Objectives: We evaluated the clinical applicability of serum concentration techniques to enhance the detection of prostate-specific antigen (PSA) in men with recurrent prostate cancer after radical prostatectomy.
Methods: We concentrated blood serum by lyophilization and ultrafiltration from female patients who had undergone cystoprostatectomy, "cured" patients who had undergone radical prostatectomy, patients without prostate cancer, and patients with prostate cancer treated with radiation or hormonal therapy. The primary study group consisted of 31 patients with recurrent disease after radical prostatectomy whose initial postoperative PSA fell to undetectable levels (less than 0.07 ng/mL) that later turned positive (0.07 ng/mL or more) by the Tosoh AIA 600 immunoassay run in the ultrasensitive mode. All serum samples of less than 0.07 ng/mL were concentrated by lyophilization or ultrafiltration.
Results: Serum concentrated by lyophilization and filtration detected PSA recurrence significantly earlier than did unconcentrated serum in 29 of 31 patients (94%) and in 28 of 31 patients (90%), respectively. The mean advantage for the 29 patients was 362 days; for the 28 patients it was 383 days. The mean native PSA was 0.04 ng/mL (range 0.00 to 0.06) at the time of earliest detection by concentration techniques. Serum from female patients who had undergone cystoprostatectomy and "cured" patients who had undergone radical prostatectomy failed to concentrate, giving a test specificity of 100%.
Conclusions: Serum concentration is a specific and sensitive technique that provides a significant lead time of an additional 12 months in detecting cancer recurrence after radical prostatectomy when compared with nonconcentrated serum. Because the Tosoh assay, when run in the ultrasensitive mode, gave an additional lead time of 9 months at a residual cancer detection limit of 0.07 ng/mL, the combination of the Tosoh assay and serum concentration allows detection of a failed radical prostatectomy about 2 years earlier than does the Hybritech Tandem-R assay, which has a residual cancer detection limit of 0.2 ng/mL.