BJUI. Follow-up of men with an elevated PCA3 score and a negative biopsy: does an elevated PCA3 score indeed predict the presence of prostate cancer?

Transcription

1 . JOURNAL COMPILATION 2010 BJU INTERNATIONAL Urological Oncology FOLLOW-UP OF MEN WITH AN ELEVATED PCA3 SCORE AND A NEGATIVE BIOPSY REMZI ET AL. BJUI BJU INTERNATIONAL Follow-up of men with an elevated PCA3 score and a negative biopsy: does an elevated PCA3 score indeed predict the presence of prostate cancer? Mesut Remzi, Alexander Haese*, Hein van Poppel, Alexandre de la Taille, Arnulf Stenzl, Jörg Hennenlotter and Michael Marberger Medical University of Vienna, Vienna, Austria, *Martini Clinic Prostate Cancer Center, University Clinic Eppendorf, Hamburg, Universitätsklinikum Tübingen, Tübingen, Germany, UZ Gasthuisberg, Leuven, Belgium, and Hôpital Henri Mondor, Créteil, France Accepted for publication 21 December 2009 Study Type Diagnosis (exploratory cohort) Level of Evidence 2b OBJECTIVE To describe the follow-up of men with an elevated Prostate CAncer gene 3 (PCA3, a promising novel tool for prostate cancer detection) and a negative repeat biopsy (Bx-), as a previous study in men with one or two negative Bx (Bx ) scheduled for repeat Bx showed that a higher PCA3 score corresponded with a higher probability of a positive repeat Bx (Bx+). PATIENTS AND METHODS This study comprised an analysis of the follow-up of men with a PCA3 score of 20 and a repeat Bx, after which a follow-up Bx was taken. The initial study data in 463 men were also analysed to compare characteristics of: (i) men with a PCA3 score of 20 and 35 and a repeat Bx+, vs those with a Bx ; and (ii) men with a repeat Bx and a PCA3 score of 20 vs <20 and a PCA3 score of 35 vs <35. RESULTS A follow-up Bx was taken in 51 selected men; the Bx+ rate was 55%. Men with a follow-up Bx+ had a higher PCA3 score (mean 69.5, median 50.4) than those with a Bx (mean 37.7, median 28.2; P < 0.001). They also more often had high-grade prostatic intraepithelial neoplasia (HGPIN) at the previous Bx (46% vs 17%; P = 0.029). Men with a PCA3 score of 35 and a repeat Bx+ had a higher PCA3 score (mean 113.9, median 75.7) than those with a Bx (mean 87.3, median 56.9; P = 0.047). Men with a repeat Bx and an elevated/high PCA3 score more often had HGPIN than men with a low PCA3 score. CONCLUSIONS An elevated/high PCA3 score can predict prostate cancer in men with one or two previous Bx. If the repeat Bx is negative, an elevated/high PCA3 score combined with HGPIN might predict prostate cancer at the follow-up Bx. KEYWORDS follow-up, high-grade prostatic intraepithelial neoplasia, PCA3, prostate, biopsy, prostate cancer INTRODUCTION Prostate cancer is common and screening might lead to a high risk of overdiagnosis, as many men will be diagnosed with indolent cancer. A recent study showed that 1410 men would need to be screened and 48 additional cases would need to be treated to prevent one death from prostate cancer [1]. An ideal diagnostic test would lead to a reduction of unnecessary biopsies (Bx) and detection of life-threatening prostate cancer. There are many men with a high suspicion of prostate cancer because of a high/increasing PSA level or a suspicious DRE and who have had one or more negative Bx (Bx ). Prostate CAncer gene 3 (PCA3) is highly prostate-cancer specific; PCA3 mrna is greatly over-expressed (median 66 times) in prostate cancer tissue relative to benign tissue [2,3]. A commercial PCA3 assay (Progensa, Gen-Probe, San Diego, CA, USA) has shown promise as a novel tool to aid in the diagnosis of prostate cancer. It measures PCA3 and PSA mrna concentrations in urine samples taken after a DRE and the PCA3 score is calculated as (PCA3 mrna)/(psa mrna) 1000 [4]. The clinical utility of the PCA3 assay in guiding repeat Bx decisions has been reported and evidence is emerging that the PCA3 score might be indicative of the significance of prostate cancer [5 9]. A European study in 463 men with one or two previous Bx and scheduled for repeat Bx showed a better diagnostic accuracy for the PCA3 score than for the %free PSA for predicting the outcome of a repeat Bx [6]. A higher PCA3 score corresponded with a higher probability of a repeat Bx+. The primary objective of the present analysis was to describe the follow-up of men with an elevated PCA3 score ( 20) and a repeat Bx. In addition, we analysed the initial study data to evaluate relationships between clinical characteristics, the PCA3 score and the presence of prostate cancer at Bx JOURNAL COMPILATION 2010 BJU INTERNATIONAL 106, doi: /j x x

