Characterization of Pyroptosis-Related Subtypes via RNA-Seq and ScRNA-Seq to Predict Chemo-Immunotherapy Response in Triple-Negative Breast Cancer

• Published in: Frontiers in genetics Vol. 13; p. 788670
• Main Authors: Li, Chenlu, Pan, Jingjing, Jiang, Yinyan, Wu, Yanzhi, Jin, Zhenlin, Chen, Xupeng
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 21.03.2022
• Subjects: Genetics; molecular subtype; prognosis; pyroptosis; single-cell RNA sequencing; triple-negative breast cancer; tumor microenvironment; single-cell RNA sequencing; pyroptosis; triple-negative breast cancer; tumor microenvironment; prognosis; molecular subtype
• ISSN: 1664-8021; 1664-8021
• DOI: 10.3389/fgene.2022.788670

Triple-negative breast cancer (TNBC) is associated with poor prognosis and invalid therapeutical response to immunotherapy due to biological heterogeneity. There is an urgent need to screen for reliable indices, especially immunotherapy-associated biomarkers that can predict patient outcomes. Pyroptosis, as an inflammation-induced type of programmed cell death, is shown to create a tumor-suppressive environment and improve the chemotherapeutic response in multiple tumors. However, the specific therapeutic effect of pyroptosis in TNBC remains unclear. In this study, we present a consensus clustering by pyroptosis-related signatures of 119 patients with TNBC into two subtypes (clusterA and clusterB) with distinct immunological and prognostic characteristics. First, clusterB, associated with better outcomes, was characterized by a significantly higher pyroptosis-related signature expression, tumor microenvironment prognostic score, and upregulation of immunotherapy checkpoints. A total of 262 differentially expressed genes between the subtypes were further identified and the Ps-score was built using LASSO and COX regression analyses. The external GEO data set demonstrated that cohorts with low Ps-scores consistently had higher expression of pyroptosis-related signatures, immunocyte infiltration levels, and better prognosis. In addition, external immunotherapy and chemotherapy cohorts validated that patients with lower Ps-scores exhibited significant therapeutic response and clinical benefit. Combined with other clinical characteristics, we successfully constructed a nomogram to effectively predict the survival rate of patients with TNBC. Finally, using the scRNA-seq data sets, we validated the landscape of cellular subtypes of TNBC and successfully constructed an miRNA-Ps-score gene interaction network. These findings indicated that the systematic assessment of tumor pyroptosis and identification of Ps-scores has potential clinical implications and facilitates tailoring optimal immunotherapeutic strategies for TNBC.