Classification of high-grade endometrium carcinomas using molecular and immunohistochemical methods.

OBJECTIVES

As a result of the integration of molecular changes into the histological classification of cancers, which increases diagnostic repeatability, the differences between the groups become more prominent and targeted therapies gain significance. The most comprehensive molecular study regarding endometrial carcinomas (EC) is The Cancer Genome Atlas (TCGA) project. According to TCGA, endometrial carcinomas are classified into four molecular prognostic subgroups: copy-number-low/p53-wild-type (p53wt), DNA polymerase epsilon (POLE)-mutated/ultramutated (POLEmt), microsatellite-instability/hypermutated (MSI), and copy-number-high/p53-mutated (p53mt). In this study, we aim to apply the molecular classification to our high-grade endometrial cancer patients, and particularly, to identify our overtreated patients.

MATERIAL AND

METHODS

Ninety-seven patients diagnosed with high-grade EC in Selcuk University, Faculty of Medicine between 2009-2018 were retrospectively evaluated and classified into four subgroups. Primary outcomes of overall and progression-free survival were evaluated for clinical, pathological, and molecular features. Further, all molecular groups were divided into endometroid and non-endometrioid groups, and disease-free survival (DFS) and overall survival (OS) were investigated across groups.

RESULTS

According to molecular classification, 23 patients (23.7%) were assigned to the MSI group, 21 (21.6%) to the POLEmt group, 40 (41.2%) to the p53mt group, and 13 (13.4%) to the p53wt group. Patients' DFS (p = 0.001) and OS rates (p = 0.001) were significantly different according to their molecular classification. The results of our analyses determined that, in the molecular classification of high-grade ECs, the p53mt group had the poorest prognosis and the POLEmt group had the best prognosis. Tumor size, myometrial invasion, lymphovascular space invasion (LVSI), lymph node metastasis, cervical invasion, ovarian invasion and stage showed statistically significant differences based on molecular classification (p < 0.05).

CONCLUSIONS

The use of molecular classification in the clinical practice will allow more accurate prognostic prediction and more appropriate treatment planning, particularly as high-grade ECs constitute a heterogenous group with poor prognosis.