A Novel Prognostic Signature of comprising Nine NK Cell signatures Based on Both Bulk RNA Sequencing and Single-Cell RNA Sequencing for Hepatocellular Carcinoma

Qi Yu; Xuefeng Shi; Hongjian Wang; Shun Zhang; Songnian Hu; Ting Cai

doi:10.7150/jca.85873

A Novel Prognostic Signature of comprising Nine NK Cell signatures Based on Both Bulk RNA Sequencing and Single-Cell RNA Sequencing for Hepatocellular Carcinoma

J Cancer. 2023 Jul 16;14(12):2209-2223. doi: 10.7150/jca.85873. eCollection 2023.

Authors

Qi Yu^{1

2

3}, Xuefeng Shi⁴, Hongjian Wang⁵, Shun Zhang^{1

3}, Songnian Hu⁶, Ting Cai^{1

2

3}

Affiliations

¹ Department of Experimental Medical Science, Ningbo No.2 Hospital, Ningbo 315010, China.
² Ningbo Institute of Life and Health Industry, University of Chinese Academy of Sciences, Ningbo 315032, China.
³ Key Laboratory of Diagnosis and Treatment of Digestive System Tumors of Zhejiang Province, Ningbo 315010, China.
⁴ Department of Pulmonary and Critical Care Medicine, Qinghai provincial people's hospital, Xining 81000, China.
⁵ College of Agricultural, Consumer and Environmental Sciences, University of Illinois at Urbana-Champaign, Champaign 61820, USA.
⁶ State Key Laboratory of Microbial Resources, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China.

Abstract

Background: Hepatocellular carcinoma (HCC) has limited prognostic prediction due to its heterogeneity. Understanding the role of natural killer (NK) cells in HCC is vital for prognosis and immunotherapy guidance. We aimed to identify NK cell marker genes through scRNA-seq and develop a prognostic signature for HCC. Methods: We analyzed scRNA-seq data (GSE149614) from 10 patients and bulk RNA-seq data from 786 patients with clinicopathological information. NK cell marker genes were identified using clustering and marker finding functions. A predictive risk signature was constructed using LASSO-COX algorithm. Functional annotations and immune cell infiltration analysis were performed, and the nomogram's performance was evaluated. Results: We identified 79 NK cell marker genes associated with NK cell-mediated cytotoxicity, apoptosis, and immune response. The multigene signature significantly correlated with overall survival (OS) in TCGA-LIHC cohort and was validated in other cohorts. Low-risk patients exhibited higher immune cell infiltration, including CD8⁺ T cells. The risk signature was an independent prognostic factor for OS (HR > 1, p < 0.001). The nomogram combining the risk signature and clinical predictors demonstrated robust prognostic ability. Conclusion: We developed a nine-gene signature prognostic model based on NK cell marker genes to accurately assess the prognostic risk of HCC. This model can be a valuable tool for personalized evaluation post-surgery. Our study underscores the potential of NK cells in HCC prognosis and highlights the importance of scRNA-seq analysis in identifying prognostic markers.

Keywords: 9-gene signature; NK cells; hepatocellular carcinoma; prognosis; scRNA-seq.