Gene Expression Signature in Diagnosis of Endometriosis

Abstract

Study Objective To develop a prototype of a complex gene expression biomarker for the diagnosis of endometriosis based on differences between molecular signatures of endometrium from women with and without endometriosis. Design Prospective observational cohort study. II-1. Evidence obtained from a well-designed, controlled trial without randomization. Setting Department of Reproductive Medicine and Surgery at the A.I. Evdokimov Moscow State Medical and Dental University. Patients or Participants 30 women with endometriosis and 15 women without endometriosis (control group). Interventions Laparoscopic excision of endometriotic foci, hysteroscopy with endometrial sampling. RNA was isolated from all samples and stored in RNA Later. RNA sequencing was performed using Illumina HiSeq 3000 equipment for single-end sequencing. Unique bioinformatics algorithms were developed and validated using experimental and public gene expression datasets. Measurements and Main Results We performed gene expression analysis of dataset containing 83 samples (30 endometrial and 53 endometriotic) and 15 samples (endometrial) of patients with and without endometriosis respectively. We extracted a complex molecular signature of 38 genes and found that the expression of 26 genes in it was significantly increased while the expression of 12 genes was significantly repressed in the endometrium of patients with endometriosis. This endometrial genetic signature successfully differentiated 53 samples of endometriotic lesions from 15 endometrial samples of healthy women (area under the ROC curve (AUC) =1. The comparison of our dataset of 83 samples of endometrial and endometriotic tissue with preexisting dataset containing 134 samples of other tissues (cervix, ovary, stomach, lung) revealed high sensitivity (94%) and specificity (97%) in the ability of studied molecular signature to equally identify endometrium and endometriotic tissue of patients with endometriosis. Conclusion We obtained a complex molecular biomarker that could be used as a basis for early diagnosis of endometriosis via utilization of endometrial biopsy. Our findings indicate that easily accessible eutopic endometrium can be potentially used as a non-surgical marker for the presence of endometriosis.