2 FOLLOW-UP OF MEN WITH AN ELEVATED PCA3 SCORE AND A NEGATIVE BIOPSY TABLE 1 Demographic and clinical characteristics of men with a repeat Bx and a PCA3 score of 20 in the repeat Bx study who had one or more follow-up Bx Mean (SD) or n (%) Men with follow-up characteristic (n) All men (51) Bx+ (28) Bx (23) P Age, years (51/28/23) 64.9 (6.3) 64.6 (6.9) 65.4 (5.6) 0.681* Total PSA, ng/ml (35/22/13) 7.2 (4.6) 7.3 (4.9) 6.9 (4.4) PCA3 score (51/28/23) 55.2 (43.7) 69.5 (46.5) 37.7 (33.1) <0.001 Median (range) 39.6 ( ) 50.4 ( ) 28.2 ( ) HGPIN at previous Bx (51/28/23) 17 (33) 13 (46) 4 (17) No. Bx cores at follow-up (27/15/12) 15.9 (5.5) 13.6 (5.4) 18.8 (4.2) Median (range) 18 (4 21) 12 (4 21) 21 (10 21) PATIENTS AND METHODS The initial study was conducted as described by Haese et al. [6]; 463 men with one or two Bx scheduled for a repeat Bx were included. In selected men with a PCA3 score of 20 and a repeat Bx, a follow-up Bx was taken. The main criteria for selecting men for follow-up Bx were the presence of high-grade prostate intraepithelial neoplasia (HGPIN) or atypical small acinar proliferation (ASAP) at the previous Bx, and a persistently elevated or rising serum total PSA level. Additional criteria were a suspicious DRE, suspicious imaging results, low %free PSA and the patient s request for a Bx. As the urologist and the patient were unaware of the study PCA3 score, the selection for follow-up was independent of the PCA3 score. Five of the six centres that participated in the initial study were involved in this follow-up analysis. Bx were taken by an experienced physician according to the specific criteria of each centre, and specimens were evaluated by the pathologist at each site. Clinical characteristics were assessed immediately before the follow-up Bx. In men undergoing a radical prostatectomy (RP) the RP specimen was analysed pathologically. The demographics and clinical characteristics were summarized for all men with a followup Bx, and for men with Bx and Bx+ separately; these two groups were compared for statistically significant differences. The demographics and clinical characteristics of men in the initial study with a PCA3 score of 20 and a repeat Bx+ vs Bx were compared for statistically significant differences, as were men with a PCA3 score of 35. Also, the characteristics of men in the initial study with a repeat Bx and a PCA3 score of 20 vs <20 were compared for statistically significant differences, and again this was also done for those with a PCA3 score of 35 vs <35. The PCA3 score thresholds were chosen based on the results of clinical studies in which a PCA3 score threshold of 35 provided the optimum balance between sensitivity and specificity for detecting prostate cancer, and a PCA3 score threshold of 20 was suggested as the best option for both avoiding unnecessary Bx and minimizing the risk of missing significant cancer [5,6]. RESULTS Of the 444 evaluable men, 155 (35%) had a PCA3 score of 20 and a repeat Bx ; 51 of these men had one or more follow-up Bx. The risk of having a follow-up Bx in patients with a repeat Bx and a PCA3 score of 20 was 33%. The (last) follow-up Bx was taken after a mean (SD, median, range) of 10.4 (6.2, 10.3, ) months). Table 1 shows the characteristics of the 51 men who had a follow-up Bx. In all, 82% of men had two Bx and 18% had three Bx; 28 (55%) of the follow-up Bx showed prostate cancer. Considering a 33% probability of having a follow-up Bx, the risk of a cancer diagnosis was 18.1%. The mean (and median) PCA3 score was statistically significantly higher in men with a follow-up Bx+ than Bx. For 35 men, information on the PSA level was available, and was comparable between men with a follow-up Bx+ and Bx. Seventeen men (33%) had HGPIN on a previous Bx; 13 (76%) had a follow-up Bx+ and four (24%) a Bx (P = 0.029). For 23 men with a follow-up Bx+, the Gleason score was known; 17 (74%) had Gleason score 6 and six (26%) Gleason