Risk profiling of endometrial cancer

This thesis consists of cohort studies to evaluate the reliability of methods that are presently accessible for the risk profiling of endometrial cancer treatment and to augment contemporary ways that might be clinically usable. Cohort studies were based on a sample of 1166 women who were surgically treated for endometrial cancer at the Department of Obstetrics and Gynecology, Helsinki University Hospital, between January 2007 and December 2013. In the first study, previously recognized risk factors for advanced stage and poor outcome were used to create a calculatory score to predict lymph node and distant metastasis in endometrial cancer (EC). The association of advanced stage disease with demographic factors, biochemical factors, preoperative histology and tumor size was examined. The model predicted stage IIIC–IV carcinomas with a sensitivity of 100%, specificity of 38.0%, PPV of 17.1% and NPV of 100%. A subgroup analyses suggests that the model predicts lymph node metastases (LNM) in addition to distant metastases. In the second study the risk scoring system developed in study I was evaluated by comparing its performance characteristics with two other risk models in predicting lymphatic dissemination. The models had similar accuracies. Survival analyses suggest that variables included in the models predict patient outcome independently of tumor stage. The third study was conducted to investigate the correlation of predictors of advanced disease and/or poor outcome with the manifestation of tumor relapses in different anatomical sites in early stage endometrioid EC. A relapse was diagnosed in 98 patients (10.5%). Of these, 15 were vaginal, 27 pelvic, 27 intra-abdominal and beyond the pelvis and 29 extra-abdominal. Poor differentiation, myometrial invasion (MI) 50% or higher, a tumor size of 3 cm or greater and abnormal peritoneal cytology were associated with an increased risk of intra-abdominal relapses. A tumor size of 3 cm or greater and abnormal peritoneal cytology predicted intra-abdominal relapses beyond the pelvis whereas poor differentiation, MI ≥ 50% and abnormal peritoneal cytology predicted extra-abdominal relapses. In the fourth study, the incidence of surgical site infection and its risk factors in EC were evaluated. In all, 78.4% had had a hysterectomy by minimally invasive surgery. Ninety‐four women (8.1%) were diagnosed with a surgical site infection. The associations clinicopathologic and surgical variables were tested. Obesity, diabetes and long operative time were associated with a higher risk of incisional infection, whereas minimally invasive surgery was associated with a smaller risk. Smoking, conversion to laparotomy, and lymphadenectomy were associated with a higher risk of organ/space infection.