SHUJİNG WANG, JİNG YANG, QİN TANG, QİANG WU
Eurasian Journal of Medicine and Oncology - 2023;7(4):350-361
Objectives: Breast cancer ranks first in morbidity and mortality among women worldwide. Herein, we constructed immune-related long non-coding RNAs (irlncRNAs) signature that could predict breast cancer prognosis. Methods: The data of breast samples were downloaded from TCGA. Differentially expressed irlncRNAs (DEirlncRNAs) compared tumor with normal samples were obtained. Established a prognostic model after DEirlncRNAs were cyclically and separately paired. The model’s accuracy was validated using ROC curve, survival analysis, clinicopathological fea tures, tumor-infiltrating immune cells, immune checkpoints, and chemotherapeutic treatment. Quantitative real-time PCR was used to analyzed the risk score of breast cancer samples. Results: Fifteen DEirlncRNAs pairs were selected to construct the prognostic model and distinguish high- or low-risk groups. Patients in high-risk group had poorer prognosis, more aggressive clinicopathologic features, lower expres sion of immune checkpoints, and higher drug sensitivity than those in low-risk group. For tumor-infiltrating immune cells, the high-risk group was positively related to cancer-promoting immune cells and negatively associated with anti cancer immune cells. In clinical samples, risk score was positively correlated with patient age and KI67 index. Conclusion: We constructed a prognostic model, which based on fifteen pairs of irlncRNAs, predicting both prognosis of breast cancer and efficacy of immunotherapy and chemotherapy as well.
