Study On The Correlation Between Endometriosis-associated Ovarian Cancer And Eutopic Endometrium And Prediction Model

cancer and eutopic endometrium and prediction model Background:Endometriosis(EMT)is a gynecological benign disease with malignant biological behavior,which can cause chronic pelvic pain and infertility in patients.According to the location of occurrence,EMT can be divided into ovarian type(Ovarian Endometriosis,OME),peritoneal type and deep infiltrating type.Ovarian type is the most common and often forms cysts with chocolate-like fluid inside,thus called chocolate cysts.Studies have shown that about 1% of OME patients may undergo malignant transformation,called Endometriosis-associated Ovarian Cancer(EAOC),whose main pathological types are endometrioid carcinoma and clear cell carcinoma.In recent years,research has found that eutopic endometrium may play an important role in the development of EMT and the malignant transformation to EAOC,but there is still no effective method to identify OME patients with malignant transformation potential,nor can the risk of future EAOC be predicted.Therefore,it is necessary to further improve the clinical prediction and diagnostic ability of EAOC,explore the origin and pathogenesis of EAOC,and provide guidance for identifying high-risk patients and early prediction and diagnosis of EAOC.Objectives:1.Clarify the role and clinical applications of biomarkers originating from eutopic endometrium in predicting the development of endometriosis(EMT).2.Screen biomarkers predictive of endometriosis-associated ovarian cancer(EAOC)from eutopic endometrium and elucidate the impact of eutopic endometrium on the development of EAOC.3.Elucidate the effects and mechanisms of the biomarkers ALDH2 and FOS protein on the proliferation and migration abilities of eutopic endometrial cells in EAOC patients.4.Construct an EAOC prediction model based on routine clinical features to provide guidance for early diagnosis of EAOC.Method:1.Use data from the Gene Expression Omnibus(GEO)to analyze the differential gene expression(DEGs)between EMT and non-EMT patients in the endometrium.The Lasso regression and real-time quantitative polymerase chain reaction(q PCR)techniques are used to further screen and validate the transcription levels of biomarker m RNAs in the endometrium of EMT patients.Gene set enrichment analysis(GSEA),Kyoto Encyclopedia of Genes and Genomes(KEGG)analysis,and Gene Ontology(GO)analysis are performed on DEGs based on R language.An EMT prediction model is constructed based on the endometrial biomarkers of EMT and non-EMT patients.2.Fresh endometrial tissue from EAOC and OME patients is collected and its gene expression levels are examined using transcriptome sequencing technology.Gene differential analysis,GO analysis,GSEA analysis,and Protein-Protein Interaction(PPI)network analysis are performed using R language.Endometrial paraffin tissue sections from EAOC and OME patients are collected and subjected to microdissection.Dataindependent acquisition(DIA)protein quantification analysis is used to detect protein expression levels,and protein differential analysis,GO analysis,KEGG analysis,and PPI analysis are performed using R language.Based on transcriptomics and proteomics analysis and pre-experimental results,biomarkers of EAOC patients in the endometrium are determined.Immunohistochemical staining(IHC)is used to detect the expression of FOS protein and ALDH2 protein in the endometrium of the two groups of patients and analyze their correlation.ALDH2,FOS,and proliferation-related protein expression were measured using the IHC method.3.Primary human endometrial stromal cells(h En SCs)were extracted and cultivated from patients with ovarian endometriosis(OME)and endometriosisassociated ovarian cancer(EAOC),respectively.The activity of FOS and ALDH2 in h En SCs was regulated using FOS inhibitors and ALDH2 agonists.FOS expression was upregulated and ALDH2 expression was downregulated in h En SCs-OME using lentivirus transfection technology.ALDH2 expression was upregulated in h En SCsEAOC using lentivirus transfection technology.The proliferation ability of h En SCs was measured using the CCK-8 assay and the migration ability was measured using the scratch assay.The expression of ALDH2,FOS,and cell cycle-and migration-related proteins in the cells was measured using Western Blot.The effect of different sources or treatments of h En SCs on the tumor volume of ES2 nude mice was examined by the nude mouse transplanted tumor model.4.The training set was conducted on patients diagnosed with OME(392 cases)and EAOC(76 cases)at The Second Hospital of Jilin University between January 2013 and December 2022.The validation set was conducted on patients diagnosed with OME(103 cases)and EAOC(31 cases)at A Hospital in Changchun and B