Clinical and Prognostic Implications of an Alternative Splicing-related Risk Model Based on TP53 Status in Breast Cancer.

Abstract

Background: Breast Cancer (BRCA) is one of the most common cancers worldwide. Abnormal Alternative Splicing (AS) is frequently observed in cancers. Understanding the intricate relationship between gene mutations and abnormal AS is vital for developing novel diagnostic and therapeutic strategies to effectively target cancer.

Objective: This study aimed to focus on the analysis of transcriptomic splicing events in patients with Breast Cancer (BRCA), particularly those with mutations in the TP53 gene. Understanding the role of AS may be helpful in revealing potential predictive indicators for survival and treatment strategies.

Methods: The splicing data were downloaded from the Cancer Genome Atlas (TCGA) breast cancer project, incorporating 972 patients in the study, classified according to TP53 mutation status. A comprehensive splicing profile of these breast tumors was outlined, and an interaction network of Alternative Splicing (AS) events and splicing factors was constructed. This allowed for the identification of specific AS events associated with TP53-mutant breast cancer. A prognostic risk model based on AS events was established, using univariate and multivariate Cox regression analyses. To understand the molecular heterogeneity, consensus clustering analyses of prognostic AS events were performed. We also investigated the association of AS patterns with the immune microenvironment and drug sensitivity.

Results: A total of 4519 significant Alternative Splicing (AS) events were distributed among 2729 genes that were altered in TP53 mutant tumors. Based on the analysis of these events, a prognostic risk model was created involving ten AS events from ten genes (such as NKTR, CD46, VCAN, etc.). The survival analysis showed that patients with high-risk scores had significantly poorer overall survival (p<0.001, HR=2.46, 95% CI 1.90-3.18) than those with low-risk scores. Furthermore, the study identified four molecular subtypes related to AS events (C1, C2, C3, and C4), which showed significant differences in immune cell infiltration, with C1 and C4 clusters having a higher degree of immune cell infiltration than C2 and C3. The chemosensitivity analysis revealed that these different AS clusters have different sensitivities to several anticancer drugs, such as docetaxel, paclitaxel, and doxorubicin, with C1 and C4 clusters being more sensitive than the other clusters.

Conclusion: We have demonstrated differential transcriptomic splicing events between TP53 mutant and wild-type cases of breast cancer, establishing an effective prognostic risk model based on AS events. These findings provide new insights that may aid in understanding the biological behavior of breast cancer and potentially in optimizing treatment strategies for breast cancer.