Identification Of Prognostic Predictors And Exosome-shuttled Biomarkers For Osteoporotic Patients

Background:Along with the aging population,the incidence of osteoporosis keeps soaring,which causes a heavy burdern to both society and individuals.The severity of osteoporosis acts in two ways including:1)The disease comes silently.Although guidelines have changed to encouraging early diagnosis and treatment.However,no distinguish and robust biomarkers in the clinical practice to help identify patients at early stage,therefore most people don't know the disease is there untifl the fracture occurs;2)High mortality rate after osteoporotic fracture.It has been reported that the mortality rate for hip fracture is as high as 36%.Given these,identify specific markers in screening out osteroprotic patients and those with fracture who are at higher risk of death is of extreme importance.Our primary experiments showed that we could lock down some markers including both clinical parameters and molecular expressions patients' in blood so as to tackled the task aforementioned.Objective:1)Identify risk predictors and buid a risk stratification model in prediction of long-term mortality for hip fracture patients;2)Identify osteoporosis biomarkers from exosome-derived protein and microRNAs.Methods:1)Recruit a large prospective cohort,follow up the participants to record their death as the primary outcome.Using Cox regression model or Random Forest algorithm to screen out the independent predictors and to build a risk stratification model.2)Collect the blood specimen from the participants,and then compare the expression differences of exosome-derived proteins and microRNAs by using proteomics and microarray analysis so as to detect specific biomarkers related to the disease.Results:1)Admission anemia is an independent predictior for hip fracture mortality,as compared to normal subjects,the mortality risk for mild anemia patients increases by 82.9%(HR 1.829 95%CI 1.250-2.675),for moderate anemia patients increases by 157.9%(HR 2.579 95%CI 1.792-3.712),and for the severe anemia patients by 184%(HR 2.84095%CI 1.407-5.731);2)Caculateed a novel risk predictor as ?RDW,the larger ?RDW patients experience,the higher mortality risk they will have;3)Constructed a risk stratification model by using machine learning algorithm,achieving a C-statistic as 0.75 for 2-year mortality risk prediction and 0.72 for 4-year;4)Identified 4 exosome-derived proteins in related to osteoporosis,including PSMB9,AARS,PCBP2,and VSIR;5)Identified 4 exosome-derived microRNAs in related to osteoporosis,namly miR-133b,miR-199a,miR-663a,and miR-4697.Conclusion:The preditors and risk stratification model shows a good performance in mortality risk prediction,albeit further validation requires.The identification of exosome-derived biomarkers for osteoporosis provide molecule cludes in early diagnosis and disease monitoring.