A novel hypoxia gene signature indicates prognosis and reveals the multi-omics molecular landscape of tumour tissue in patients with hepatocellular carcinoma

Qiangnu Zhang,Juan Liao,Lijun Qiao,Quan Liu,Pengyu Liu,Mengting Xia,Xiaotao Huang,Liping Liu

DOI: https://doi.org/10.1101/2020.07.10.198176

2020-01-01

Abstract:AbstractBackgroundPrevious studies on the impact of hypoxia on hepatocellular carcinoma (HCC) mostly focused on in vitro and animal models. The clinical value of assessing the degree of hypoxia in in vivo tissues and hypoxia-related molecular landscapes in HCC remain poorly defined.MethodsA novel hypoxia gene signature was extracted from hypoxia-treated HCC cells using microarray analysis and a robust rank aggregation algorithm and was verified using public data. Next, the relationships between the hypoxia gene signature and the clinical characteristics and prognoses of HCC patients were analysed. Based on the multi-omics data from The Cancer Genome Atlas-Liver Hepatocellular Carcinoma (TCGA-LIHC) and 10 independent HCC cohorts from the Gene Expression Omnibus (GEO), a comprehensive analysis was conducted using the hypoxia gene signature to describe hypoxia-associated multi-omic molecular landscapes and immune microenvironments in HCC tissues.ResultsA 21-gene hypoxia signature was constructed to effectively indicate the exposure of hypoxia in HCC tissues. This novel hypoxia signature and the hypoxia scores calculated based on this signature were closely correlated to some clinical characteristics of HCC patients and can be used to evaluate prognosis. HCC tissues with different hypoxia scores differed significantly in transcriptomic, genomic, epigenomic, and proteomic alterations and immune microenvironments, some of which were related to the clinical prognoses of patients.ConclusionThe 21-gene signature can effectively estimate hypoxia exposure of HCC tissues and has clinical value in the assessment and prediction of the prognosis of HCC patients. Using this 21-gene signature, hypoxia-associated molecular landscapes were described at the tissue level. This comprehensive molecular-level understanding can help to further elucidate the mechanism of hypoxia in tumours and guide clinical cancer therapy. The assessment of the degree of hypoxia is strongly recommended in the personalized treatment of HCC patients to benefit specific patient groups.