Zhou Wei, Chi Hao, Zhao Xiaohu, Tao Guangrong, Gan Jianhe
Department of Infectious Diseases, The First Affiliated Hospital of Soochow University, Suzhou, 215006, Jiangsu, China.
Department of Medical Laboratory, Huaian Hospital of Huaian City, Huaian, China.
Clin Transl Oncol. 2025 Mar;27(3):1166-1175. doi: 10.1007/s12094-024-03669-0. Epub 2024 Aug 23.
The prognosis of hepatocellular carcinoma (HCC) is poor and there is no stable and reliable molecular biomarker for evaluation. This study attempted to find reliable prognostic markers from tumor mutational profiles.
A total of 362 HCC samples with whole-exome sequencing were collected as discovery datasets, and 200 samples with targeted sequencing were used for validation of the relevant results. All HCC samples were obtained from previously published studies. Bayesian non-negative matrix factorization was used to extract mutational signatures, and multivariate Cox regression models were utilized to identify the prognostic role of mutational factors. Gene set enrichment analysis was employed to discover potential signaling pathways associated with specific mutational groups.
In the HCC discovery dataset, a total of four mutational signatures (i.e., signatures 4, 6, 16, and 22) were extracted, of which signature 16 characterized by T>C mutations was observed to be associated with favorable HCC prognosis, and this correlation was also found in the validation dataset. Further analysis showed that patients with ARID1A mutations exhibited inferior survival outcomes in both discovery and validation datasets. Mechanistic exploration revealed that the presence of signature 16 was associated with better immune infiltration and tumor immunogenicity, while patients with ARID1A mutations were away from these favorable immunological features.
By integrating somatic mutation data and clinical information of HCC, this study identified that signature 16 and ARID1A mutations were associated with better and worse outcomes respectively, providing a basis for prognosis prediction and clinical treatment strategies of HCC.
肝细胞癌（HCC）预后较差，且尚无稳定可靠的分子生物标志物用于评估。本研究试图从肿瘤突变谱中寻找可靠的预后标志物。
收集362例进行全外显子测序的HCC样本作为发现数据集，200例进行靶向测序的样本用于验证相关结果。所有HCC样本均取自先前发表的研究。采用贝叶斯非负矩阵分解提取突变特征，并利用多变量Cox回归模型确定突变因素的预后作用。运用基因集富集分析来发现与特定突变组相关的潜在信号通路。
在HCC发现数据集中，共提取出四个突变特征（即特征4、6、16和22），其中以T>C突变为特征的特征16被观察到与HCC良好预后相关，且在验证数据集中也发现了这种相关性。进一步分析表明，在发现数据集和验证数据集中，携带ARID1A突变的患者生存结局较差。机制探索显示，特征16的存在与更好的免疫浸润和肿瘤免疫原性相关，而携带ARID1A突变的患者则缺乏这些有利的免疫特征。
通过整合HCC的体细胞突变数据和临床信息，本研究确定特征16和ARID1A突变分别与较好和较差的结局相关，为HCC的预后预测和临床治疗策略提供了依据。
