Hierarchical Clustering of Cutaneous Melanoma Based on Immunogenomic Profiling

Cutaneous melanoma is an aggressive malignancy with high heterogeneity. Several studies have been performed to identify cutaneous melanoma subtypes based on genomic profiling. However, few classifications based on assessments of immune-associated genes have limited clinical implications for cutaneous melanoma. Using 470 cutaneous melanoma samples from The Cancer Genome Atlas (TCGA), we calculated the enrichment levels of 29 immune-associated gene sets in each sample and hierarchically clustered them into Immunity High (Immunity_H, n=323, 68.7%), Immunity Medium (Immunity_M, n=135, 28.7%), and Immunity Low (Immunity_L, n=12, 2.6%) based on the ssGSEA score. The ESTIMATE algorithm was used to calculate stromal scores (range: -1,800.51–1,901.99), immune scores (range: -1,476.28–3,780.33), estimate scores (range: -2,618.28–5,098.14) and tumor purity (range: 0.216–0.976) and they were significantly correlated with immune subtypes (Kruskal–Wallis test, P < 0.001). The Immunity_H group tended to have higher expression levels of HLA and immune checkpoint genes (Kruskal–Wallis test, P < 0.05). The Immunity_H group had the highest level of naïve B cells, resting dendritic cells, M1 macrophages, resting NK cells, plasma cells, CD4 memory activated T cells, CD8 T cells, follicular helper T cells and regulatory T cells, and the Immunity_L group had better overall survival. The GO terms identified in the Immunity_H group were mainly immune related. In conclusion, immune signature-associated cutaneous melanoma subtypes play a role in cutaneous melanoma prognosis stratification. The construction of immune signature-associated cutaneous melanoma subtypes predicted possible patient outcomes and provided possible immunotherapy candidates.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.