BACKGROUND: Although CA 19-9 is the primary marker used in the diagnosis and treatment of pancreatic cancer, other serum tumor markers have also been utilized in the follow-up of pancreatic cancer. We investigated the clinical utility of CYFRA 21-1, AFP, CEA, CA 19-9, CA 125, NSE, and combinations of these markers in patients with pancreatic cancer. METHODS: We enrolled patients with primary pancreatic cancer and benign pancreatic cystic disease (n = 163). We performed sensitivity tests for multiple tumor markers, plotted receiver operating characteristic curves, and conducted multivariate analysis using the Cox proportional hazard method. Survival data were evaluated using Kaplan-Meier analysis of overall survival. RESULTS: Among multiple tumor markers assessed in this study, CA 19-9 showed good diagnostic performance, with an area under the curve of 0.86 ± 0.04 in ROC analysis. Based on two different cutoff values, CYFRA 21-1 (≥ 2.0 and 1.83 ng/mL) had a respective sensitivity of 80.4% and 82.3% and was also more significant than the other tumor markers in a parallel test. There was a weak significant relationship between tumoral fluorodeoxyglucose uptake and CYFRA 21-1 or CA 19-9. Initial CA 125, CYFRA 21-1, and CEA could be utilized to categorize subgroups with different overall survival. In multivariate analyses, CA 125 (HR 18.8, p < 0.001) and CYFRA 21-1 levels (HR 0.962, p = 0.006) demonstrated independent prognostic significance for predicting overall survival. CONCLUSIONS: In addition to CA 19-9, the present study suggested that various tumor markers could be used in the diagnosis and prognosis of pancreatic cancer. Further studies are warranted to confirm the clinical usefulness of diverse biological markers in pancreatic cancer.
BACKGROUND: Although CA 19-9 is the primary marker used in the diagnosis and treatment of pancreatic cancer, other serum tumor markers have also been utilized in the follow-up of pancreatic cancer. We investigated the clinical utility of CYFRA 21-1, AFP, CEA, CA 19-9, CA 125, NSE, and combinations of these markers in patients with pancreatic cancer. METHODS: We enrolled patients with primary pancreatic cancer and benign pancreatic cystic disease (n = 163). We performed sensitivity tests for multiple tumor markers, plotted receiver operating characteristic curves, and conducted multivariate analysis using the Cox proportional hazard method. Survival data were evaluated using Kaplan-Meier analysis of overall survival. RESULTS: Among multiple tumor markers assessed in this study, CA 19-9 showed good diagnostic performance, with an area under the curve of 0.86 ± 0.04 in ROC analysis. Based on two different cutoff values, CYFRA 21-1 (≥ 2.0 and 1.83 ng/mL) had a respective sensitivity of 80.4% and 82.3% and was also more significant than the other tumor markers in a parallel test. There was a weak significant relationship between tumoral fluorodeoxyglucose uptake and CYFRA 21-1 or CA 19-9. Initial CA 125, CYFRA 21-1, and CEA could be utilized to categorize subgroups with different overall survival. In multivariate analyses, CA 125 (HR 18.8, p < 0.001) and CYFRA 21-1 levels (HR 0.962, p = 0.006) demonstrated independent prognostic significance for predicting overall survival. CONCLUSIONS: In addition to CA 19-9, the present study suggested that various tumor markers could be used in the diagnosis and prognosis of pancreatic cancer. Further studies are warranted to confirm the clinical usefulness of diverse biological markers in pancreatic cancer.