A Novel Prognostic Prediction Model Based on Pyroptosis-Related Clusters for Breast Cancer

Abstract

Breast cancer (BC) is the most common cancer affecting women and the leading cause of cancer-related deaths worldwide. Compelling evidence indicates that pyroptosis is inextricably involved in the development of cancer and may activate tumor-specific immunity and/or enhance the effectiveness of existing therapies. We constructed a novel prognostic prediction model for BC, based on pyroptosis-related clusters, according to RNA-seq and clinical data downloaded from TCGA. The proportions of tumor-infiltrating immune cells differed significantly in the two pyroptosis clusters, which were determined according to 38 pyroptosis-related genes, and the immune-related pathways were activated according to GO and KEGG enrichment analysis. A 56-gene signature, constructed using univariate and multivariate Cox regression, was significantly associated with progression-free interval (PFI), disease-specific survival (DSS), and overall survival (OS) of patients with BC. Cox analysis revealed that the signature was significantly associated with the PFI and DSS of patients with BC. The signature could efficiently distinguish high- and low-risk patients and exhibited high sensitivity and specificity when predicting the prognosis of patients using KM and ROC analysis. Combined with clinical risk, patients in both the gene and clinical low-risk subgroup who received adjuvant chemotherapy had a significantly lower incidence of the clinical event than those who did not. This study presents a novel 56-gene prognostic signature significantly associated with PFI, DSS, and OS in patients with BC, which, combined with the TNM stage, might be a potential therapeutic strategy for individualized clinical decision-making.