Collagen Family Genes Associated with Risk of Recurrence after Radiation Therapy for Vestibular Schwannoma and Pan-Cancer Analysis

Abstract

Background: The safety of radiotherapy techniques in the treatment of vestibular schwannoma (VS) shows a high rate of tumor control with few side effects. Neuropeptide Y (NPY) may have a potential relevance to the recurrence of VS. Further research is still needed on the key genes that determine the sensitivity of VS to radiation therapy.

Materials and methods: Transcriptional microarray data and clinical information data from VS patients were downloaded from GSE141801, and vascular-related genes associated with recurrence after radiation therapy for VS were obtained by combining information from MSigDB. Logistics regression was applied to construct a column line graph prediction model for recurrence status after radiation therapy. Pan-cancer analysis was also performed to investigate the cooccurrence of these genes in tumorigenesis.

Results: We identified eight VS recurrence-related genes from the GSE141801 dataset. All of these genes were highly expressed in the VS recurrence samples. Four collagen family genes (COL5A1, COL3A1, COL4A1, and COL15A1) were further screened, and a model was constructed to predict the risk of recurrence of VS. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses revealed that these four collagen family genes play important roles in a variety of biological functions and cellular pathways. Pan-cancer analysis further revealed that the expression of these genes was significantly heterogeneous across immune phenotypes and significantly associated with immune infiltration. Finally, Neuropeptide Y (NPY) was found to be significantly and negatively correlated with the expression of COL5A1, COL3A1, and COL4A1.

Conclusions: Four collagen family genes have been identified as possible predictors of recurrence after radiation therapy for VS. Pan-cancer analysis reveals potential associations between the pathogenesis of VS and other tumorigenic factors. The relevance of NPY to VS was also revealed for the first time.

Figure 1

Flowchart of the analysis process. ARGS: angiogenesis-related genes; DEGs: differentially expressed genes; TME: tumor microenvironment.

Figure 2

Results of differential and intersection analysis. (a) Venn diagram showing eight genes after taking intersection of DEGs and ARGs. (b) Volcano diagram showing differential and intersection genes. (c) Heat map showing expression of intersecting genes in tumor tissue and relationship to clinical traits.

Figure 3

(a) Nomogram showing the column line graph prediction model for recurrence after radiotherapy. (b) Calibration graph showing the calibration of the prediction model.

Figure 4

(a) ROC curves showing the classification and predictive efficacy of the predictive model. (b) DCA curves showing the range of clinical predictive safety.

Figure 5

Box plot showing the expression of the four genes in the pan-cancerous tissue and its paracancerous tissues.

Figure 6

(a) Heat map showing the expression of the four genes in pan-cancerous tumor tissue. (b) Correlation heat map showing the correlation results of the expression of the four genes in pan-cancerous tissue. (c) Cox analysis showing the results of the four genes in pan-cancerous survival analysis. Intersection with the midline represents no statistical significance.

Figure 7

Survival curves showing the results of four genes in MESO, KIRP, and LGG (KM method).

Figure 8

The results of the analysis of variance and correlation analysis demonstrate the relationship between the four genes and (a) the tumor immune subtype and (b–d) the tumor microenvironment score. Blanks in the heat map represent no statistically significant differences in correlation analysis.

Figure 9

Correlation of NPY with collagen family genes and tumor recurrence after radiotherapy. Correlation analysis of NPY, COL3A1, COL4A1, COL5A1, and COL15A1 with each other (a). Differential expression of these genes (NPY, COL3A1, COL4A1, COL5A1, and COL15A1) in different ages and genders.