Clinical Utility of the PCA3 Urine Assay in European Men Scheduled for Repeat Biopsy

european urology 54 (2008) 1081 1088 available at www.sciencedirect.com journal homepage: www.europeanurology.com Prostate Cancer Clinical Utility of the PCA3 Urine Assay in European Men Scheduled for Repeat Biopsy Alexander Haese a, *, Alexandre de la Taille b, Hendrik van Poppel c, Michael Marberger d, Arnulf Stenzl e, Peter F.A. Mulders f, Hartwig Huland g, Clément-Claude Abbou b, Mesut Remzi d, Martina Tinzl d, Susan Feyerabend e, Alexander B. Stillebroer f, Martijn P.M.Q. van Gils f, Jack A. Schalken f a Department of Urology, University Clinic Hamburg, Hamburg, Germany b Hopital Henri Mondor, Créteil, France c Universitair Ziekenhuis Gasthuisberg, Leuven, Belgium d University of Vienna, Vienna, Austria e Uniklinikum Tübingen, Tübingen, Germany f Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands g Martini-Clinic Prostate Cancer Center, University Clinic Hamburg, Hamburg, Germany Article info Abstract Article history: Accepted June 16, 2008 Published online ahead of print on June 26, 2008 Keywords: Diagnostic methods PCA3 Prostate biopsy Prostate cancer Prostate specific antigen Sensitivity Specificity Please visit www.eu-acme.org/ europeanurology to read and answer questions on-line. The EU-ACME credits will then be attributed automatically. Background: The Prostate CAncer gene 3 (PCA3) assay has shown promise as an aid in prostate cancer (pca) diagnosis in identifying men with a high probability of a positive (repeat) biopsy. Objective: This study evaluated the clinical utility of the PROGENSA PCA3 assay. Design, setting, and participants: This European prospective, multicentre study enrolled men with one or two negative biopsies scheduled for repeat biopsy. Measurements: After digital rectal examination (DRE), first-catch urine was collected to measure PCA3 mrna concentration and to calculate the PCA3 score. The PCA3 score was compared to biopsy outcome. The diagnostic accuracy of the PCA3 assay was compared to percent of free prostate-specific antigen (%fpsa). Results and limitations: In 463 men, the positive repeat biopsy rate was 28%. The higher the PCA3 score, the greater the probability of a positive repeat biopsy. The PCA3 score (cut-off of 35) had a greater diagnostic accuracy than %fpsa (cut-off of 25%). The PCA3 score was independent of the number of previous biopsies, age, prostate volume, and total prostate-specific antigen (PSA) level. Moreover, the PCA3 score was significantly higher in men with high-grade prostate intraepithelial neoplasia (HGPIN) versus those without HGPIN, clinical stage T2 versus T1, Gleason score 7 versus <7, and significant versus indolent (clinical stage T1c, PSA density [PSAD] <0.15 ng/ml, Gleason score in biopsy 6, and percent positive cores 33%) pca. Conclusions: The probability of a positive repeat biopsy increases with rising PCA3 scores. The PCA3 score was superior to %fpsa for predicting repeat prostate biopsy outcome and may be indicative of clinical stage and significance of pca. # 2008 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. University Clinic Hamburg, Department of Urology, Martinistraße 52, 20246 Hamburg, Germany. Tel. +49 40 428033442; Fax: +49 40 428032444. E-mail address: haese@uke.uni-hamburg.de (A. Haese). 0302-2838/$ see back matter # 2008 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2008.06.071

1082 european urology 54 (2008) 1081 1088 1. Introduction Prostate cancer (pca) is one of the most common male cancers in the Western world [1]. Currently, early detection of pca relies primarily on an abnormal digital rectal examination (DRE) and an elevated prostate-specific antigen (PSA) level leading to a prostate biopsy. However, because of low positive predictive values, up to 75% of men with PSA values in the 2.5 10-ng/ml range and/or suspicious DRE have a negative first biopsy. Further, 10 35% of these men have pca detected on repeat biopsy [2,3]. In men with a negative first biopsy but persistent suspicion of pca, the European Association of Urology (EAU) guidelines recommend a repeat biopsy [4]. However, in approximately 80% of men, these repeat biopsies are negative. Not only economic aspects but also anxiety, discomfort, and sometimes severe complications are associated with prostate biopsies [2,3]. Thus, there is a need for additional tests to increase the probability of detecting pca at repeat biopsy and reduce the number of unnecessary biopsies. In this respect, the Prostate CAncer gene 3 (PCA3) assay, a new pca gene-based