A Prognostic Model for Breast Cancer Survival Based on PCD and m6A Gene Interactions

Weimiao Li¹Haocheng Bai¹Jiajing Yang¹Meng Zhan¹Shuqun Zhang¹Guoxu Zheng^2*

• ¹Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China
• ²Department of Immunology, Fourth Military Medical University, Xi’an, China

Background: Breast cancer (BC) is the most prevalent malignancy in women, with patient outcomes heavily influenced by complex molecular mechanisms like programmed cell death (PCD) and RNA methylation. While some studies have investigated how specific PCD types and N6-methyladenosine-related genes (m6A-RGs) are associated with breast cancer, the research on combined PCD mechanisms and their role in breast cancer development is limited. This study integrates PCD-related genes (PCD-RGs) and m6A-RGs to offer new insights for breast cancer clinical treatment. Methods: Transcriptomic data and related genes were respectively retrieved from public databases and published literature. First, PCD-m6A genes identified through the correlation scoring and differentially expressed genes were intersected to obtain candidate genes. Furthermore, to infer potential causal relationships between gene expression and survival, we applied a two-sample Mendelian randomization approach using summary-level data from public databases. Therefore, prognostic genes were further obtained through Mendelian randomization and regression analyses, and a prognostic model was then constructed. Additionally, functional enrichment, immune infiltration, and drug sensitivity analyses were conducted. Finally, the expression intensity of prognostic genes was verified by RT-qPCR and IHC. Results: Through a series of analyses, seven prognostic genes were identified. Following this, the prognostic model has been demonstrated to have a certain degree of accuracy as indicated by both transcriptomic public sets. Successively, enrichment analysis revealed numerous pathways, among which herpes simplex virus 1 infection was notable; its relevance lies in overlapping immune evasion pathways with BC, a core focus of our investigation. Immune cell infiltration analysis revealed that 11 immune cell types, including M1 macrophages, exhibited significant differences between high and low groups. A key finding from drug sensitivity analysis was that the high-risk group exhibited significantly increased sensitivity to several drugs, including CCT018159, rapamycin, vinblastine, metformin, and roscovitine. The expression levels of MYD88, DAXX and ANXA5 were significantly upregulated in the control samples compared to breast cancer samples. Moreover, the expression levels of SESN3, CRIP1, DPP4 and PIK3CA were significantly upregulated in breast cancer samples compared to control samples. Conclusion: This study constructed a risk model based on seven prognostic genes, offering new potential strategies for breast cancer therapy.