基于3种基因启动子甲基化联合端粒长度构建肺癌诊断支持向量机模型

Establishment of support vector machine model of lung cancer screening based on three gene promoter methylation combined with relative telomere length

目的：：探讨基于外周血白细胞 DNA FHIT、RASSF1A 和 p16基因启动子甲基化水平以及 DNA 端粒长度等4项生物标志建立的支持向量机模型在肺癌诊断中的意义。方法：通过实时定量甲基化特异性 PCR 法,测定200例正常人（对照组）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支持向量机模型。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）,27.62（9.09～52.86）和17.17（3.86～50.87）,0.59（0.16～4.50）和0.36（0.06～4.00）,端粒长度分别为（0.93±0.32） kb 和（1.16±0.57） kb,两组间差异有统计学意义（Z =3.044、2.075、2.641和4.972,P 均〈0.05）。基于上述4项生物标志建立的判别分析和支持向量机模型对预测集的 ROC 曲线下面积及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 p16、RASSF1A、FHIT 基因启动子甲基化和端粒长度的肺癌诊断支持向量机模型。

Aim： To explore significance of support vector machine（SVM） model for diagnosis of lung cancer by de-tections of fragile histidine traid（FHIT）,RASSF1A and p16 gene promoter methylation status and relative telomere length of DNA in white blood cells（WBCs） from peripheral blood. Methods： The status of p16,RASSF1A and FHIT gene promoter methylation and relative telomere length of DNA in WBCs from peripheral blood of 200 healthy individuals（control group） and 200 patients with lung cancer were measured by SYBR green-based quantitative methylation-specific PCR and quantita-tive PCR, respectively. Then SVM model based on the above 4 biomarkers was developed. Results： The status of FHIT, RASSF1A and p16 gene promoter methylation was 3. 33（1. 86 - 6. 40） and 2. 85（1. 39 - 5. 44）,27. 62（9. 09 - 52. 86） and 17. 17（3. 86 - 50. 87）,0. 59（0. 16 - 4. 50） and 0. 36（0. 06 - 4. 00） in lung cancer group and control group, and rel-ative telomere length was （0. 93 ± 0. 32） kb and （1. 16 ± 0. 57） kb. There were significant differences in the 4 biomarkers （Z = 3. 044,2. 075,2. 641, and 4. 972,P 〈 0. 05）. Areas of discrimination analysis and SVM model based on the 4 biomar-kers under receiver operating curve and the 95% CI were 0. 670（0. 569 - 0. 761） and 0. 810（0. 719 - 0. 882）, respective-ly. Conclusion： The SVM model for lung cancer diagnosis based on the 4 biomarkers as p16, RASSF1A, FHIT promoter methylation and relative telomere length has been successfully established.