score 7. For seven men having a follow-up Bx+ information on the subsequent RP was known (Table 2). One patient had pt2a disease, one pt3a and one pt4; all others were pt2c. Four of these cancers were Gleason 7. The subanalysis of initial study data comprised 463 men, of whom 28% had a repeat Bx+; 259 had a PCA3 score of 20, of whom 94 (36%) had a repeat Bx+; 154 had a PCA3 score of 35 of whom 60 (39%) had a repeat Bx+ (Table 3). Men with a PCA3 score of 20 (and a PCA3 score of 35) and a repeat Bx+ had a statistically significantly higher PSA level, a higher PSA density (PSAD), and more often a suspicious DRE than men with a Bx (Table 3). Men with a PCA3 score of 35 and a repeat Bx+ had a statistically significantly higher PCA3 score than those with a Bx. Men with a repeat Bx and a PCA3 score of 20 had a statistically significantly higher PCA3 score and age, and more often had HGPIN, than men with a PCA3 score of <20 (Table 4). This was also the case in the group of men with a repeat Bx and a PCA3 score of 35 vs <35. DISCUSSION In the present sample of men with a PCA3 score of 20 and two or more repeat Bx, the risk of an additional follow-up Bx within another 10 months was 33%, and 55% had a JOURNAL COMPILATION 2010 BJU INTERNATIONAL 1139

3 REMZI ET AL. TABLE 2 Characteristics of the seven men who had a follow-up Bx+ and were treated with RP Characteristic Patient Total PSA, ng/ml Free PSA, ng/ml Suspicious DRE Y N N N N N Y PCA3 score Follow-up Bx Total cores No. of +ve cores Gleason score 6 (3 + 3) 6 (3 + 3) 6 (3 + 3) 6 (3 + 3) 6 (3 + 3) 6 (3 + 3) 6 (3 + 3) Clinical stage T1c T1c T1c T1c RP Gleason score 7 (3 + 4) 6 (3 + 3) 7 (3 + 4) 7 (3 + 4) 6 (3 + 3) 6 (3 + 3) 7 (4 + 3) RP pathological stage T3aN0M0 T2aN0 T2cNXMX T4NXMX T2cN0MX T2cN0MX T2cN0M0 RP +ve margins, yes/no N N N N N TABLE 3 Clinical characteristics of men with a PCA3 score of 20 or 35 and Bx+ vs Bx Mean (SD) or PCA3 score 20 PCA3 score 35 median, (n) Bx+ (94) Bx (165) P Bx+ (60) Bx (94) P Age, years (94/163/60/93) 66.4 (6.3) 65.1 (6.3) 0.116* 67.0 (6.4) 65.9 (6.5) 0.300* Total PSA, ng/ml (93/162/59/92) 11.4 (12.6) 8.0 (5.1) (14.9) 8.4 (5.9) PSAD, ng/ml/ml (90/157/57/90) 0.26 (0.32) 0.16 (0.10) (0.37) 0.16 (0.10) PCA3 score 82.5 (100.8) 61.5 (80.6) (115.1) 87.3 (99.3) Suspicious DRE, n (%) (93/164/60/93) 30 (32) 27 (17) (37) 18 (19) TABLE 4 The clinical characteristics of men with repeat Bx and a PCA3 score of 20 vs <20 and those with a PCA3 score of 35 vs <35 Mean (SD) or PCA3 score PCA3 score median, (n) 20 (165) <20 (170) P 35 (94) <35 (241) P Age, years (170/163/240/93) 65.1 (6.3) 62.4 (6.6) <0.001* 65.9 (6.5) 62.9 (6.4) <0.001* Total PSA, ng/ml (170/162/240/92) 8.0 (5.1) 8.0 (5.5) (5.9) 7.9 (5.0) PSAD, ng/ml/ml (165/157/232/90) 0.16 (0.10) 0.15 (0.08) (0.10) 0.15 (0.09) PCA3 score 61.5 (80.6) 10.2 (4.9) < (99.3) 15.3 (9.1) < Suspicious DRE, n (%) (169/164/240/93) 27 (17) 22 (13) (19) 31 (13) HGPIN, n (%) 46 (28) 30 (18) (33) 45 (19) follow-up Bx+. Of men with a first Bx, 10 30% have a repeat Bx+ and detection rates are even lower for a third or fourth Bx [10 12]. In the present sample the repeat Bx+ rate was substantially higher, suggesting that men with an elevated PCA3 score have a greater probability of a repeat Bx+, and an elevated/high PCA3 score can predict the presence of prostate cancer. If, of the 60% of men who had a PCA3 score 35 and a repeat Bx in the initial study [6], a third have a follow-up Bx, of whom 55% would be positive, the total percentage of men with a PCA3 score 35 and who have prostate cancer would increase to 50%. However, it must be considered that this was a selected sample of men. There might be bias, in that particularly those men were included who had a high probability of a follow-up Bx+ based on the presence of risk factors for prostate cancer. The mean PSA level in these 1140 JOURNAL COMPILATION 2010 BJU INTERNATIONAL

