胃癌阿霉素敏感性预测模型的建立与验证

发布时间：2018-08-18 14:31

【摘要】： 化疗是治疗进展期胃癌的重要手段,阿霉素作为最常用的化疗药物之一,临床上常与其他化疗药物(如5-氟尿嘧啶、顺铂)配伍用于胃癌的化疗,但是临床上胃癌化疗的效果并不满意,有文章报道胃癌对阿霉素的敏感性仅为30%,其中最主要的原因是胃癌细胞具有天然耐药或者在化疗过程通过接触化疗药物中获得耐药性。因此,提前预测胃癌患者是否对阿霉素敏感对于化疗效果非常重要,然而目前临床并未常规开展药物敏感性检测的细胞学实验,主要原因在于其细胞培养周期长、易污染、难于标准化。利用基因芯片技术预测化疗药物敏感性已见诸报道,但此方法耗资巨大,对实验室硬件条件要求较高,同样也难以临床常规开展。为此我们在本研究中探讨是否可能利用基因芯片提供的信息,获得一些能够预测胃癌细胞阿霉素敏感性的关键分子,建立阿霉素耐药预测模型。【目的】利用基因芯片信息建立胃癌阿霉素敏感性多分子预测模型,用于辅助临床化疗决策。【方法】1. cDNA微阵列分析技术检测胃癌细胞及其阿霉素耐药细胞SGC7901与SGC-7901/ADR全基因表达谱间的差异基因;2.从斯坦福大学基因芯片数据库(http://genome-www5.stanford.edu)引入另外一株胃癌阿霉素耐药细胞EPG-257RDB及其亲本细胞EPG-257P的全基因表达谱并分析其差异性基因;3.比较两个细胞系间的共同差异性基因;4.利用cDNA微阵列分析软件PAM筛选后续实验的目的基因;5.采用RT-PCR技术检测人胃癌组织标本中目的基因的表达情况;6.采用胃癌原代细胞培养法测定人胃癌标本阿霉素敏感性;7.利用统计软件对人胃癌组织标本中目的基因的表达情况及阿霉素敏感性作多元回归分析,建立多分子预测模型;8.对验证组的胃癌标本作相同的处理并统计分析多元回归模型的可靠性。【结果】1.胃癌细胞及其耐药细胞SGC7901与SGC-7901/ADR对比共发现差异性基因471个;EPG-257RDB与其亲本细胞EPG-257P对比共发现差异性基因10,144个;2.通过对比两个细胞系的胃癌细胞及其阿霉素耐药细胞的全基因表达谱,得到90个在表达上具有相同趋势的差异性基因,其中仅PJA1在耐药细胞低表达,余基因均在耐药细胞中高表达;3.利用cDNA微阵列分析软件PAM筛选后得到7个与阿霉素化疗敏感性相关的候选基因,分别是ADAM22、CYR61、IFITM1、FN1、SPHK1、G1P2、GNAI1,均在耐药细胞中高表达;4.利用NLReg统计软件对20例人胃癌组织标本中候选基因的表达情况及阿霉素敏感性作多元回归分析,建立多分子阿霉素敏感性预测模型;此模型中IFITM1和G1P2这两个基因被排除,预测模型的相关系数达0.8838;5. 19例人胃癌组织标本给予同上的处理,对目的基因的表达情况及阿霉素敏感性结果进行验证,证明建立的多分子阿霉素敏感性预测模型有效(r = 0.7316);6.对胃癌新鲜组织标本进行HDRA检测发现阿霉素的敏感性低于30%。【结论】1.建立了基于ADAM22、CYR61、FN1、SPHK1、GNAI1这5个基因的多分子阿霉素敏感性预测模型,并经验证可用于临床胃癌患者化疗前的阿霉素耐药的评估;2.胃癌对阿霉素耐药情况普遍,利用预告模型对阿霉素的耐药性进行预测有助于提高化疗疗效;3.在对耐药细胞的全基因组筛选中,预先进行脱药培养有助于减少细胞凋亡途径相关基因的干扰,获得影响其敏感性的关键基因。
[Abstract]:Chemotherapy is an important method for the treatment of advanced gastric cancer. As one of the most commonly used chemotherapeutic drugs, adriamycin is often used in combination with other chemotherapeutic drugs (such as 5-fluorouracil, cisplatin) in clinic. However, the effect of chemotherapy for gastric cancer is unsatisfactory. It has been reported that the sensitivity of gastric cancer to adriamycin is only 30%, the most important of which is adriamycin. The reason is that gastric cancer cells have natural resistance or acquire resistance through exposure to chemotherapy drugs. Therefore, it is very important to predict whether gastric cancer patients are susceptible to adriamycin in advance. However, there is no routine clinical cytological test for drug sensitivity testing, mainly because of its cell culture. Gene chip technology has been reported to predict chemosensitivity, but this method is expensive, requires high hardware conditions in the laboratory, and is also difficult to carry out routine clinical practice. A predictive model for adriamycin resistance was established to predict the sensitivity of gastric cancer cells to adriamycin.
[Objective] To establish a multi-molecular prediction model of adriamycin sensitivity in gastric cancer by using gene chip information for assisting clinical chemotherapy decision-making.
