Two types of immune infiltrating cells and six hub genes can predict the occurrence of myasthenia gravis in patients with thymoma

ABSTRACT

Thymoma is the most common primary mass in anterior mediastinum. Although associated with low malignancy, it is often accompanied by myasthenia gravis resulting in poor prognosis. Due to the dual factors of tumor immune tolerance and autoimmune reaction, it is urgent to understand the immune status of MG with thymoma. In this study, RNA sequencing data were obtained from the TCGA and GEO cohorts to identify differentially expressed messenger RNAs and infiltrated immune cells. A total of 121 samples in TCGA and 43 samples in GEO were screened out. The infiltrated immune cells were identified by CIBERSORT, in which Tfh cells and activated DC cells were abnormal in thymoma patients. The differently expressed genes were performed by package LIMMA. The functional characteristics of differently expression genes were analyzed by GO and KEGG; one GO and seven KEGG pathways were both found in both TCGA and GEO cohorts. Meanwhile, 27 common differently expressed genes were obtained and were displayed by a Venn diagram. The TRRUST was used to screen the hub genes for the common 27 different genes and 6 genes were found. Then, PPI networks were constructed. Subsequently, the relationship between SCNAs of common genes and related immune cells tested by TIMER. Kaplan–Meier plots, ROC curve and Cox’s expression model for immune infiltration and hub genes were also tested. In conclusion, we found that two types of immune infiltrated cells and six hub genes can predict the occurrence of myasthenia gravis in thymoma patients.

KEYWORDS:

• Thymoma
• myasthenia gravis
• infiltrated immune cells
• Tfh cell
• activated DC cells

Highlights

1. The infiltration of Tfh cells and activated DC cells is abnormal in thymoma patients.

2. Six hub genes were obtained by analysis of thymoma patient’s data from TCGA and GEO cohorts.

3. The infiltration of Tfh cells, activated DC cells and six hub genes are predictors of the occurrence of MG in thymoma patients.

Acknowledgements

We gratefully thank the contributions from the websites of TCGA, GEO, Genemania, TRRUST, and TIMER.

Date Availability

All data can be obtained from TCGA and GEO cohorts.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This work was supported by the National Science foundation of China under Grant [number 81571590]; National Science foundation of Tianjin under Grant [number 19JCZDJC35500]; and Beijing, Tianjin, Hebei basic research cooperation project under Grant [number 19JCZDJC64400(Z)].