marker, has shown promising results. The PCA3 gene is highly overexpressed (median: 66- fold) in >95% of malignant (ie, tumour or metastatic) prostate tissue compared to benign and normal prostate tissue [5 8]. The PROGENSA PCA3 assay measures PCA3 and PSA mrna concentrations in post-dre urine [9]. In a study of 226 American men with PSA values 2.5 ng/ml and 1 negative biopsy, the PCA3 score increased with the probability of a positive repeat biopsy [10]. PCA3 was superior to PSA for pca detection at repeat biopsy. A PCA3 score of 35 provided a specificity of 72% and a sensitivity of 58%. The primary objective of this study is to determine the performance characteristics and clinical utility of the PCA3 assay in detecting pca at repeat biopsy in European men. 2. Methods 2.1. Study design This was a prospective, multinational, multicentre European study. Men with one or two previous negative prostate biopsies (6 cores performed at 3 mo prior to enrolment) scheduled for repeat biopsy were enrolled in six European centres. The respective Independent Ethics Committees (IECs) approved the study protocol, and informed consent was obtained. Men with more than two previous negative biopsies receiving medical therapy known to affect serum PSA with symptoms of urinary tract infection (UTI) and a history of pca or invasive treatment for benign prostatic hyperplasia (BPH) were excluded. Men with high-grade prostate intraepithelial neoplasia (HGPIN) or atypical small acinar proliferation (ASAP) at prior biopsy were classified as negative for pca and included in the study. 2.2. Specimen collection, processing, and PCA3 assay procedure Blood and urine specimens were collected between August 2006 and July 2007 immediately before repeat biopsy. Total and free PSA (fpsa) were assessed simultaneously at each participating site. First-catch urine samples were collected following an attentive DRE (three strokes per lobe) as described by Groskopf et al [9]. The urine sample was processed and tested to quantify PCA3 and PSA mrna concentrations using the PROGENSA PCA3 assay [9]. Operators performing the PCA3 assay were unaware of the subject s clinical status. The PCA3 score was calculated as [PCA3 mrna]/[psa mrna] 1000. At least 10 standardized peripheral zone biopsy cores were taken by an experienced physician. The specimens were evaluated by the pathologist at each site. 2.3. Statistical analyses A sample size of 97 evaluable subjects with positive biopsy results and 73 evaluable subjects with negative biopsy results was required to demonstrate a sensitivity of 50% and a specificity of 75% within a 10% confidence interval (CI) width (2-sided) at a 0.05 significance level [10]. Assuming that pca was detected in 15% of the study population and 10% of the subjects were nonevaluable, approximately 730 subjects had to be enrolled; however, an interim analysis performed in February 2007 showed that the pca detection rate was 25%. When maintaining the other assumptions, this would mean that a total of 432 subjects needed to be enrolled. The performance of the PCA3 assay was evaluated in terms of sensitivity and specificity by comparing the PCA3 score to biopsy results. The diagnostic accuracy of the PCA3 assay was compared to that of percent of free PSA (%fpsa), which was chosen as a comparator because the majority of men had a serum PSA of 4 10 ng/ml, in which range the use of %fpsa improves the specificity of PSA for pca detection [11,12]. Performance analyses of the PCA3 assay were also conducted by number of previous biopsies and total PSA ranges. The relationship between PCA3 score and prostate volume, clinical stage, Gleason sum score, significant versus indolent pca, and no HGPIN versus HGPIN at repeat biopsy were also assessed. Indolent pca was defined according to the Epstein criteria as clinical stage T1c, PSA density (PSAD) <0.15 ng/ml, Gleason score in biopsy 6, and percent of positive cores 33% [13,14]. The Epstein criteria use the parameter number of positive cores <3 ; however, this is based on a six-core biopsy. We applied the parameter percent of positive cores 33%, as this is appropriate with biopsy schemes sampling more than six cores. The nonparametric Wilcoxon rank sum test or the Kruskal-Wallis test was used to assess significance levels. Pearson correlation coefficients examined the relationship between the PCA3 score and prostate volume or PSA level. The nonparametric method of DeLong was used to compare the area under the curves (AUC) of the receiver operating characteristics (ROC) of the PCA3 score to %fpsa.

