P53 SEQUENCE-ANALYSIS PREDICTS TREATMENT RESPONSE AND OUTCOME OF PATIENTS WITH ESOPHAGEAL-CARCINOMA

BACKGROUND. The ability to predict biologic behavior and treatment res ponsiveness would be a valuable asset in the multimodality approach to esophageal carcinoma. The authors examined whether alterations of the p53 gene correlate with clinicopathologic parameters, response to pre operative chemotherapy/radiotherapy, and outcome in patients with esop hageal carcinoma. METHODS. Histopathologic/genetic analysis of p53 was performed on formalin fixed, paraffin embedded tissues. Tissue sectio ns were stained immunohistochemically for p53 protein followed by topo graphic genotyping comprised of polymerase chain reaction amplificatio n and direct sequencing of p53 exons 5-8. All patients received induct ion chemotherapy (5-fluorouracil, cisplatin, and cu-interferon) and co ncurrent external beam radiotherapy (4500 centigrays) followed by rese ction. RESULTS, p53 analysis performed on 42 tumors from patients with potentially resectable esophageal carcinoma revealed 25 of the 42 tum ors (59.5%) to be p53 immunopositive; however, only 17 of the 42 tumor s (40.5%) were proven to contain p53 point mutational damage in exons 8 (n = 5), 5 (n = 5), 7 (n = 4), and 6 (n = 3). Eight cases were weakl y immunopositive and had no genotype mutation suggesting hyperexpressi on of normal wild-type p53. Genotyping also identified two immunonegat ive cases with deletion-type mutations (exons 5 and 6). Tissue samples collected before and after chemotherapy/radiotherapy exhibited fideli ty in p53 mutational genotype in all cases. The presence of a p53 poin t mutation positively correlated with pTNM stage (P = 0.003) and resid ual disease in the resected specimen (P = 0.01). Moreover, survival of patients with p53 mutations was significantly lower than that of pati ents without mutations (overall survival of 21.6 months vs. 40 months; P = 0.0038; and disease free survival of 14.1 months vs. 38 months; P = 0.0004). CONCLUSIONS. Histopathologic/genetic analysis is a better determinant of p53 mutational damage than immunohistochemistry alone a nd can be used as a prognostic marker for esophageal carcinoma. p53 ge notyping may define a subset of patients who respond to chemotherapy/r adiotherapy and may predict who potentially benefits from multimodalit y therapy. (C) 1998 American Cancer Society.