Predicting Recurrence In Endometrial Cancer Based On A Combination Of Classical Parameters And Immunohistochemical Markers

Background:Endometrial cancer is one of the common gynecological tumors that endanger women's life and health.Incidence and mortality have increased year by year.At present,the choice of parameters for assessing the prognosis of endometrial cancer at home and abroad is still mostly limited to classic clinical parameters,such as FIGO staging,pathological classification and muscular infiltration.The number of patients with recurrence gradually increases,even in early low-risk endometrial cancer,some patients relapse after surgery.Therefore,it is urgent to establish a new prediction model to evaluate the prognosis of patients with recurrence,and provide reference for clinical prognosis management.Objective : The aim of this study was to establish a nomogram to predict the recurrence of endometrial cancer(EC)by adding immunohistochemical markers to the traditional clinical and pathological parameters.Methods: The archived data of 537 patients with stage I-IIIendometrial cancer who received primary surgical treatment between October 2013 and May 2018 were retrieved and analyzed.Data of 473 included cases were randomly split into two sets: training and validation in a predefined ratio of 7:3.A univariate regression was performed to screen factors associated with recurrence in EC in the training cohort(n=332),and a Cox proportional hazards multivariate model of selected prognostic features was applied to develop a nomogram,which was further validated in the validation cohort(n=141).The prediction capabilities of different combinations of parameters were also compared to confirm the capacity of this proposed model in clinical utility.Results: There were 47 recurrent cases in the training cohort and 20 cases in the validation cohort.Based on univariate and multivariate Cox regression analysis as well as the clinical relevance,FIGO stage(P = 0.029),pathological type(P = 0.062),ER(P = 0.048)and P53(P = 0.030)were finally included in the nomogram model.Therefore,recurrence-free survival(RFS)was best predicted by the proposed nomogram with a C-index of 0.88(95% confidence interval(CI),0.84–0.92),and the validation set confirmed the findings with a c-index of 0.79(95% CI,0.66–0.92).Conclusions: Immunohistochemical markers in addition to traditional clinicopathological parameters can best predict the recurrence in FIGO stage I-III EC.This nomogram model was demonstrated to be a robust toolfor predicting recurrence free survival rate.