european urology 54 (2008) 1081 1088 1083 Table 1 Characteristics of the study population Men with negative biopsy n = 335 Men with positive biopsy n = 128 p value All evaluable men n = 463 Median Mean SD/ number (%) Median Mean SD/ number (%) Median Mean SD/ number (%) Age (yr; n = 333/128/461) 64 63.7 6.6 66 66.1 6.5 0.0005 * 65 64.4 6.6 One previous negative 239 (72%) 92 (73%) 0.8625 & 331 (72%) biopsy (%) Two previous negative 92 (28%) 34 (27%) 126 (28%) biopsies (%) Number of cores taken 12 16.1 6.6 12 15.7 6.6 0.3829 # 12 16.0 6.6 (n = 334/127/461) Serum total PSA (ng/ml; n = 332/126/458) 6.7 8.0 5.3 8.2 11.2 11.4 0.0004 # 7.0 8.9 7.6 Men with serum total PSA (%) <4 ng/ml 31 (9%) 12 (10%) 43 (9%) 4 10 ng/ml 236 (71%) 67 (53%) 303 (66%) >10 ng/ml 65 (20%) 47 (37%) 112 (25%) Prostate volume 51 58.7 31.0 45 51.2 24.5 0.0203 # 50 56.6 29.5 (ml; n = 323/123/446) Men with suspicious DRE 49 (15%) 39 (31%) 0.0001 & 88 (19%) (number [%]; n = 333/127/460) PCA3 score 19.5 35.5 62.1 33.7 63.8 91.8 <0.0001 # 23.8 43.3 72.6 Note: Patient numbers in subcategories do not always add up to the total number of patients because of missing data. Abbreviations: SD, standard deviation; PSA, prostate-specific antigen; DRE, digital rectal examination; PCA3, Prostate CAncer gene 3. * t test. # Wilcoxon rank sum test. & x 2 test. Uni- and multivariable logistic regression models to predict the presence of pca at repeat biopsy were fitted using DRE, total PSA, %fpsa, prostate volume, and PCA3 score. Bootstrapcorrected accuracy was quantified using AUC estimates in models with and without the PCA3 score. PCA3 was coded as a cubic spline to allow nonlinear effects and to obviate the limitations associated with the use of categorical cut-offs. Differences in predictive accuracy were compared using the Mantel-Haenszel test. 3. Results 3.1. Informative rate respectively; 92% had a biopsy Gleason score of 6 7; and 78% had 33% of the total number of cores positive for cancer. 3.3. PCA3 score An increasing PCA3 score corresponded with an increasing probability of a positive repeat biopsy (Fig. 1). Men with a PCA3 score 35 had a 39% probability of a positive repeat biopsy compared to a 22% probability in men with a PCA3 score <35 ( p = 0.0001, Pearson s x 2 test). The mean PCA3 score Among 470 subjects, a total of 467 urine samples (99.4%) had adequate concentrations of PCA3 and PSA mrna to calculate the PCA3 score. 3.2. Study population Of 467 men with a PCA3 score, 463 had conclusive biopsy results, and 128 (28%) had cancer on repeat biopsy (Table 1). Men with a positive biopsy had a statistically significant higher age, higher total PSA, lower prostate volume, more often a suspicious DRE, and a higher mean PCA3 score than men with a negative biopsy. Men with a positive biopsy had clinical stage T1c and T2 in 67% and 26% of cases, Fig. 1 The higher the PCA3 score, the greater the probability of a positive repeat prostate biopsy. Abbreviation: PCA3, Prostate CAncer gene 3.

