Recurrence in Oral Premalignancy: Clinicopathologic and Immunohistochemical Analysis

Abstract

Oral leukoplakia (OL) has a propensity for recurrence and malignant transformation (MT). Herein, we evaluate sociodemographic, clinical, microscopic and immunohistochemical parameters as predictive factors for OL recurrence, also comparing primary lesions (PLs) with recurrences. Thirty-three patients with OL, completely removed either by excisional biopsy or by laser ablation following incisional biopsy, were studied. Selected molecules associated with the STAT3 oncogenic pathway, including pSTAT3, Bcl-xL, survivin, cyclin D1 and Ki-67, were further analyzed. A total of 135 OL lesions, including 97 PLs and 38 recurrences, were included. Out of 97 PLs, 31 recurred at least once and none of them underwent MT, during a mean follow-up time of 48.3 months. There was no statistically significant difference among the various parameters in recurrent vs. non-recurrent PLs, although recurrence was most frequent in non-homogeneous lesions (p = 0.087) and dysplastic lesions recurred at a higher percentage compared to hyperplastic lesions (34.5% vs. 15.4%). Lower levels of Bcl-xL and survivin were identified as significant risk factors for OL recurrence. Recurrences, although smaller and more frequently homogeneous and non-dysplastic compared to their corresponding PLs, exhibited increased immunohistochemical expression of oncogenic molecules, especially pSTAT3 and Bcl-xL. Our results suggest that parameters associated with recurrence may differ from those that affect the risk of progression to malignancy and support OL management protocols favoring excision and close monitoring of all lesions.

Keywords: Bcl-xL; Ki-67; STAT3; cyclin D1; laser ablation; oral leukoplakia; oral potentially malignant disorders; predictive biomarkers; recurrence; survivin.

Figure 1

Flowchart of all 135 oral leukoplakia (OL) lesions, developing primary lesions (PLs) and recurrences (Rs), in all 33 patients.

Figure 2

Kaplan–Meier analysis of first recurrence in oral leukoplakia (OL) lesions (n = 31).

Figure 3

Representative photomicrographs of hematoxylin and eosin (H&E) and immunohistochemical stains of all studied molecules (cyclin D1, pSTAT3, Bcl-xL, survivin and Ki-67) in a non-recurrent primary oral leukoplakia lesion (PL)–1st column, in a recurrent PL–2nd column and in its corresponding recurrence–3rd column (magnification 100×, scale bar 100 μm; for inserts magnification 200×, scale bar 50 μm).

Figure 4

Comparison of mean immunohistochemical scores according to the presence and degree of dysplasia. (A) Cyclin D1 percentage, intensity and total scores in primary oral leukoplakia lesions (PLs), (B) Ki-67 total score in PLs and (C) Ki-67 total score in recurrences.

Figure 4

Comparison of mean immunohistochemical scores according to the presence and degree of dysplasia. (A) Cyclin D1 percentage, intensity and total scores in primary oral leukoplakia lesions (PLs), (B) Ki-67 total score in PLs and (C) Ki-67 total score in recurrences.

Figure 5

Comparison of immunohistochemical mean scores between recurrent and non-recurrent primary oral leukoplakia lesions (PLs). (A) Bcl-xL percentage, intensity and total scores and (B) survivin percentage, intensity and total scores.

Figure 6

Comparison of immunohistochemical mean scores of (A) pSTAT3 and (B) Bcl-xL, between primary oral leukoplakia lesions (PLs) that recurred and their corresponding first recurrences.

Figure 6

Comparison of immunohistochemical mean scores of (A) pSTAT3 and (B) Bcl-xL, between primary oral leukoplakia lesions (PLs) that recurred and their corresponding first recurrences.