Prognosis and stratification of breast cancer immune infiltration status based on a mitochondria-related gene signature.

Abstract

Abstract Background Tumor metabolic reprogramming has attracted extensive attention, and mitochondria play a vital role in this process as a metabolic hub. Despite increasing evidence that mitochondria are involved in breast tumorigenesis, the impact of mitochondria-related genes on breast cancer remains unclear. In this study, we aimed to construct a novel mitochondria-related gene signature through bioinformatic strategies to predict and stratify the prognosis, immune infiltration, and treatment response of breast cancer patients. Methods The transcriptomic data and clinical features of breast cancer samples were extracted from The Cancer Genome Atlas and Molecular Taxonomy of Breast Cancer International Consortium databases. We constructed a mitochondria-related gene signature to predict survival, and Cox regression and receiver operating characteristic were used to evaluate the overall predictive performance. Subsequently, we combined the risk scores with corresponding clinical features to construct a nomogram model and evaluated the model’s accuracy by clinical calibration curve and decision curve analysis. After dividing patients based on risk scores, several immune-related analyses were performed to compare the immune infiltration status between different groups. Finally, we compared the mutation status and drug sensitivity to explore the treatment response and underlying mechanism of the difference in prognosis. Results We constructed an eight mitochondria-related gene risk signature by bioinformatic strategies and verified it by Cox regression, receiver operating characteristic, calibration curve and decision curve analyses. Patients with low-risk score have a better prognosis, enhanced immune infiltration, significantly different mutation landscapes, and a more sensitive response to antitumor drugs, which may account for the favorable survival. Conclusion The mitochondria-related gene signature is a novel prognostic risk signature that can be used as a predictor for patient stratification in breast cancer. In addition, this signature can effectively distinguish the immune infiltration and drug sensitivity status of patients. This may provide a new perspective for the treatment of breast cancer focusing on mitochondria.