Evaluation of pro-carboxypeptidase A and carboxypeptidase A as serologic markers for adenocarcinoma of the pancreas

Abstract

Background: A serological marker for pancreatic cancer may allow for early detection and potentially more effective treatments. Pro-carboxypeptidase A (pro-CPA) is produced exclusively in the pancreas and converted to its active form, CPA, in the intestinal lumen. We hypothesized that alterations in serum pro-CPA and/or CPA may be useful as a diagnostic test for pancreatic cancer. Patients and methods: Serum samples obtained from 34 patients with pancreatic adenocarcinoma prior to surgical intervention and 64 control patients were assayed for pro-CPA and CPA. A variety of statistical methods was used to evaluate the utility of these measurements individually and in combination to classify the samples with respect to the presence or absence of pancreatic adenocarcinoma. Results: Because of positive skewing of the data in some populations, transformation of the data to natural logarithmic scales was used and resulted in normal distributions. All pancreatic cancer patients had ln(CPA) levels within or below the normal range defined as two standard deviations from the control group mean (-2.714±0.413). Ln(pro-CPA) levels in 24 of 34 cancer patients were outside the normal range of the control group (0.306±0.33). Pancreatic cancer patients with ln pro-CPA values within the control range had low ln CPA, advanced stage and/or evidence of pancreatic insufficiency. While each of these individual values (ln pro-CPA or ln CPA) does not adequately separate all control from cancer patients, a bivariate classification rule is presented that uses both ln pro-CPA and ln CPA simultaneously to predict the presence of pancreatic cancer with a sensitivity of 91% and a specificity of 95%. Conclusions: The data presented suggest that abnormalities in serum pro-CPA and CPA levels are associated with the presence of pancreatic cancer.