Identification of the pyroptosis-related prognostic gene signature and characterization of tumor microenvironment infiltration in triple-negative breast cancer

Background: Triple-negative breast cancer remains a highly deleterious disease due to the lack of specific targeted therapy and immunotherapy. A growing body of evidence supports pyroptosis' role in tumorigenesis and prognosis, but it still needs further exploration in our understanding of the tumor microenvironment in triple-negative breast cancer. Methods: : Consensus clustering analysis was performed to construct pattern clusters. A correlation analysis was conducted between pattern clusters and TME using GSVA, ESTIMATE, and CIBERSORT. Then, risk score and nomogram for predicting overall survival were constructed and verified. Results: : We found distinct clusters that differed significantly in terms of prognosis, biological process, clinical features, and tumor microenvironment. Risk scores could predict patient prognosis and response to immunotherapy. Comparatively to other models, the nomogram is extremely effective at predicting prognosis. Conclusions: : In immune microenvironment landscapes, pyroptosis-mediated pattern clusters differed markedly. Risk score and nomogram were effective predictive models. These findings could aid in the development of customized treatment for triple-negative breast cancer patients.