RESUMO
BACKGROUND: Numerous clinical parameters have been suggested as predictors of outcome for patients with head and neck carcinoma treated with radiation therapy, but their applicability remains controversial. Inactivation of the p53 tumor suppressor results in radioresistance in experimental systems and might predict treatment failure in human patients. We have tested this hypothesis by comparing the predictive power of nuclear accumulation of p53 protein with that of clinical and histopathologic markers in patients with glottic carcinoma treated with primary radiotherapy. METHODS: Clinical charts were reviewed for 165 patients with glottic squamous cell carcinoma treated with radiation therapy. One hundred and twenty-one patients with T1 or T2 classified tumors were determined to have received adequate treatment and to have adequate follow-up data for further study. Archival pretreatment tumor biopsies from a subpopulation of patients were examined for p53 protein by immunohistochemistry. The influence of clinical and histopathologic variables and p53 nuclear protein on tumor recurrence was studied by bivariate and multivariate analysis. RESULTS: The recurrence rate was lowest for patients with moderately to poorly differentiated tumors (P < 0.05). This was the only significant predictor of outcome in this patient population. The presence of immunohistochemically detectable p53 antigen was not predictive of tumor recurrence in 70 patients for whom there was both p53 and sufficient follow-up data. CONCLUSIONS: Histologic differentiation was prognostic for tumor recurrence in this population of patients with glottic carcinoma treated with radiation therapy. In contrast, nuclear accumulation of p53 protein was not predictive of tumor response or recurrence in this population.
