Bioinformatic analysis and in vitro validation of a five-microRNA signature as a prognostic biomarker of hepatocellular carcinoma

Wang Li; Xiangshuo Kong; Tao Huang; Lujun Shen; Peihong Wu; Qi-Feng Chen

doi:10.21037/atm-20-2509

Bioinformatic analysis and in vitro validation of a five-microRNA signature as a prognostic biomarker of hepatocellular carcinoma

Ann Transl Med. 2020 Nov;8(21):1422. doi: 10.21037/atm-20-2509.

Authors

Wang Li¹, Xiangshuo Kong², Tao Huang^{1

3

4}, Lujun Shen^{1

3

4}, Peihong Wu¹, Qi-Feng Chen^{1

3

4}

Affiliations

¹ Department of Medical Imaging and Interventional Radiology, Sun Yat-sen University Cancer Center, Guangzhou, China.
² Department of Oncology, Yantai Yuhuangding Hospital, Yantai, China.
³ State Key Laboratory of Oncology in South China, Guangzhou, China.
⁴ Collaborative Innovation Center for Cancer Medicine, Guangzhou, China.

Abstract

Background: Existing research has identified correlations between numerous microRNAs (miRNAs) and the prognosis of hepatocellular carcinoma (HCC). However, the role of a combination of miRNAs in predicting HCC survival requires further elucidation.

Methods: miRNA expression profiles and clinical data from HCC patients were downloaded from The Cancer Genome Atlas (TCGA). Differentially expressed (DE) miRNAs in tumor versus normal samples were identified. All HCC patients were randomly assigned to a training cohort or a validation cohort at a ratio of 1 to 1. A least absolute shrinkage and selection operator (LASSO) Cox regression model was subsequently employed to establish the miRNA signature. The constructed miRNA signature was then developed and validated.

Results: In total, 127 DE miRNAs were detected between HCC and paracancerous tissue using HCC RNA sequencing (RNA-Seq) data extracted from TCGA database. LASSO Cox regression generated a five-miRNA signature consisting of has-mir-105-2, has-mir-9-3, has-mir-137, has-mir-548f-1, and has-mir-561 in the training cohort. This risk model was significantly related to survival (P=5.682e-6). Log-rank tests and multivariate Cox regression analyses revealed the five-miRNA signature as an independent prognostic indicator [HR =3.285, 95% confidence interval (CI): 1.737-6.213], with the area under curve (AUC) of the miRNA signature being 0.728. The effects of the miRNA signature were further confirmed in the validation cohort and in the OncomiR Cancer Database and Gene Expression Omnibus (GEO) dataset. Functional enrichment analysis revealed the potential effects of the five-miRNA signature in tumor-related biological pathways and processes. Cell Counting Kit-8, Transwell, and wound healing assays, were used to evaluate the role of has-mir-137 in HCC cell proliferation and migration in vitro.

Conclusions: We established a novel five-miRNA signature which reliably predicted prognosis in HCC patients and which could be used to assist in both strategic counseling and personalized management in HCC.

Keywords: Hepatocellular carcinoma (HCC); least absolute shrinkage and selection operator (LASSO); microRNAs (miRNAs); prognostic signature.