Triple-negative breast cancer (TNBC) accounts for 15-30% of all breast cancer cases and is clinically difficult to treat due to the lack of hormone or human epidermal growth factor receptor 2 receptors, which are usually targeted by the most successful therapeutic approaches. Immune checkpoint inhibitors (ICIs) have offered long-term survival benefits in several types of solid tumors, however with low response rates. Thus, there is an urgent need to develop feasible biomarkers for identifying patients with TNBC, who are responsive. The present study demonstrated that the immune microenvironment of TNBC has the highest expression of immunoregulatory molecules among all pathologic types. The tumor mutation burden (TMB) of TNBC was not strongly correlated with cytolytic activity and showed no significant associations with different degrees of immune cell infiltration and TMB. The machine learning method divided patients with TNBC into two groups characterized by 'hot' and 'cold' tumors, according to whether immune-associated genes were highly expressed, and different responses to immunotherapy were seen between these two groups. Furthermore, patients with a TP53MutPIK3CAWild genotype demonstrated favorable immunotherapy-responsive signatures and may have improved outcomes with ICIs. In conclusion, the present study revealed that TP53 and PIK3CA may be appropriate biomarkers to screen for patients who would benefit most from ICIs, which could guide precise immunotherapy for patients with TNBC.
Keywords: TP53MutPIK3CAWild; biomarker; immune checkpoint inhibitors; triple-negative breast cancer.
Copyright: © Cheng et al.