Objective: Evaluation of the analytical performance of the Hybritech Tandem-E PSA assay as applied to the Rotterdam part of the European Randomized Study of Screening for Prostate Cancer (ERSPC) during 1994-1997.
Design and methods: For assessment of test performance imprecision and accuracy contributing to clinical decision making was examined. Pre-analytical variables (specimen handling), long-term reagents stability and calibration check with the Stanford 90:10 PSA Calibrator were studied.
Results: Total prostate-specific antigen proved to be a stable marker. Provided correct storage at 4 degrees C, we were not able to find differences between samples analyzed directly after blood withdrawal and those analyzed the following day, with or without centrifugation. Day-to-day precision, we found during 1997 a coefficient of variation of 4.9% for concentration 1.8 micrograms/L and 4.2% for 4.8 micrograms/L (n = 182). The variation among the 21 Tandem-E reagent batches used during 1994-1997 proved to be small. Application of the Stanford 90:10 PSA Calibrator revealed 3% higher values for Tandem-E.
Conclusion: The overall, long-term picture of Tandem-E PSA is reliable in our hands. Possible differences from the true PSA values, not specific for Tandem-E, need to be elaborated.