DETERMINING PROGNOSIS OF CLINICALLY LOCALIZED PROSTATE-CANCER BY IMMUNOHISTOCHEMICAL DETECTION OF MUTANT P53

Objectives. Mutations of the p53 tumor suppressor gene can result in u nregulated cellular growth and have been implicated in numerous malign ancies. The objective of this study was to determine whether the detec tion of mutant p53 by immunohistochemical staining is predictive of pr ogression in clinically localized adenocarcinoma of the prostate. Meth ods. Immunohistochemical staining for mutant p53 was performed on 40 f ormalin-fixed radical prostatectomy specimens. Benign glands in the se ctions sewed as controls. Immunoreactivity (IR) was categorized semi-q uantitatively from 0 to 4+ (0 = no IR, 1+ = 1% to 10%, 2+ = 1% to 40%, 3+ = 41% to 70%, 4+ = 71% to 100%). Results were then compared to Gle ason score, Stage (T2 versus T3), surgical margins, lymph node and sem inal vesicle involvement, age, race, preoperative prostate-specific an tigen (PSA), and biochemical progression. Biochemical progression was defined as a persistently elevated postoperative PSA of 0.2 ng/ml or g reater. Results. Thirty-two of the 40 tumors (80%) stained for mutant p53. None of the tumors that did not stain progressed, whereas 20 of 3 2 (62.5%) of the tumors that did stain progressed, with an overall mea n follow-up of 50.8 months. Immunoreactivity did not correlate with an y of the known prognostic variables but did have statistically signifi cant correlation with progression by all three statistical methods use d (Fisher's exact test, logistic regression, and log-rank test). Concl usions. Strict quality control and newer antigen retrieval techniques reveal p53 abnormalities in many prostate cancers. Immunohistochemical detection of mutant p53 appears to be an independent predictor of pro gression. These data suggest potential utility of p53 as a preoperativ e prognostic indicator in localized prostate cancer.