Background: Breast cancer is one cancer that develops from breast tissue and one of the major reasons of deaths in of women all over the world. The tumor-infiltrating lymphocytes in tumor immune microenvironment are correlated with the prognosis in breast cancer patients, and play an important role in the occurrence and development of breast cancer.
Methods: In this study, by integrating the immune gene expression of 20 breast cancer cohorts from the public dataset, an immune-based prognostic score was established. This immune-based prognostic score was found to be correlated with prognosis, stromal score, tumor purity, three famous immune checkpoints, and immune escape mechanism in breast cancer patients.
Results: The clinical application of the prognostic score was verified by the breast cancer patients treated with chemotherapy, and good therapeutic benefit of the prognostic score was obtained. In addition, the XGBoost classifier was used to construct for predicting the high and low prognostic score subtypes, and the predictive results indicated that the XGBoost was suitable to predict these two subtypes in breast cancer patients.
Conclusion: Based on these results, we believed that the prognostic score may be used as an effective prognostic marker and may provide great help for chemotherapy treatment of breast cancer patients.
[http://dx.doi.org/10.1016/S1470-2045(17)30904-X] [PMID: 29233559]