中文摘要：

目的探讨由外周血白细胞DNA中FHIT、RASSF1A和p16基因启动子甲基化水平以及DNA相对端粒长度4项生物标志建立的决策树模型（Decision tree,DT）在肺癌诊断中的意义。方法通过实时甲基化特异性PCR（Real-time methylation specific PCR）法,测定200例正常对照和200例肺癌患者外周血白细胞DNA中FHIT、RASSF1A和p16基因启动子甲基化水平,实时荧光定量PCR方法测定外周血DNA相对端粒长度,基于上述4种生物标志构建肺癌诊断决策树模型。结果肺癌组和对照组中FHIT、RASSF1A和p16基因启动子甲基化水平分别为3.33（1.86~6.40）和2.85（1.39~5.44）（P=0.002）,27.62（9.09~52.86）和17.17（3.86~50.87）（P=0.038）,0.59（0.16~4.50）和0.36（0.06~4.00）（P=0.008）,端粒相对长度分别为0.93±0.32和1.16±0.57（P〈0.001）,这4项生物标志在两组间差异有统计学意义。基于4项生物标志建立的判别分析和决策树模型对预测集的的ROC曲线下面积（AUC）及95%CI分别为0.670（0.569~0.761）和0.810（0.719~0.882）。结论成功构建基于FHIT、RASSF1A、p16基因启动子甲基化和端粒损伤4种生物标志的肺癌诊断决策树模型。

英文摘要：

Objective To Identify significances of Decision tree model for diagnosis of lung cancer by detections of fragile histidine traid（ FHIT）,RASSF1 A and p16 genes promoter methylation status and relative telomere length of white blood cells in peripheral blood DNA. Methods The status of FHIT RASSF1 A and p16,genes promoter methylation and relative telomere length in peripheral blood DNA of 200 healthy individuals and 200 patients with lung cancers were measured by SYBR green- based quantitative methylation- specific PCR and quantitative PCR respectively,and then Decision tree model which was based on the four biomarkers was developed. Results The status of FHIT,RASSF1 A and p16 genes promoter methylation,respectively,were 3. 33（ 1. 86 ~ 6. 40）,2. 85（ 1. 39 ~ 5. 44）（ P = 0. 002）,27. 62（ 9. 09 ~ 52. 86）,17. 17（ 3. 86 ~ 50. 87）（ P = 0. 038）,0. 59（ 0. 16 ~ 4. 50）,0. 36（ 0. 06 ~ 4. 00）（ P = 0. 008）. Relative telomere length,respectively,were（ 0. 93 ± 0. 32）,（ 1. 16 ± 0. 57）（ P〈0. 001）. There were statistically significant differences in these four biomarkers. Areas of discrimination analysis and Decision tree model which were based on four biomarkers under receiver operating curve were 0. 640 and 0. 830,respectively. Conclusion Decision tree model of lung cancer diagnosis which was based on four biomarkers of FHIT,RASSF1 A,p16 promoter methylation and telomere damage was successfully builded.