1084 european urology 54 (2008) 1081 1088 Table 2 Sensitivity and specificity of the PCA3 assay and %fpsa Sensitivity Specificity PCA3 score cut-off 20 73% 51% PCA3 score cut-off 35 47% 72% PCA3 score cut-off 50 35% 82% %fpsa cut-off 25% 83% 23% Abbreviations: PSA, prostate-specific antigen; %fpsa, percent-free PSA; PCA3, Prostate CAncer gene 3. was statistically significantly higher in men with a positive biopsy (63.8; 95% CI, 47.7 79.8) than in men with a negative biopsy (35.5; 95% CI, 28.8 42.2; p < 0.0001). The median PCA3 score was 33.7 in men with a positive biopsy and 19.5 in men with a negative biopsy. It should be noted that as the PCA3 score had a skewed distribution function, the use of arithmetic median values may be more appropriate (eg, for determining cut-off values) than mean values. The AUC ROC was 0.658 for the PCA3 score and 0.578 for %fpsa ( p = 0.0804). The performance characteristics at different PCA3 score cut-off points are presented in Table 2. The PCA3 score cut-off of 35 provided the optimal balance between sensitivity (47%) and specificity (72%). In comparison, the specificity of %fpsa (cut-off 25%) was 23%. At a sensitivity of 80%, the PCA3 score specificity of 46% was higher than the 27% for %fpsa. The same applied for a sensitivity of 90%: The specificity was 25% for the PCA3 score compared with 16% for %fpsa. 3.4. Impact of clinical variables on PCA3 score 3.4.1. Number of previous biopsies The performance of the PCA3 score was comparable in men with one or two previous negative biopsies (Fig. 2). The mean/median PCA3 score was similar in both groups (39.5:23.6 vs 50.4:24.9, p = 0.3622), respectively. The sensitivity and specificity of the PCA3 score at a cut-off of 35 in men with one versus two previous negative biopsies was comparable with 46% versus 50% for sensitivity and 73% versus 70% for specificity. 3.4.2. Age The PCA3 assay had similar performance characteristics in younger (<65 yr) and older men (65 yr): AUC ROC 0.634 versus 0.639, respectively. 3.4.3. Prostate volume and total PSA Fig. 3 shows the relationship between prostate volume and PCA3 score or total PSA. Men with a higher prostate volume had higher levels of total PSA (Fig. 3a). In contrast, the mean PCA3 score was Fig. 2 The diagnostic accuracy of the PCA3 assay is independent of the number of previous negative biopsies. Abbreviations: AUC, area under the curve; ROC, receiver operating characteristics. independent of prostate volume (Fig. 3b) and total PSA (Fig. 4 and Table 3). The Pearson correlation coefficient confirmed the lack of relationship between PCA3 score and prostate volume (r = 0.03547, p = 0.4549). The small correlation coefficient (r = 0.10235, p = 0.0285) showed that the relationship between PCA3 score and total PSA is very weak. 3.4.4. Clinical stage, Gleason score, and the significance of pca Men with clinical stage T2 had a statistically significantly higher mean PCA3 score than men with clinical stage T1c (Table 4). Similarly, the mean PCA3 score was significantly higher in Gleason score 7 versus Gleason score <7 cancers, 72 of 87 (83%) men with complete data had significant pca. In these men, the mean PCA3 score was significantly higher than in men with indolent pca. 3.4.5. HGPIN When comparing the mean PCA3 score in men with HGPIN (n = 106) and without HGPIN (n = 357) before enrolment, men with HGPIN had a statistically significantly ( p = 0.0094) higher mean PCA3 score (49.3; 95% CI, 39.0 59.6) than men without HGPIN (41.5; 95% CI, 33.5 49.6). Similarly, in men with a negative biopsy during the study, those with HGPIN had statistically significantly higher mean PCA3 scores (47.9; 95% CI, 36.1 59.8) than those without (31.8; 95% CI, 24.0 39.7, p = 0.016). Of 94 men with a PCA3 score 35 who had a negative biopsy, 34 men (36%) had HGPIN on biopsy.

