而单因素分析中，肿瘤大小为显著预测因子)。另外，7版中将阳性淋巴结个数≥7个定义为N3没有必要，这是因为，此亚组患者的生存率与N2组患者无显著差异(其5年生存率分别为2.9%和0)。

 

据6版分期标准绘制的Kaplan-Meier生存曲线中，ⅡB期、Ⅱ期和Ⅳ期患者的曲线重叠。仅对患者进行Ⅰ、Ⅱ、Ⅲ、Ⅳ期的基本分期(无亚组)发现，Ⅲ、Ⅳ期的生存曲线相同。依据7版分期系统，应用所有8个亚组分期发现，ⅡA期与ⅡB期、ⅢA期与ⅢB期的生存曲线相同。但仅应用4期基本分期，Kaplan-Meier生存曲线显示，7版分期系统很好地区分了Ⅰ至Ⅳ期患者。

 

总之，与6版分期系统相比，7版系统具有较好的同质性、区别能力、梯度单调性及更小的赤池信息量准则(AIC，一种更佳的预测模型标志)值。然而，本研究发现，组织学分级无预测作用，与为制定7版分期系统而进行的研究结果相异。

 

研究者们在解释其中差异时说：“虽然与全球食管协作数据库相比，我们的研究样本量较小，但我们介绍的是单中心的经验。整个研究期间，患者接受的外科手术、病理检查和随访均保持不变。相比之下，之前发表的全球食管癌协作的资料来自13个中心，而时间跨越近30年；其研究偏倚不可避免。”

 

另外，研究者指出，鳞状细胞癌组织类型在全球数据库中占不足一半，而本研究中均为鳞状细胞癌，故需进行进一步的外部验证。

 

两篇文章的所有作者均声明，没有与其评论或分析相关的利益冲突。

 

爱思唯尔  版权所有