The Study Of Long Non-coding RNA In Plasma Extracellular Vesicles In The Diagnosis Of Breast Cancer

Background:Breast cancer is a malignant tumor that seriously endangers women’s health.How to detect and diagnose breast cancer as early as possible has been an important research topic for clinical and medical researchers.Extracellular vesicles(EVs)are widely involved in the material exchange and information exchange between cells through their contents,thus to regulate various physiological and pathological processes of organisms.In addition,EVs have a high abundance in blood,and the extracellular vesicles secreted by tumors may contain tumor-specific nucleic acids and protein molecules,which are of great value for the diagnosis of tumors.At present,the diagnostic value of long non-coding RNA(lncRNA)as tumor markers has been widely concerned by many researchers.However,its long length,low intracellular abundance and the homology of mRNA are main problems in the field of its detection and diagnosis.In the humoral environment,these RNAs are found to be highly concentrated in extracellular vesicles,making them of high detection value.However,the in-depth application of high-throughput sequencing technology enables rapid and efficient screening of large-scale RNA,which makes the study of long non-coding RNA in plasma extracellular vesicles in the diagnosis of breast cancer have a high clinical practical value.Methods:(1)By comparing alcohol-induced sedimentation method and columnar membrane affinity adsorption method,the plasma-derived extracellular vesicle RNA extraction method suitable for this study was selected.(2)High-throughput whole-transcriptome assays were performed on 6/8 breast cancer samples(preinvasive carcinoma/invasive carcinoma)and 6 non-cancerous samples to screen candidate lncRNAs molecules with differences between groups.(3)RT-qPCR was used to verify the expression level of candidate lncRNAs in large-scale clinical plasma samples.(4)According to the lncRNA expression of samples,the breast cancer tumor classifier CLnc was constructed by combining machine learning algorithm.Results:(1)The column extraction method based on affinity adsorption produced fewer exosome impurities,maintained better morphology and was more suitable for extraction and identification in clinical testing than the deposition method based on alcohol substances.(2)A total of thousands of differentially expressed lncRNAs were screened for comparison between the breast cancer group and the non-cancer group,and 17 lncRNAs with significant expression differences were finally identified as candidate molecules after cluster analysis,functional annotation and target pathway analysis.(3)17 differentially expressed lncRNAs were also differentially expressed in large-scale samples.(4)The machine learning algorithm was used to construct the breast cancer tumor classifier,and finally the classifier CLnc composed of LR-02,LR-04,LR-05,LR-08,LR-09,LR-10,LR-11,a total of 7 lncRNAs was obtained.Conclusion:In this study,lncRNAs with significant differences in plasma-derived extracellular vesicles of breast cancer and non-cancer group were screened by high-throughput sequencing detection and calculation,and then were verified by RT-qPCR of large-scale samples.Combined with machine learning algorithm,CLnc,a tumor classifier device for breast cancer detection composed of 7 lncRNAs molecules,was selected.