european urology 54 (2008) 1081 1088 1085 Fig. 4 The diagnostic accuracy of the PCA3 assay is independent of serum total PSA level. Abbreviations: PCA3, Prostate CAncer gene 3; PSA, prostate-specific antigen; AUC, area under the curve; ROC, receiver operating characteristics. Fig. 3 (a) Serum total PSA level is dependent on prostate volume, whereas (b) the PCA3 score is independent of prostate volume. Median PCA3 scores: 24.4, 24.0, and 23.0 for prostate volumes <30 ml, 30 50 ml, and >50 ml, respectively. Bars represent mean W 95% CI. Abbreviations: PSA, prostate-specific antigen; PCA3, Prostate CAncer gene 3; CI, confidence interval. 3.5. Uni- and multivariable analysis Table 5 shows that the PCA3 score represented a statistically significant and independent predictor of pca at repeat biopsy ( p 0.006). In addition, the PCA3 score represented the most informative univariable predictor and was capable of increasing predictive accuracy in multivariable models by 4.2%, which was highly significant ( p < 0.001). 3.6. Potential reduction in number of biopsies versus characteristics of missed pca At a PCA3 cut-off of 20, a 44% reduction of repeat biopsies could be achieved, while 11 biopsy Gleason score 7 and one Gleason score 8 pca would have been missed (Table 6). Of these, two were clinical stage T2, and one had 33% of cores positive for cancer. At a PCA3 score cut-off of 35, 67% of biopsies could have been avoided, while 24 cancers with a Gleason score of 7, two with a Gleason score of 8, and one with a Gleason score of 9 would have been missed. A total of seven of these were stage T2. In five cases, the percentage of positive biopsy cores was 33%. Table 3 Mean and median PCA3 score and diagnostic accuracy of PCA3 score (cut-off 35) at different total PSA levels PCA3 score in: n Median Mean 95% CI p value * Sensitivity Specificity All 463 23.8 43.3 36.7 49.9 47% 72% PSA <4 ng/ml 43 27.7 41.8 28.0 55.7 0.7282 50% 65% PSA 4 10 ng/ml 303 23.8 41.6 32.6 50.5 49% 74% PSA >10 ng/ml 112 22.6 47.3 35.6 59.0 43% 69% Abbreviations: CI, confidence interval; PSA, prostate-specific antigen; PCA3, Prostate CAncer gene 3. * Between group analysis using Kruskal Wallis test.

