Comprehensive Analysis of TRP Channel–Related Genes in Patients With Triple-Negative Breast Cancer for Guiding Prognostic Prediction
Haojie Zhang,Xiangsheng Zhang,Xiaohong Wang,Hongguang Sun,Changran Hou,Yue Yu,Song Wang,Fangxu Yin,Zhenlin Yang
DOI: https://doi.org/10.3389/fonc.2022.941283
IF: 4.7
2022-07-07
Frontiers in Oncology
Abstract:Background: Triple-negative breast cancer (TNBC) is a special subtype of breast cancer. Transient Receptor Potential (TRP) channel superfamily has emerged as a novel and interesting target in a variety of tumors. However, the association of TRP channel–related genes with TNBC is still unclear. Methods: The The Cancer Genome Atlas (TCGA)-TNBC and GSE58812 datasets were downloaded from the public database. The differentially expressed TRP channel–related genes (DETGs) were screened by limma package, and mutations of the above genes were analyzed. Subsequently, new molecular subtypes in TNBC-based DETGs were explored by consensus clustering analysis. In addition, Lasso–Cox regression analysis was used to divide it into two robust risk subtypes: high-risk group and low-risk group. The accuracy and distinguishing ability of above models were verified by a variety of methods, including Kaplan–Meier survival analysis, ROC analysis, calibration curve, and PCA analysis. Meanwhile, CIBERSORT algorithm was used to excavate status of immune-infiltrating cells in TNBC tissues. Last, we explored the therapeutic effect of drugs and underlying mechanisms of risk subgroups by pRRophetic package and GSEA algorithm, respectively. Results: A total of 19 DETGs were identified in 115 TNBC and 113 normal samples from TCGA database. In addition, missense mutation and SNP were the most common variant classification. According to Lasso–Cox regression analysis, the risky formula performed best when nine genes were used: TRPM5, TRPV2, HTR2B, HRH1, P2RY2, MAP2K6, NTRK1, ADCY6, and PRKACB. Subsequently, Kaplan–Meier survival analysis, ROC analysis, calibration curve, and Principal Components Analysis (PCA) analysis showed an excellent accuracy for predicting OS using risky formula in each cohort (P < 0.05). Specifically, high-risk group had a shorter OS compared with low-risk group. In addition, T-cell CD4 memory activated and macrophages M1 were enriched in normal tissues, whereas Tregs were increased in tumor tissues. Note that the low-risk group was better therapeutic effect to docetaxel, doxorubicin, cisplatin, paclitaxel, and gemcitabine than the high-risk group (P < 0.05). Last, in vitro assays, Quantitative Real-time PCR (qRT-PCR) indicated that TRPM5 was significantly highly expressed in MDA-MB-231 and MDA-MB-468 cells compared with that in MCF-10A cells (P < 0.01). Conclusion: We identified a risky formula based on expression of TRP channel–related genes that can predict prognosis, therapeutic effect, and status of tumor microenvironment for patients with TNBC.
oncology