The Study On Endometriosis Related Biomarkers And Establishment Of Multi-parameter Diagnostic Model

Background and Objective:Endometriosis (EMs) is a kind of benign disease of women，but it has potential risk tobecome malignant tumor. Our studies try to explore the possibility to establish highefficient diagnostic model to distinguish malignant tumor from EMs by using serumbiochemical parameters, tumor markers, cytokines and serum proteomics respectively, inorder to provid more accurate laboratory informations to help early diagnosis and therapyto malignant tumor and EMs.Material and Methods:Subjects were divided into healthy control group (85healthy women), EMs group (80EMs patients) and disease control group (61patients with ovarian cancer and endometrialcancer).38serum biochemical markers were evaluated on Cobas8000automatic analyzer.14serum tumor markers were evaluated on Cobas6000ECL analyzer and i2000SRimmunoassay analyzer.15cytokines expression profile were detected by Luminex200liquid chip analyzer. Make use of MALDI-TOF/MS technology to analyze protein massspectrum and logistic regression analysis to establish diagnostic model.Results:1. Two diagnostic models were set up with serum biochemical markers.0.924AUC (areaunder curve) of ROC analysis for discriminating EMs group from healthy controlgroup with sensitivity of84.60%and specificity of92.90%. And0.934AUC of ROCanalysis for discriminating EMs group from disease control group with sensitivity of91.00%and specificity of80.30%.2. Two diagnostic models were set up with serum tumor markers.0.992AUC of ROCanalysis for discriminating EMs group from healthy control group with sensitivity of97.40%and specificity of95.30%. And0.871AUC of ROC analysis for discriminatingEMs group from disease control group with sensitivity of76.90%and specificity of 83.60%.3. A diagnostic model was set up with serum cytokines.0.789AUC of ROC analysis fordiscriminating EMs group from disease control group with sensitivity of78.20%andspecificity of63.90%.4. A diagnostic model was set up by associate with serum biochemical markers, tumormarkers, and cytokines.0.911AUC of ROC analysis for discriminating EMs groupfrom disease control group with sensitivity of89.70%and specificity of80.30%.5. Two diagnostic models were set up with protein mass spectrum.0.854AUC of ROCanalysis for discriminating EMs group from healthy control group with sensitivity of78.10%and specificity of85.70%. And0.740AUC of ROC analysis for discriminatingEMs group from disease control group with sensitivity of62.80%and specificity of77.40%.Conclusion:In conclusion, several diagnostic models have been set up by using the serum multiplebiomarkers, which could discriminate EMs from healthy and malignant tumor populationmore efficiently than individual biomarker. This study was a new exploration of EMsdiagnose that would improve the early diagnose and therapy of EMs and tumor.