辅助放化疗对早期胃癌预后影响:基于人群倾向性评分匹配分析

Effect of adjuvant chemoradiotherapy on prognosis of early gastric cancer: a propensity score matching analysis

目的探讨放化疗对早期胃癌(early gastric cancer,EGC)患者术后总生存期(overall survival,OS)和肿瘤特异性生存期(cancer-specific survival,CSS)的影响。方法提取SEER数据库中2010年至2015年诊断的EGC患者信息,根据纳入与排除标准筛选患者,通过Cox回归分析和竞争风险回归分析分别探讨放化疗对OS和CSS的影响。采用倾向性评分匹配分析(propensity score matching,PSM)校正混杂因素后进一步验证。结果纳入研究的1 262例EGC患者,研究截止时共有238例死亡,其中因胃癌相关原因死亡121例。放化疗组(187例)与未放化疗组(1 075例)OS比较,差异无统计学意义(P>0.05),单因素Cox回归分析表明,放化疗对OS无影响(P=0.433),多因素Cox回归分析提示,种族、性别、年龄、肿瘤大小、浸润深度(T)和淋巴结受累(N)情况是影响EGC患者术后OS的独立风险因素(P<0.05)。虽然放化疗组肿瘤特异性死亡累积发生率显著高于未放化疗组(P=0.015),但竞争风险回归分析表明,放化疗并未对CSS产生影响(P=0.820)。在T1N1亚组中,放化疗同样未对OS和CSS产生影响(P>0.05)。PSM校正混杂因素后,两组间比较OS和CSS差异均无统计学意义(P>0.05)。进一步亚组分析,无论肿瘤是局限于黏膜层(T1a),还是浸润黏膜下层(T1b),亦或淋巴结是否受累,放化疗组与未放化疗组比较,OS和CSS差异均无统计学意义(P>0.05)。结论 EGC患者术后辅助放化疗并不能使患者OS及CSS获益,需要进行前瞻性随机对照研究进一步证实。

Objective To investigate the effect of adjuvant chemoradiotherapy on the overall survival(OS) and cancer-specific survival(CSS) of patients with early gastric cancer(EGC) after surgery. Methods Patients with EGC diagnosed from 2010 to 2015 in SEER database were screened according to inclusion and exclusion criteria, and Cox regression and competitive risk regression were used to explore the effects of chemoradiotherapy on OS and CSS, respectively. Propensity score matching(PSM) was applied to adjust confounding factors for further validation. Results Among the 1 262 EGC patients involved in the study, a total of 238 patients died at the last follow-up, including 121 patients who died from gastric cancer-specific causes. There was no difference in OS between chemoradiotherapy group(187 cases) and non-chemoradiotherapy group(1 075 cases). Univariate Cox regression analysis showed that chemoradiotherapy had no influence on OS(P=0.433). Multivariate Cox regression analysis showed that race, gender, age, tumor size, invasion depth(T) and lymph node(N) involvement were independent factors affecting OS in postoperative EGC patients(P<0.05). Although the cumulative incidence of cancer-specific death in chemoradiotherapy group was significantly higher(P=0.015), the competing risk regression analysis indicated that chemoradiotherapy was unrelated to CSS(P=0.820). In the T1 N1 subgroup, chemoradiotherapy also had no effect on OS and CSS(P>0.05). After PSM adjusted for confounding factors, there was no significant difference between two groups in OS and CSS(P>0.05). Further subgroup analysis showed that there was no difference between chemoradiotherapy group(113 cases) and non-chemoradiotherapy group(113 cases) in OS and CSS, regardless of whether the tumor was infiltrated to the mucosa(T1 a) or submucosa(T1 b), or whether lymph nodes were involved(P>0.05). Conclusion Postoperative adjuvant chemoradiotherapy can not benefit OS and CSS in EGC patients, which needs further validation in a prospective randomized controlled study.