Mixed endometrial carcinoma: is it time to profile the two components separately?

Objective: Advances in molecular profiling have significantly altered the approach to endometrial cancer (EC). In clinical practice, the assessment of mismatch repair (MMR) proteins and p53 status, combined with the detection of pathogenic POLE mutations, currently categorizes EC into four molecular subgroups with prognostic implications, particularly in early-stage disease: POLE-mutated, MMR-deficient (MMRd), p53-abnormal (abn), and no specific molecular profile (NSMP). However, the current approach is to assess mixed endometrial carcinomas (MEEC) as a single entity without specific molecular evaluation of individual histological components. The present study was designed as a hypothesis-generating analysis to explore this heterogeneity. Methods: The present analysis was conceived to evaluate whether profiling histological components of MEEC separately could provide additional prognostic information. MEEC with an endometrioid and a serous or clear cell component underwent immunohistochemical analysis of p53 and MMR proteins and POLE sequencing on the undissociated part and then on separate components. Results: Eight MEEC were included. Six cases of endometrioid endometrial carcinoma (EEC) and clear cell carcinoma (CCC) showed that the same mutations were detected in the undissociated tumor and in separate components. Two cases consisted of EEC with serous carcinoma (SC). Both had pathogenic POLE mutations, normal p53 expression, and pMMR status and, therefore, were potentially at low risk. Further analysis revealed differences in the histological components. In particular, in one case (case 8), the serous component was p53-abn and POLE-mutated, whereas the endometrioid component (55% of the tumor and high-grade) was POLE wild-type, representing a potential intermediate-high risk profile. It must be noted that no clinical follow-up data are available for this specific case to confirm whether this finding would have definitively altered the clinical outcome. Conclusion: Despite the retrospective nature and limited number of cases, a discrepancy was identified in a case of MEEC with a serous component when compared to molecular analysis of the tumor as a single entity, as per current guidelines. While our findings necessitate evaluation of the current molecular profiling method, the number of tumors analyzed was very restricted, and our observations pertain solely to a single sample. Therefore, these findings ought to be interpreted judiciously and are merely for the purpose of generating hypotheses.