Multi-region sequencing with spatial information enables accurate heterogeneity estimation and risk stratification in liver cancer

• Published in: Genome medicine 2022-12, Vol.14 (1), p.142-142, Article 142
• Main Authors: Yang, Chen, Zhang, Senquan, Cheng, Zhuoan, Liu, Zhicheng, Zhang, Linmeng, Jiang, Kai, Geng, Haigang, Qian, Ruolan, Wang, Jun, Huang, Xiaowen, Chen, Mo, Li, Zhe, Qin, Wenxin, Xia, Qiang, Kang, Xiaonan, Wang, Cun, Hang, Hualian
• Format: Article
• Language: English
• Subjects: Biopsy; Carcinoma, Hepatocellular - genetics; Correlation analysis; Genetic aspects; Genetic transcription; Health aspects; Hepatocellular carcinoma; High-Throughput Nucleotide Sequencing; Hospitals; Humans; Immunotherapy; Intra-tumor heterogeneity; Liver cancer; Liver Neoplasms - genetics; Localization; Medical research; Medicine, Experimental; Metastasis; Microenvironments; Patients; Phenotypes; Prognostication; Regions; Risk Assessment; Risk factors; Sampling; Software; Spatial discrimination; Spatial distribution; Tumor Microenvironment - genetics; Tumors
• ISSN: 1756-994X; 1756-994X
• DOI: 10.1186/s13073-022-01143-6

Numerous studies have used multi-region sampling approaches to characterize intra-tumor heterogeneity (ITH) in hepatocellular carcinoma (HCC). However, conventional multi-region sampling strategies do not preserve the spatial details of samples, and thus, the potential influences of spatial distribution on patient-wise ITH (represents the overall heterogeneity level of the tumor in a given patient) have long been overlooked. Furthermore, gene-wise transcriptional ITH (represents the expression pattern of genes across different intra-tumor regions) in HCC is also under-explored, highlighting the need for a comprehensive investigation. To address the problem of spatial information loss, we propose a simple and easy-to-implement strategy called spatial localization sampling (SLS). We performed multi-region sampling and sequencing on 14 patients with HCC, collecting a total of 75 tumor samples with spatial information and molecular data. Normalized diversity score and integrated heterogeneity score (IHS) were then developed to measure patient-wise and gene-wise ITH, respectively. A significant correlation between spatial and molecular heterogeneity was uncovered, implying that spatial distribution of sampling sites did influence ITH estimation in HCC. We demonstrated that the normalized diversity score had the ability to overcome sampling location bias and provide a more accurate estimation of patient-wise ITH. According to this metric, HCC tumors could be divided into two classes (low-ITH and high-ITH tumors) with significant differences in multiple biological properties. Through IHS analysis, we revealed a highly heterogenous immune microenvironment in HCC and identified some low-ITH checkpoint genes with immunotherapeutic potential. We also constructed a low-heterogeneity risk stratification (LHRS) signature based on the IHS results which could accurately predict the survival outcome of patients with HCC on a single tumor biopsy sample. This study provides new insights into the complex phenotypes of HCC and may serve as a guide for future studies in this field.