Pan-cancer analysis and KNSTRN as a potential prognosis biomarker.

Abstract

e15180 Background: KNSTRN (Kinetochore-Localized Astrin-Binding) protein participates in the biological processes of chromosome alignment and segregation and is required for accurate mitosis. However, only a few researches have focused on the functional roles of KNSTRN in various cancers. Pan-cancer analysis facilitates better understanding the KNSTRN functions in tumor development and progression. Methods: In this study, RNA-Seq data of the tumor and the adjacent tissue, along with the patient information, were collected from 30 solid tumor types in TCGA database. For each tumor type, the tumor samples were stratified into High_KNSTRN (HK) or Low_KNSTRN (LK) based on the threshold of median KNSTRN expression. As an important prognostic biomarker in pan cancers, tumor mutation burdens (TMB) were evaluated between HK and LK groups. Highly and unqiuely expressed genes in each group were included for functional enrichment analysis. To further investigate the prognostic value of KNSTRN, Kaplan-Meier survival analyses were carried out based on the overall survival (OS) data. Moreover, the mutation types such as SNV, indel, CNV, and rearrangement ooccured in those tumor types were obtained from cBioPortal of Cancer Genomics platform. We also used TIMER2 database to evaluate the associations between KNSTRN expression and CAFs and CD8+ T cells infiltrating levels. Results: We found KNSTRN was significantly up-regulated in 20 cancer types and significantly down-regulated in THCA. Consistent with the previous studies, higher KNSTRN expression levels were significantly associated with shorter OS in most cancers. The most frequently detected mutations were found in MESO, SKCM and UCS. The “deep deletion” CNV was the most frequently occurred type in MESO and UCS, and the SNV mutations highly occurred in SKCM, which meant the KNSTRN mutations were quite diverse across various cancer types. In addition, KNSTRN expression and TMB were positively correlated, and the higher TMB indicated worse prognoses in 13 cancers. Highly expressed genes in the LK group mainly participated in immune response and a large family of environmental information processing pathways. Compared to highly expressed genes related pathways in HK group, KNSTRN may likely be involved in pro-tumor regulator pathways. Moreover, KNSTRN was positively correlated with immunosuppressive genes and CAFs cell infiltration levels but negatively correlated to CD8+ T cells in most cancers. Conclusions: Most of the cancers had up-regulated KNSTRN expressions compared with adjacent tissues. Higher KNSTRN expression was significantly associated with worse clinical outcomes in those cancers. Comprehensive downstream analysis revealed the underlying mechanism of KNSTRN in tumor development and its potential clinical value as a prognostic biomarker.