The expression of CCL17 and potential prognostic value on tumor immunity in thyroid carcinoma based on bioinformatics analysis

Abstract

Although CCL17 has been reported to exert a vital role in many cancers, the related studies in the thyroid carcinoma have never reported. As a chemokine, CCL17 plays a positive role by promoting the infiltration of immune cells into the tumor microenviroment (TME) to influence tumor invasion and metastasis. Therefore, this study is aimed to investigate the association of CCL17 level with potential prognostic value on tumor immunity in the thyroid carcinoma (THCA) based on the bioinformatics analysis. GEPIA database was applied to analyze CCL17 mRNA expression in THCA data from TCGA database. Through the collection of the data, totally 500 tumor and 57 normal tissue samples were taken for the study. According to survival status and survival time in 500 tumor samples and CCL17 expression from RNA-seq data, all patients were categorized as high- expression (n = 64) and low-expression (n = 436) groups using X-tile program. Next, the association of CCL17 with survival in the thyroid carcinoma patients was examined by using the Kaplan-Meier plotter database. Then, weighted gene co-expression network (WGCNA) was employed to analyze the 1424 DEGs to classify 9 modules. Besides, STRING database was used to obtain the hub genes. GO and KEGG database were employed to explore blue module genes enrichment situations. In addition, TISIDB was used to analyze the relationship of CCL17 expression with tumor-infiltrating lymphocytes proportion, immunostimulators, and major histocompatibility complexes in THCA. The correlation of CCL17 with 22 TIIC subtypes was evaluated by ESTIMATE and CIBERSORT databases. The association of CCL17 level with gene marker of immune cells in THCA was analyzed by GEPIA and TIMER databases. Finally, immunohistochemistry was applied to validate CCL17 expression in 21 tumor and para-carcinoma tissue samples. CCL17 expression in tumors was significantly up-regulated relative to non-carcinoma samples. Patients from CCL17 high-expression group had significantly decreased overall survival compared with low-expression group, which has a significantly importantly potential prognostic value. Moreover, CCL17 and clinical characteristics were analyzed, suggesting that CCL17 expression significantly increased among patients of advanced stage, with advanced T classification, advanced N classification, and higher CCR4 expression. Based on WGCNA, expression of 1424 DEGs in blue module with 258 genes was negatively related to dismal survival and clinical lymph node metastasis in THCA patients. Moreover, CCR4 and CCL17 genes were identified as hub genes within blue module. CCL17 high-expression had greater ImmuneScore, StromalScore and ESTIMATEScore, while lower TumorPurity compared to the CCL17 low-expression. Then, GO and KEGG database were used to analyze blue module genes enrichment situations. The result showed that genes in blue module were associated with cytokine-cytokine receptor interaction, chemokine, and PI3K - Akt pathways. The results of tumor-infiltrating lymphocytes proportion, immunostimulators, and major histocompatibility complexes were significantly positive in CCL17 high-expression. Our findings showed that B cells naïve, T cells CD4 memory resting, T cells CD8, T cells regulatory (Tregs), and dendritic cells resting were the main immune components of THCA tumor microenvironment (TME). CCL17 high-expression in TC was significantly positively related to expression of immune cell gene markers. The result of immunohistochemistry demonstrated that CCL17 expression in tumor tissues significantly increased compared with para-carcinoma tissues. CCL17 high-expression was significantly positively associated with age and advanced N classification, suggesting that CCL17 could accelerate tumor progression by promoting the lymph node metastasis. CCL17 high-expression in THCA tumor microenvironment (TME) accelerates local infiltration of immune cells and enhances anticancer immunity, resulting in worse survival of patients and exerting potential prognostic value on tumor immunity in THCA.

Keywords: Biomarker; Chemokine CCL17; Prognosis; Thyroid papillary carcinoma.

Fig. 1

Diagnostic and prognostic value of CCL17 in PTC. (A) Comparison of CCL17 mRNA expression levels between PTC and normal tissues in GEPIA. (B) K-M survival analysis was performed to determine differences in OS between the CCL17 high-expression and low-expression groups.

Fig. 2

Correlation between the expression of CCL17 and clinical characters in PTC. (A) Pairing diagram. (B) Group comparison chart. (C) The correlation between the CCL17 and stage. (D) The correlation between the CCL17 and T.(E) The correlation between the CCL17 and N.(F) The correlation between the CCL17 and M. (G)The correlation between the CCL17 and CCR4.

Fig. 3

Weighted co-expression network construction and identification of the key module containing CCL17. (A) Calculation of the scale-free fit index of various soft. (B) Analysis of the mean connectivity of various soft-thresholding powers (β). (C)A total of 1424 DEGs were clustered based on the dissimilarity measure (1-TOM) and were divided into 9 modules. (D) Clustering dendrogram of 497 PTC patients. (E) A correlation heatmap between module eigengenes and clinical parameters of PTC. (F–G) Scatter plot of blue module eigengenes. (H-M) Scatter plot of GS for clinical stages, TNM stage, and fustat vs. MM in the blue module.

