Identification of a gene set correlated with immune status in ovarian cancer by transcriptome-wide data mining

Background: Checkpoint block-mediated tumor immunotherapy has obtained huge success in treating various malignancies. However, the efficacy of immunotherapies against ovarian cancer (OC) was low largely due to limited information on the tumor immune microenvironment (TIME) of OC. This results in lacking reliable biomarkers to select the right patients for immunotherapy and prognosis prediction. Published studies have reported that high tumor mutation burden (TMB) can generate many neoantigens resulting in a higher degree of an anti-tumor immune response. However, the correlation between TMB and TIME of OC remains controversial let along a collection of TMB information is time and resource consuming. In this study, we considered the TMB and TIME of OC comprehensively to categorize patients based on their tumor local immune status and provide information that could guide further OC immunotherapy trials as well as the prognosis prediction. Method: OC RNAseq data were downloaded from The Cancer Genome Atlas (TCGA) and divided into TMB high and TMB low subgroups according to TMB scores for comparison. OC cases were also grouped into immunity high or immunity low based on their profiles of tumor-infiltrating leukocytes (TILs) analyzed by ssGSEA (Single Sample Gene Set Enrichment Analysis). Besides, the function of genes enriched in both TMB high and immunity high was analyzed by Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes pathway (KEGG), and protein-protein interaction (PPI) network. The correlation of each gene with overall survival was also analyzed. Results: : Positive association was observed between the OC grade and the TMB. Both TMB high and immunity high were significantly correlated with a better prognosis. However, different from the findings of other studies, TMB of OC didn’t correlated with preferable TIME in our analysis. By comparing the up-regulated signature genes in TMB high and immunity high cases, we found 14 overlapped genes that were mainly involved in immune response-related pathways. We further analyzed the prognostic value of these 14 genes and found the upregulation of 4 of them, CXCL13, FCRLA, PLA2G2D, and MS4A1 are significantly associated with better survival. With available data collected form a melanoma cohort, we also found that these four genes are positively associated with better response to immune checkpoint blockade-based immunotherapy. Conclusion: By comparing signature genes enriched in TMB high and immunity high subgroups, four genes, CXCL13, FCRLA, PLA2G2D, and MS4A1, were found positively correlated with both OC immune infiltration and better prognosis. Additionally, these four genes predicted the response to checkpoint blockade-based immunotherapy in a melanoma cohort indicating that they can also be used as biomarkers for further immunotherapy clinical trials of OC.