Fisher s exact test for contingency tables. A two-tailed p-value <0.05 was accepted as statistically significant.

BJUI A prospective, randomized trial comparing the Vienna nomogram to an eight-core prostate biopsy protocol Angus Lecuona and Chris F. Heyns Department of Urology, Tygerberg Hospital and University of Stellenbosch, Western Cape, South Africa Accepted for publication 21 July 2010 Study Type Diagnostic (RCT) Level of Evidence 1b OBJECTIVE To compare prostate cancer detection rates using the Vienna nomogram versus an 8-core prostate biopsy protocol. To compare the complication rates of transrectal prostate biopsy in the two groups. PATIENTS AND METHODS In a prospective randomized trial, men with a serum PSA 2.5 ng/ml were stratified according to serum PSA (I = PSA 2.5 10; II = PSA 10.1 30; III = PSA 30.1 50 ng/ml) and were then randomized to group A (number of cores determined according to the Vienna nomogram) or group B (8-core prostate biopsy). Statistical analysis was performed using Student s t-test for parametric data, Mann- Whitney test for nonparametric data and What s known on the subject? and What does the study add? Several studies have shown that increasing the number of prostate biopsy cores will increase the detection rate of prostate cancer, but also risks overdiagnosing insignificant cancer, particularly in the elderly. Our study suggests that there is no significant advantage in using the Vienna nomogram to determine the number of prostate biopsies to be taken, compared to an eight-core biopsy protocol. Fisher s exact test for contingency tables. A two-tailed p-value <0.05 was accepted as statistically significant. RESULTS In the period July 2006 to July 2009, 303 patients were randomized to group A (n = 152) or group B (n = 151). There were no significant differences in serum PSA, prostate volume, PSA density or post-biopsy complications between the groups. The cancer detection rate was lower in group A than in group B for the whole study cohort (35.5% vs 38.4%), for those with PSA < 10 ng/ml (28.1% vs 33%) and for those with prostate volume >50 ml (22% vs 25.8%). These differences were not statistically significant (NSS). CONCLUSION These findings suggest that there is no significant advantage in using the Vienna nomogram to determine the number of prostate biopsy cores to be taken, compared to an 8-core biopsy protocol. KEYWORDS prostate cancer, prostate biopsy, Vienna nomogram INTRODUCTION With the introduction of PSA measurements and the availability of TRUS the optimal number of prostate biopsy cores has become controversial. Several studies have been performed to determine the best prostate biopsy protocol for detecting clinically significant cancer. Increasing the number of biopsy cores increases detection rates [1,2], but risks over-diagnosing insignificant cancers, and it may increase the risk of complications, whereas taking too few biopsy cores will lead to significant cancers being missed. Prostate volume is an important consideration, because too few biopsies in large prostates may fail to diagnose significant cancer [3 5]. Studies have shown that standard sextant (six-core) biopsies may be inadequate [1,3,5 7], because repeat biopsies yield cancer in a further 10 27% of cases [3,6]. A recent study from Vienna [8] proposed using a nomogram to determine the optimal number of TRUS-guided biopsy cores in men with a serum PSA level of 2 10 ng/ml, by taking into account the patient s age and prostate volume. This study compared a group of men who underwent TRUS-guided prostate biopsies according to the Vienna nomogram with a historical group who underwent eight-core biopsies. The group that underwent biopsies according to the Vienna nomogram yielded a significantly higher prostate cancer detection rate of 36.7%, compared with the historical eightcore biopsy group, in which 22% were diagnosed with prostate cancer [8]. The study from Vienna has a few shortcomings. First, it was not a prospective, randomized trial, but compared prospective use of the nomogram with eight-core biopsies obtained in a historical comparison group. The two study groups were not completely comparable with regard to mean age and serum PSA. Second, the study did not compare the risk of complications or adverse events in the two groups, i.e. it is not known whether the greater number of biopsy cores required by the Vienna nomogram results in a higher complication rate. 204 2010 108, 204 208 doi:10.1111/j.1464-410x.2010.09887.x

