Free and extracellular vesicle-associated microRNAs from endometrial fluid as non-invasive diagnostic biomarkers of implantative endometrium

Abstract Study question Is it possible to use free and extracellular vesicle-associated microRNAs from endometrial fluid as non-invasive biomarkers for implantative endometrium? Summary answer It is feasible to use free and extracellular vesicle-associated miRNAs as non-invasive tools for the detection of an implantative endometrium. What is known already MicroRNAs (miRNAs) and extracellular vesicles (EVs) from endometrial fluid (EF) have been described as mediators of embryo-endometrial crosstalk. Therefore, the miRNA analysis of EF could provide a non-invasive technique for recognizing an implantative endometrium and consequently improve implantation rates. Study design, size, duration A cohort of 162 women who assisted the human reproduction unit from January 2018 to February 2021. Of them, 72 participated in the setup and samples were collected before starting any fertility treatment in natural cycles. For the discovery of the predicted models (n = 30) and validation (n = 60), the EF was obtained from women undergoing frozen embryo transfer on day 5 and the sample was collected immediately before embryo transfer. Participants/materials, setting, methods We compared five different methodologies, two of which consisted of direct RNA extraction while the other three had EV enrichment prior to RNA extraction. A smallRNA-seq was performed to determine the most efficient method and to find a predictive model to differentiate between implantation and non-implantation endometrium. The models were confirmed by qPCR in two set of samples, discovery and validation cohorts with a different implantation outcome. Main results and the role of chance Our results showed that EV-enrichment protocols permit detection of a higher number of miRNAs. In addition, we obtained two predictive models based on three miRNAs that allow us to differentiate between an implantative and non-implantative endometrium. Results for model 1 in discovery cohort were: AUC=0.93; p-value = 0.003 and in validation cohort were: AUC= 0.69; p-value = 0.019. Results for model 2 in discovery cohort were: AUC=0.92; p-value = 0.0002 and in validation cohort were: AUC= 0.78; p-value= 0.0002. Limitations, reasons for caution One limitation to consider is the inherent variability of both the women involved in the trial and the embryos transferred. In our study, we have previously selected embryos based on morphology but we lacked genetic and molecular studies, a good complement that will certainly improve our test’s accuracy. Wider implications of the findings This study introduces new protocols to analyze miRNAs from very small volumes of EF, which could be implemented in clinical practice for the assessment of the endometrial status using miRNA-based non-invasive tools. Our results suggest that with model-2 it is possible to identify a non-implantative endometrium with 0.6-sensitivity and 0.93-specificity. Trial registration number not applicable