J Cancer 2021; 12(3):703-716. doi:10.7150/jca.50277

Research Paper

Decoding Immune Heterogeneity of Melanoma and identifying immune-prognostic hub genes

Yu Zhang, Siyu Hao, Yingli Gao, Weina Sun, Yuzhen Li

Department of Dermatology, the Second Affiliated Hospital of Harbin Medical University, No.248 Xuefu Road, Nangang District, Harbin. 150081, P. R. China.

This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/). See http://ivyspring.com/terms for full terms and conditions.
Citation:
Zhang Y, Hao S, Gao Y, Sun W, Li Y. Decoding Immune Heterogeneity of Melanoma and identifying immune-prognostic hub genes. J Cancer 2021; 12(3):703-716. doi:10.7150/jca.50277. Available from https://www.jcancer.org/v12p0703.htm

File import instruction

Abstract

Melanoma is an aggressive skin cancer that has gained attention worldwide. Growing evidence has highlighted that the tumor microenvironment (TME) is an important feature of carcinogenesis and contributes to therapeutic efficacy in melanoma. However, additional advances in melanoma immuno-oncology are necessary to achieve a comprehensive knowledge of the immune infiltrate population and to identify accurate and readily measurable biomarkers. In this study, we analyzed gene expression of 468 melanoma cases from the TCGA database, which led to the identification of three melanoma clusters (representedby low, median and high infiltration) that display unique immune features. We found that the microenvironment clusters had substantial prognostic efficacy. The median cluster was characterized by an inability to draw immune cells, highlighting possible immune escape mechanisms, and lower CXCL9 and CXCL10 expression, which was correlated to poor prognosis. Deep molecular characterization of immune cells, cytolytic-activity and tumor-inflammatory status revealed diversity of the local immune infiltration landscape in the melanoma clusters. Differentially expressed genes related to TME were extracted from each infiltration cluster. Functional annotations revealed that these genes were mainly related to immune system activation and the processes of immunoreaction. The top ten hub genes in immune infiltration-related protein-protein interaction (PPI) networks were selected for further prognostic investigation. Further validation showed that five of ten hub genes were good prognostic biomarkers for melanoma in two independent groups from the Gene Expression Omnibus database. In brief, these data highlight that systemic characterization of melanoma could uncover tumor infiltrate characteristics, which can help select the most adequate treatment and identify consistent and important indicators of the local immune tumor microenvironment in melanoma patients.

Keywords: melanoma, tumor microenvironment, immune heterogeneity, prognostic biomarkers