Comparison of a peanut agglutinin test and an immunochemical faecal occult blood test in detecting colorectal neoplasia in symptomatic patients.

BACKGROUND: Currently available methods for detection of early-stage colorectal cancer are reliant on faecal occult blood (FOB) tests. Bleeding, however, is not specific for colorectal neoplasia. Enzymatically detected or peanut agglutinin (PNA)-detectable galactose-beta1-3-N-acetyl-galactosamine residues found in rectal mucus have been used to detect colorectal cancer. METHODS: The sensitivity and specificity of the PNA rectal mucus test were compared with those of an immunological test for faecal occult blood (Hemolex) in 199 symptomatic patients referred for colorectal investigations. All patients also underwent a colonoscopy. SDS-PAGE and PNA-overlay were used to characterize PNA-binding proteins in normal and malignant colorectal tissue. RESULTS: The PNA test had a similar sensitivity to that of Hemolex for colorectal carcinoma (83% vs. 72%), adenomas (55% vs. 50%), inflammatory bowel disease (52% vs. 48%) and hyperplastic polyps (48% vs. 25%). The sensitivity of the PNA test and Hemolex for colorectal neoplasia was 69% vs. 59% and specificity 68% vs. 86% (p=0.002). SDS-PAGE and PNA-overlay showed some commonly expressed PNA-binding proteins in both normal mucosa and colorectal cancer and a higher and even selective expression of 160 kD PNA-binding protein in colorectal cancer. CONCLUSIONS: A single PNA test in its present form is as sensitive an indicator of colorectal neoplasia as Hemolex completed over three days, but lacks specificity. The 160 kD cancer-associated antigen we have identified is under further characterization for development of a more specific PNA test.