A Decision-Aid Model for Predicting Triple-Negative Breast Cancer ICI Response Based on Tumor Mutation Burden

Background: Tumor mutation burden (TMB), a genomic biomarker, has proven to be a strong predictor of immunotherapy response but is not widely adopted. This study investigates the association between TMB and immune checkpoint inhibitors (ICIs) response in TNBC patients. Methods: From the TCGA database, patients were stratified into two levels based on TMB and validated using survival analysis. Then, four machine learning models were trained to classify TNBC patients based on histological features into high and low TMB. To further validate our approach, we compared the genomic landscapes of both groups, identified differentially expressed genes (DEGs), and performed pathway enrichment analysis. Results: Our findings revealed a significant association between TMB and ICI response in TNBC. Random forest model effectively classified TNBC patients based on the representative histological features and clinical data with an accuracy of 0.82 on the validation set. The genomic analysis revealed that FAT3, TTN, and DYNC2H1 had a significantly high mutation rate in the TMB groups. Genes impacting cancer progression and immunogenicity were identified in the DEG analysis as IGF2, CLEC3A, and CASC9. Conclusions: This study constructs a model to identify suitable TNBC patients for immunotherapy and highlights the potential role of TMB associated with genomic alterations in predicting immune response in TNBC.