Introduction/Background Traditional risk classification in endometrial cancer was based on clinicopathological data. Preoperative assessment is used to tailor the extend of surgery. The TCGA surrogate has been recently incorporated into risk stratification by ESGO-ESTRO-ESP guidelines. The aim of this study is to evaluate if preoperative determination of molecular classification is feasible and can improve preoperative estimation of risk group.

Methodology In this retrospective cohort study, we identified all newly clinical early-stage endometrial cancer cases operated between January 2021 and April 2022. All cases had preoperative MRI and endometrial biopsy where molecular classification was done. Patients were allocated to a risk group based on 2021 ESGO-ESTRO-ESP guidelines by three Methods using only preoperative clinicopathological data, using preoperative molecular and clinicopathological data, using postoperative molecular and clinicopathological data.

Results 55 cases were included. In all cases molecular classification was done preoperative while the patient was in waiting list. In figure 1, concordance between preoperative and postoperative assessment is shown.

• Download figure
• Open in new tab
• Download powerpoint

Abstract 2022-RA-888-ESGO Figure 1

Relation between preoperative and postoperative risk assessment

When molecular classification is only taken account postoperatively, the Cohen’s kappa coefficient for the concordance is 0.37 (95% CI= 0.20–0.54) and risk was underestimated in 16/55 (29.1%) and overestimated in 8/55 (14.5%). If molecular classification is added to preoperative assessment, the Cohen’s kappa coefficient is 0.54 (95% CI= 0.38–0.70) and risk was underestimated in 12/55 (21.9%) and overestimated in 5/55 (10.0%). Overall agreement between preoperative and postoperative assessment for hystotype was 89.1%, for grade was 74.5%, and between radiologic and definitive stage was 74.5%.

Conclusion Preoperative determination of molecular classification is feasible and seems to increase the reliability of preoperative risk stratification. However, in around 22% of cases risk is still underestimated leading to inadequate surgery strategy. Sentinel lymph node biopsy can elegantly overcome this problem by providing information on the lymph node status with minimal morbidity and its implementation should be encouraged.