500例中晚期食管癌单纯放疗的多因素分析

Prognostic factors of locally advanced esophageal carcinoma treated by radiotherapy alone

目的分析中晚期食管癌单纯放疗的预后因素,为个体化治疗方案制定和准确评估预后提供参考标准。方法单纯放疗的中晚期胸段食管癌500例中,男318例,女182例,中位年龄56.5岁。胸上段癌195例,胸中段癌277例,胸下段癌28例。X线片病变长度≤3.0cm者36例,3.1～7.0cm者318例,7.1～10.0cm者132例,>10.0cm者14例。胸部CT扫描显示病变长度≤3.0cm者14例,3.1～7.0cm者130例,7.1～10.0cm者298例,>10.0cm者58例。病变最大层面上测量肿瘤最大直径≤2.0cm者87例,2.1～3.0cm者153例,>3.0cm者260例。应用加速器6MVX线给予一前、两后斜野等中心或源皮距照射,常规分割或后程加速超分割,总剂量60～66Gy,30～39分次,6.0～6.3周完成。结果全组1、3、5年生存率分别为71.1%、32.6%、20.8%。单因素分析显示传统预后因素性别、声哑、病变部位、病变长度、软组织阴影、溃疡和食管腔扭曲成角均明显影响预后;CT扫描相关因素如病变长度、最大外侵深度、最大直径、气管支气管受侵、降主动脉受侵、椎前三角消失和淋巴结转移也显著影响长期生存;治疗相关因素如前垂直野宽、野长、后斜野宽、野长、总剂量、总疗程时间、即时疗效和近期疗效对生存也有显著影响。Cox多因素分析显示淋巴结转移、CT显示的病变最大直径、声哑、CT显示病变长度、病变部位、病变最大外侵程度、后斜野宽度、总疗程时间、即时疗效和近期疗效均为独立性预后因素。综合多种因素的预后指数PI值越低,预后越好。结论中晚期食管癌单纯放疗后,除传统预后因素外,胸部CT扫描能提供更准确的评估预后的指标,预后指数模型能够更敏感地预测疗后生存。

Objective To evaluate the prognostic factors of locally advanced esophageal carcinoma treated by radiotherapy alone and provide a proposal for individualized treatment decision making and prognosis estimating. Methods 500 patients with locally advanced esophageal carcinoma treated by radiotherapy alone were analyzed retrospectively. There were 318 male and 182 female, with median age of 56.5 years. The lesions in upper-, middle-, and lower thoracic esophagus were 195, 277, 28 patients, respectively. The length of the lesion shown by barium esophagogram was: ≤3.0?cm 36; 3.1-7.0?cm 318; 7.1- 10.0?cm 132 and >10.0?cm 14 patients, whereas the length of lesion shown by CT scan was:≤3.0?cm 14; 3.1-7.0?cm 130; 7.1-10.0?cm 298 and > 10.0?cm 58 patients respectively. The largest diameter shown by CT scan was: ≤2.0?cm 87; 2.1-3.0?cm 153; and > 3.0?cm 260 patients. Radiotherapy was delivered by conventional fractionation or accelerated hyperfractionated irradiation with 6?MV linear accelerator X-ray, the treatment regimen consisted of one anterior field and two posterior oblique fields, SSD or SAD technique, to a total dose of 60-66?Gy by 30-39 fractions within 6.0- 6.3 w. Kaplan-Meier method was used to analyze the survival and log-rank test was used to evaluate the differences between the groups. Cox regression model and prognosis index (PI) model were used to analyze the prognostic factor. Results The 1-,3-,and 5-year overall survivals were 71.1%, 32.6% and 20.8%,respectively. By univariate analysis, it was proved that gender, hoarseness, lesion characteristics shown by barium esophagogram such as lesion location ,soft tissue shadow, tumor ulcer and esophagus angularity; lesion characteristics shown by CT scan such as length of lesion, depth of tumor invasion, largest diameter of lesion, invasion of trachea and/or bronchus, invasion of descending aorta, disappearance of pre-vertebral trigone and lymph node metastasis; treatment relevant factors such as the width and length of anterior irradiation field and posterior oblique field, total dose, total treatment time, immediate response were prognostic factors for the survival. By multivariate analysis, it was proved that lymph node metastasis, hoarseness, lesion characteristics shown by CT scan such as length of lesion, location of lesion, largest diameter of lesion, largest depth of tumor invasion and the treatment relevant factors such as the width of posterior oblique field, total treatment time, immediate response were independent prognostic factors for the survival. The prognosis index (PI) score sensitively predicted the long term survival of esophageal carcinoma. Conclusion Esophagus CT scan can provide more accurate information for evaluating the prognosis of esophageal carcinoma treated by radiotherapy alone.