Oral Dysplasia Adjacent to Small T1 Oral Squamous Cell Carcinomas with Superficial Invasion
Sirisa Kandel, Vijay Jayaprakash, Mihai Merzianu. Roswell Park Cancer Institute, Buffalo, NY
Background: Oral squamous dysplasia (OD) grade is important for management decision making but is an imperfect predictor of malignant potential in certain lesions. In other anatomic sites (e.g., uterine cervix), grading of dysplasia relies heavily on the presence and localization of mitotic activity. We aim to assess squamous dysplasia, analyze the presence of mitoses and other histologic parameters in mucosa adjacent to small superficially invasive oral squamous cell carcinoma (OSCC).
Design: Pathology database was reviewed retrospectively for T1 OSCC with available slides from resection and excisional biopsies specimens only. Tumor width and depth were measured on the slides. Areas immediately adjacent to the invasive tumors was evaluated for dysplasia, presence and localization of mitoses and presence of erosion. Any mitotic figure irrespective of atypia were recorded as follows: present in basal/ parabasal, mid and superficial layers. Dysplasia was graded on both sides of the invasive tumor and highest grade of dysplasia was recorded using WHO criteria as mild, moderate, severe and carcinoma in-situ. A binary system of low (LGOD; mild) and high grade dysplasia (HGOD; moderate, severe OD and CIS) was also used. Fisher exact test was used for statistical analysis.
Results: Twenty-six patients, 17 men and 9 women (mean age 58, range 38-80) with T1 OSCC from tongue (n=23), palate (n=2) and gingiva (n=1) were included. Clinical presentation was a mass in 8/26, ulceration/erosion in 2/26 and leukoplakia in 9/26. Tumors width ranged from 0.1 to 1.5 cm (mean 0.6; median 0.5) and its depth from 0.05 to 0.8 cm (mean 0.18; median 0.1). Highest grade of dysplasia was mild in 9/26 (35%), moderate in 7/26 (27%) and severe OD/CIS in 10/26 (38%) patients. Mitoses were present in mid and superficial layers exclusively in HGOD (7/17) and not seen in LGOD (0/9) (p-value 0.03). However, mitoses were absent all together in 5/9 LGOD and 5/17 HGOD. No significant relationship between presence of mitoses, tumor depth or width and erosion in the dysplastic mucosa was present.
Conclusions: Mucosa immediately adjacent to these small OSCC were associated with mild dysplasia in one third and moderate dysplasia in one fourth of cases and mid/superficial layers mitoses were absent in most cases. These findings provide indirect evidence that criteria for grading dysplasia used by surgical pathologists in other epithelia will underestimate the malignant potential of a significant subset of oral dysplastic lesions.
Category: Head & Neck
Wednesday, March 6, 2013 9:30 AM
Poster Session V # 200, Wednesday Morning