Hospital in Changchun between January 2013 and December 2022.Patient medical records were collected and a table of patient demographics was created.A binary logistic regression model was used to construct a prediction model for EAOC,and the discriminative ability of the model for distinguishing between EAOC and OME patients was evaluated using the ROC curve.The reliability of the model was assessed using a calibration curve based on the Hosmer-Lemeshow test.Internal validation of the prediction model was completed using 10-fold cross-validation.The clinical decision curve and nomogram were used to evaluate the practical application value and visualization of the model.Results:1.Through bioinformatics analysis,96 genes related to malignant biological characteristics such as inflammation,cell proliferation,and cell adhesion associated with EMT were screened out in situ endometrial tissue.Lasso regression further selected 4 m RNAs(EGR1,FOS,NUP62 CL,and CSF1R)in situ endometrial tissue as biomarkers for predicting EMT.Based on these 4 genes,an EMT prediction model with good discriminative ability(AUC = 0.9124)was constructed.The model passed the Hosmer-Lemeshow test(P> 0.05),and the AUC after 5-fold cross-validation was0.8969.2.Transcriptomic and proteomic analysis showed that cell mitosis,migration,invasion,and other related pathways were further activated in the in situ endometrial tissue of EAOC patients compared to OME patients,and ALDH2 protein expression was downregulated while FOS protein expression was upregulated.The protein expression of ALDH2 and FOS in the in situ endometrial tissue of EAOC patients showed a negative correlation.3.Compared with h En SCs-OME,ALDH2 protein expression was significantly downregulated while FOS protein expression was significantly upregulated in h En SCsEAOC.FOS inhibitor significantly inhibited the proliferation and migration ability of h En SCs-EAOC.Overexpression of FOS protein enhanced the proliferation and migration ability of h En SCs-OME but had no effect on ALDH2 protein expression.Downregulation of ALDH2 protein expression promoted the proliferation and migration ability of h En SCs-OME and upregulated FOS protein expression.ALDH2 agonist inhibited the proliferation and migration ability of h En SCs-EAOC and downregulated FOS protein expression.FOS inhibitor reversed the enhanced proliferation ability of h En SCs-OME caused by downregulation of ALDH2.Overexpression of FOS protein restored the decreased proliferation ability of h En SCsEAOC mediated by ALDH2 agonist.h En SCs-EAOC promotes the proliferation of ovarian clear cell carcinoma,and overexpression of ALDH2 reverses this effect.4.Increasing age at onset,increasing maximum tumor diameter,menopause,and the presence of blood flow within the tumor are independent risk factors for EAOC.Increased number of deliveries was an independent protective factor for EAOC occurrence.The optimal cutoff value for age is 47 years and for maximum tumor diameter is 8.75 cm.Based on the routine clinical characteristics of OME and EAOC patients and binary logistic regression,a predictive model for EAOC was successfully constructed with good discriminative ability(Training set AUC=0.9717;Validation set AUC=0.9361),and the model passed the Hosmer-Lemeshow test(P>0.05).After 10-fold cross-validation,the AUC was 0.9667.Conclusion:1.Gene expression changes associated with malignant biological characteristics in endometrium may be related to the occurrence and development of EMT.A multi-gene prediction model based on m RNA expression levels of four genes(EGR1,FOS,NUP62 CL,and CSF1R)in endometrium has good discriminative ability,stability,and clinical usability,and can effectively identify EMT patients.It may be used for early prediction of EMT in the future.2.Gene and/or protein expression changes associated with malignant tumors in endometrium may be the reason why patients are more likely to develop EAOC,and low expression of ALDH2 and high expression of FOS are risk factors for EAOC.3.The expression of ALDH2 protein was negatively correlated with the expression of FOS protein.Low expression of ALDH2 protein and high expression of FOS protein predicted that h En SCs might have the potential for malignant transformation.h En SCs in EAOC patients could promote the proliferation of ovarian clear cell carcinoma,while overexpression of ALDH2 could reverse this effect.4.OME patients over 47 years old,with a maximum tumor diameter exceeding8.75 cm,menopause,the number of births is less than equal 1,and presence of blood flow in the tumor,should be vigilant for EAOC occurrence.An EAOC prediction model based on conventional clinical features and binary logistic regression of OME and EAOC patients has good discriminative ability,stability,and clinical usability,and can effectively identify EAOC patients.