The Construction And Validation Of Prognostic MiRNA Signature From The Integrated Analysis Of Patients With Invasive Breast Cancer

Objectives:Despite the advances in early detection and treatment method,breast cancer still has a high mortality rate,even in the patients that were supposed to have a good prognosis.The purpose of this study is to identify a panel of miRNA signature that could predict prognosis in breast cancer,adding new insights to current classification criteria and treatment therapy.Methods: We downloaded miRNA sequencing data along with their clinicopathological data of 1098 breast cancer patients from The Cancer Genome Atlas,analyzed 253 patients(153 as a training set and 100 as a validation set)after a selection process for human miRNA.A 3-miRNA signature was generated by analysis in the training set.Then the performance of the signature was confirmed by the validation set.After the construction of the signature,we combined training set and validation set together to do subgroup analysis.We also did additional experimental procedures including quantitative RT-PCR,flow-cytometric,CCK8 to illustrate the correlation between the signature and breast cancer cell cycle,apoptosis and proliferation.Results: 3 miRNAs(hsa.mir.31,hsa.mir.16.2,hsa.mir.484)were identified to be significantly correlated with patient prognosis and performed good stability.Our results suggest higher expression of hsa.mir.484 indicated worse prognosis,while higher expression of hsa.mir.31 and hsa.mir.16.2 indicated better prognosis.The risk score =(0.494*status of hsa.mir.484)–(0.786* status of hsa.mir.16.2)-(0.620* status of hsa.mir.31).In the subgroup analysis,the results showed that the signature performed good stability in young(?45y),basal-like and hormone receptor positive female breast cancer patients,while the predcitation was not very accurate in Her2-enrichment subtype or patients over 45 y.What's more,the proliferation rate of the cells was significantly increased in high risk breast cancer cells,along with the activation of cell cycle and slight decreasion of apoptosis rate,while the low risk group had opposite results.Conclusion: Our result indicated that the three-miRNA signature can predict the prognosis of breast cancer accurately,especially in young,basal-like and hormone receptor positive breast cancer subtype,so we recommend that patients in high risk group should be treated with more aggressive treatment therapy and shorter follow-up period.