Abstract
BackgroundThe prognosis of high grade serous ovarian cancer (HGSOC) patients is closely related to the immune microenvironment and immune response. Based on this, the purpose of this study was to construct a model to predict chemosensitivity and prognosis, and provide novel biomarkers for immunotherapy and prognosis evaluation of HGSOC.MethodsGSE40595 (38 samples), GSE18520 (63 samples), GSE26712 (195 samples), TCGA (321 samples) and GTEx (88 samples) were integrated to screen differential expressed genes (DEGs) of HGSOC. The prognosis related DEGs (DEPGs) were screened through overall survival analysis. The DEGs-encoded protein–protein interaction network was constructed and hub genes of DEPGs (DEPHGs) were generated by STRING. Immune characteristics of the samples were judged by ssGSEA, ESTIMATE and CYBERSORT. TIMER was used to analyze the relationship between DEPHGs and tumor-infiltrating immunocytes, as well as the immune checkpoint genes, finally immune-related DEPHGs (IDEPHGs) were determined, and whose expression in 12 pairs of HGSOC tissues and tumor-adjacent tissues were analyzed by histological verification. Furthermore, the chemosensitivity genes in IDEPHGs were screened according to GSE15622 (n = 65). Finally, two prediction models of paclitaxel sensitivity score (PTX score) and carboplatin sensitivity score (CBP score) were constructed by lasso algorithm. The area under curve was calculated to estimate the accuracy of candidate gene models in evaluating chemotherapy sensitivity.Results491 DEGs were screened and 37 DEGs were identified as DEPGs, and 11 DEPHGs were further identified. Among them, CXCL13, IDO1, PI3, SPP1 and TRIM22 were screened as IDEPHGs and verified in the human tissues. Further analysis showed that IDO1, PI3 and TRIM22 could independently affect the chemotherapy sensitivity of HGSOC patients. The PTX score was significantly better than TRIM22, PI3, SPP1, IDO1 and CXCL13 in predicting paclitaxel sensitivity, so was CBP score in predicting carboplatin sensitivity. What’s more, both of the HGSOC patients with high PTX score or high CBP score had longer survival time.ConclusionsFive IDEPHGs identified through comprehensive bioinformatics analysis were closely related with the prognosis, immune microenvironment and chemotherapy sensitivity of HGSOC. Two prediction models based on IDEPHGs might have potential application of chemotherapy sensitivity and prognosis for patients with HGSOC.
Highlights
Ovarian cancer (OC) is a common gynecological malignancy, The 5-year survival rate of OC patients remains low because of the high heterogeneity of the tumor and high incidence of drug resistance during chemotherapy [1,2,3,4,5]
The results showed that low expression of CXCL13, CXCR4, FGF13, IDO1, SPP1 and TRIM22 in high grade serous ovarian cancer (HGSOC) was associated with decreased overall survival rate of patients, and the high expression of KIT, LYVE1, PI3, SLC2A1 and SNCA had a connection with the decreased overall survival rate of patients
We found that the infiltration of dendritic cells (DCs), activated dendritic cells, immature dendritic cells, plasmacytoid dendritic cells, CD8 + T cells and Th1 cells generally increased while the infiltration of B cells decreased when further compared the immune infiltrating cells, immune functions and activities of pathways between normal samples and HGSOC samples in two cohorts (Fig. 3b)
Summary
Ovarian cancer (OC) is a common gynecological malignancy, The 5-year survival rate of OC patients remains low because of the high heterogeneity of the tumor and high incidence of drug resistance during chemotherapy [1,2,3,4,5]. The standard treatment for OC is maximal cytoreductive surgery and postoperative paclitaxel-platinum combined chemotherapy, with a higher survival rate than other regimens, including cyclophosphamide-platinum/ gemcitabine and so on [6]. Most patients develop treatment resistance after prolonged treatment, and almost all patients with recurrent ovarian cancer (ROC) eventually develop platinum resistance [7, 8]. New immunotherapy strategies like immune checkpoint inhibitors therapy, cancer vaccine inoculation and adoptive immunotherapy have had a broad prospect for OC patients with platinum resistance in clinical treatment. The purpose of this study was to construct a model to predict chemosensitivity and prognosis, and provide novel biomarkers for immunotherapy and prognosis evaluation of HGSOC
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
