The pyroptosis-related signature predicts prognosis and influences the tumor immune microenvironment in dedifferentiated liposarcoma

• Published in: Open medicine (Warsaw, Poland) Poland), 2024-01, Vol.19 (1), p.20230886-20230886
• Main Authors: Chen, Wenjing, Cheng, Jun, Cai, Yiqi, Wang, Pengfei, Jin, Jinji
• Format: Article
• Language: English
• Subjects: Cluster analysis; dedifferentiated liposarcoma; Genes; immune infiltrates; immunotherapy; Liposarcoma; Medical prognosis; pyroptosis; tumor microenvironment
• ISSN: 2391-5463; 2391-5463
• DOI: 10.1515/med-2023-0886

Dedifferentiated liposarcoma (DDL), a member of malignant mesenchymal tumors, has a high local recurrence rate and poor prognosis. Pyroptosis, a newly discovered programmed cell death, is tightly connected with the progression and outcome of tumor. The aim of this study was to explore the role of pyroptosis in DDL. We obtained the RNA sequencing data from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression databases to identify different pyroptosis-related genes (PRGs) expression pattern. An unsupervised method for clustering based on PRGs was performed. Based on the result of cluster analysis, we researched clinical outcomes and immune microenvironment between clusters. The differentially expressed genes (DEGs) between the two clusters were used to develop a prognosis model by the LASSO Cox regression method, followed by the performance of functional enrichment analysis and single-sample gene set enrichment analysis. All of the above results were validated in the Gene Expression Omnibus (GEO) dataset. Forty-one differentially expressed PRGs were found between tumor and normal tissues. A consensus clustering analysis based on PRGs was conducted and classified DDL patients into two clusters. Cluster 2 showed a better outcome, higher immune scores, higher immune cells abundances, and higher expression levels in numerous immune checkpoints. DEGs between clusters were identified. A total of 5 gene signatures was built based on the DEGs and divided all DDL patients of the TCGA cohort into low-risk and high-risk groups. The low-risk group indicates greater inflammatory cell infiltration and better outcome. For external validation, the survival difference and immune landscape between the two risk groups of the GEO cohort were also significant. Receiver operating characteristic curves implied that the risk model could exert its function as an outstanding predictor in predicting DDL patients' prognoses. Our findings revealed the clinical implication and key role in tumor immunity of PRGs in DDL. The risk model is a promising predictive tool that could provide a fundamental basis for future studies and individualized immunotherapy.