Expanding the Transcriptome of Head and Neck Squamous Cell Carcinoma Through Novel MicroRNA Discovery.

Leigha D Rock,Wan L Lam,Adam P Sage,Brenda C Minatel,Mateus Camargo Barros-Filho,Florian Guisier,Cathie Garnis,Greg L Stewart,Erin A Marshall

doi:10.3389/fonc.2019.01305

Abstract

Head and neck squamous cell carcinoma (HNSCC) has a poor survival rate mainly due to late stage diagnosis and recurrence. Despite genomic efforts to identify driver mutations and changes in protein-coding gene expression, developing effective diagnostic and prognostic biomarkers remains a priority to guide disease management and improve patient outcome. Recent reports of previously-unannotated microRNAs (miRNAs) from multiple somatic tissues have raised the possibility of HNSCC-specific miRNAs. In this study, we applied a customized in-silico analysis pipeline to identify novel miRNAs from raw small-RNA sequencing datasets from public repositories. We discovered 146 previously-unannotated sequences expressed in head and neck samples that share structural properties highly characteristic of miRNAs. The combined expression of the novel miRNAs revealed tissue and context-specific patterns. Furthermore, comparison of tumor with non-malignant tissue samples (n = 43 pairs) revealed 135 of these miRNAs as differentially expressed, most of which were overexpressed or exclusively found in tumor samples. Additionally, a subset of novel miRNAs was significantly associated with HPV infection status and patient outcome. A prognostic-model combining novel and known miRNA was developed (multivariate Cox regression analysis) leading to an improved death and relapse risk stratification (log rank p < 1e-7). The presence of these miRNAs was corroborated both in an independent dataset and by RT-qPCR analysis, supporting their potential involvement in HNSCC. In this study, we report the discovery of 146 novel miRNAs in head and neck tissues and demonstrate their potential biological significance and clinical relevance to head and neck cancer, providing a new resource for the study of HNSCC.

Highlights

Head and neck squamous cell carcinoma (HNSCC) is the eighth most common cancer worldwide [1] and has a poor survival rate, mainly due to late stage diagnosis, and frequent disease recurrence [2]
A discovery cohort consisting of publicly available highthroughput raw small-RNA sequencing data from 523 tumors along with 43 paired non-malignant samples was retrieved from The Cancer Genome Atlas (TCGA) on the cgHUB data repository, available at: https:// cancergenome.nih.gov/
Available small-RNA sequencing data from an independent cohort (n = 20) of oral squamous cell carcinoma samples were obtained from the Gene Expression Omnibus (GEO) repository (Accession GSE52663) [36]

Summary

Introduction

Head and neck squamous cell carcinoma (HNSCC) is the eighth most common cancer worldwide [1] and has a poor survival rate, mainly due to late stage diagnosis, and frequent disease recurrence [2]. MiRNA dysregulation has been shown as a frequent and important event across all stages of cancer [5,6,7,8], as well as in many different cancer types [9,10,11,12,13,14,15]. Their stability in biofluids and tissue biopsies presents opportunities for biomarker discovery [4, 16] and subsequently drug target detection [17,18,19]. Under-expression of let-7d and miR-205 are associated with poor survival in HNSCC [26], and circulating miR-142, miR-186, miR195, miR-374b, and miR-574 have been shown to be promising markers for monitoring therapy in HNSCC patients [27]

Methods

Results

Conclusion