Purpose: To evaluate a new immunoassay for identification and quantitation of MUC1 in the sera of patients with pancreatic cancer or pancreatitis. The sensitivity and specificity of the assay are examined and compared to results from a CA19-9 immunoassay.
Methods: An in vitro enzyme immunoassay was established with monoclonal antibody PAM4 as the capture reagent, and a polyclonal anti-MUC1 antibody as the probe. Patient sera were obtained from healthy, adult patients with acute and chronic pancreatitis, and those with pancreatic and other forms of cancer, and were measured for PAM4-reactive MUC1.
Results: At a cutoff of 10.2 units/mL, 41 (77%) of 53 pancreatic cancer patients, none of the healthy individuals (n = 43), and only four (5%) of 87 patients with pancreatitis were positive above this value. Among nonpancreatic cancers investigated, colorectal cancers gave the highest percentage of positives (14%; five of 36). Overall, the sensitivity and specificity of the immunoassay for pancreatic cancer were 77% and 95%, respectively. Receiver operator characteristic analyses for discrimination of pancreatic cancer from pancreatitis provided an area under the curve of 0.89 (95% CI, 0.82 to 0.93), with a specificity of 95.4% and a positive likelihood ratio of 16.8. A direct pair-wise comparison of PAM4 and CA19-9 immunoassays for discrimination of pancreatic cancer and pancreatitis resulted in a significant difference (P < .003), with the PAM4 immunoassay demonstrating superior sensitivity and specificity. CONCLUSION The high sensitivity and specificity observed suggest that the PAM4-based immunoassay of circulating MUC1 may be useful in the diagnosis of pancreatic cancer.