Elevated preoperative levels of CA 19-9 and CA 125 predicts overall survival time in the pancreatic adenocarcinoma. Single institution series.

Abstract

Background: Pancreatic cancer is a devastating disease, being the fourth cause of cancer-related death worldwide. Several studies have investigated the use of multiple cancer biomarkers, such as C 19-9, CA 125, and CEA, as prognostic factors for overall survival in pancreatic cancer. CA 125 seems to have superior predictive utility in selected groups of PDAC patients. Material and methods: We retrospectively analyzed data collected from 129 patients admitted to our Department due to diagnosis with pancreatic cancer. Prior to the survival analysis, the preliminary assessment of pre-treatment levels of biomarkers was carried out. The overall survival time was defined as that elapsing from the admission date to the date of death. Results: The patients mean age was 62 +/- 9.5 years, while the median overall survival (OS) was 7mo 12d. As for tumor localization, most of the patients had PDAC within the head of the pancreas (n=93), followed by PDAC of the pancreatic body (n=15), pancreatic tail (n=14) and both pancreatic body and tail (n=7). Ninety-five patients had an unresectable tumor and 34 were diagnosed with a resectable tumor (tab.1). The statistically significant correlation was found for CA 125 (ρ=-0.355 p<0.001) and CA 19-9 (ρ=-0.225 p=0.012). We chose the following cut-off points: CA 125>=20 IU/mL was considered as high, and CA 19-9>=200 IU/mL as significantly elevated. In the univariate analysis in the Kaplan-Meier survival model, adjusted for age, both elevated biomarkers were statistically significant prognostic factors of OS (CA 125<20 median OS- 10mo 3d vs. CA 125>=20- 4mo 17d p=0.001) and (CA19-9<200 median OS- 8mo 3d vs. CA 19-9>=200- 4mo 20d p=0.001). Patients’ gender and, PDAC resectability and its localization were not statistically significant prognostic factors (log rank test p=0.8; p=0.108 and p=0.578 respectively). In the age-adjusted multivariate analysis, both biomarkers remained significant- CA 125>=20 (HR: 1.73 95%CI 1.27-2.58 p=0.006) CA 19-9>=200 (HR: 1.78 95%CI 1.19-2.66 p=0.005) Conclusions: Our study proves the utility of the pretreatment assessment of CA 125 because its level is tightly correlated with OS. It may be hypothesized that the pretreatment measurement of both CA 19-9 and CA 125 can provide the valuable information about patients’ prognosis.