RCT COMPARING THE VIENNA NOMOGRAM TO 8-CORE PROSTATE BIOPSIES Age (years) Prostate volume (ml) 50 51 60 61 70 70 0 30 8 8 8 6 31 40 12 10 8 6 41 50 14 12 10 8 51 60 16 14 12 10 61 70 18 16 14 12 >71 18 18 16 14 TABLE 1 The Vienna nomogram [8]: number of biopsy cores based on patient age and total prostate volume were performed under local anaesthesia (periprostatic infiltration of 2% lignocaine). A Toshiba diagnostic ultrasound machine with a 6-MHz transrectal probe was used to instil the anaesthetic and to perform the prostate biopsies. Antibiotic prophylaxis was given 1 h before biopsy using 1 g ciprofloxacin followed by two 500-mg doses at 12-h intervals. TABLE 2 Analysis of whole cohort All study subjects Vienna nomogram Eight-core biopsy P value Patients (n) 152 151 Age (years) 65.1 (45 82) 63.4 (40 81) 0.0435* PSA level (ng/ml) 9.4 (2.2 46) 9.2 (2.6 48) 0.8109 Prostate volume (ml) 47.4 (11 220) 51.5 (10 194) 0.2126 PSA density 0.27 0.23 0.2900 Biopsy cores (n) 10.2 (6 18) 8 Complications, n (%) Any 45 (48.4) 43 (48.9) 1.000 None 48 (51.6) 45 (51.1) Fever 4 (0.04) 5 (0.06) Urinary retention 1 (0.01) 0 (0) DRE findings Benign 124 (83.2) 115 (76.2) 0.1518 Suspicious or malignant 25 (16.8) 36 (23.8) Histology No prostate cancer 98 (64.5) 93 (61.6) Cancer detected 54 (35.5) 58 (38.4) 0.6351 Data are shown as mean and range (minimum and maximum values), unless otherwise stated. *Statistically significant; fever requiring systemic antibiotics; urinary retention secondary to gross haematuria requiring catheterization and irrigation. Statistical analysis was performed using GRAPHPAD INSTAT software with the unpaired Student s t test for parametric data, Mann Whitney U test for non-parametric data and Fisher s exact test for contingency tables. A two-tailed P-value <0.05 was accepted as significant with a power of 80%. Data are shown as mean and range (minimum and maximum values). Sample size calculation was performed using GRAPHPAD STATMATE software. The study protocol was approved by the Human Research Ethics Committee of the Faculty of Health Sciences of the University of Stellenbosch. RESULTS In the period July 2006 to July 2009, 926 patients underwent prostate biopsies at our unit. After the exclusion criteria were applied, 303 patients were randomized into group A (number of biopsy cores determined according to the Vienna nomogram, n = 152) or group B (eight-core biopsy, n = 151). The aim of our prospective, randomized trial was to determine if the Vienna nomogram would yield a higher cancer detection rate than an eight-core biopsy protocol, without incurring a higher complication rate, in a study population referred to a tertiary urology unit. PATIENTS AND METHODS Patients included in this study were men with a PSA level >2.5 ng/ml undergoing TRUSguided prostate biopsy in the Department of Urology, Tygerberg Hospital, University of Stellenbosch, South Africa. Exclusion criteria included previous prostate biopsy or surgery, previous diagnosis of prostate cancer, a history of urinary retention, previous histological evidence of prostatitis and confirmed urinary tract infection. Patients were stratified according to their serum PSA measurement into those with a low, intermediate or high risk of having prostate cancer (group I, PSA level 2.5 10 ng/ ml; group II, 10.1 30 ng/ml; group III, 30.1 50 ng/ml) and then randomized using computer-generated random numbers: group A underwent TRUS-guided prostate biopsies with the number of cores determined according to the