A novel signature incorporating genes related to lipid metabolism and immune for prognostic and functional prediction of breast cancer.

Purpose: Breast cancer prognosis and functioning have not been thoroughly examined in relation to immunological and lipid metabolism. However, there is a lack of prognostic and functional analyses of the relationship between lipid metabolism and immunity in breast cancer.

Methods: DEGs in breast cancer were obtained from UCSC database, and lipid metabolism and immune-related genes were obtained from GSEA and Immune databases. A predictive signature was constructed using univariate Cox and LASSO regression on lipid metabolism and immune-related DEGs. The signature's prognostic significance was assessed using Kaplan-Meier, time-dependent ROC, and risk factor survival scores. Survival prognosis, therapeutic relevance, and functional enrichment were used to mine model gene biology. We selected IL18, which has never been reported in breast cancer before, in the signature to learn more about its function, potential to predict outcome, and immune system role. RT-PCR was performed to verify the true expression level of IL18.

Results: A total of 136 DEGs associated with breast cancer responses to both immunity and lipid metabolism. Nine key genes (CALR, CCL5, CEPT, FTT3, CXCL13, FLT3, IL12B, IL18, and IL24, < 1.6e) of breast cancer were identified, and a prognostic was successfully constructed with a good predictive ability. IL18 in the model also had good clinical prognostic guidance value and immune regulation and therapeutic potential. Furthermore, the expression of IL18 was higher than that in paracancerous tissue.

Conclusions: A unique predictive signature model could effectively predict the prognosis of breast cancer, which can not only achieve survival prediction, but also screen out key genes with important functional mechanisms to guide clinical drug experiments.