Abstract Background Measurement of free prostate specific antigen (fPSA) in serum is used to aid in distinguishing prostate cancer (PC) from benign prostate conditions in men 50 years or older with hsPSA of 4.0-10.0 ng/mL (µg/L) and digital rectal exam (DRE) nonsuspicious for PC. Studies were performed to evaluate clinical performance for the Atellica® IM fPSAII assay* [fPSAII/hsPSAx100 ratio]. Methods Serum samples were prospectively collected from 1250 patients referred to an urologist for PC evaluation at 26 sites throughout the United States. At diagnostic cutoffs for %fPSAII spanning the spectrum of %fPSAII values, the percentage of PCs detected (sensitivity) and the corresponding percentage of biopsies avoided in men without PC (specificity) were tabulated. At a pre-specified sensitivity near 90%, %fPSAII specificity (relative to biopsy result) was compared to hsPSA specificity. The mean PC probabilities on biopsy and nonparametric 95% CI (2.5th and 97.5th percentiles) were calculated. The difference between these proportions was evaluated to estimate the improvement in specificity of %fPSAII relative to hsPSA alone. Difference in specificity was significant if the zero value was not bounded by 2.5th and 97.5th percentile values of the improvement estimate with a minimum of 5% improvement in specificity. Results Sensitivity and specificity results are presented in the table. The cutoffs provided a range of sensitivity estimates from 86%-100%. At 22% cutoff, sensitivity (90.3%) plus specificity (21.4%) is greater than 110.0%. Significant improvement in %fPSAII specificity over hsPSA alone was 8.9% with 95% CI of 6.8%-11.5%. Conclusions The clinical performance of the Atellica IM Analyzer fPSAII assay was acceptable and demonstrated an improvement in detection of PC than using hsPSA alone in this study. *CAUTION: Investigational Device. Not available for sale. Future availability cannot be guaranteed. The results are preliminary and were achieved in unique setting with no expected timeline for completion.
Read full abstract