Chansky Kari, Sculier Jean-Paul, Crowley John J, Giroux Dori, Van Meerbeeck Jan, Goldstraw Peter
Statistics Department, Cancer Research And Biostatistics, Seattle, Washington 98101, USA.
J Thorac Oncol. 2009 Jul;4(7):792-801. doi: 10.1097/JTO.0b013e3181a7716e.
To assess the impact of cell type, age, and gender in addition to pathologic tumor, node, metastasis (TNM) stage in surgically managed stage I-IIIA non-small cell lung cancer (NSCLC) cases from the international staging database of the International Association for the Study of Lung Cancer.
From the 67,725 cases of NSCLC submitted to the staging database, 9137 surgically managed cases were selected for which all the following variables were available: pathologic stage, age, gender, and specific histologic cell type. Performance status and smoking history were examined in subsets. Methods used were Cox proportional hazards regression and recursive partitioning and amalgamation (RPA) analyses.
Pathologic TNM stage, age, and gender were all independently prognostic for survival. The bronchioloalveolar carcinoma (BAC) subtype had superior survival over other cell types despite the potential for heterogeneity in this group. Adjusted comparisons revealed a small survival advantage for squamous cell carcinomas over non-BAC adenocarcinoma histology and also over large cell, though the effect appeared to be limited to the male patients. RPA revealed the importance of TNM stage primarily, and age was prognostic within stage groups. Cell type was not found to add prognostic value in the RPA analysis. Prognostic groups were formed based on the RPA output, and the prognostic value of these groupings was validated using the North American Surveillance, Epidemiology, and End Results Registries. Performance status and smoking history were prognostic in the subsets where data were available. Effects of other variable were not influenced by the inclusion of smoking status in regression models.
Age and gender are confirmed as important prognostic factors in surgically resected NSCLC. Cell type is less important, although the small population of cases classified as BAC have a survival advantage over other histologies, and there may be a small survival advantage for squamous cell carcinomas over non-BAC adenocarcinomas. Imbalances between stage, gender, and cell type at presentation may lead to a misleading result with respect to cell type in unadjusted analyses. Pathologic TNM category is the most important prognostic factor in this analysis.
除了病理肿瘤、淋巴结、转移(TNM)分期外,评估细胞类型、年龄和性别对国际肺癌研究协会国际分期数据库中接受手术治疗的Ⅰ-ⅢA期非小细胞肺癌(NSCLC)病例生存情况的影响。
从提交至分期数据库的67725例NSCLC病例中,选取9137例接受手术治疗的病例,这些病例具备以下所有变量:病理分期、年龄、性别和特定组织学细胞类型。在亚组中检查了体能状态和吸烟史。使用的方法是Cox比例风险回归和递归划分与合并(RPA)分析。
病理TNM分期、年龄和性别均为生存的独立预后因素。细支气管肺泡癌(BAC)亚型的生存率高于其他细胞类型,尽管该组可能存在异质性。校正后的比较显示,鳞状细胞癌相对于非BAC腺癌组织学以及大细胞癌有小的生存优势,不过这种影响似乎仅限于男性患者。RPA分析显示TNM分期最为重要,年龄在各分期组中具有预后价值。在RPA分析中未发现细胞类型增加预后价值。根据RPA结果形成了预后组,并使用北美监测、流行病学和最终结果登记处对这些分组的预后价值进行了验证。在有数据的亚组中,体能状态和吸烟史具有预后价值。回归模型中纳入吸烟状态并未影响其他变量的作用。
年龄和性别被确认为手术切除的NSCLC的重要预后因素。细胞类型的重要性较低,尽管归类为BAC的少数病例比其他组织学类型具有生存优势,并且鳞状细胞癌相对于非BAC腺癌可能有小的生存优势。在未校正的分析中,就诊时分期、性别和细胞类型之间的不平衡可能导致关于细胞类型的误导性结果。在此分析中,病理TNM分类是最重要的预后因素。