Background: Pancreatitis-associated protein (PAP) is known as a marker for pancreatitis and cystic fibrosis. The aim of our study was to evaluate PAP in patients with pancreatic cancer, to assess its correlation to the extent of the disease and to compare it to CA19-9.
Patients and methods: This prospective study comprised 75 individuals. Thirty had pancreatic cancer, 30 were healthy controls and 15 had benign lesions of the pancreas. PAP was determined by enzyme-linked immunosorbent assay. Statistical analysis was by Wilcoxon test and Spearman correlation coefficients.
Results: As compared to healthy individuals and using a cut-off of 18 micrograms/l corresponding to a sensitivity of 90%, the specificity of PAP for pancreatic cancer was 82.8%. PAP elevation in cancer patients could not be explained by concomitant pancreatitis (p = 0.649). PAP did not show correlation to tumour size (p = 0.14), T-stages (p = 0.706) or tumour grading (p = 0.105), but was significantly correlated to the overall extent of the disease according to the UICC stages (p = 0.002). No correlation between PAP and CA19-9 was seen. Jaundice was not found to influence PAP values (p = 0.4).
Conclusion: Elevation of PAP in patients with pancreatic cancer is not merely explainable by concomitant pancreatitis, but seems to be due to increased PAP production by the cancer cells and is also correlated to tumour load as expressed by the UICC stages.