DNA content and clinicopathological features aid in distinguishing ameloblastic carcinoma from ameloblastoma

Background

Ameloblastoma and ameloblastic carcinoma are epithelial odontogenic tumors that can be morphologically similar. In the present study, we evaluated the DNA content and Ki-67 index in the two tumors.

Methods

The paraffin blocks of the tumors were selected to obtain sections for the immunohistochemical reactions and preparation of the cell suspension for acquisition in a flow cytometer. The Random Forest package of the R software was used to verify the contribution of each variable to classify lesions into ameloblastoma or ameloblastic carcinoma.

Results

Thirty-two ameloblastoma and five ameloblastic carcinoma were included in the study. In our sample, we did not find statistically significant differences in Ki-67 labeling rates. A higher fraction of cells in 2c (G1) was correlated with the diagnosis of ameloblastoma, whereas higher rates of 5c-exceeding rate (5cER) were correlated with ameloblastic carcinoma. The Random Forest model highlighted histopathological findings and parameters of DNA ploidy study as important features for distinguishing ameloblastoma from ameloblastic carcinoma.

Conclusion

Our findings suggest that the parameters of the DNA ploidy study can be ancillary tools in the classification of ameloblastoma and ameloblastic carcinoma.