| 摘要: |
| 目的 利用指纹基因群(GES)分析技术挖掘现有公共芯片数据揭示锯齿型结直肠癌的预后和耐药特性。方法 从基因表达谱数据储存公共平台(GEO)中下载4个有预后信息的结直肠癌表达谱芯片数据(GSE14333、GSE17538、GSE33113、GSE37892),用批间误差校正和单基因信息抽提等信息技术获得1个整合的表达谱队列(n=600)。获得锯齿型结直肠癌表达谱有关数据及相应特异性指纹基因群,并与前面的表达谱队列合并,建立指纹基因群评分系统,依托已知锯齿型结直肠癌样本的基因群评分确定分组界值。根据分组界值,将600例结直肠癌分为锯齿型、过渡型和传统型结直肠癌组,利用Kaplan-Meier 和Cox模型分析锯齿型结直肠癌的预后特征。同时,收集并下载与不可切除结直肠癌姑息治疗有关的表达谱数据(GSE28702、GSE5851),经表达谱数据抽提整理后,获得每个样本的锯齿型指纹基因群评分,通过比较治疗有效组和无效组之间评分的统计学差异来揭示锯齿型结直肠癌与药物耐药的关系。结果 根据结直肠癌亚型评分界值,600个结直肠癌中有50个被分为锯齿型,126个被分类为过渡型,424个被分为传统型。锯齿型和过渡型结直肠癌与传统型结直肠癌相比具有几乎完全相同的预后能力,多因素Cox模型发现锯齿型和过渡型结直肠癌是患者术后复发的独立危险因素。单纯比较锯齿型和传统型结直肠癌,多因素Cox模型发现锯齿型结直肠癌是患者预后的不良因素(AHR=1.792,95%CI 1.011~3.177)。西妥昔治疗有效组的指纹基因群评分显著低于无效组(P=0.017),但FOLFOX治疗有效组和无效组的基因群评分无差异。结论 结直肠癌锯齿型亚型是结直肠癌患者术后复发的独立危险因素,且与西妥昔单抗治疗耐药相关。 |
| 关键词: 锯齿型结直肠癌 指纹基因群 预后 肿瘤抗药性 |
| DOI：10.3724/SP.J.1008.2014.01247 |
| 投稿时间：2014-01-12修订日期：2014-09-25 |
| 基金项目: |
|
| Integrated gene expression signature analysis of prognosis and drug-resistance in patients with serrated colorectal adenocarcinoma |
| MIAO Hua1△,CAO Fu-ao2△,LI Xu2,MIAO Zong-yuan1,YE Chun1,ZHANG Chang2,WANG Han-tao2* |
(1. Department of General Surgery, The First People's Hospital of Pinghu, Pinghu 314200, Zhejiang, China;
2. Department of Colorectal Surgery, Changhai Hospital, Second Military Medical University, Shanghai 200433, China
△Co-first authors.
*Corresponding authors) |
| Abstract: |
| Objective To conduct a integrated analysis of gene expression signature and gene expression profile, so as to provide reference for the prognosis and drug-resistance of serrated colorectal adenocarcinoma (SCA). Methods We downloaded four gene expression datasets (GSE14333, GSE17538, GSE33113, and GSE37892) of colorectal carcinoma with the follow-up survival data from GEO database, and then integrated them into a entire expression profile (n=600) with batch adjustment and gene expression value extraction. An SCA gene signature and the corresponding expression profile were integrated with the previous expression dataset. Using the gene signature score model we assigned score to each patient in the gene expression dataset and classified the patients into serrated, transitional, or conventional subtypes. Kaplan-Meier analysis and Cox model were used to compare the risks of cancer recurrence between three subtypes of colorectal carcinoma. Gene signature model were also used to generate the score for each patient in the datasets associated with alleviative therapy of colorectal cancer(GSE28702, GSE5851). The SCA drug-resistance was analyzed by observing the therapeutic effective group and non-effective group. Results According to cut-off value of CRC subtypes, 600 patients in the combined dataset were classified as 50 with serrated subtype, 126 with transitional subtype, and 424 with conventional subtype. Survival analysis showed the serrated and transitional subtypes had very similar scores to predict patient survival, and they were also independent risk factors for postoperative recurrence.When comparing serrated and conventional subtypes, multivariate Cox model analysis indicated the patients with serrated subtype was an unfavorable independent risk factor for prognosis(AHR=1.792, 95%CI 1.011-3.177). Responders to cetuximab treatment had significantly higher signature scores than non-responders (P=0.017), while responders to FOLFOX treatment had similar signature scores with the non-responders. Conclusion Serrated subtype is an independent risk factor of postoperative recurrence in SCA patients, and is related to the treatment with cetuximab. |
| Key words: serrated colorectal adenocarcinoma gene expression signature prognosis neoplasm drug resistance |
