The strategy based on gene expression profiling using microarray for tailor-made treatment in patients with lung cancer

目的淋巴结转移和肿瘤复发是影响恶性肿瘤预后的主要因素，但其分子机制尚不清楚。本研究的目的是鉴定正常肺组织和肺癌组织之间差异表达的基因，并研究100例原发性肺癌的基因表达谱，以选择一组用于肺癌临床特征的基因预测因子。并通过评估所选基因表达水平的统计学显著性来构建预测集。结果在候选基因的搜索中，腺癌中有530个基因表达差异，鳞癌中有519个基因表达差异。94个基因仅在有淋巴结转移的癌组织中表现出不同的表达模式，60个基因表现出与肿瘤复发有关。使用最合适的基因组，可以预测肺癌患者的临床未来。预测评分系统预测腺癌的准确率为90.9%，预测鳞状细胞癌的准确率为91.5%，预测淋巴结转移的准确率为71.4%，结论基因表达分析结合统计学分析可有效区分乳腺癌的组织病理学类型和临床特征，如淋巴结转移和淋巴结转移。复发本研究的结果显示了基于病理诊断和分子分类相结合的肺癌亚组的可能性。

Purpose Lymph node metastasis and tumor recurrence are major factors associated with poor prognosis in the cancer, but little is known of their molecular mechanisms. The aim of this study was to identify genes differentially expressed between normal and cancerous lung tissues, and to investigate the gene-expression profiles of 100 primary lung cancers to select a set of gene predictors for the clinical features of lung cancer.Experimental Design Gene expression profiles were obtained using an oligonucleotide microarray, and the construction of predictor sets was performed by evaluating the statistical significance of the expression levels of selected genes.Results In the search for candidate genes, 530 genes showed differential expression in adenocarcinoma and 519 genes in squamous cell carcinoma. Ninety-four genes showed a distinct expression pattern exclusively in cancer tissues with lymph-node metastasis and 60 genes showed involvement with tumor recurrence. Using the most suitable set of genes, it was possible to predict the clinical futures of patients with lung cancer. The prediction scoring system yielded 90.9% accuracy for forecasting adenocarcinoma and 91.5% accuracy for squamous cell carcinoma, 71.4% accuracy for forecasting lymph node metastasis and 84.6% accuracy for tumor recurrence in independent cases.Conclusions The gene expression analysis and the combination of statistical analysis successfully distinguished histopathological types and clinical features such as lymph node metastasis and recurrence. The findings of this study show the possibility of subgrouping lung cancer based on the combination of pathological diagnosis and molecular classification.