Cytodiagnosis of endometrial carcinoma and hyperplasia on imprint smears with additional immunocytochemistry using Ki-67 and p53 biomarkers

Objective: It is said to be difficult to interpret the different endometrial lesions by cytomorphology; however, evaluation of the microarchitecture of the cell clumps and application of immunocytochemistry can improve diagnostic accuracy. The aim of the present study was to evaluate cytolomorphological features and correlate them with the histological diagnosis of benign and malignant endometrial lesions, and to investigate certain immunocytochemical biomarkers to achieve a more accurate cytodiagnosis.

Methods: In the present study, we graded the cytomorphology on imprint smears of 35 low-grade endometrial endometrioid carcinomas compared with 23 cases of endometrium ranging from disordered proliferative to benign hyperplastic. Additionally, 10 cases of high-grade endometrial carcinoma and 11 cases of atrophic endometrium were evaluated. Ki-67 and p53 biomarkers were applied to the cytological smears.

Results: A total cytological score less than six, resulting from nuclear overlapping, nuclear/cytoplasmic ratio, the presence of a branched pattern, vesicular cytoplasm and loss of cohesiveness, distinguished all the cases of disordered proliferative and benign hyperplastic endometrium from low-grade endometrioid carcinomas of endometrium (P < 0.0001). The application of different cut-off values for Ki-67 and p53 helped differentiate certain endometrial lesions in our study. The integration of the immunocytochemical score of Ki-67 and p53 into the cytological score resulted in a final score that was also diagnostically useful.

Conclusions: The results of the present study demonstrated that evaluation of certain cytological features along with specific immunocytochemical findings could improve the accuracy of endometrial cytodiagnosis but our findings need to be tested in a routine clinical situation, using pre-operative cytological samples, to ascertain whether the diagnostic criteria are reproducible.