人乳头状瘤病毒感染与肺癌预后相关性的Meta分析

目的 采用Meta分析方法探索人乳头瘤状病毒(Human papillomavirus, HPV)感染与肺癌预后的相关性。方法 利用主题词与自由词结合的方式, 检索PubMed、Embase和Cochrane文献数据库。截至2018年10月, 共检索获得123篇相关文献。根据纳入和排除标准筛选文献后, 摘录研究基本信息、HPV检测方法、肺癌患者、风险比(Hazard ratio, HR)值和95%可信区间(Confidence interval, CI)等指标。利用随机效应模型的Meta分析, 评价HPV感染与肺癌患者预后相关性。利用Q检验和I²统计量评价异质性, 利用Egger's和Begg's法检验发表偏倚。结果 本研究最终纳入文献11篇(亚洲地区9篇, 欧洲地区和美国2篇), 纳入肺癌患者1501例。使用随机效应模型的Meta分析结果显示, HPV感染与肺癌预后无统计学关联, 合并HR=0.90, 95% CI:0.71~1.13。针对肺癌病理亚型进行分层分析, 结果提示, HPV感染肺腺癌患者的预后明显优于无HPV感染肺腺癌患者, 合并HR=0.65, 95% CI:0.49~0.85。通过依次剔除纳入的研究, 进行敏感性分析, 结果无统计学变化。Egger's检验(P=0.708)和Begg's检验(P=0.784)的结果提示, 本研究不存在发表偏倚。结论 HPV感染与肺腺癌患者预后可能相关, 未来仍需更多基础和临床研究进一步探索HPV感染和肺腺癌的关联及其背后的机理机制。

Relationship between human papillomavirus infection and prognosis of lung cancer:A meta-analysis

Objective The Objective of this study was to explore the association between human papillomavirus(HPV)and prognosis of lung cancer by meta-analysis. Methods The PubMed, Embase and Cochrane literature databases studies were searched using a combination of subject terms and free words.As of October 2018, a total of 123 related documents were obtained.After screening the literature according to inclusion and exclusion criteria, the basic information of the study, HPV detection methods, lung cancer patients, hazard ratio(HR)values and 95% confidence interval(CI)were extracted from each study.The meta-analysis of random effects models was used to evaluate the correlation between HPV infection and prognosis in patients with lung cancer.Heterogeneity was assessed using the Q test and I² statistics, and publication bias was tested using Egger's linear regression test and Begg's rank correlation test. Results The study finally included 11 articles(9 in Asia, 2 in Europe and US), and 1439 patients with lung cancer.Meta-analysis using a random-effects model showed no significant association between HPV infection and prognosis of lung cancer(HR=0.90, 95% CI:0.71~1.13).A stratified analysis of lung cancer pathological subtypes showed that the prognosis of patients with HPV-infected lung adenocarcinoma was significantly better than that in patients without HPV-infected lung adenocarcinoma(HR=0.65, 95% CI:0.49~0.85).Sensitivity analysis was performed by sequentially removing the included studies, and the results were not statistically significant.The results of Egger's test(P=0.708)and Begg's test(P=0.784)suggest that there is no publication bias in this study. Conclusion HPV infection may be related to the prognostic of patients with lung adenocarcinoma.More basic and clinical studies are needed to further explore the association between HPV infection and lung adenocarcinoma as well as the corresponding mechanisms in the future.