Suprabasal p53 immunoexpression is strongly associated with high grade dysplasia and risk for malignant transformation in potentially malignant oral lesions from Northern Ireland

Abstract

Aims: No good predictive marker for the malignant transformation of potentially malignant oral lesions (PMOLs) is currently available. This study re-evaluated the value of p53 immunoexpression to predict malignant transformation of PMOLs after discounting possible confounding factors.

Methods: PMOLs from 18 patients who showed progression to carcinoma, 16 of the respective carcinomas, and PMOLs from 18 matched controls were evaluated by immunohistochemistry (IHC) for p53 expression. A mouse monoclonal antibody that detects wild-type and mutant forms of human p53 was used. The p53 immunostaining pattern was also correlated with the degree of dysplasia.

Results: Suprabasal p53 staining was significantly associated with high grades of dysplasia (p < 0.01). The specificity and positive predictive value (PPV) for malignant transformation of suprabasal p53 staining were superior to the assessment of dysplasia, but sensitivity was inferior. All carcinomas derived from PMOLs with suprabasal p53 showed strong p53 immunostaining. However, the absence of suprabasal p53 staining and/or dysplastic changes did not preclude malignant transformation in a considerable proportion of PMOLs.

Conclusions: This study confirms and extends previous findings that suprabasal p53 immunoexpression has a high PPV for malignant transformation of PMOLs and can be used as a specific marker for lesions that are at high risk for malignant transformation. The absence of suprabasal p53 staining (that is, absence of, or basal, p53 staining) is non-informative for prognostic purposes. Because of its limited sensitivity, p53 IHC is not a substitute for the assessment of dysplasia in the evaluation of PMOLs. Instead, p53 IHC emerges as a clinically useful supplement of histopathological assessment in the prognosis of PMOLs.

Figure 1

Flow diagram depicting the most relevant characteristics and natural history of the potentially malignant oral lesions analysed. The p53 immunoexpression patterns of the lesions are shown in relation to their histopathological characteristics and behaviour upon follow up. Lesions that did not progress to carcinoma (controls) were selected by minimisation to match the group of lesions that progressed to carcinoma (cases) in variables such as patients' sex, age, and smoking habits, location of the lesion, and time of follow up. p53 immunostaining: −, negative; Bas, basal; SB, suprabasal. No/M Dyspl, no or mild dysplasia; Mod/Sev Dyspl, moderate or severe dysplasia; †suprabasal p53 extended into the margin of the section in five cases; p53 margin, p53 immunostaining at the margin of the section.

Figure 2

Biopsy of oral leukoplakia. Suprabasal p53 immunoexpression is seen in the centre of the histological section but is absent from its margins. This lesion did not progress to oral squamous cell carcinoma during follow up (p53 immunohistochemistry, counterstained with haematoxylin).

Figure 3

Biopsy of recurrent leukoplakia. Both the primary and recurrent lesions showed suprabasal p53 immunoexpression extending to the margin of the histological section. This lesion progressed to oral squamous cell carcinoma during follow up (p53 immunohistochemistry, counterstained with haematoxylin).

Figure 4

Biopsy of oral squamous cell carcinoma from a patient who presented previously with leukoplakia, in which suprabasal p53 staining was demonstrated, showing p53 immunoexpression in almost all tumour cells (p53 immunohistochemistry, counterstained with haematoxylin).

Figure 5

Protocol proposed for the evaluation of potentially malignant oral lesions. ICH, immunohistochemistry.