Zheng Jie, Zhu Ming-Yu, Wu Fei, Kang Bin, Liang Ji, Heskia Fabienne, Shan Yun-Feng, Zhang Xin-Xin
Department of Interventional Radiology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou 325000, China.
Department of Gastroenterology, Ruijin Hospital North, School of Medicine, Shanghai Jiao Tong University, Shanghai 201800, China.
Ann Transl Med. 2020 Mar;8(5):195. doi: 10.21037/atm.2020.01.93.
Hepatocellular carcinoma (HCC) is one of the most common and lethal malignancies. Early detection of HCC could largely reduce mortalities. Ultrasonography (US) and serum Alpha Fetoprotein (AFP) test are the screening methods that are most frequently applied to high-risk populations. Due to the poor performance of AFP testing, and the highly operator-dependent nature of US, a biomarker for HCC early diagnosis is highly sought after. We developed a method for HCC screening using a 22-gene expression signature.
Peripheral whole blood of 98 patients were processed through microarrays for the first round of feature selection via two strategies, Minimal Redundancy Maximal Relevance and Least Absolute Shrinkage and Selection Operator combined with Support Vector Machine (SVM). Candidate genes were combined for further validation through qPCR in an enlarged population with 316 samples with 104 chronic hepatitis, 112 liver cirrhosis (LC), and 100 HCC.
A 22-gene signature was established in classifying HCC and non-cancer samples with good performance. The area under curve reached 0.94 in all of the samples and 0.93 in the AFP -negative samples.
We have established a blood mRNA signature with high performance for HCC screening. Our results show transcriptome of peripheral blood could be valuable source for biomarkers.
肝细胞癌（HCC）是最常见且致命的恶性肿瘤之一。早期检测肝细胞癌可大幅降低死亡率。超声检查（US）和血清甲胎蛋白（AFP）检测是最常用于高危人群的筛查方法。由于AFP检测效果不佳，且超声检查高度依赖操作者，因此迫切需要一种用于肝细胞癌早期诊断的生物标志物。我们开发了一种利用22个基因表达特征进行肝细胞癌筛查的方法。
对98例患者的外周全血进行微阵列处理，通过最小冗余最大相关和最小绝对收缩与选择算子结合支持向量机（SVM）这两种策略进行第一轮特征选择。将候选基因组合起来，通过qPCR在一个扩大的人群中进一步验证，该人群包括316个样本，其中有104例慢性肝炎、112例肝硬化（LC）和100例肝细胞癌。
建立了一个22个基因的特征，用于对肝细胞癌和非癌样本进行分类，性能良好。在所有样本中曲线下面积达到0.94，在AFP阴性样本中达到0.93。
我们建立了一种用于肝细胞癌筛查的高性能血液mRNA特征。我们的结果表明外周血转录组可能是生物标志物的宝贵来源。