Fig. 3

Weighted co-expression network construction and identification of the key module containing CCL17. (A) Calculation of the scale-free fit index of various soft. (B) Analysis of the mean connectivity of various soft-thresholding powers (β). (C)A total of 1424 DEGs were clustered based on the dissimilarity measure (1-TOM) and were divided into 9 modules. (D) Clustering dendrogram of 497 PTC patients. (E) A correlation heatmap between module eigengenes and clinical parameters of PTC. (F–G) Scatter plot of blue module eigengenes. (H-M) Scatter plot of GS for clinical stages, TNM stage, and fustat vs. MM in the blue module.

Fig. 3

Weighted co-expression network construction and identification of the key module containing CCL17. (A) Calculation of the scale-free fit index of various soft. (B) Analysis of the mean connectivity of various soft-thresholding powers (β). (C)A total of 1424 DEGs were clustered based on the dissimilarity measure (1-TOM) and were divided into 9 modules. (D) Clustering dendrogram of 497 PTC patients. (E) A correlation heatmap between module eigengenes and clinical parameters of PTC. (F–G) Scatter plot of blue module eigengenes. (H-M) Scatter plot of GS for clinical stages, TNM stage, and fustat vs. MM in the blue module.

Fig. 4

PPI interaction network of hub genes. (A) Interaction network constructed with the nodes with interaction confidence value > 0.95. (B) The top 30 genes ordered by the number of nodes.

Fig. 5

Analysis of blue module genes 258 hub genes. (A) The GO analysis of hub genes. (B–D) KEGG enrichment analysis of blue module genes. (E–F) GSEA analysis of CCL17 high-expression versus CCL17 low-expression.

Fig. 5

Analysis of blue module genes 258 hub genes. (A) The GO analysis of hub genes. (B–D) KEGG enrichment analysis of blue module genes. (E–F) GSEA analysis of CCL17 high-expression versus CCL17 low-expression.

Fig. 6

Further analysis of the correlation between CCL17 expression and immune infiltration. (A–D) ESTIMATE analysis suggested that the CCL17 high -expression group had a higher ImmuneScore and StromalScore than the CCL17 low-expression group, while the TumorPurity score was lower. (E–P) ImmuneScore, StromalScore and ESTIMATEScore were significantly positively associated with stage, T classification, and N classification of TMN stages.

Fig. 7

Correlation between CCL17 expression and immune cell infiltration. (A) Violin plot of CCL17 high- and low-expression groups in THCA patients. (B–C) Proportion of 22 immune cells in THCA in the CCL17 high-expression group. (D) Correlation between CCL17 expression and immune cells by ESTMATE. (E) Correlation of CCL17 with TILs, immunostimulators, and MHCs in TISIDB. A. (E) Correlation analysis between the expression of CCL17 and 28 types of TILs across human cancers by TISIDB. (F) Correlation analysis between the expression of CCL17 and immunostimulators across human cancers by TISIDB. (G) Correlation analysis between the expression of CCL17 and MHCs across human cancers by TISIDB. (H)TIMER analysis of purity-corrected partial Spearman’s correlation between the expression of CCL17 and six immune cells in THCA.

Fig. 7

Correlation between CCL17 expression and immune cell infiltration. (A) Violin plot of CCL17 high- and low-expression groups in THCA patients. (B–C) Proportion of 22 immune cells in THCA in the CCL17 high-expression group. (D) Correlation between CCL17 expression and immune cells by ESTMATE. (E) Correlation of CCL17 with TILs, immunostimulators, and MHCs in TISIDB. A. (E) Correlation analysis between the expression of CCL17 and 28 types of TILs across human cancers by TISIDB. (F) Correlation analysis between the expression of CCL17 and immunostimulators across human cancers by TISIDB. (G) Correlation analysis between the expression of CCL17 and MHCs across human cancers by TISIDB. (H)TIMER analysis of purity-corrected partial Spearman’s correlation between the expression of CCL17 and six immune cells in THCA.

Fig. 7

Correlation between CCL17 expression and immune cell infiltration. (A) Violin plot of CCL17 high- and low-expression groups in THCA patients. (B–C) Proportion of 22 immune cells in THCA in the CCL17 high-expression group. (D) Correlation between CCL17 expression and immune cells by ESTMATE. (E) Correlation of CCL17 with TILs, immunostimulators, and MHCs in TISIDB. A. (E) Correlation analysis between the expression of CCL17 and 28 types of TILs across human cancers by TISIDB. (F) Correlation analysis between the expression of CCL17 and immunostimulators across human cancers by TISIDB. (G) Correlation analysis between the expression of CCL17 and MHCs across human cancers by TISIDB. (H)TIMER analysis of purity-corrected partial Spearman’s correlation between the expression of CCL17 and six immune cells in THCA.

Fig. 8

CCL17 expression in PTC tissues. The levels of CCL17 expression in 21 pairs of PTC and para-carcinoma tissues are determined by immunohistochemistry using anti-CCL17 antibodies. The immunohistochemistry staining show that there are 6 sample staining score was moderate, 15 sample was strong in PTC, 19 sample staining was weak and 2 sample was moderate in para-carcinoma tissues.