Yanshuo Han

Identifying the Role of Oxidative Stress‐Related Genes as Prognostic Biomarkers and Predicting the Response of Immunotherapy and Chemotherapy in Ovarian Cancer

Background. Ovarian cancer (OC) is one of the most frequently seen and fatal gynecological malignancies, and oxidative stress (OS) plays a critical role in the development and chemoresistance of OC. Materials and Methods. OS‐related genes (OSRGs) were obtained from the Molecular Signatures Database. Besides, gene expression profiles and clinical information from The Cancer Genome Atlas (TCGA) were selected to identify the prognostic OSRGs. Moreover, univariate Cox regression, LASSO, and multivariate Cox regression analyses were conducted sequentially to establish a prognostic signature, which was later validated in three independent Gene Expression Omnibus (GEO) datasets. Next, gene set enrichment analysis (GSEA) and tumor mutation burden (TMB) analysis were performed. Afterwards, immune checkpoint genes (ICGs) and the tumor immune dysfunction and exclusion (TIDE) algorithm, together with IMvigor210 and GSE78220 cohorts, were applied to comprehensively explore the role of OSRG signature in immunotherapy. Further, the CellMiner and Genomics of Drug Sensitivity in Cancer (GDSC) databases were also applied in investigating the significance of OSRG signature in chemotherapy. Results. Altogether, 34 prognostic OSRGs were identified, among which 14 were chosen to establish the most valuable prognostic signature. The Kaplan‐Meier (KM) analysis suggested that patients with lower OS‐related risk score had better prognosis. The area under the curve (AUC) values were 0.71, 0.76, and 0.85 in 3, 5, and 7 years separately, and the stability of this prognostic signature was confirmed in three GEO datasets. As revealed by GSEA and TMB analysis results, OC patients in low‐risk group might have better immunotherapeutic response, which was consistent with ICG expression and TIDE analyses. Moreover, both IMvigor210 and GSE78220 cohorts demonstrated that patients with lower OS‐related risk score were more likely to benefit from anti‐PD‐1/L1 immunotherapy. In addition, the association between prognostic signature and drug sensitivity was explored. Conclusion. According to our results in this work, OSRG signature can act as a powerful prognostic predictor for OC, which contributes to generating more individualized therapeutic strategies for OC patients.

CircRNA-Based Cervical Cancer Prognosis Model, Immunological Validation and Drug Prediction

Background: Cervical cancer (CC) is a common cancer in female, which is associated with problems like poor prognosis. Circular RNA (circRNA) is a kind of competing endogenous RNA (ceRNA) that has an important role in regulating microRNA (miRNA) in many cancers. The regulatory mechanisms of CC immune microenvironment and the transcriptome level remain to be fully explored. Methods: In this study, we constructed the ceRNA network through the interaction data and expression matrix of circRNA, miRNA and mRNA. Meanwhile, based on the gene expression matrix, CIBERSORT algorithm was used to reveal contents of tumor-infiltrating immune cells (TIICs). Then, we screened prognostic markers based on ceRNA network and immune infiltration and constructed two nomograms. In order to find immunological differences between the high- and low-risk CC samples, we examined multiple immune checkpoints and predicted the effect of PD-L1 ICI immunotherapy. In addition, the sensitive therapeutics for high-risk patients were screened, and the potential agents with anti-CC activity were predicted by Connective Map (CMap). Results: We mapped a ceRNA network including 5 circRNAs, 17 miRNAs and 129 mRNAs. From the mRNA nodes of the network six genes and two kind of cells were identified as prognostic makers for CC. Among them, there was a significant positive correlation between CD8+ T cells and SNX10 gene. The results of TIDE and single sample GSEA (ssGSEA) showed that T cells CD8 do play a key role in inhibiting tumor progression. Further, our study screened 24 drugs that were more sensitive to high-risk CC patients and several potential therapeutic agents for reference. Conclusions: Our study identified several circRNA-miRNA-mRNA regulatory axes and six prognostic genes based on the ceRNA network. In addition, through TIIC, survival analysis and a series of immunological analyses, T cells were proved to be good prognostic markers, besides play an important role in the immune process. Finally, we screened 24 potentially more effective drugs and multiple potential drug compounds for high- and low-risk patients.