Analysis of the tumor microenvironment and mutation burden identifies prognostic features in thymic epithelial tumors

Abstract

Thymic epithelial tumors (TETs) are one of the rarest adult malignancies in the anterior mediastinum. Thymic carcinomas (TCs) are less prevalent among TETs, but they are more clinically aggressive. Immunotherapy has emerged as a promising therapeutic approach for refractory TETs, even though chemotherapy remains the conventional treatment for the advanced disease. However, limited attention has been paid to the features of the tumor microenvironment (TME) which might provide clinically relevant information and guide treatment regimen design. Especially, to date, there have been only a few studies focusing on the differences between the TME and genomic features preserved by TETs and TCs. We analyzed the TME and genomic characteristics of TETs using RNA sequencing and whole-exome sequencing, finding that distinct characteristics of TME in different pathogenic subtypes of TETs. According to those findings, we found that thymic carcinomas had significantly lower expression of HMGB1, a pro-inflammatory cytokine-related gene, than thymomas, and low HMGB1 expression was linked to a poor prognosis. Additionally, higher mutation burdens were significantly associated with the later stage and more advanced pathological types. Thymoma patients with lower mutation burdens tended to relapse within 3 years. In summary, different characteristics of TME and genomic features between thymoma and thymic carcinoma were associated with clinical outcomes of TETs and presented promisingly predictive value for efficacy and toxicity of immunotherapy.

Keywords: Thymic epithelial tumors; genomic features HMGB1; the tumor microenvironment; thymic carcinomas; thymomas.

Figure 1

Immune and stromal scores of thymomas and thymic carcinomas. A. Immune scores of thymomas and thymic carcinomas (P=0.1498, Mann-Whitney U test); B. Immune scores of different subtypes of TETs (One-way ANOVA followed by Tukeys post hoc test). C. Stromal scores of thymomas and thymic carcinomas (P=0.0027, Mann-Whitney U test). D. Stromal scores of different subtypes of TETs (One-way ANOVA followed by Tukeys post hoc test). *P<0.05, **P<0.01, ***P<0.001; T: thymoma; TC: thymic carcinoma.

Figure 2

Immune infiltration in thymomas and thymic carcinomas. A. The classification of tumor immune infiltration by cell clusters in all samples; B. The different proportion of immune cells evaluated by ssGSEA between thymomas and thymic carcinomas (P<0.01, paired Wilcoxon test); C-H. Expression levels of cytokines related gene in immune response among different subtypes of TETs (One-way ANOVA followed by Tukeys post hoc test); I. HMGB1 expression levels among different subtypes of TETs (One-way ANOVA followed by Tukeys post hoc test); J. Lower expression levels of HMGB1 was associated with significantly worse OS (P=0.0245, HR=0.21, 95% CI, ranged from 0.054 to 0.81) in this TCGA-TETs cohort; K. T cell receptor (TCR) repertoire analysis showed Shannon index among different subtypes of TETs (One-way ANOVA followed by Tukeys post hoc test). *P<0.05, **P<0.01, ***P<0.001; T: thymoma; TC: thymic carcinoma.

Figure 3

Genomic features and mutation burdens of TET patients. A. Somatic mutation profile in TETs; B. Significantly higher mutation burden of thymic carcinomas than that of thymomas; C. Mutation burdens of different subtypes of thymic epithelial tumors. One-way ANOVA followed by Tukeys post hoc test was applied (*P<0.05, **P<0.01); D. In thymomas cohort of TCGA-TETs, patients who suffered recurrent disease within 3 years presented lower mutation burden than those without relapse in 3 years.