Research On The Pathogenesis Of Endometriosis Based On Single-cell Transcriptomics And Construction Of A Prediction Model For Postoperative Recurrence Of Ovarian Endometriosi

PART 1:Research on the pathogenesis of endometriosis based on single-cell RNA sequencingObjective:Endometriosis(EMS)is a chronic condition characterized by the presence of functional endometrial tissue outside of uterine cavity,which is associated with pelvic pain and infertility.Although the hormonal responses of endometriotic lesions,the molecular changes in the lesions across menstrual cycle have not been fully delineated.This study aimed to identify the transcriptome characteristics of the ectopic leisions through single-cell RNA sequencing and compare its molecular changes with eutopic endometrium at single-cell level.Methods:We performed single-cell transcriptome sequencing with 12 endometrial samples and 5 endometriomas.For the 12 endometrial samples.6 were from EMS patients(3 from proliferative phase and 3 from secretory phase)and 6 were from non-EMS controls(3 from proliferative phase and 3 from secretory phase).For the 5 endometriomas,2 of them derived from the patients at proliferative phase and 3 at secretory phase.Cells were clustered by the uniform manifold approximation and projection algorithm.Pseudo-time analysis was applied to delineate the cells state transition along the menstrual phases for epithelial and stromal cells.Results:For the endometrium,we defined two groups of menstrual feature genes corresponding to the estrogen-dominated proliferative phase and progestogen-dominated secretory phase of the endometrium,and confirmed the gene lists well evaluated the menstrual stages for single-cell RNA datasets.For the endometrioma,we identified the ectopic endometrial "seed" cells and revealed them exhibited similar molecular dynamics along the menstrual cycle as the eutopic endometrium.We also identified WNT5A upregulation in stromal cells and the enrichment of CCL21-expressing pericytes in endometriomas.Conclusions:Our data provided a new way to assess the menstrual phase of the endometrium,and revealed the previously unexplored menstrual-phase-dependent changes of endometrial "seed" cells in the ectopic lesion from single-cell data.We also identified WNT5A upregulation may promote the formation of ectopic lesions in EMS.In addition,compared with the endometrial tissue,pericytes in ovarian endometrioma was significantly increased.PART 2:The culture and preliminary exploration of endometrial organoidsObjective:In this study,the eutopic endometrial organoids model of patients with endometriosis(EMS)and endometrial organoids model of patients without EMS were cultured,and a series of phenotypic characteristics were verified.On this basis,the endometrial organoids biobank was constructed and the transcriptomic differences between endometrial glandular epithelium in patients with EMS and patients without EMS were compared from the perspective of endometrial organoids.Methods:Eutopic endometrium from patients with EMS who were treated by laparoscopic surgery and confirmed by postoperative pathology,and endometrium from patients without EMS who were determined by laparoscopic surgery and postoperative pathology were collected to culture endometrial organoids.The morphology of endometrial organoids was observed under light microscope,and the distribution of living and dead cells was observed by Calcein-AM/PI Double Staining,the proliferative activity of organoid was testing by BdU markers,glandular epithelial markers were testing by immunofluorescence,the stemness of endometrial organoid cells was detected by immunohistochemistry,and the internal ultrastructure of organoid cells was observed by transmission electron microscope.In addition,RNA was extracted from endometrial organoids and transcriptome sequencing was used to compare the transcriptomic differences of endometrial organoids between patients with EMS and patients without EMS.Results:We successfully cultured endometrial organoids in patients with EMS and patients without EMS,and the stable passage,cryopreservation and resuscitation of endometrial organoid were achieved.A series of phenotypic tests showed that:the organoids were spheroids,living cells were mainly distributed in the periphery,and dead cells were mainly located in the interior.BdU testing indicated that the proliferation activity of organoid was positive.The results of immunofluorescence indicated that organoid expressed EpCAM and Keratin 7,which were both glandular epithelial cell markers.Immunohistochemical results indicated that the expression of stemness markers including Sox9,LGR5 and SSEA1 were positive in organoids.The ultrastructure of organoid cells including cilia and close intercellular connections was observed under transmission electron microscopy.Transcriptome sequencing results suggest that organoids are used as experimental objects,compared with the endometrial organoids of the control group,the GO analysis of genes with low expression in the eutopic endometrial organoids of the patients with endoheterosis was mainly concentrated in the regulation of ion transmembrane transport,extracellular matrix organization,extracellular structure organization,hormone transport,and hormone secretion,etc.Conclusions:Endometrial organoids are an ideal research model for the study of endometria-related diseases.The establishment of endometrial organoids biobank in patients with EMS and in patients without EMS is of great significance for the later study of the pathogenesis of endometriosis.PART 3:The development of predictive nomogram of recurrence for patients with endometrioma after cystectomyObjective:This study aimed to establish an effective prognostic nomogram for the postoperative recurrence of endometrioma or endometriosis-related pain for patients with endometrioma after long-term follow-up,who were younger than 45 years old and received postoperative therapy.Methods:The predictive nomogram was based on 323 patients who underwent cystectomy for endometrioma at Perking Union Medical College Hospital from January 2009 to April 2013,and the last follow-up occurred in September 2018.We collected information on all included patients,including preoperative data,intraoperative data,and long-term follow-up data after surgery.The Cox proportional hazards regression model was used to evaluate the prognostic effects of multiple clinical parameters on recurrence.The survival curve was depicted based on Kaplan-Meier method and compared by logrank method.The Index of concordance(C-index)and calibration curves were used to access the discrimination ability and predictive accuracy of the nomogram respectively,and the results were further validated via bootstrap resampling.In addition,calculating the area under the curve(AUC)via risk scores of patients aimed to further access the prediction ability of the model.Results:On multivariate analysis of derivation cohort,independent factors for recurrence such as dysmenorrhea degree,sum of both cyst diameters,presence of adenomyosis,and other essential factors for recurrence such as age at surgery,presence of uterine fibroids were all selected into the nomogram.The C-index of the nomogram for predicting recurrence was 0.683(95%CI,0.610-0.755).The calibration curve for probability of recurrence for 7 years and 9 years showed great agreement between prediction by nomogram and actual observation.Furthermore,the AUCs of risk score for 7-year and 9-year were 0.680 and 0.790 respectively.Conclusions:This research tried to develop the predictive nomogram of recurrence for patients with endometrioma after cystectomy.The C-index and calibration curve of nomogram,as well as the AUC of the nomogram was potential to predict the recurrence probability.In addition,this predictive nomogram needs external data sets to further validate its prognostic accuracy in the future.