Multivariate Survival Analysis And Molecular Classification Of Endometrial Cancer

Part 1 Multivariate survival analysis and quality of life of the patients with ECSection 1 Prognostic factors in endometrial cancer Objective: The aim of this study was to identify the multivariate independent prognostic factors of endometrial cancer. Methods: The retrospective analysis included cases with EC who received the initial surgical treatment from Jun 2000 to Apr 2014 in department of gynecology, Beijing Chao Yang hospital of capital medical university. Data of demographics and clinicopathology were collected. Survival analysis was performed by Kaplan-Meier curves and log-rank test; the independent risk factors for survival were identified by Cox proportional hazard analysis. Results: A total of 314 patients with complete clinicopathologic and follow-up data were enrolled. The median follow-up time was 43 months and the 5-year survival rate was 91.8%. Univariate analysis revealed that age, menopause, tumor markers, cervical stroma invasion, tumor grade, lymph node metastasis, vascular cancer embolus and adjuvant therapy were significantly associated with overall survival(P<0.05). Cox regression analysis identified FIGO stage, depth of myometrial invasion and medical disease as independent prognostic factors for overall survival(P<0.05). Conclusion: FIGO stage, depth of myometrial invasion and medical disease are the independent prognostic factors for overall survival of EC. These risk factors need to be studied in greater detail.Section 2 Quality of life and sexual function in endometrial cancer survivorsObjective: This study was performed to evaluate the quality of life(QOL) and sexual function and identify their associated factors in endometrial cancer(EC) survivors. Methods: A cross-sectional study was conducted. The participants in this study were EC survivors after surgery who visited the gynecological outpatient department for routine surveillance from May 2014 to May 2015. QOL and sexual function was measured using the Assessment of Cancer Therapy-General(FACT-G) and Female Sexual Functioning Index(FSFI) questionnaire. A score below 26.55 was defined as female sexual dysfunction(FSD). Multivariate analysis and logistic regression was used to identify the factors associated with QOL and sexual function. Results: A total of 118 women completed valid questionnaires. The mean score of FACT-G was 84.08±14.83. Chemotherapy and marital status significantly impaired the physical wellbeing domain and social/family wellbeing scores, respectively(P<0.05). Moreover, monthly income was a significant factor that affected the total FACT-G scores. 68.6% patients had FSD and 55.9% patients never had sexual intercourse with their partners after surgery. Age, following time after surgery, radiotherapy and consultation significantly correlated with FSD. Conclusion: The risk factors associated with QOL and sexual function need to be studied in greater detail. Prospective studies that evaluate the effects of clinical psychological intervention on sexual function and quality of life are needed in the future.Part 2 Molecular classification of endometrial cancer based on text miningSection 1 Systematic analysis of endometrial cancer-associated hub proteins Objective: The aim of this study was to systematically characterize the expression of endometrial cancer(EC)-associated genes, and to analysis the functions, pathways and networks of EC-associated hub proteins. Methods: Gene data for EC were extracted from the Pub Med(MEDLINE) database using text mining(TM) based on natural language processing(NLP). Protein-protein interaction(PPI) networks and pathways were integrated and obtained from The Kyoto Encyclopedia of Genes and Genomes(KEGG) and other databases. Proteins that interacted with at least 10 other proteins were identified as the hub proteins o f the EC-related genes network. Results: A total of 489 genes were identified as EC-related with P<0.05, and 32 pathways were identified as significant(P<0.05, FDR<0.05). A network of EC-related proteins that included 271 interactions was constructed. The 22 proteins that interact with 10 or more other proteins(P<0.05, FDR<0.05) were identified as the hub proteins of this PPI network of EC-related genes. These 22 proteins are EGFR, IGFIR, MET, PDGFRβ, CCND1, JUN, FGFR1, MYC, PIK3 CA, PIK3 CG, FGFR2, KRAS, MAPK1, MAPK3, CTNNB1, PIK3R1, PIK3R2, RELA, FOS, JAK2, AKT1 and AKT2. Conclusion: Our data may help to reveal the molecular mechanisms of EC development and provide implications for targeted therapy for EC. However, corrections between certain proteins and EC continue to require additional exploration.Section 2 Molecular classification of endometrial cancer by hub proteins Objective: The aim of this study was to create a molecular classification for endometrial cancer(EC) by hub proteins. Methods: A tissue microarray(TMA) containing 109 endometrial cancers(from Jan 2000 to Apr. 2011) was prepared. The expression of 22 different protein biomarkers which were obtained from text mining was assessed by immunohistochemistry and the data was analyzed using discriminant analysis by steps. Results: The high expression of c-MET、FGFR2、CTNNB1、AKT1 and AKT2 was associated with poor overall survival(P<0.05). The discriminant analysis showed that AKT1 and FGFR2 had the most closely relationship with prognostic, and the accuracy of discriminant function was 76.1%. Conclusion: c-MET、FGFR2、CTNNB1、AKT1 and AKT2 may play an important role in the progression of EC and should be studied further as prognostic and therapeutic tools. Joint detection of the expression of AKT1 and FGFR2 has the value in the evaluation of prognosis.