Identification of anoikis-related molecular patterns to define tumor microenvironment and predict immunotherapy response and prognosis in soft-tissue sarcoma.

Abstract

Background: Soft-tissue sarcoma (STS) is a massive threat to human health due to its high morbidity and malignancy. STS also represents more than 100 histologic and molecular subtypes, with different prognosis. There is growing evidence that anoikis play a key role in the proliferation and invasion of tumors. However, the effects of anoikis in the immune landscape and the prognosis of STS remain unclear. Methods: We analyzed the genomic and transcriptomic profiling of 34 anoikis-related genes (ARGs) in patient cohort of pan-cancer and STS from The Cancer Genome Atlas (TCGA) database. Single-cell transcriptome was used to disclose the expression patterns of ARGs in specific cell types. Gene expression was further validated by real-time PCR and our own sequencing data. We established the Anoikis cluster and Anoikis subtypes by using unsupervised consensus clustering analysis. An anoikis scoring system was further built based on the differentially expressed genes (DEGs) between Anoikis clusters. The clinical and biological characteristics of different groups were evaluated. Results: The expressions of most ARGs were significantly different between STS and normal tissues. We found some common ARGs profiles across the pan-cancers. Network of 34 ARGs demonstrated the regulatory pattern and the association with immune cell infiltration. Patients from different Anoikis clusters or Anoikis subtypes displayed distinct clinical and biological characteristics. The scoring system was efficient in prediction of prognosis and immune cell infiltration. In addition, the scoring system could be used to predict immunotherapy response. Conclusion: Overall, our study thoroughly depicted the anoikis-related molecular and biological profiling and interactions of ARGs in STS. The Anoikis score model could guide the individualized management.