Epitope analysis of a prostate-specific antigen (PSA) C-terminal-specific monoclonal antibody and new aspects for the discrepancy between equimolar and skewed PSA assays.

Background: Immunoassays to measure prostate-specific antigen (PSA) often give different values for the same patient samples, and the calibrators among commercial immunoassays are not interchangeable. We developed three novel assays to quantify the free and complexed forms of PSA in serum. Methods: We synthesized 46 peptides, which encompassed the entire PSA molecule, and determined the interactions between selected monoclonal antibodies (MAbs) and those peptides or the intact PSA molecule. Results: MAb PA313 did not cross-react with human glandular kallikrein (hK2), which has 78% amino acid homology to PSA. This MAb bound with K D = 40 nmol/L to the C-terminal peptide of PSA and distinguished between a synthetic peptide derived from PSA (PSA46A: NH2-C-R226KWIKDTIVANP237-COOH) that differed from one derived from hK2 (PSA46B: NH2-C-R226KWIKDTAANP237-COOH) by a single amino acid. Only the MAb combination of PA313/PA121 showed equimolar reactivity with PSA and with PSA complexed with α1-antichymotrypsin (PSA-ACT). The free form of PSA (F-PSA) was determined by MAbs PA313/FPA503, and the amount of complexed PSA (C-PSA) in PSA-ACT was determined by αACT/PA313. The total PSA (T-PSA) measured by either of the equimolar assays (PA313/PA121 or Tandem-R) was consistent with the sum of F-PSA and C-PSA. In contrast, T-PSA by a skewed assay (IMx) was higher than F-PSA + C-PSA when the ratio of F-PSA to T-PSA (F/T) was >0.15. T-PSA measured by IMx was nearly equal to F-PSA/0.55 + C-PSA. The coefficient 0.55 reflected different reactivities of the IMx assay with PSA-ACT and PSA. Conclusion: The discrepancy between the values measured by equimolar and skewed assays depends on the ratio of free to total PSA in the sample.