MULTIVARIATE-ANALYSIS OF THE PRONOSTIC FA CTORS OF SURVIVAL AND THE RESPONSE TO TREATMENT OF SQUAMOUS-CELL ESOPHAGEAL CANCER BY CONCOMITANTRADIOCHEMOTHERAPY - VALUE OF P53 IMMUNODETECTION

Citation
Jf. Seitz et al., MULTIVARIATE-ANALYSIS OF THE PRONOSTIC FA CTORS OF SURVIVAL AND THE RESPONSE TO TREATMENT OF SQUAMOUS-CELL ESOPHAGEAL CANCER BY CONCOMITANTRADIOCHEMOTHERAPY - VALUE OF P53 IMMUNODETECTION, Gastroenterologie clinique et biologique, 19(5), 1995, pp. 465-474
Citations number
65
Categorie Soggetti
Gastroenterology & Hepatology
ISSN journal
03998320
Volume
19
Issue
5
Year of publication
1995
Pages
465 - 474
Database
ISI
SICI code
0399-8320(1995)19:5<465:MOTPFC>2.0.ZU;2-C
Abstract
Objective. - The purpose of this study was to determine the prognostic value of the expression of protein p53, EGF receptors (EGR-R), cell p roliferation antigen (Ki67) and DNA analysis by flow cytometry on per- endoscopic biopsies as well as the ability of these factors to predict response to concomitant chemoradiation in patients with squamous cell oesophageal carcinoma. Methods. - Sixty-two patients with squamous ce ll oesophageal carcinoma were prospectively included in this study. Fo r 58 patients (51 men, 7 women; mean age: 59.1 +/- 9.3 years), clinica l response to chemoradiation was correlated with the findings of flow cytometry (ploidy, % of cells in S-phase) and immunohistochemistry (p5 3, EGFR, Ki67). There were 4 patients in stage I, 14 in stage II, 27 i n stage III, and 13 in stage IV. Chemoradiation (2 cycles associating continuous 5FU 800 mg/m(2)/24 h from D1 to 5 and from D22 to 26, Cispl atyl 70 mg/m(2) on DI and D22; 15 Gy/5d from DI to 5 and from D22 to 2 6), was performed prior to surgery in 19 patients (group I) and as the only treatment in 39 patients (group II), with a third cycle from D43 . Clinical response was defined as complete or incomplete, ascertained by endoscopy and biopsy, 2 to 3 weeks after the end of chemoradiation . Results. - Mean survival in all 58 patients was 13.0 months. Surviva l was significantly longer in responders than in non responders (14.7 vs 9.6 months; P = 0.03). Among MO patients, survival was not differen t in case of exclusive chemoradiation therapy or chemoradiation therap y followed by surgical excision (17.6 vs 13.0 months; NS). Monofactori al analysis showed that, in addition to response, the variables relate d to survival were stage, non-metastatic status, and absence of p53 su rexpression. After multifactorial analysis according to the Cox model, the remaining variables were non-metastatic status, and absence of p5 3 surexpression. A complete response with negative biopsies was observ ed in 39 out of 58 patients, i.e. 67.3 +/- 12.1 % (group I: 12 out of 19; group II: 27 out of 39; NS). According to monofactorial analysis, 3 factors were predictive of complete response, i.e. non surexpression of p53 (P < 0.05) and tumour diameter (P = 0.04). After step-by-step logistic regression, non surexpression of p53 and tumour diameter cont inued to be predictive. The relative risk of a noncomplete response wa s 5.46 if p53 was detected and 1.84 for each cm of added tumour diamet er. These two factors were independent. Conclusions. - In this study t he predictors of complete response were absence of p53 surexpression a nd tumour diameter ascertained by CT-scan. Flow cytometry and Ki67 ant igen had no prognostic value and were not predictors of response.