Interassay Variability and Clinical Implications of Five Different Prostate-specific Antigen Assays
Basil Kaufmann,Paloma Pellegrino,Laura Zuluaga,Reuben Ben-David,Michael Müntener,Etienne X. Keller,Katharina Spanaus,Arnold von Eckardstein,Michael A. Gorin,Cédric Poyet
DOI: https://doi.org/10.1016/j.euros.2024.03.008
IF: 3
2024-03-22
European Urology Open Science
Abstract:Background and objective Prostate-specific antigen (PSA) remains a critical marker for prostate cancer (PCa) detection and monitoring. Recognising historical variability in PSA assays and the evolution of assay technology and calibration, this study aims to reassess interassay variability using the latest generation of five assays in a contemporary cohort of men undergoing prostate biopsy. Methods Five different commercially available PSA assays were tested in a blood sample of 76 men before undergoing a prostate biopsy. Total PSA (tPSA) and free-to-total PSA ratio (%fPSA) were compared across assays, using Roche (Basel, Switzerland) as the benchmark, and correlated with biopsy outcome to analyse the impact on PCa diagnosis. The statistical analysis included Passing-Bablok regression and Bland-Altman plots, with a p value threshold of <0.05 for significance. Key findings and limitations Among the 76 men, 28 (36.8%) were diagnosed with significant PCa (defined as International Society of Urological Pathology grade ≥2). A high correlation was observed between tPSA and %fPSA values among the different PSA assays tested ( r 2 ≥ 0.9). The Passing-Bablok analysis showed that tPSA results varied substantially among the assays, with slopes ranging between 0.78 and 1.04. Compared with the tPSA of Roche, tPSA values were on average 20.7% lower by Beckman (Oststeinbeck, Germany), 15.2% lower by Abbott (Chicago, IL, USA), 6.1% lower by Diasorin (Saluggia, Italy), and 9.6% higher by Brahms (Hennigsdorf, Germany; p < 0.001 for all). The %fPSA values by Abbott and Brahms were higher at 15.7% and 10.6%, respectively ( p < 0.001), while the Beckman and Diasorin values had minimal differences of -0.3% and 2.3%, respectively ( p > 0.05). The variability across assays would have resulted in discrepancies in both the sensitivity and the specificity for tPSA and %fPSA by at least 14%, depending on the cut-offs applied. Conclusions and clinical implications Despite the use of the latest PSA assays, relevant variability of tPSA and %fPSA results can be observed among different assays. There is an urgent need for standardised calibration methods and greater awareness among practitioners concerning interassay variability. Clinicians should acknowledge that clinically relevant thresholds may depend on the specific PSA assay and that ideally the same assay is applied over time for better clinical decision-making. Patient summary Prostate-specific antigen (PSA) is a critical marker for prostate cancer (PCa) detection and monitoring. However, significant variations were observed in the results of the latest PSA assays. Thus, standardised calibration methods and greater awareness among practitioners concerning interassay variability are needed.
urology & nephrology