Objective: To verify the value of whole genomic copy number variation (WGCNV) detection and scoring system in the diagnosis and prognosis of lung adenocarcinoma. Methods: Seventy-six lung adenocarcinoma specimens including ninety-one tumor samples and twenty adjacent non-tumor lung tissue samples were collected using Laser capture microdissection (LCM). Whole genomic amplification (WGA) was used to enrich DNA and construct a sequencing library for next generation sequencing (NGS). Changes of larger than 5Mb CNV in this study were analyzed and scored. The nuclear grading and score of tumor cells in the surgery and pleural effusion cytology of lung adenocarcinoma specimens were evaluated separately. For each case, we evaluated (1) nuclear size, (2) mitotic counts, (3) nuclear atypia, (4) atypical mitoses. The data of disease-free survive (DFS) and overall survive (OS) were collected for assessing the prognostic value of WGCNV score. Meanwhile, receiver operating characteristic (ROC) and area under curve (AUC) were used to define a cut-off value and evaluate the diagnostic significance in lung adenocarcinoma. Results: The WGCNV scores of twenty adjacent non-tumor lung tissue samples were treated as normal control and all of WGCNV scores of tumor samples range from 0 to 9.95, the median score was 2.7. The WGCNV scores were divided into three groups: low score group <1.74, medium score grade 1.74~4.23, high score grade >4.23. The WGCNV score was positively associated with the nuclear grade scoring (r=0.780 90, P<0.001). The result for evaluation of prognostic value of the WGCNV scores showed that comparing with low WGCNV score group, Hazard Ratio (HR) of medium score group was 4.11 (95%CI=0.72~23.57) and high score group was 2.07 (95%CI=0.30~14.12). These results suggested that the risks of the medium and high WGCNV score group elevated. According to the analysis results of ROC curve, when the cut off value was 0.01, the sensitivity and specificity for lung adenocarcinoma diagnosis were 97.8% and 95.0% respectively, the positive predictive value (PPV) and negative predictive value (NPV) were 99.0% and 90.1%, respectively, the AUC was 0.981. In the differentiation of adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA) group and invasive adenocarcinoma group, when the cut off value was 1.8, the sensitivity and specificity between the two groups were 78.1% and 94.4%, and the PPV and NPV were 98.0% and 52.0%, respectively, the AUC was 0.896. Conclusion: This study verifies that WGCNV scoring system has a potential diagnostic and prognostic value in lung adenocarcinoma.
目的： 探讨全基因组拷贝数变异(WGCNV)检测和评分系统在肺腺癌诊断和预后预测中的价值。 方法： 应用显微微切割技术获取76例肺腺癌患者肿瘤组织标本91份和癌旁正常对照组织样本20份，通过全基因组扩增(WGA)富集DNA并构建测序文库，进行二代测序。评估肺腺癌手术和恶性胸水细胞学标本的肿瘤细胞核级别改变，在肿瘤核细胞大小、核分裂象计数、细胞核不典型性和不典型核分裂4个方面进行分级。评价WGCNV评分的预测预后的价值，根据受试者工作特征(ROC)曲线分析WGCNV评分在肺腺癌中的诊断价值。 结果： 20例癌旁正常肺组织样品WGCNV改变作为正常对照，所有肿瘤标本WGCNV评分为0～9.95分，中位WGCNV评分为2.7分。核级别评分与WGCNV评分之间存在正相关(r＝0.780 90，P<0.000 1)。中评分组(1.74～4.23分)相对低评分组(<1.74分)的风险比为4.11(95% CI为0.72～23.57)，高评分组(>4.23分)相对低评分组的风险比为2.07(95% CI为0.30～14.12)，中、高评分组风险呈上升趋势。ROC曲线分析显示，当临界值为0.01时，诊断肺腺癌的灵敏度和特异度分别为97.8%和95.0%，阳性预测值(PPV)和阴性预测值(NPV)分别为99.0%和90.1%，ROC曲线下面积(AUC)为0.981，WGCNV评分具有较好的诊断肺腺癌的价值。当临界值为1.8时，鉴别浸润性腺癌与原位腺癌及微浸润性腺癌的灵敏度和特异度分别为78.1%和94.4%，PPV和NPV分别为98.0%和52.0%，AUC为0.896。 结论： WGCNV评分系统在肺腺癌标本中有一定的诊断和预测预后价值。.
Keywords: Adenocarcinoma; Diagnosis; Gene copy number variation; Lung neoplasms; Nucleus; Prognosis.