G. Charriere et al., IMMUNORADIOMETRIC ASSAYS OF PROSTATE-SPECIFIC ANTIGEN AND PROSTATIC ACID-PHOSPHATASE - ANALYTICAL AND CLINICAL-EVALUATION IN 702 SUBJECTS, Journal of clinical ligand assay, 18(2), 1995, pp. 107-111
We evaluated analytical and clinical performances of Hybritech monoclo
nal immunoradiometric assays for prostate specific antigen (PSA) and p
rostatic acid phosphatase (PAP). Two-hundred eighteen patients with pr
ostate cancer and 263 with benign diseases were studied and compared w
ith 221 healthy men, Assay coefficients of variation were below 5% and
the minimum detectable mass concentration was 0.2 mu g/L for PSA and
0.05 mu g/L for PAP. Dilution linearity was good but a high dose hook
was observed for PSA concentrations over 7000 mu g/L. Results were com
pared with those obtained with radioimmunoassays using polyclonal anti
bodies. For PSA, the correlation was good (r(2) = 0.998), but Hybritec
h results were consistently lower by a factor of 1.6; for PAP, the cor
relation was poor (r(2) = 0.773) in low values (< 10 mu g/L). The uppe
r limit of reference values (95th percentile) in 221 healthy men was f
ound to be 1.9 mu g/L for PSA and 2.6 mu g/L for PAP. The diagnostic s
ensitivity of the PSA assay in 218 patients with prostate cancer was e
xcellent, but the specificity in benign prostatic disease was rather l
ow. A higher specificity was observed with the PAP assay but sensitivi
ty was good only in stage D cancers.