4 FOLLOW-UP OF MEN WITH AN ELEVATED PCA3 SCORE AND A NEGATIVE BIOPSY men was 7.2 ng/ml and 33% had HGPIN on a previous Bx, suggesting that this was indeed a selected sample at increased risk of prostate cancer. Therefore the Bx+ rate might have been higher than in a population in whom a routine follow-up Bx would have been taken. However, it was also shown that the PCA3 score was statistically significantly higher in men with a follow-up Bx+ than in those Bx. Moreover, the subanalysis showed that men with a PCA3 score of 35 and a repeat Bx+ had a statistically significantly higher PCA3 score than those with a Bx. This confirms the observation that men with a high PCA3 score have an increased probability of a repeat Bx+. This is in accordance with previous studies in which the PCA3 score was also statistically significantly higher in men with a repeat Bx+ than Bx, and in which an increasing PCA3 score corresponded with an increasing probability of a repeat Bx+ [5,6]. The PCA3 assay might thus aid the urologist and patients in guiding the decision as to whether an immediate repeat Bx is needed or can be delayed, and might as such prevent unnecessary Bx. A community-based study reported the good performance of the PCA3 assay in clinical practice [13]. The optimum PCA3 score threshold for use in clinical practice is under discussion [6,14,15]. The PCA3 score is a continuous variable; an increasing PCA3 score corresponds to an increasing probability of a (repeat) Bx+ [5,6]. However, for use in clinical practice the following PCA3 score thresholds can be proposed for use in combination with other risk factors, such as the PSA level, age, prostate volume, and family history of prostate cancer: (i) a PCA3 score of 35, the threshold with the best diagnostic accuracy (balance between specificity and sensitivity) which avoids 67% of repeat Bx [5,6]; (ii) a PCA3 score of 20, a threshold to assure a very low risk of missing clinically significant disease (9%) while still avoiding 44% of repeat Bx [6]; (iii) a PCA3 score of 50, a threshold to assure a high probability of detecting (significant) cancer on (repeat) Bx or scheduling at short notice an extended Bx if the Bx is negative [8,9]. Further research should be conducted to confirm the optimum PCA3 score thresholds for the different patient populations in clinical practice. The subanalysis showed that in men with a repeat Bx and an elevated/high PCA3 score, HGPIN was present more often than in those with a low PCA3 score. This suggests that an elevated/high PCA3 score but a repeat Bx might indicate HGPIN. A previous study showed that the PCA3 score in men with a repeat Bx was statistically significantly higher in those with HGPIN than in those without [6]. Another study showed that PCA3 mrna is expressed by >90% of HGPIN tissue [16]. Although this needs to be further evaluated, these data suggest that an elevated/high PCA3 score might identify men with prostate cancer or those with HGPIN (at risk of developing the disease). In the follow-up of men with an elevated PCA3 score and a previous Bx, more men with HGPIN at a previous Bx had a follow-up Bx+ than a Bx, suggesting that the combination of an elevated/high PCA3 score and the presence of HGPIN is predictive of prostate cancer. However, a potential overestimation of the follow-up Bx+ rate might have occurred due to selection bias. There is an ongoing debate as to whether the presence of HGPIN is associated with an increased risk of prostate cancer at repeat Bx and further investigation of this matter is necessary [17 19]. Men with an elevated/high PCA3 score and a repeat Bx+ had a higher PSA level, a higher PSAD, and more often a suspicious DRE, than men with a Bx, which could therefore be considered prognostic factors for a Bx+. A high PSA level and suspicious DRE are wellestablished prognostic factors for a Bx+ [20]. Examination of RP specimens showed a significant upgrading and upstaging from Bx to RP specimens. Nearly a third of patients with prostate cancer will have a significant Gleason score upgrading between Bx and RP [21]. In conclusion, about a third of men with a PCA3 score of 20 and a repeat Bx will have a third (or fourth) Bx independently of the PCA3 score within 10 months of follow-up. In 55% of these patients the Bx shows prostate cancer. While considering the possibility of selection bias, the high follow-up Bx+ rate in men with an elevated PCA3 score and a previous Bx suggests that an elevated/high PCA3 score can predict the presence of (HGPIN and) prostate cancer. This is confirmed by the observation that men with a (followup) Bx+ had a higher PCA3 