The landscape of tumors-infiltrate immune cells in papillary thyroid carcinoma and its prognostic value

Abstract

Introduction: Thyroid cancer is a very common malignant tumor in the endocrine system, while the incidence of papillary thyroid carcinoma (PTC) throughout the world also shows a trend of increase year by year. In this study, we constructed two models: ICIscore and Riskscore. Combined with these two models, we can make more accurate and reasonable inferences about the prognosis of PTC patients.

Methods: We selected 481 PTC samples from TCGA and 147 PTC samples from GEO (49 samples in GSE33630, 65 samples in GSE35570 and 33 samples in GSE60542). We performed consistent clustering for them and divided them into three subgroups and screened differentially expressed genes from these three subgroups. Then we divided the differential genes into three subtypes. We also distinguished the up-regulated and down-regulated genes and calculated ICIscore for each PTC sample. ICIscore consists of two parts: (1) the PCAu was calculated from up-regulated genes. (2) the PCAd was calculated from down-regulated genes. The PCAu and PCAd of each sample were the first principal component of the relevant gene. What's more, we divided the patients into two groups and constructed mRNA prognostic signatures. Additionally we also verified the independent prognostic value of the signature.

Results: Though ICIscore, we were able to observe the relationship between immune infiltration and prognosis. The result suggests that the activation of the immune system may have both positive and negative consequences. Though Riskscore, we could make more accurate predictions about the prognosis of patients with PTC. Meanwhile, we also generated and validated the ICIscore group and Riskscore group respectively.

Conclusion: All the research results show that by combining the two models constructed, ICIscore and Riskscore, we can make a more accurate and reasonable inference about the prognosis of patients with clinical PTC patients. This suggests that we can provide more effective and reasonable treatment plan for clinical PTC patients.

Keywords: Papillary thyroid carcinoma; Prognostic model; Tumors-infiltrate immune cells.

Figure 1. Flow chart of this study.

Overview of research design

Figure 2. Exploration and validation the correlation between differentiation of ICIclusters and immune cell infiltration.

Through ssGSEA, 28 immune-infiltrating cells were enriched. (A) The heat map is included the age, pStage, gender, tumor purity, estimate score, immune score, stromal score and ICIclusters. (B) The correlation matrice of tumor-infiltrating immune cells. (C) The comparation of enrichment score among three ICIclusters. (D) The expression level of PD-L1 in each combination was compared. (E) The expression level of CTLA4 in each combination was compared.

Figure 3. Prognostic correlation analysis of three ICIclusters.

(A and B) Enrichment plots showing different enrichment of different diseases and pathways in the Rank in Ordered Dataset. (C) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in Total set (D) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in T3 + T4

Figure 4. The GO and KEGG enrichment analysis of Up-regulated Gene and Down-regulated Gene.

(A) The GO enrichment analysis of Up-regulated Gene. (B) The GO enrichment analysis of Down-regulated Gene. (C) The KEGG enrichment analysis of Up-regulated Gene. (D) The KEGG enrichment analysis of Down-regulated Gene.

Figure 5. Establishment and verification of Geneclusters.

(A) The heat map is included the tumor purity, estimate score, immune score, stromal score, age, pStage, gender, ICIcluster and Genecluster. (B) Comparison of infiltration levels of immune cells in three Geneclusters. (C) The expression level of immune checkpoint related gene PD-L1. (D) The expression level of immune checkpoint related gene CTLA4.

Figure 6. Prognostic differences of the three Geneclusters.

(A) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in total set. (B) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in female. (C) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in age <45.(D) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in age >45. (E) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in N1. (F) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in Stage1 + Stage2. (G) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in T1 + T2.

Figure 7. Construction of ICIscore and its clinical prognostic analysis.

(A) The construction of ICIscore with high and low grouping model. (B) The scatter plot showing the relationship between ICIscore and Immunesocre. (C) Alluvial diagram of Genecluster distribution in groups with different Geneclusters, ICIscores, and survival outcomes. (D) The Kaplan-Meier (K–M) survival curves in ICIscore high and low grouping. (E) Differences in immune cell infiltration expressed in high and low ICIscore subgroups. (F) Immune-checkpoint-relevant genes and immune-activation-relevant genes expressed in high and low ICIscore subgroups. (G and H) Enrichment plots showing different enrichment of different pathways in the ICIscore high and low grouping.

Figure 8. Construction of Riskscore.

(A) The construction of a 5-mRNA prognostic signature. (B) Alluvial diagram of Genecluster distribution in groups with different gene clusters, risk, and Disease-free survival (DFS) outcomes. (C) Differences in immune cell infiltration expressed in high and low Riskscore.(D) Immune-checkpoint-relevant genes and immune-activation-relevant genes expressed in high and low Riskscore. (E and F) Enrichment plots showing different enrichment of different diseases in the Riskscore high and Riskscore low.

Figure 9. Validation of the prognostic signature of Riskscore.

(A) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in the training set. (B) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in the testing set. (C) The Kaplan-Meier (K-M) curves of Disease-free survival (DFS) in the total set. (D–F) Time-dependent ROC curves in the training set, testing set and the total set at 1-year, 3-year and 5-year.

Figure 10. Independent prognostic analysis of ICIscore and Riskscore.

(A) Univariate COX analysis of ICIscore about PTC prognostic signatures and clinical characteristics. (B) Multivariate COX analysis of ICIscore about PTC prognostic signatures and clinical characteristics. (C) Univariate COX analysis of Riskscore about PTC prognostic signatures and clinical characteristics. (D) Multivariate COX analysis of Riskscore about PTC prognostic signatures and clinical characteristics. (E) The nomogram of different clinical traits of the patients without scores. (F) The nomogram of different clinical traits of the patients with ICIscore. (G) The nomogram of different clinical traits of the patients with Riskscore.