| Objective:Breast cancer has a high incidence among women in the world.It is considered to be an important factor affecting women’s daily life,physical and mental health.In recent years,the number of confirmed cases and deaths of breast cancer has increased continuously,becoming the main cause of death for women.The prognosis of patients withlymph node metastasis is generally poor.However,a combined genetic biomarker associated with lymph node metastasis that can effectively predict the prognosis of breast cancer has not been reported.Methods:The data of gene expression and clinical of 981 breast cancer tissues and 109 normal tissues were acquired from TCGA database.The DESeq2 package was used to screen differential genes between breast cancer tissues and normal breast tissues,breast cancer LNM tissuesand NLNM tissues.Venn diagrams were used to determine the intersection between the two groups of upregulated genes to obtain DEGs in common.Univariate Cox regression followed by multivariate Cox regression was performed for crossover genes to identify independent prognostic factors and build a predictive model.For each sample,a risk score was computed.and then Kaplan Meier curves were used to compare the prognosis between the two groups.the effect evaluation of the polygenic prognostic model was evaluated using the ROC curve.According to the median risk score,981 breast cancer patients were divided into low-and high-risk groups.Cox regression models were developed to evaluate the impact of polygenic risk score and clinical-pathologic characteristics on prognosis of breast cancer patients.Kaplan Meier survival curves of all factors and their subgroups were drawn to verify the predictive effect of polygene model in the prognosis of breast cancer.Finally,the potential molecular mechanisms and characteristics of these genes was explored by functional and pathway enrichment analysis.Results:A total of 5023 differential genes were screened between breast cancer tissues and normal breast tissues,of which 2082 were upregulated genes.A total of 178 differential genes were screened between breast cancer LNM and NLNM tissues,among which 47 were upregulated genes.The two sets of upregulated genes were intersected to obtain 39 intersected genes.Univariate and multivariate Cox regression analyses were performed on the 39 intersection genes,and seven genes（IAPP,LHX1,NDST4,LYZL2,HUS1B,XKR7,and PSCA）associated with LNM were selected to construct a polygenic prognostic model for breast cancer.The risk assessment score formula was as follows:risk score=(0.24469*Exp LHX1)+(0.42797*Exp XKR7)+(0.43192*Exp LYZL2)+(0.22742*Exp NDST4)+(0.13228*Exp PSCA)+(0.35249*Exp I APP)-(0.52426*Exp HUS1B).981 breast cancer patients were divided into low-and high-risk groups.Kaplan-Meier survival curve showed that the survival rate of high-risk group was significantly lower than that of low-risk group.The area under the ROC curve（AUC）was 0.714,indicating that the multi gene model had good predictive ability.Cox regression analysis of clinical data showed that age,cancer status,and the 7-gene model risk score were independent prognostic factors in patients with breast cancer.Kaplan Meier survival analysis showed that the multigene prognostic model had better prognostic predictive value in patients with all subgroups of breast cancer（except N0,M1 and associated tumors）.Functional annotation in high-risk patients suggests that this risk model was mainly associated with pathways such as neuroactive ligand receptor interaction,steroid hormone biosynthesis,and retinol metabolism.Conclusions:1.A seven-genes model associated with lymphatic metastasis was able to better predict the prognosis of breast cancer patients and may serve as a combined biomarker for breast cancer prognosis.2.The gene model can be used to predict the prognosis of advanced breast cancer patients with positive lymph nodes. |
PDF Full Text Request