Mutant p53 protein overexpression is associated with poor outcome in patients with well or moderately differentiated ovarian carcinoma

BACKGROUND: It has been shown that the p53 gene is mutated in 30-80% of ovarian carcinomas and that the genetic alterations most often manifest as an accumulation of mutant p53 protein in tumor tissue. The prognostic significance of these findings for patients with ovarian cancer, however, must be established clearly. METHODS: Mutant p53 protein in 90 consecutive epithelial ovarian carcinomas was quantitatively analyzed using a time-resolved immunofluorometric procedure. In contrast to immunohistochemical techniques, this method uses two anti-p53 antibodies. The Cox model was used to evaluate the strength of the associations between the prognostic markers and disease relapse or death at univariate and multivariate levels. Kaplan-Meier survival curves were calculated for patients who were p53-positive or negative and for subgroups with a different clinical stage, histologic grade, or residual postsurgical tumor. RESULTS: The positivity rates for p53 included 1/21 (5%) with Stage I disease, 1/6 (17%) with Stage II, 29/51 (57%) with Stage III, and 8/12 (67%) with Stage IV (total = 39/90, 43%). Patients with p53-negative tumors had a significantly longer disease free survival than did patients with p53-positive tumors (P = 0.03); these results were similar for overall survival (P = 0.06). Multivariate analysis revealed that the presence of postsurgical residual tumor was the only predictor significantly associated with poor patient outcome. However, when patients were divided into groups based on histologic grade, patients with well (G1) and moderately (G2) differentiated tumors had a significantly higher risk of cancer relapse and death if mutant p53 protein was present in their tumors compared with patients who were negative for mutant p53 protein (< 0.01). CONCLUSIONS: The immunofluorometric measurement of mutant p53 protein accumulation in epithelial ovarian carcinomas of a low histologic grade was associated significantly with an increased risk for cancer relapse and death. A similar trend also was suggested for early stage disease and in the absence of residual tumor after surgery. These increased risks, however, were not found for patients with high grade or advanced stage cancer or for those with residual tumor. To the authors' knowledge, this is the first report suggesting that p53 tumor protein accumulation is a marker of poor prognosis in a subset of patients with ovarian cancer.