Abstract
BACKGROUND: Carbohydrate antigen 72-4 (CA72-4) is a tumor marker for gastric cancer however its role in esophageal cancer (EC) is still controversial. The aim of this study is to determine the prognostic value of CA72-4 in patients with esophageal squamous cell carcinoma (ESCC).
METHODS: From January 2006 to December 2007 we conducted a retrospective analysis of 192 consecutive patients with ESCC. A receiver operating characteristic (ROC) curve for survival prediction was plotted to verify the optimum cut-off point for CA72-4. Univariate and multivariate analyses were performed to evaluate the prognostic parameters.
RESULTS: The positive rate for CA 72-4 in our study was 18.8% (36/192). The ROC curve for survival prediction showed the optimum cut-off point for CA 72-4 to be 3.95 U/mL. Patients with CA 72-4 ≤3.95 U/mL had a significantly better five-year overall survival (51.4% vs. 13.6%; P<0.001) and relapse-free survival (49.5% vs. 19.8%; P < 0.001) than those with CA 72-4 levels >3.95 U/mL. Multivariate analyses showed that CA 72-4 was a significant predictor of both overall survival and relapse-free survival. CA 72-4 levels >3.95 U/mL had a hazard ratio (HR) of 2.129 [95% confidence interval (CI): 1.436-3.155; P < 0.001] for overall survival and 2.151 (95% CI: 1.449-3.192; P < 0.001) for relapse-free survival.
CONCLUSIONS: CA 72-4 is an independent predictive factor for long-term survival in ESCC. We conclude that 3.95 U/mL may be the optimum cut-off point for CA72-4 in predicting survival in ESCC. Although CA 72-4 shows significant association with poorer prognosis, its low sensitivity limits clinical application.