Purpose: We assess whether the complex between prostate specific antigen (PSA) and alpha1-protease inhibitor in serum can be used to reduce further the number of false-positive PSA screen results independent of total and free PSA.
Materials and methods: Sera from 304 consecutive screen positive subjects, including 78 with and 226 without prostate cancer, and serum PSA of 4 to 10 microg./l. or higher in the Finnish, randomized, population based prostate cancer screening trial were analyzed for PSA-alpha-protease inhibitor, and total and free PSA. Main outcome measures were specificity, sensitivity and area under receiver operating characteristics curve for proportions of free PSA and PSA-alpha 1-protease inhibitor, and for a combination of these among screen positive cases.
Results: The proportion of serum PSA-alpha 1-protease inhibitor of total PSA was lower in cancer cases than in controls (0.9% versus 1.6%, p <0.001). Logistic regression analysis of total PSA, free PSA and PSA-alpha 1-protease inhibitor showed that PSA-alpha 1-protease inhibitor in serum was an independent variable for discrimination between subjects with and without prostate cancer (p = 0.006) in the PSA range of 4 to 10 microg./l. The proportion of PSA-alpha 1-protease inhibitor alone improved specificity less than the proportion of free PSA but when these were combined by logistic regression they performed better than the proportion of free PSA alone at sensitivities of 85% to 95% (p <0.001).
Conclusions: Serum PSA-alpha 1-protease inhibitor improves the specificity of total and free PSA in a screening population with total PSA 4 to 10 microg./l.