1086 european urology 54 (2008) 1081 1088 Table 4 Mean and median PCA3 score according to clinical stage, Gleason score in biopsy, and clinical significance of the cancer PCA3 score in: n Median Mean 95% CI p value * All men with positive repeat biopsy 128 33.7 63.8 47.7 79.8 T1c 81 26.8 56.0 34.5 77.5 0.005 T2 30 61.7 88.4 56.9 119.9 Gleason score <7 70 28.1 62.1 35.6 88.6 0.0401 Gleason score 7 52 45.3 68.6 50.8 86.5 Indolent pca ** 15 21.4 24.5 14.9 34.0 0.0059 Significant pca 72 42.1 75.9 50.3 101.5 Note: Patient numbers in subcategories do not always add up to the total number of patients because of missing data. Abbreviations: PCA3, Prostate CAncer gene 3; CI, confidence interval; pca, prostate cancer; PSA, prostate-specific antigen. * Wilcoxon rank sum test. ** Defined as clinical stage T1c, PSA density <0.15, Gleason score biopsy 6, percent of positive cores 33%. Table 5 Uni- and multivariable regression analysis Univariable analysis Multivariable analysis Base model Base model + PCA3 score OR p value PA (%) OR p value OR p value Age 1.051 0.004 0.578 1.042 0.035 1.024 0.243 Serum total PSA 1.063 0.001 0.600 1.07 0.003 1.064 0.007 %fpsa 0.974 0.063 0.578 0.992 0.637 0.988 0.477 DRE 2.610 <0.001 0.577 2.473 0.001 2.263 0.006 Prostate volume 0.990 0.024 0.563 0.982 0.002 0.985 0.015 PCA3 score * <0.001 0.663 0.006 PA (%) 66.8 71.0 Increment in PA (%) +4.2 p value <0.001 Abbreviations: PCA3, Prostate CAncer gene 3; PA, predictive accuracy; OR, odds ratio; PSA, prostate-specific antigen; %fpsa, % free PSA; DRE, digital rectal examination. * Coded as cubic spline to allow nonlinear effects. Table 6 Characteristics of missed cancers with a Gleason score I7 at a PCA3 score cut-off of 20 PCA3 score Gleason score % cores positive Clinical stage Total PSA PSAD 19.42 8 10.00 T2 6.6 0.198 17.17 7 50.00 T1c 10.4 0.224 17.25 7 27.59 T2 6 0.143 12.87 7 25.00 29.4 0.896 3.22 7 20.00 T1c 18.3 0.333 14.93 7 16.67 T1c 10.1 0.362 15.73 7 12.50 T1c 11 0.383 17.39 7 12.50 T1c 20.5 0.488 14.29 7 10.00 6 0.345 9.1 7 6.67 T1c 6.6 0.099 19.41 7 5.26 T1c 12.2 0.163 2.65 7 Note: Empty cells denote missing data. Abbreviations: PCA3, Prostate CAncer gene 3; PSA, prostate-specific antigen; PSAD, PSA density. 4. Discussion This study investigated the PCA3 assay as a diagnostic tool to aid in the decision to perform repeat biopsy in men with one or two previous negative biopsies. The informative rate was >99%, showing that the specimen-collection procedure is robust. The performance characteristics of the PCA3 assay in this European study are comparable to those of a North American PCA3 repeat biopsy study

european urology 54 (2008) 1081 1088 1087 involving 233 men with elevated serum PSA levels and at least one previous negative biopsy [10]. In both studies, the probability of a positive repeat biopsy increased with higher PCA3 scores; the PCA3 score cut-off of 35 provided the optimal balance between sensitivity and specificity, with a specificity of 72% for pca detection. In this study, the PCA3 score (cut-off 35) had a better diagnostic accuracy than %fpsa (cut-off 25%) for predicting repeat biopsy outcome. Furthermore, the diagnostic accuracy of the PCA3 score was not affected by age, number of previous negative biopsies, different total PSA ranges, or prostate volume, which is consistent with results in 570 North American men scheduled for (first or repeat) biopsy [15]. This study demonstrates for the first time that the PCA3 score was significantly higher in men with clinical stage T2, biopsy Gleason score 7, and significant pca compared to men with clinical stage T1, biopsy Gleason score <7, and indolent pca. This may indicate that the PCA3 score is indicative of the clinical stage and aggressiveness of pca. It was also recently shown that the PCA3 score is associated with tumour volume and Gleason score in prostatectomy