The absolute tissue specificity of prostate specific antigen (PSA) allows the use of PSA test not only for detecting recurrence or metastasis at an early stage after radical prostatectomy but also for screening prostate cancer if combined with digital rectal examination. There is also a need to improve the current PSA test to better differentiate between prostate cancer and benign prostate hyperplasia (BPH). Because of these clinical applications, a much greater demand was placed on PSA test for extra sensitivity, accuracy, and precision even within the normal PSA concentration range. However, the current commercial assay kits for PSA do not provide correct PSA values. Many factors contributing to the problem include the specificity of the anti-PSA antibodies, the composition of the calibrator, the PSA values assigned to the calibrator, the PSA isoform used for anti-PSA antibody preparation, the test design, and the composition of the diluent. Most problems were derived from the failure of realizing earlier that the majority of the PSA exists in serum not as free PSA but as complexes with protease inhibitors. Other problems, such as constantly changing composition of various forms of PSA in serum specimens, and different clearance rates for various forms of PSA make almost impossible to develop an ideal assay for PSA. Therefore, we suggest that test should be designed for measuring PSA-ACT (PSA-alpha 1-antichymotrypsin) complex only. Changing the focus from the measurement of total PSA of various forms to the PSA-ACT complex alone may improve the differentiation between prostate cancer and BPH but may also simplify the selection of anti-PSA antibodies and the preparation of calibrator for the assay.