Prostate biopsy is routinely recommended for suspicious DRE results regardless of PSA.3
Biopsy is also recommended using PSA thresholds ranging from 2.5 to 4.0 ng/mL.1, 2, 15
However, this has led to unnecessary biopsies and possible over-detection of some cancers.15–17
To elucidate whether phi
PSA-isoform measurement can improve PCa early detection, we examined a large, prospective cohort to predict biopsy findings in patients with moderate PSA elevations (2.0–10.0 ng/mL) and benign DRE findings. Such men are at higher risk of PCa (25% cancer detection rate compared with 4% in the general male population aged ≥50 years).3
Our bootstrapped population was designed to mirror this 25% incidence of PCa on biopsy.
Prostate biopsy may be associated with discomfort, anxiety, and financial costs. Minor complications occur frequently, and major complications are possible, underscoring the need for more specific markers to reduce unnecessary biopsies. We sought to determine the utility of p2PSA and phi for this clinical goal.
Precursor forms of PSA have been shown to improve the accuracy of PSA for detecting PCa.5, 6, 9–12, 28, 29
Specifically, preliminary reports suggest that p2PSA may be useful at PSA concentrations from 2.0–10.0 ng/mL.6, 9–12, 28, 29
Some, but not all, studies have suggested an association between proPSA and PCa aggressiveness.10, 12, 20
Thus, p2PSA and phi
are being investigated in active surveillance programs to help overtreatment of insignificant PCa.19, 30
Catalona et al. previously reported in the PSA range of 2.0–10.0 ng/mL, the proPSA-to-fPSA ratio (%proPSA) yielded a higher specificity than %fPSA.9
Results from a separate multi-site study also supported the role of p2PSA, in combination with PSA and fPSA, in reducing unnecessary biopsies.12, 13
In the current study, the specificity for phi
was higher than %fPSA at all pre-specified sensitivities, and PCa risk increased directly with increasing phi
values. This suggests a role for phi
as a patient monitoring tool, since increasing phi
values reflect PCa risk.19
For example, at 95% sensitivity, the specificity of phi
was 16.0% compared to 8.4% for %fPSA. Moreover, at lower sensitivities (90%, 85%, and 80%) for PCa detection that might be preferred to reduce the detection of possibly “insignificant” tumors, phi
had a significantly greater specificity than %fPSA. These results were consistent across age groups, PSA concentrations, and ethnic groups, suggesting that they are representative of the intended-use population.
For individual risk assessment, the probability of PCa varied considerably based upon phi values. For example, a man with a phi ≥ 55 (13% of the study population) had a > 52% probability of PCa and 4.7-fold increased relative risk of positive biopsy. In contrast, at approximately 90% sensitivity, a patient with a phi < 25 had an 11% probability of PCa.
For the PCa group, higher phi values were also significantly associated with a higher percentage of biopsy Gleason grade ≥ 7, ranging from 26% to 42% for phi concentrations < 25 and ≥ 55, respectively. For the entire study population, the AUC for phi (0.724) exceeded that of %fPSA (0.670) in discriminating Gleason ≥ 4+3 PCa vs. lower Gleason grade PCa or negative biopsies. Using a phi cutoff of 21.3 (95% sensitivity), 25% of missed cancers were Gleason score ≥7; therefore, careful surveillance is necessary. The AUCs for phi also exceeded those of %fPSA in all three prostate volume tertiles, suggesting that phi provides better discrimination of PCa from benign disease than %fPSA across the spectrum of prostate volumes. Because phi did not differ by age and race these results suggest that phi may be applicable to a broad spectrum of men as an adjunct to predict clinically-significant PCa.
The large number of subjects in the present validation study provides confidence in the phi cutoffs determined. Phi is highly effective when used in patients with moderately elevated PSA concentrations who may be most likely to benefit from early diagnosis and curative PCa treatment. A physician might recommend biopsy for a patient with a phi ≥ 55.0 (risk = 52.1%) and surveillance for some men with a phi <25.0 (risk = 11.0%). For patients reluctant to undergo prostatic biopsy, a high phi might increase compliance with the appropriate follow-up.
We conclude that the phi
measurement ([−2]proPSA / fPSA) × PSA1/2
) may be useful to reduce unnecessary biopsies with improved specificity at various sensitivities for PCa detection in men age ≥50 years with PSA concentrations from 2.0–10.0 ng/mL, and negative DRE findings.****