samples [16]. Although this should be further evaluated, these data suggest that the PCA3 score may also be used to select those patients who are candidates for active surveillance. It has been previously shown that HGPIN is associated with an increased risk of pca at repeat biopsy [17 19]. Here, the PCA3 score in men with a negative biopsy was significantly higher in those with than without HGPIN, suggesting that the PCA3 score identifies men at risk of developing pca. In addition, a recent study demonstrated that PCA3 mrna is expressed by >90% of HGPIN tissue [20]. Further prospective studies should evaluate whether the PCA3 score may be used to monitor men with chronically elevated serum PSA levels at regular intervals for the development of clinically significant pca. The repeat biopsy was positive in 28% of men. This is at the high range of previously reported data (10 35%) [2,3,10] but similar to the 27% positive biopsy rate found in the North American PCA3 repeat biopsy study [10]. Joniau et al [21] found 33.8% pca at repeat biopsies in patients with HGPIN at first biopsy. With regard to pca detection, both the North American and the present study indicate that a PCA3 score cut-off of 35 provides an optimal balance between sensitivity and specificity. This cut-off may therefore identify men with an increased probability of having a positive repeat biopsy [10]. However, in clinical practice, urologists want to avoid unnecessary biopsies while minimizing the risk of missing significant cancers. In this dataset, a PCA3 score cut-off of 20 would reduce the number of repeat biopsies by 44% while missing only 12 (9%) cancers with a Gleason score of 7 8, of whom two were clinical stage T2 and one had >33% of cores positive for cancer either parameter eliminating those cancers from Epstein s indolent criteria. At a PCA3 score cut-off of 35, 67% of biopsies would have been avoided, while 27 (21%) cancers with a Gleason score of 7 9 would have been missed. A total of seven of these were stage T2. In five cases, the percentage of positive biopsy cores was >33%. Therefore, a PCA3 score cut-off of 20 that still avoids 44% of unnecessary biopsies while missing only 12:128 (9%) of significant cancers may have the highest utility regarding the decision of whether the cancer found on repeat biopsy is clinically insignificant. This needs to be further evaluated in prospective clinical trials. The development of a nomogram incorporating the PCA3 score and other diagnostic variables may further improve the predictive accuracy of the PCA3 score for its use in clinical practice. This study showed that the inclusion of the PCA3 score in a multivariable model significantly increased its predictive accuracy. The inclusion of the PCA3 score in nomograms should therefore be further evaluated. It should be noted that a low PCA3 score indicates a low probability of cancer on biopsy but, as with most diagnostic tests, does not exclude pca in the biopsy or a clinically significant pca. 5. Conclusions This European study in men with one or two previous negative biopsies shows that the PCA3 assay may aid in deciding which patients need a repeat biopsy. An increasing PCA3 score corresponds with an increasing probability of a positive repeat biopsy. In this study, the PCA3 score had greater diagnostic accuracy than %fpsa for predicting repeat prostate biopsy outcome, even at a sensitivity of 80 90%. The utility of the PCA3 score is independent of the number of previous biopsies, prostate volume, and total PSA. The PCA3 score may also be indicative of the significance of pca. Author contributions: Alexander Haese had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Haese, de la Taille, van Poppel, Marberger, Schalken, Huland, Stenzl, Abbou. Acquisition of data: Haese, Mulders, Remzi, Tinzl, Feyerabend, Stillebroer, van Gils.

