Serum levels of prostatic acid phosphatase (PAP), γ-seminoprotein (γ-Sm) and prostate-specific antigen (PSA) were determined simultaneously in 57 patients with benign prostatic hyperplasia (BPH) and in 50 untreated patients with prostatic cancer (adenocarcinoma, N=47 and non-adenocarcinoma, N=3). The correlations between the serum levels of γ-Sm and PSA in these patients were assessed by linear regression analysis. Some fundamental studies were added for explaining the causes of discrepancy between the serum levels of γ-Sm and PSA.
All of BPH group underwent transurethral resection of the prostate (TURP) and the sera were obtained for measurements before, immediately after and 18 hours after TURP. The γ-Sm correlated well with the PSA in the sera obtained before (r=0.76) and 18 hours after (r=0.73) TURP. However, there was no correlation (r=0.26) between them in the sera obtained immediately after TURP.
In 47 untreated patients with adenocarcinoma of the prostate, no significant correlation (r=0.19) between serum levels of γ-Sm and PSA was observed, although there was correlation (r=0.51) between those of PAP and PSA. When these patients were classified into two groups, M
0 (stage A-C; N=26) and M
1 (stage D; N=21), however, the serum γ-Sm correlated with the serum PSA in M
0 group (r=0.57), but didn't in M
1 group (r=0.11). Furthermore, the differences in the means of PAP (p<0.05) and PSA (p<0.001) between M
0 group and M
1 group were statistically significant, although the serum γ-Sm failed to distinguish M
0 from M
1.
The anti-PSA antibody of “PSA Kit” reacted against the standard γ-Sm adopted from “γ-Sm Kit”. Surprisingly, the anti-γ-Sm antibody of “γ-Sm Kit” also reacted against the standard PSA adopted from “PSA Kit”. The γ-Sm and PSA apparently cross-reacted each other. The quantitative analyses with serial dilution of the sera were done by using each assay in 3 patients whose serum levels of γ-Sm were markedly different from those of PSA. The dilution curve for PAP appeared to be rectilineal, and that for PSA also appeared to be approximately rectilineal. However, the γ-Sm assay failed to be proportional.
In conclusion, the correlation between serum levels of γ-Sm and PSA was absent in certain circumstances, when the true values of them were expected to be much higher than those determined. At present, the role of available assay for serum γ-Sm appears to be limited for determining the true value.
View full abstract