OV is one of the most common oncogenic malignancies among women. Despite multiple progress in OV treatment, 70–80% of patients with OV who initially respond to treatment eventually relapse and die (32). Immunotherapy as a cancer treatment has become a growing field, but reports of success in treating epithelial OV are limited (33). Tumor-infiltrating T cells in the tumor microenvironment are reported to be the biological basis of immunotherapy and promising predictors of cancer prognosis (14). Therefore, investigation of immune-related RNAs associated with tumor-infiltrating T cells in OV patients could be helpful in finding prognostic markers for OV.
In the present study, we calculated the immune and stromal scores on the basis of gene expression data of patients with OV, and the immune-related DEGs associated with tumor-infiltrating T cells were analyzed. The results showed that 37 and 49 immune-related DEGs of CD4+ and CD8+ T cells, respectively, were screened. Subsequently, the PPI network showed that GZMB was a hub node in both PPI networks constructed by immune-related DEGs of CD4+ and CD8+ T cells. Moreover, the ceRNA chr22-38_28785274-29006793.1/ has-miR-1249-5p/CD3E was obtained from the constructed ceRNA networks associated with CD4+ and CD8+ T cells. Furthermore, survival analysis results revealed that key immune-related DEGs of CD4+ and CD8+ T cells, such as GZMB and CD3E, were positively correlated with patient prognosis. These results merit further discussion.
GZMB is a serine protease that is primarily expressed by multiple immune cells, including macrophages, lymphocytes, and mast cells (34). It is reported that GZMB is used by activated cytotoxic T lymphocytes to induce apoptosis of target cells (35). In tumor-infiltrating cytotoxic T lymphocytes, type I interferon is found to inhibit tumor growth by inducing STAT3 activation to promote the expression of GZMB (36). Moreover, Prizment et al. has demonstrated that GZMB expression is correlated with the improved survival of patients with colorectal cancer and might be used as a useful prognostic factor (37). In our study, GZMB was a hub node in the two PPI networks constructed by immune-related DEGs of CD4+ and CD8+ T cells. Moreover, GZMB was found to be positively correlated with patient prognosis. Thus, we speculate that GZMB could be used as a potential prognostic marker of tumor-infiltrating T cells for OV.
CD3E is one of the four invariant chains physically associated with the T-cell receptor (TCR), and the TCR–CD3 complex has been reported to exert a significant function in the immune system (38). The combined immunoscore of CD103 and CD3 may be used to identify patients with early relapse or long-term survival in serous OV (39). We found that CD3E was targeted by the greatest number of miRNAs in the miRNA-target regulatory network, and CD3E showed significant positive correlation with patient OS. Therefore, CD3E may also be a promising prognostic marker of tumor-infiltrating T cells for OV. In addition, CD3E was found to be targeted by several chemical small molecules, such as 1,3-butadiene and benzene. It is reported that 1,3-butadiene and benzene exposure can increase lifetime risk of cancer (40). Thus, we speculate that monitoring the expression of CD3E could be used to predict lifetime risk of cancer caused by exposure to chemical small molecules.
Furthermore, miRNAs able to target CD3 subunit molecules and affect the immune system are considered as promising biomarkers for early cancer detection (41). In our study, we found that CD3E was targeted by multiple miRNAs, such as has-miR-1249-5p. miR-1249 is found to act as either an oncogene or tumor suppressor in different tumor types. For instance, miR-1249 is upregulated in patients with small cell carcinoma of the esophagus, and its enhanced expression is associated with tumor relapse (42). Nevertheless, miR-1249 is shown to be markedly downregulated in colorectal cancer tissues, and it may serve as a novel therapeutic target (43). Our data predicted the target relationship between miR-1249-5p and CD3E. Although little is known about miR-1249 in OV, our result prompts us to speculate that miR-1249-5p may be a potential regulator mediating the key role of tumor-infiltrating T lymphocytes in OV via the targeting of CD3E. In addition, existing evidence has confirmed that lncRNAs function as ceRNAs to competitively bind to the miRNA binding sites, consequently reducing their regulatory effect on target mRNAs (44, 45). Our results showed that the ceRNA chr22-38_28785274-29006793.1/has-miR-1249-5p/CD3E was obtained from the two constructed ceRNA networks related to CD4+ and CD8+ T cells. Therefore, we speculate that ceRNAs, such as chr22-38_28785274-29006793.1/has-miR-1249-5p/CD3E, could be a target of tumor-infiltrating T cells and are involved in the immune escape of tumors in OV. It would be interesting to further investigate the clinical significance of ceRNA chr22-38_28785274-29006793.1/has-miR-1249-5p/CD3E in OV.
In conclusion, our data reveal that tumor-infiltrating T cells may contribute to the immune escape of tumor cells in OV by regulating GZMB and ceRNAs, such as chr22-38_28785274-29006793.1/has-miR-1249-5p/CD3E. GZMB and CD3E may serve as promising T cell immune-related biomarkers with prognostic value in OV.