A replication study of plasma microRNA expression in adolescents with endometriosis

Micro-ribonucleic acids (miRNAs), stable noncoding RNAs involved in gene regulation, have diagnostic potential. Our purpose was to validate 59 plasma miRNAs identified within a discovery study of adolescents with endometriosis vs. controls. Nested case-control study. Adolescents (<26 y) with surgically-visualized endometriosis (n=54, 96% rASRM Stage I-II) and controls (n=10, no endometriosis history), matched on age (±2 yrs, mean 21 y) and hormone use (y/n), were included. WERF EPHect protocols were used. miRNAs included 48 from a discovery study of plasma of adolescent cases vs. controls (n=10 each, via qt-PCR) plus 11 previously published miRNAs. For this replication study, plasma miRNAs were quantified on a bead-based platform (FireflyTM, Abcam). Wilcoxon signed-rank tests compared miRNA expression between cases/controls. Stepwise logistic regression models were built using miRNAs with p-values <0.10 (p=0.20 for entry and stay). Overall, 9 miRNAs were chosen: 4 among hormone users, 2 among non-users, and 4 in all participants. AUCs were similar for the three models: 0.74, 0.77 and 0.74, respectively (Table). In models restricted to pathology-confirmed cases (26 hormone users, 7 non-users) AUCs improved slightly (0.75, 0.86, 0.83). These miRNAs are associated with cell adhesion and proliferation, tumor metastasis, and embryonic neuro- and myogenesis. Several miRNAs were significantly dysregulated in plasma from adolescents with rASRM stage I/II endometriosis vs. controls, and together suggest fair to good (AUC 0.74-0.86) discrimination. The small number of miRNAs that replicated among the groups reinforces the need for multi-phased discovery and targeted replication. Hormonal medications appear to impact miRNA expression differently in cases v. controls. Future studies of miRNA variation by stage, hormonal and non-hormonal treatments and among disease phenotypes are needed.Tabled 1miRNAs selectedDiscovery StudyValidation StudyHormone usersNot on HormonesAll (adjusted for hormone use)(42 cases, 84 controls)(12 cases, 24 controls)(54 cases, 108 controls)OR (95% CI)pOR (95% CI)pOR (95% CI)pAUC = 0.74AUC = 0.77AUC = 0.74slnhsa_mir_376b_3pHormone Users0.38 (0.18, 0.81)0.01slnhsa_mir_1908_5pAdult Literature0.26 (0.08, 0.86)0.030.41 (0.13, 1.25)0.12slnhsa_mir_542_3pHormone Users0.67 (0.37, 1.19)0.17slnhsa_let_7g_3pAdult Literature2.09 (1.01, 4.34)0.05slnhsa_mir_296_3pNot on Hormones1.72 (1.08, 2.75)0.02slnhsa_mir_154_5pHormone Users15.8 (1.10, 227.6)0.04slnhsa_mir_219a_5pBoth Horm Groups2.65 (0.76, 9.20)0.12slnhsa_mir_124_3pHormone Users1.55 (0.84, 2.83)0.16slnhsa_mir_23b_5pHormone Users0.61 (0.35, 1.07)0.08 Open table in a new tab