Identification of potential key genes of TGF-beta signaling associated with the immune response and prognosis of ovarian cancer based on bioinformatics analysis

Xiaoxue Zhang; Liping Han; Huimin Zhang; Yameng Niu; Ruopeng Liang

doi:10.1016/j.heliyon.2023.e19208

Identification of potential key genes of TGF-beta signaling associated with the immune response and prognosis of ovarian cancer based on bioinformatics analysis

Heliyon. 2023 Aug 19;9(8):e19208. doi: 10.1016/j.heliyon.2023.e19208. eCollection 2023 Aug.

Authors

Xiaoxue Zhang¹, Liping Han², Huimin Zhang², Yameng Niu¹, Ruopeng Liang³

Affiliations

¹ Department of Physical Examination, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, PR China.
² Department of Obstetrics and Gynecology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, PR China.
³ Department of Hepatobiliary and Pancreatic Surgery, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, PR China.

Abstract

Background: TGF-beta signaling is a key regulator of immunity and multiple cellular behaviors in cancer. However, the prognostic and therapeutic role of TGF-beta signaling-related genes in ovarian cancer (OV) remains unexplored.

Methods: Data of OV used in the current study were sourced from TCGA and GEO databases. Consensus clustering was applied to classify OV patients into different clusters using TGF-beta signaling-related genes. Differentially expressed genes (DEGs) between different clusters were screened by the "limma" R package. Prognostic genes were screened from DEGs by univariate Cox regression, followed by the construction of the TGF-beta signaling-related score. The prognostic value of TGF-beta signaling-related score was evaluated in both training and testing OV cohorts. Moreover, the immune status, GSEA and therapeutic response between low- and high-score groups were performed to further reveal the potential mechanisms.

Results: By consensus clustering, OV patients were classified into two clusters with different tumor immune environments. After differential expression and univariate Cox regression analyses, GMPR, PIEZO1, EMP1, CXCL13, GADD45B, SORCS2, FOSL2, PODN, LYNX1 and SLC38A5 were selected as prognostic genes. Using PCA algorithm, the TGF-beta signaling-related score of OV patients was calculated based on prognostic genes. Then OV patients were divided into low- and high-TGF-beta signaling-related score groups. We observed that the two score groups had significantly different survivals, tumor immune environments and expressions of immune checkpoints. In addition, GSEA results showed that immune-related pathways and biological processes, like chemokine signaling pathway, TNF signaling pathway and T cell migration were significantly enriched in the low-score group. Moreover, patients in the low- and high-score groups had remarkably different sensitivity to chemo- and immunotherapy.

Conclusion: For the first time, our study identified ten prognostic genes associated with TGF-beta signaling, constructed a prognostic TGF-beta signaling-related score and investigated the effect of TGF-beta signaling-related score on OV immunity and therapy. These findings may enrich our knowledge of the TGF-beta signaling in OV prognosis and help to improve the prognosis prediction and treatment strategies in OV.

Keywords: Ovarian cancer; Prognosis; TGF-beta signaling.