Prognostic Influence of p5 3 Nuclear Overexpression in Colorectal Carcinoma M. Galindo Gallego, M.D.,* M. J. Fern:indez Acefiero, M.D.,t J. Sanz Ortega, M.D.,~: A. Aljama Delgado,t J. L. Balibrea Cantero, M.D.~ From the *Hospital de la Fuenfda, Cercedilla, and Departments of t Surgical Pathology, Hospital de M6stoles, and ~Surgical Pathology and ~Surgery, Hospital Clfnico de Madrid, Madrid, Spain PURPOSE: The aim of this study was to test the prognostic influence of p53 nuclear overexpression in colorectal car- cinoma. METHODS: We performed an analysis of the prog- nostic influence of the nuclear overexpression of p53 with irmnunohistochemistry in 126 cases of colorectal carcinoma operated on in our hospital between 1987 and 1992, with a minimum follow-up time of 60 months (5 years). RESULTS: Our results show a statistically significant prognostic influ- ence of p53 overexpression on disease-free survival time, but not on the overall survival time, in univariate analysis, but this influence is lost in multivariate analysis. CONCLU- SIONS: Our results confirm recent reports by other authors that failed to show the independent prognostic value of p53 in colorectal carcinoma. [Key words: Colorectal carcinoma; p53; Prognosticators] Galindo Gallego M, Fernandez Acefiero MJ, Sanz Ortega J, Aljama Delgado A, Balibrea Cantero JL. Prognostic influence of p53 nuclear overexpression in colorectal carcinoma. Dis Colon Rectum 2000;43:971-975. C olorectal carcinoma accounts for 15 percent of all malignancies in the United States, and it is, excluding the cutaneous malignancies, the second- most common tumor after lung carcinoma in males and breast carcinoma in females and the second- leading cause of death from cancer in both genders in the United States.* The incidence of this neoplasm ranges from 3.4 cases per 100,000 per year in Nigeria to 35.8 cases per 100,000 per year in Connecticut. 2 Despite multiple studies tumor stage remains the best prognosticator for colorectal cancer. Tumor stage considers depth of the infiltration and presence of regional lymph node metastasis, both factors with maximal prognostic influence. Tumor stage is the only criterion used for the selection of patients with a poor prognosis to receive adjuvant chemoradiation therapy. Nevertheless, it is a well-known fact that overall survival rates after curative resection for colo- rectal carcinoma ranges from 40 to 60 percent in five years in the various studies. 3 This fact makes it impor- tant to find new tumor markers that can define subsets No reprints are available. 971 of patients with poor prognosis who should receive more aggressive therapies to improve survival. The p53 gene is a suppressor gene that undergoes mutations or deletions in a high percentage of pa- tients with malignancies, including colorectal carci- noma. This gene is located on 17p13.1 and encodes a nuclear phosphoprotein of 53 kd composed of 393 amino acids, which forms tetramers and binds to the T transforming protein of SV40 virus (antigen SV40 T). 4 The functional alteration of the p53 gene is a bio- logically important fact in the genetic alterations im- plicated in colorectal carcinogenesis. These abnor- malities range from complete deletions to point mutations and are one of the most frequent genetic abnormalities found in human tumors. >7 This group of changes leads to the production of a mutated p53 protein with an abnormal structure and a longer half- life inside the nuclei of cells, which allows detection by immunohistochemistry. Therefore, p53 nuclear overexpression can be considered an indicator of mutations and functional alteration of the p53 gene. Immunohistochemical detection of intranuclear p53 has been shown as a negative prognosticator in breast, s lung,9, 1° renal, 11 bladder, 12 endometrial, 13 ovarian, ~4 and gastric carcinomas ~5 and also in soft tissue sarcomas. ~6 In breast cancer it has even been used to define a subset of high-risk patients who need adjuvant therapy. In advanced colorectal carcinoma p53 mutations and nuclear overexpression have been shown to have an independent prognostic influence; however, fur- ther studies have failed to show this prognostic influ- ence in all patients but have defined subsets of pa- tients in whom this factor can be useful. In this report we analyze the prognostic influence of this factor and its relationship to other factors with proven prognos- tic influence in colorectal carcinoma.