论文部分内容阅读
目的通过对行根治性放疗的鼻咽癌 (NPC)病例资料的分析 ,探讨影响NPC预后的临床因素及预后指数模型在NPC中的应用价值。方法选择 1992年 9月～ 1993年 2月间采用根治性放射治疗的Ⅰ～Ⅳa期鼻咽低分化鳞癌共 84例 ,分析年龄、性别、临床分期、放疗分割方法、化疗与否、治疗前血清VCA IgA、细胞免疫功能状态等因素与鼻咽癌预后的关系 ,生存统计采用Kaplan Meier法和Log rank检验 ,多因素分析采用Cox逐步回归模型 ,并以Cox模型所得各因素的变异系数乘以分组值计算预后指数。结果全组中位生存期 72个月 ,1、3、5年生存率分别为 83 2 %、73.2 %、6 9.6 %。单因素分析显示临床分期、放疗分割方法、治疗前血清VCA -IgA、细胞免疫功能状态与预后有关 ,多因素分析显示仅临床分期、放疗分割方法、细胞免疫功能状态为NPC的显著预后因素。综合以上三种因素的预后指数模型能较好地区分不同的预后亚组。结论临床分期、放疗分割方法、细胞免疫功能状态是NPC的独立预后因素。预后指数模型能够较TNM临床分期等单个因素更好地反映预后。
Objective To investigate the clinical factors affecting the prognosis of NPC and the prognostic index model in NPC by analyzing the data of cases with nasopharyngeal carcinoma (NPC) undergoing radical radiotherapy. Methods A total of 84 patients with stage Ⅰ ~ Ⅳa nasopharyngeal poorly differentiated squamous cell carcinoma treated with radical radiotherapy from September 1992 to February 1993 were enrolled in this study. Age, gender, clinical stage, radiotherapy and chemotherapy, chemotherapy or not, Serum VCA IgA, cellular immune function and other factors and the prognosis of nasopharyngeal carcinoma, survival statistics using Kaplan Meier method and Log rank test, multivariate analysis using Cox stepwise regression model, and Cox model derived coefficient of variation of each factor times Grouping values were calculated for the prognostic index. Results The median survival time was 72 months. The 1, 3, 5-year survival rates were 83.2%, 73.2% and 69.6% respectively. Univariate analysis showed that clinical staging, radiotherapy and radiotherapy methods, pretreatment serum VCA -IgA and cellular immune function were related to prognosis. Multivariate analysis showed that clinical stage, radiotherapy alone and cellular immune function were significant prognostic factors in NPC. Prognosis index model based on the above three factors can better distinguish different prognosis subgroups. Conclusion Clinical staging, radiotherapy method, and cellular immune function status are independent prognostic factors of NPC. The prognostic index model better reflects the prognosis than a single factor such as TNM clinical stage.