Prediction of chemosensitivity for patients with advanced gastric cancer by DNA misroarray analysis, using specimens obtained from fine needle aspiration biopsy.

微阵列的全球基因表达谱分析已被用作鉴定预后和预测标记基因的有价值的工具。尽管这项技术现在已经广泛普及并且相对标准化，但只有很少的数据可以比较化疗中的反应者和无反应者。我们在不可切除或复发的胃癌患者中进行了每周一次紫杉醇和5'-DFUR的I期和II期临床研究。该方案有效率为50%。预测化疗敏感性是很重要的，因为一半的患者对这些药物没有反应。因此，我们分析了胃癌患者预处理活检的基因表达数据，并比较了反应者和无反应者。为了建立一种预测胃癌患者化疗反应的方法，我们使用全基因组DNA微阵列分析了20例晚期胃癌患者的24份活检样本，这些患者接受了相同的pa…More clitaxel和5'-DFUR治疗。24例中有8例为超声内镜下细针穿刺活检。分离RNA，使用Affymetrix U133 plus2 Array进行表达谱分析。三分之二的活检产生了足够量的RNA用于表达谱分析，并为14例患者的16个样本获得了高质量的数据。在细针穿刺活检获得的样本中，8个样本中有6个（75%）产生了足够量的RNA。我们分析了14例患者的14个样本，其中7例有反应，7例无反应。为了从56475个基因中筛选基因，我们首先调查了所有存在p-call的基因，并选择了958个基因。接下来，我们研究了958个基因，这些基因在应答者和无应答者之间的平均差异是50倍。我们使用Fisher比率来评估反应者和无反应者之间的可分离性。在7名应答者和7名无应答者之间鉴定出39个表达差异显著的基因。应答者中有37个基因上调，2个基因下调。少

Global gene expression profiling by microarrays has been used as a valuable tool for the identification of prognostic and predictive marker genes. Even though this technology is now wide spread and relatively standardized, there are only few data available which compare responders with non-responders in chemotherapy. We conducted phase I and II clinical studies of weekly paclitaxel and 5'-DFUR in patients with unresectable or recurrent gastric cancer. Response rate of this regimen was 50%. It is important to predict the chemosensitivity, because half of the patients showed no effect to these agents. Therefore we analyzed gene expression data of pretreatment biopsies of gastric cancer patients and compared responders with non-responders. To establish a method to predict the response of gastric cancer patients to chemotherapy, we used a genome-wide DNA microarray to analyze 24 biopsy samples of advanced gastric cancer from 20 patients who had been treated with an identical protocol of pa … More clitaxel and 5'-DFUR. Eight of 24 samples were obtained from fine needle aspiration biopsy by endoscopic ultrasonography.RNA was isolated and expression profiling performed using Affymetrix U133 plus2 Array. Two thirds of the biopsies yielded sufficient amounts of RNA for expression profiling and high quality data were obtained for 16 samples of 14 patients. In the samples obtained from fine needle aspiration biopsy, 6 (75%) of 8 samples yielded sufficient amounts of RNA. We analyzed 14 samples of 14 patients who were consisted of 7 responders and 7 non-responders. To filter genes out of the 54675,we first investigate all genes for which p-call were present, and selected 958 genes. We next investigate 958 genes for which mean average differences were>2-fold between responders and non-responders. We used the Fisher ratio to evaluate separatability between responders and non-responders. Thirty-nine genes whose expression differed significantly between 7 responders and 7 non-responders were identified. Thirty-seven genes were up-regulated and two genes were down-regulated in responders. Less