论文部分内容阅读
目前,对于肝癌(hepatocellular carcinoma,HCC)早期检测中放射成像技术无法确定的结节,通常需要通过活检取样来进行组织病理学鉴定。然而,一些早期肝癌组织中所呈现的病理改变微小难辨,活检取样时也经常会出现未取到癌组织而漏检的情况。针对这些临床上肝癌早期诊断的困难,同时为了应对基因表达谱检测批次效应等对分类结果的影响,作者提出了利用癌旁组织中基因表达的秩序关系来鉴别早期肝癌及癌前病变的方法。首先找出在肝癌患者的肝硬化组织(cirrhosis tissue in patients with HCC,wHCC)和非肝癌患者的肝硬化组织(cirrhosis tissue in patients without HCC,woHCC)间基因表达值的相对秩序发生逆转的基因对,然后,借用大样本的肝癌组织表达谱数据筛选出HCC及wHCC中表达相对秩序一致的基因对,再基于这些基因对发展能将HCC、wHCC与woHCC进行鉴别的分类器。结果显示,该分类器对来自不同平台、不同实验室的验证集都有稳健的分类能力,可以用于早期肝癌的辅助鉴别诊断。
At present, the nodules undetermined by radiography in the early detection of hepatocellular carcinoma (HCC) usually need to be identified by biopsy for histopathology. However, some of the early liver cancer pathological changes presented in the tiny hard to distinguish, biopsy samples are often taken without cancer tissue missing the case. In response to the difficulties in the early diagnosis of these clinical liver cancers, and in response to the detection of the batch effect of the gene expression profile on the classification results, the authors propose a method for identifying early hepatocarcinoma and precancerous lesions by using the relationship of the order of gene expression in adjacent tissues . First, find the gene pair that reverses the relative order of gene expression values between cirrhosis tissue in patients with HCC (wHCC) and cirrhosis tissue in patients without HCC (woHCC) Then we screened a large number of hepatocellular carcinoma tissue profile data to screen out the gene pairs with the same relative expression order in HCC and wHCC and then classify these genes for the development of HCC, wHCC and woHCC. The results show that this classifier has robust classification ability for validation sets from different platforms and different laboratories and can be used for the differential diagnosis of early-stage liver cancer.