Y. Wei1,2, J. Han3, Y. Liu4, L. Wang1, and C. Han1; 1Department of Radiation Oncology, the Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei, China, 2Hebei General Hospital, Shijiazhuang, Heibei, China, 3Department of Medical Oncology, the Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei, China, 4Department of Pathology, the Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei, China
Purpose/Objective(s):To map the coexpressed genes and transcriptional regulatory network of programmed cell death ligand 1 (PD-L1) in oesophageal squamous cell carcinoma (ESCC), screen for potential biomarkers coexpressed with PD-L1, and search for target genes that may be involved in PD-L1-mediated regulation of tumour immune status. Histological validation of the target genes was performed to analyse the relationship between target genes and prognosis and immune cell infiltration in patients with ESCC. Materials/
Methods: The Cancer Genome Atlas (TCGA) ESCC dataset was used to identify genes coexpressed with PD-L1 via the cBioPortal data analysis platform. Venny analysis at the whole-transcriptome level was performed on the genes coexpressed with PD-L1 to screen for target genes. Multiple clustering and molecular relationship network analyses were conducted, and a PD-L1 regulatory network was established. Cancer and adjacent tissue samples were collected from ESCC patients undergoing surgical treatment, a tissue microarray was constructed, and immunohistochemical methods were used to detect the expression of target genes and PD-L1 in the tissue microarray. The relationship between target genes and patient survival was evaluated. An immunofluorescence dual-labelling localization method was used to detect the localization and distribution of target genes and PD-L1. The differential expression of target genes and their relationship with the immune microenvironment were evaluated using TIMER and TISIDB. Results: A total of 95 genes coexpressed with PD-L1 were identified. The transcription factors STAT2, IRF-8, IRF-2, IRF-4, and IRF-7 were involved in the transcriptional regulation of more than 30% of the genes in this gene set. After screening for hub molecules and analysing the network node connectivity of the genes coexpressed with PD-L1, we identified the 10 node genes with the highest connectivity—CXCL10, B2M, CD80, GBP1, CXCL9, DDX58, CD86, CXCL11, GBP2, and CMPK2. After histological validation, it was found that B2M is highly expressed in ESCC tissue and is also expressed in interstitial cells and that high expression of B2M is associated with a poor prognosis. There was a correlation between B2M and PD-L1 expression, and the two genes were colocalized. Through online database analysis, it was found that B2M was highly expressed in ESCC tissue and closely related to immune cell infiltration. Conclusion: The hub nodes coexpressed with PD-L1 in ESCC are all immune-related molecules, among which CXCL10, B2M, CD80, GBP1, CXCL9, DDX58, CD86, CXCL11, GBP2, and CMPK2 have an extensive regulatory effect on the network of PD-L1-coexpressed genes. B2M is a potential prognostic indicator for ESCC patients and an important immune-related factor. The expression of B2M may have important implications for guiding immunotherapy in patients with ESCC.