Vienna nomogram (Table 1) and group B underwent an eight-core TRUSguided prostate biopsy. All biopsy cores in both groups were taken from the lateral peripheral zone, evenly distributed from cranial to caudal. The prostate volume was measured by TRUS and biopsies The results of all study subjects and relevant subgroups are summarized in Tables 2 and 3. The mean patient age was slightly but significantly higher in the Vienna nomogram group, but there were no significant differences in serum PSA level, prostate volume, PSA density or post-biopsy complications between the groups. The overall cancer detection rate was higher in the eight-core group (38.4%) than in the Vienna nomogram group (35.5%) but this difference was not significant (NS) (Table 2). The proportion of patients with DRE findings suspicious of malignancy was slightly higher in the eight-core group (23.8% vs 16.8%, NS). This discrepancy may be ascribed to the fact that study subjects were not stratified according to DRE findings before randomization. When controlling for this difference in the subgroup 2010 205

LECUONA and HEYNS of subjects with PSA level <10 ng/ml and benign DRE the cancer detection rate was still higher in the eight-core group (26.9% vs 21.6%; NS). Analysis of different subgroups (PSA level <10 ng/ml, DRE benign plus PSA level <10 ng/ml, prostate volume >50 ml, age <60 or >70 years) showed no significant differences in the cancer detection rates between the Vienna nomogram and eight-core biopsy groups (Table 3) There were no significant differences between the two groups (overall or in any subgroup) as far as complications experienced were concerned. Only minor complications were reported, with self-limiting macroscopic haematuria being the most common. Other complications included haematospermia, haematochezia, dysuria, painful ejaculation and fever. DISCUSSION McNeal [9] described the different zones of the prostate in 1969 and commented on the origin and spread of prostate cancer in the peripheral zone. Hodge et al. [10] introduced systematic sextant biopsy of the prostate under TRUS guidance in 1989. These six biopsies were distributed evenly, from cranial to caudal in the mid-lobar, parasagittal plane. Stamey [11] suggested placing these biopsy cores more laterally to better sample the peripheral zone. Several authors have shown that biopsies of the transition zone do not increase cancer detection rates [12]. In 2005, Remzi and colleagues developed the Vienna nomogram (Table 1), using data from the European Prostate Cancer Detection study [3,8]. The nomogram indicates the optimum number of cores to be taken during prostate biopsy, based on patient age and total prostate volume. The objective of the nomogram is to increase prostate cancer detection rates in younger men and those with larger prostates (so decreasing the number of repeat biopsies) and to avoid detection of insignificant cancers in older patients. In a prospective study of 502 men (PSA level 2 10 ng/ml) the overall cancer detection rate was 36.7% [8]. The mean number of cores taken, patient age, total prostate volume and TABLE 3 Analysis of subgroups Vienna nomogram Eight-core biopsy P value Subgroup PSA level <10 ng/ml Age (years) 64.9 (45 79) 62.7 (40 78) 0.0163* PSA level (ng/ml) 5.8 (2.2 9.9) 5.7 (2.6 9.7) 0.7534 Prostate volume (ml) 47.4 (11 220) 46.0 (10 194) 0.9959 PSA density 0.16 (0.03 0.51) 0.16 (0.03 0.57) 0.7812 DRE findings, n (%) Benign 102 (91.1) 93 (83.0) Suspicious or malignant 10 (8.9) 19 (17.0) 0.1102 Biopsy cores (n) 10.3 (6 18) 8 No prostate cancer 82 (71.9) 75 (67) Cancer detected 32 (28.1) 37 (33) 0.4710 Subgroup DRE findings benign and PSA level <10 ng/ml No prostate cancer 80 (78.4) 68 (73.1) Cancer detected 22 (21.6) 25 (26.9) 0.4065 Subgroup: prostate volume >50 ml No prostate cancer 39 (78) 46 (74.2) Cancer detected 11 (22) 16 (25.8) 0.6642 Complications, n (%) Any 16 (59.3) 14 (37.8) None 11 (40.7) 23 (62.2) 0.1286 Subgroup: age <60 years No prostate cancer 26 (63.4) 32 (61.5) Cancer detected 15 (36.6) 20 (38.5) 0.1000 Biopsy cores (n) 10.8 8 Subgroup: age >70 years No prostate cancer 29 (67.4) 17 (53.1) Cancer detected 14 (32.6) 15 (46.9) 0.2376 Biopsy cores (n) 8.5 8 Data are shown as mean and range (minimum and maximum values) unless otherwise stated. *Statistically significant. PSA measurement were 10, 65 years, 43 ml and 5.5 ng/ml, respectively. On multivariate analysis, prostate volume was not an independent predictor of prostate cancer detection, so eliminating the sampling error caused by large prostates by taking more cores. This prospective study also compared well against data from the European Prostate Cancer Detection study [3], where the cancer detection rate was 22% on first biopsy and 10% on repeat biopsy, using an eight-core biopsy. During the study, 926 patients underwent prostate biopsies at our unit, of whom 303 were included. The rationale for excluding patients with UTI, urinary retention or previous prostatic procedures was to exclude those with a falsely raised PSA level. The fact that 623 patients were excluded from the study reflects our referral patient profile, where most men present with advanced prostate cancer disease or complications of BPH such as UTI or retention. In the present study the cancer detection rate was higher in the eight-core biopsy group than in the Vienna nomogram group for the whole study cohort (38.4% vs 35.5%), for those with a PSA reading <10 ng/ml (33% 206 2010

RCT COMPARING THE VIENNA NOMOGRAM TO 8-CORE PROSTATE BIOPSIES versus 28.1%), for those with prostate volume >50 ml (25.8% vs 22%) and for those with both PSA level <10 ng/ml and clinical BPH (26.9% vs 21.6%) (Tables 2 and 3). However, these differences were small and not significant. The mean PSA values of the two groups were 9.4 ng/ml (Vienna nomogram group) vs 9.2 ng/ml (Control group), and reflect all patients with PSA levels between 2.5 and 50 ng/ml. Most patients had a PSA level <10 ng/ml (n = 226; 75%) with a mean of 5.8 ng/ml in the Vienna nomogram group compared with 5.7 ng/ml in the control group. The two groups are comparable in all aspects except age, for which the Vienna nomogram group has a mean age of 64.9 years vs 62.7 years in the control group. This subgroup (PSA level 2.5 10 ng/ml) is also comparable to the patients in the original Vienna nomogram article (Table 3). Findings on DRE were more often suspicious of malignancy in the eight-core group than in the Vienna nomogram group for the whole cohort and subgroups. This discrepancy between the two groups occurred by chance, despite stratification according to serum PSA, and probably explains the higher cancer detection rate in group B. However, even when controlling for this, the eightcore biopsy protocol still detected more prostate cancer than the Vienna nomogram group, although the difference was not significant. Several authors have shown that increasing the number of prostate biopsies increases the cancer detection rate [1,13 15]. The use of extended biopsy protocols eventually led to saturation biopsy schemes [6,7], which involve taking 22 24 cores in one sitting. These biopsy schemes particularly target the lateral peripheral zone and apex of the prostate. As an initial biopsy strategy, Jones et al. [16] did not find saturation biopsies useful. In a cohort of 139 patients who underwent saturation biopsies (24 cores), 44.6% were diagnosed with prostate cancer, compared with a 51.7% detection rate in a historical comparison group (10 cores) consisting of 87 patients. Guichard et al. [17] published a series of 1000 consecutive patients. The cancer detection rates of the 6-, 12-, 18- and 21-core biopsy strategies were 31.7%, 38.7%, 41.5% and 42.5%, respectively, in patients who underwent first-time biopsies. Other authors [18] had similar results of a 40% cancer detection rate, using saturation biopsy techniques for firsttime biopsies. However, saturation biopsies have proven effective when used in the repeat biopsy scenario [19,20]. