Keratoacanthoma of the gingiva – A pathological conundrum

Keratoacanthoma (KA) is an epithelial neoplasm which occurs on sun-exposed hair-bearing skin. The term was suggested by Rook and Whimster1 to distinguish these lesions from squamous cell carcinomas. Cutaneous KAs may be solitary or multiple and present as keratotic, painless nodules that rapidly develop over 4–6 weeks and resolve spontaneously after a few months, with scarring. The cell of origin is assumed to be within the pilo-sebaceous unit.2 Solitary intraoral KAs in children are particularly rare and controversial lesions, with only 4 cases previously reported in the English literature.3, 4, 5, 6 We present a further case of solitary intraoral KA of the maxillary gingiva in a child.

A seven year old Caucasian girl was referred by her dental practitioner for assessment of a gingival swelling in the region of tooth 22, which had not resolved following attempted incision and drainage. The lesion had rapidly increased in size over one month and was painful on contact with cold. Clinical examination showed a 20mm×10mm firm, sessile swelling on the attached gingiva of the partially erupted 22. No other lesions were visible, and radiographic examination revealed nil of note. Excisional biopsy showed a largely endophytic squamous epithelial lesion, with no significant cellular atypia, centered round a keratin plug (Fig. 1). From this keratin filled cavity nests with variable central keratinization radiated into the lamina propria. The lesion was well circumscribed, but undermined the adjacent epithelium and extended to the deep margin. An associated inflammatory infiltrate was present. On further review a small recurrence was noted which was re-excised and the surrounding bone and tooth 22 root surface curetted. The patient was followed up for another two years with no further recurrence and was discharged to the care of her dental practitioner.

View Large Image Download to PowerPoint

Figure 1. Cross-section of the lesion on gingiva related to tooth 22, demonstrating the lesion proliferating round a central cavity. H&E, Overall×20 magnification.

Simple molecular analysis was not helpful in further characterization of this lesion. Ploidy analysis (DNA image cytometry, Prof E Odell, Department of Oral Pathology, Guy’s and St Thomas’ Dental Institute) demonstrated that both lesions were diploid. Expression of the cell cycle marker MiB-1 was very similar to that of the overlying gingival epithelium whilst that of p53 protein was elevated in cells in the basal compartment of the lesion. In situ hybridization for Human Papilloma Virus and Epstein Barr Virus did not demonstrate any evidence of infection.

The architecture of this lesion resembles cutaneous keratoacanthoma. However, given the proposed cell of origin of cutaneous KA, this diagnosis within the oral cavity is problematic, and the differential diagnosis includes a low grade squamous cell carcinoma and carcinoma cuniculatum. Others have suggested that it is more likely that such lesions are hamartomatous proliferations of odontogenic epithelium perhaps from remnants of dental lamina4. The infrequency of occurrence, unclear natural history and differing age profile of intraoral lesions in comparison with cutaneous KAs may indicate that intraoral lesions have a different origin, aetiology and clinical course from cutaneous KAs, warranting further investigation.