score than men with a Bx. This contributes to the increasing scientific evidence that the PCA3 assay is a useful additional tool (to age, PSA level, PSAD and DRE) in clinical practice for guiding Bx decisions, and can as such prevent a considerable proportion of unnecessary Bx. Men with a high PCA3 score but a Bx containing HGPIN might be followed more intensively to reassure them that potentially aggressive cancers are not missed. ACKNOWLEDGEMENTS The authors are grateful to Ismar Healthcare NV for their assistance in writing of the manuscript. The publication of this article was supported by General-Probe Incorporated. CONFLICT OF INTEREST Alexandre de la Taille and Arnulf Stenzl are study investigators for General-Probe Incorporated; Hein van Poppel is a paid consultant for General-Probe Incorporated; Michael Marberger is a paid consultant and study investigator for MSD, GSK, G-P Pharm and R Wolf Medical Instruments. REFERENCES 1 Schröder FH, Hugosson J, Roobol MJ et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med 2009; 360: Bussemakers MJ, van Bokhoven A, Verhaegh GW et al. DD3: a new prostatespecific gene, highly overexpressed in prostate cancer. Cancer Res 1999; 59: Hessels D, Klein Gunnewiek JM, van Oort I et al. DD3 PCA3 -based molecular urine analysis for the diagnosis of prostate cancer. Eur Urol 2003; 44: 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: Marks LS, Fradet Y, Deras IL et al. PCA3 molecular urine assay for prostate cancer in men undergoing repeat biopsy. Urology 2007; 69: Haese A, de la Taille A, Van Poppel H et al. Clinical utility of the PCA3 urine assay in European men scheduled for repeat biopsy. Eur Urol 2008; 54: Deras IL, Aubin SM, Blase A et al. PCA3: a molecular urine assay for predicting prostate biopsy outcome. J Urol 2008; 179: JOURNAL COMPILATION 2010 BJU INTERNATIONAL 1141

5 REMZI ET AL. 8 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: Whitman EJ, Groskopf J, Ali A et al. PCA3 score before radical prostatectomy predicts extracapsular extension and tumor volume. J Urol 2008; 180: 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: Seitz C, Palermo S, Djavan B. Prostate biopsy. Minerva Urol Nefrol 2003; 55: Djavan B, Ravery V, Zlotta A et al. Prospective evaluation of prostate cancer detected on biopsies1,2,3 and4: when should we stop? J Urol 2001; 166: Shappell SB, Fulmer J, Arguello D, Wright BS, Oppenheimer JR, Putzi MJ. PCA3 urine mrna testing for prostate carcinoma. Patterns of use by community urologists and assay performance in reference laboratory setting. Urology 2009; 73: Chun FK, de la Taille A, Van Poppel H et al. Prostate cancer gene3 (PCA3): development and internal validation of a novel biopsy nomogram. Eur Urol 2009; 56: Hessels D, Schalken JA. The use of PCA3 in the diagnosis of prostate cancer. Nat Rev Urol 2009; 6: Popa I, Fradet Y, Beaudry G, Hovington H, Beaudry G, Tetu B. Identification of PCA3 (DD3) in prostatic carcinoma by in situ hybridization. Mod Pathol 2007; 20: Brawer MK. Prostatic intraepithelial neoplasia: an overview. Rev Urol 2005; 7 (Suppl. 3): S Schoenfield L, Jones JS, Zippe CD et al. The incidence of high-grade prostatic intraepithelial neoplasia and atypical glands suspicious for carcinoma on firsttime saturation needle biopsy, and the subsequent risk of cancer. BJU Int 2007; 99: Epstein JI, Herawi M. Prostate needle biopsies containing prostatic intraepithelial neoplasia or atypical foci suspicious for carcinoma: implications for patient care. J Urol 2006; 175: European Association of Urology. EAU guidelines on prostate cancer Available at: fileadmin/tx_eauguidelines/2009/full/ Prostate_Cancer.pdf. Accessed July Chun FKH, Briganti A, Shariat SF et al. Significant upgrading affects a third of men diagnosed with prostate cancer: predictive nomogram and internal validation. BJU Int 2006; 98: Correspondence: Mesut Remzi, Medical University of Vienna, Department of Urology, Waehringer Guertel 18-20, 1090 Vienna, Austria. mesut.remzi@meduniwien.ac.at Abbreviations: PCA3, Prostate CAncer gene 3; Bx (+)( ), biopsy (positive) (negative); HGPIN, high-grade prostatic intraepithelial neoplasia; RP, radical prostatectomy; ASAP, atypical small acinar proliferation JOURNAL COMPILATION 2010 BJU INTERNATIONAL