Related Experiment Videos
Clinicopathologic spectrum and developmental reactivation in adenomatoid odontogenic tumour: A 35-year retrospective
Kochli Channappa Niranjan1, Nitya Krishnasamy2, Shriya Gaonkar1
1Department of Oral and Maxillofacial Pathology and Oral Microbiology, SDM College of Dental Sciences and Hospital, Shri Dharmasthala Manjunatheshwara University, Dharwad- 580 009, Karnataka, India.
Background:
Adenomatoid odontogenic tumour (AOT) is an uncommon benign epithelial odontogenic tumour characterised by distinctive clinicopathological and histopathological features. This study evaluated the clinical, radiographic, and histopathological spectrum of AOT, compared intra-follicular (IF-AOT) and extra-follicular (EF-AOT) variants, and assessed the potential utility of machine learning for variant classification.
Materials And Methods:
A retrospective analysis of 29 histopathologically confirmed intraosseous AOTs diagnosed over a 35-year period (1989-2024) was performed. Demographic, clinical, radiographic, histopathological, treatment, and follow-up data were evaluated. Cases were classified as IF-AOT or EF-AOT based on their relationship with impacted teeth. Associations between variant type and clinical characteristics were assessed using Fisher's exact test. A decision-tree model incorporating clinical variables was used to classify AOT variants.
Results:
The mean age at diagnosis was 19.07 years, with a female-to-male ratio of 3:1. The maxilla was affected in 58.6% of cases, with a predominance of lesions in the anterior region. IF-AOT accounted for 69.0% of cases and EF-AOT for 31.0%. Histopathological examination demonstrated duct-like structures, whorled epithelial nests, sheet-like and rosette-like arrangements, together with less common findings, including CEOT-like calcifications, stellate reticulum-like cells, squamous metaplasia, and dentigerous cyst-associated changes. No statistically significant associations were identified between AOT variant and the clinical features evaluated (p > 0.05). The decision-tree model achieved an overall classification accuracy of 88.9%. Postoperative follow-up ranged from 6 months to 2 years, with no recurrences documented during the follow-up period.
Conclusion:
AOT demonstrates considerable clinicopathological and histopathological diversity while retaining its characteristically favourable biological behaviour. Integration of clinical, radiographic, and histopathological findings facilitates accurate recognition of its variants and uncommon morphological presentations. Machine learning may provide an adjunctive approach to variant classification; however, validation in larger datasets is required.