Predict The Prognosis Of Breast Cancer And Lung Metastasis Gene Expression Profile

Partâ… :Integrated gene expression profile predicts prognosis of breast cancer patientsObjective:To develop a gene expression profiling for predicting the risk of recurrence and metastasis in breast cancer by screening a set of specific genetic markers using low density microarray.Methods:The expression of genes in breast cancer tissue samples was detected by a low density microarray to screen specific gene markers and develop a new gene expression profiling.An unsupervised hierarchical cluster was performed by Cluster 3.0 and Treeview 1.6 software.COX proportional-hazards regression and Kaplan-Meier survival analyses were used to identify the prognosis significance of the gene expression profiling.The comparison with other signatures was performed by ROC analyze.Results:24 specific genetic markers were selected from 248 candidate genes to develop a new gene expression profiling.The gene expression signature stratified correctly samples of training set and test set into the good prognosis group and the poor one.The difference of disease-free survival between two groups is highly significant.Other than gene expression profiling,the prognosis was only significantly associated with axillary lymph node status in univariate and multivariate Cox's proportional-hazards regression analyses.The gene signature also showed good predictive effect in lymph node positive and negative subgroup,especially in lymph node negative subgroup.When compared with other signatures,the integrated signature showed more predominant power of predication in breast cancer patients receiving chemotherapy.Conclusions:The genes expression profiling is the strongest independent prognostic factor in breast cancer patients receiving chemotherapy and will critically aid clinical decision making at the level of individualization for most breast cancer. Partâ…¡:Gene expression profile predict lung metastasis of breast cancer patientsObjective:To establish a human breast cancer cell line with highly spontaneous lung metastatic capability so as to provide suitable model for the study of molecular mechanism related to metastasis and screen a set of specific genetic markers for predicting the risk of lung metastasis in breast cancer patients.Methods:MDA-MB-231HM,a highly lung metastatic variant of parental MDA-MB-231 cell,was derived by six cycles of selection in vivo.The original of the highly lung metastatic cell was demonstrated using microsatellite DNA.The difference between MDA-MB-231HM and parental cell was tested in vitro by proliferation assays,flow cytometry as well as transwell and in vivo by comparing the weight of orthotopic tumor and the number of metastasis in lung of nude mice.The expression of genes in two breast cancer cell lines was detected by a low density microarray to screen specific gene markers and develop a new gene expression profiling.An unsupervised hierarchical cluster was performed in 63 breast cancer patients to identify the prognosis significance of the gene expression profiling.The results of microarray were partly validated by Real-time PCR.Results:The two cell lines all originated from human breast cancer.The capability of MDA-MB-231HM was dramatically increased in proliferative and invasive study in vitro.The weight of orthotopic tumor and the number of lung metastasis in MDA-MB-231HM were large when comparing to parental cell. Ninety-two genes were significantly differentially expressed between the two cell line and 16 specific genetic markers were selected to develop a new gene expression profiling by compared to poor prognosis signature.The results of microarray were validated correctly by Real-time PCR in 11 of 12 genes.The gene expression signature stratified correctly samples into the lung metastatic group and the non-lung metastatic one.The difference of lung metastasis-flee survival between two groups is highly significant.Other than gene expression profiling,the prognosis was only significantly associated with ER status in univariate Cox's proportional-hazards regression analyses.The hazard ratio of gene signature was higher than ER status,suggesting that the signature was the strongest prognostic factor. Conclusions:We established a human breast carcinoma cell line with High Capability of Spontaneous lung Metastasis and develop a new gene expression profiling for predicting the risk of lung metastasis in breast cancer patients.