[Methods] 1. Differential gene expression profiles between gastric cancer cells and their doxorubicin-resistant cells SGC7901 and SGC-7901/ADR were detected by cDNA microarray analysis; 2. Another doxorubicin-resistant cell line EPG-257RDB and its parent cell EPG-257P were introduced from the Stanford University Gene Chip Database (http://genome-ww5.stanford.edu). Total gene expression profiles and differential genes were analyzed; 3. Comparing the common and differential genes between the two cell lines; 4. Screening the target genes of the follow-up experiment by using cDNA microarray analysis software PAM; 5. Detecting the expression of target genes in human gastric cancer tissues by RT-PCR; 6. Detecting the expression of target genes in human gastric cancer by primary cell culture method. Specimens sensitivity to adriamycin; 7. Using statistical software for human gastric cancer tissue specimens in the expression of the target gene and adriamycin sensitivity multivariate regression analysis, to establish a multimolecular prediction model; 8. To verify the same treatment of gastric cancer specimens and statistical analysis of the reliability of the multivariate regression model.
[Results] A total of 471 differentially expressed genes were found in gastric cancer cells and their drug-resistant cells SGC7901 and SGC-7901/ADR; 10,144 differentially expressed genes were found in EPG-257RDB and its parental cells EPG-257P; and 90 differentially expressed genes were obtained by comparing the whole gene expression profiles of gastric cancer cells and their doxorubicin-resistant cells. Among them, only PJA1 was low expressed in drug-resistant cells, and the remaining genes were high expressed in drug-resistant cells. 3. Seven candidate genes related to chemotherapy sensitivity of adriamycin, ADAM22, CYR61, IFITM1, FN1, SPHK1, G1P2, GNAI1, were high expressed in drug-resistant cells after screening by cDNA microarray analysis software PAM. The expression of candidate genes and their sensitivity to adriamycin in 20 human gastric cancer tissues were analyzed by using NLReg statistical software and a multivariate regression model was established to predict the sensitivity to adriamycin. The expression of the target gene and the results of adriamycin sensitivity were validated, which proved that the multi-molecule adriamycin sensitivity prediction model was effective (r = 0.7316). 6. The sensitivity of adriamycin was less than 30% in fresh gastric cancer tissues detected by HDRA.
[Conclusion] 1. Established the multi-molecule adriamycin sensitivity prediction model based on ADAM22, CYR61, FN1, SPHK1, GNAI1 genes, and proved that it can be used to evaluate the adriamycin resistance of gastric cancer patients before chemotherapy. 2. The drug resistance of gastric cancer to adriamycin is common, and the prediction model is helpful to improve the adriamycin resistance. Chemotherapeutic effect; 3. In genome-wide screening of drug-resistant cells, pre-drug-free culture is helpful to reduce the interference of apoptosis pathway-related genes and obtain the key genes affecting their sensitivity.
【学位授予单位】：第四军医大学
【学位级别】：博士
【学位授予年份】：2008
【分类号】：R735.2;R-332

【相似文献】