Clinical role of PCA3 in repeat biopsy
Several studies have demonstrated that PCA3 score is independent of age, PSA level, or prostate size,27-29 and may be more effective in PCa detection compared with PSA or percent-free-PSA alone.16,27,29-31. In an early clinical study, Hessells et al. investigated PCA3 measurement (using RT-PCR assay) in urine sediments of 108 men who had serum PSA value above 3 ng/mL. In 24 men who had PCa on biopsy, the assay was positive for the PCA3 gene in 67%, indicating a negative predictive value of 90% and a specificity of 83%.32 These fundamental findings led to further studies by others showing PCA3 over expression in PCa appears to be reproducible and found in up to 95% of tested samples.24,32-34
Ruiz-Aragón and Márquez-Peláez performed a meta-analysis of the published literature (2003-2009) investigating the role of PCA3 in screening for PCa.35 Fourteen studies met inclusion criteria and they included more than 3,400 men with an average age ranging between 62-65 years and mean PSA levels from 2.5 to 8.7 ng/mL. All patients underwent a PBx as a reference test to compare to the antigen determination.
The overall sensitivity and specificity for PCA3 reported in this meta-analysis was 63% (range 46.8%-82.3%) and 75% (range 56.3%-89%), respectively. Due to the heterogeneity of these data, the authors concluded that PCA3 test should be used in conjunction with PSA and physician’s clinical judgment to determine which patients can be spared from repeat PBx.
Several large studies have confirmed the association between PCA3 value and PCa detection. A prospective multicenter study29 evaluated the utility of the PCA3 assay in 463 European men with history of at least one prior negative PBx scheduled to undergo repeat biopsy. When comparing men with PCa to those with negative repeat biopsies, the mean PCA3 scores were significantly higher in those with detected cancers, (63.8 vs. 35.5 p <0.0001).
Further, the authors reported that in patients with significant PCa (Gleason score 7 or higher on repeat biopsy), the PCA3 score was significantly higher than those in patients with indolent or less clinically significant disease (stage T1c, Gleason score less than 7, and PSA density less than 0.15), (mean 68.6 vs. 24.5 p = 0.006).
The clinical utility of the PCA3 assay in predicting repeat biopsy outcome was further validated in the placebo arm of REDUCE (Reduction by Dutasteride of
Prostate Cancer Events (REDUCE) study).28 The PCA3 assay was used in 1,140 men within the placebo arm of the study (mean PCA3 34). The probability of having a repeat biopsy was 57% in men who had PCA3 score over 100 compared to 6% among those who had PCA3 score below 5.
In addition, the role of PCA3 in predicting PBx results was studied in a prospective U.S. multicenter trial in patients undergoing repeat PBx.36 The area under the curve (AUC)—whereby a predictive accuracy of 0.5 is equivalent to a coin flip and 1.0 is perfect predictive value—for PCA3 in predicting outcome on initial repeat biopsy was reported as 0.65.
A community-based prospective clinical trial by Crawford et al.37 conducted at 50 urology practices within the U.S. evaluated the role of PCA3 test in 1,962 men with PSA levels above 2.5 ng/ml and/or abnormal DRE undergoing biopsy. The AUC for PSA and PCA3 alone were calculated as 0.569 and 0.706, respectively. The AUC increased to 0.720 when PCA3 was measured in conjunction with PSA. In this cohort, using a PCA3 cutoff of 35 reduced the number of false-positives from 1,089 to 249 (a 77% reduction) at the expense of missing 413 cancers.
The authors concluded that increase in PCA3 score correlates with increased risk of diagnosis PCa on repeat biopsy and should be used in combination with serum PSA and other clinical information to guide PBx decisions.
The clinical study leading to FDA approval of the PCA3 test in February of 2012 was a multicenter study from 14 clinical sites. The study included 466 men aged 50 years or older who had at least one negative biopsy and were offered repeat biopsy.38 Using a PCA3 score cutoff of 25 the negative PPV was 90%, 50% of repeat biopsies were avoided, and only 2% of Gleason score 7 or higher tumors were missed.
This study showed that men with a PCA3 score below 25 were 4.6 times less likely to have a positive repeat biopsy than men with score of 25 or higher (P < 0.0001). Based on these findings and others,27,29,37 most experts currently advocate use PCA3 thresholds ranging from 25-35, which appears to achieve the optimal balance between sensitivity and specificity in guiding repeat biopsy.
In early 2013, Tombal and colleagues reported the results of an interesting study assessing the value of best clinical judgment (BCJ) and the PCA3 assay in guiding the decision to perform a repeat PBx. In this case, BCJ refers to recommendations established using the RAND/UCLA Appropriateness Method for the appropriateness of repeat biopsy according to the PSA level, DRE findings, number of previous negative biopsies, prostate volume, and life expectancy, with and without consideration of PCA3.
These recommendations were applied to 1,024 subjects receiving placebo in REDUCE trial,28 and three scenarios (BCJ alone, BCJ with PCA3, and the PCA3 score alone [using a threshold value of 20]) were tested for their ability to reduce the repeat biopsy rate versus missing Gleason sum 7 or higher PCa.
They found that the diagnostic accuracy for Gleason sum 7 or higher PCa in the BCJ with PCA3 arm was superior to that of the other scenarios, with a NPV of 99%. In addition, 64% of repeat biopsies would have been avoided.39 These findings highlight that the simple addition of another diagnostic test does not replace the role of clinical judgment.