1088 european urology 54 (2008) 1081 1088 Analysis and interpretation of data: Haese, van Poppel. Drafting of the manuscript: Haese. Critical revision of the manuscript for important intellectual content: Haese, de la Taille, van Poppel, Marberger, Stenzl, Mulders, Huland, Abbou, Remzi, Tinzl, Feyerabend, Stillebroer, van Gils, Schalken. Statistical analysis: Haese. Obtaining funding: Haese, de la Taille, van Poppel, Marberger, Schalken, Huland, Stenzl, Abbou. Administrative, technical, or material support: Haese. Supervision: Haese. Other (specify): none. Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: van Poppel, advisory board member, Gen-Probe Inc; Mulders, consultant, Gen- Probe Inc; Schalken, consultant, Gen-Probe Inc. Funding/Support and role of the sponsor: Gen-Probe Inc. provided funding for the design and conduct of the study, collection of data, management of the data, analysis, interpretation of the data, preparation, review, and approval of the manuscript. Acknowledgment statement: The authors acknowledge Labor Limbach, Heidelberg, Germany, for performing PCA3 testing. The authors are grateful to Ismar Healthcare NV for their assistance in writing the manuscript and to Dr F. Chun for assistance with the statistical analyses. References [1] Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43 66. [2] Seitz C, Palermo S, Djavan B. Prostate biopsy. Minerva Urol Nefrol 2003;55:205 18. [3] Raja J, Ramachandran N, Munneke G, Patel U. Current status of transrectal ultrasound-guided prostate biopsy in the diagnosis of prostate cancer. Clin Radiol 2006;61: 142 153. [4] Heidenreich A, Aus G, Bolla M, et al. EAU guidelines on prostate cancer. Eur Urol 2008;53:68 80. [5] Bussemakers MJ, van Bokhoven A, Verhaegh GW, et al. DD3: a new prostate-specific gene, highly overexpressed in prostate cancer. Cancer Res 1999;59:5975 9. [6] Hessels D, Klein Gunnewiek JMT, van Oort I, et al. DD3 PCA3 -based molecular urine analysis for the diagnosis of prostate cancer. Eur Urol 2003;44:8 16. [7] Shaw G, Oliver RTD, Purkiss T, Prowse DM. Re: Christine McKillop. Interview with Jack Schalken: PCA3 and its use as a diagnostic test in prostate cancer. Eur Urol 2006;50:153 4, Eur Urol 2007;51:860 2. [8] McKillop C. Interview with Jack Schalken: PCA3 and its use as a diagnostic test in prostate cancer. Eur Urol 2006; 50:153 4. [9] Groskopf J, Aubin SM, Deras IL, et al. APTIMA PCA3 molecular urine test: development of a method to aid in the diagnosis of prostate cancer. Clin Chem 2006;52: 1089 95. [10] Marks LS, Fradet Y, Deras IL, et al. PCA3 molecular urine assay for prostate cancer in men undergoing repeat biopsy. Urology 2007;69:532 5. [11] Catalona WJ, Partin AW, Slawin KM, et al. Use of the percentage of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. JAMA 1998;279:1542 7. [12] Djavan B, Zlotta A, Remzi M, et al. Optimal predictors of prostate cancer on repeat prostate biopsy: a prospective study of 1,051 men. J Urol 2000;163:1144 8. [13] Epstein JI, Walsh PC, Carmichael M, Brendler CB. Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA 1994;271: 368 74. [14] Bastian PJ, Mangold LA, Epstein JI, Partin AW. Characteristics of insignificant clinical T1c prostate tumors. A contemporary analysis. Cancer 2004;101:2001 5. [15] Deras IL, Aubin SM, Blase A, et al. PCA3: a molecular urine assay for predicting prostate biopsy outcome. J Urol 2008;179:1587 92. [16] Nakanishi H, Groskopf J, Fritsche HA, et al. PCA3 molecular urine assay correlates with prostate cancer tumor volume: implication in selecting candidates for active surveillance. J Urol 2008;179:1804 9. [17] Brawer MK. Prostatic intraepithelial neoplasia: an overview. Rev Urol 2005;7(Suppl 3):S11 8. [18] Herawi M, Kahane H, Cavallo C, Epstein JI. Risk of prostate cancer on first re-biopsy within 1 year following a diagnosis of high grade prostatic intraepithelial neoplasia is related to the number of cores sampled. J Urol 2006; 175:121 4. [19] Schoenfield L, Jones JS, Zippe CD, et al. The incidence of high-grade prostatic intraepithelial neoplasia and atypical glands suspicious for carcinoma on first-time saturation needle biopsy, and the subsequent risk of cancer. BJU Int 2007;99:770 4. [20] Popa I, Fradet Y, Beaudry G, et al. Identification of PCA3 (DD3) in prostatic carcinoma by in situ hybridization. Mod Pathol 2007;20:1121 7. [21] Joniau S, Goeman L, Roskams T, et al. Effect of nutritional supplement challenge in patients with isolated highgrade prostatic intraepithelial neoplasia. Urology 2007; 69:1102 6.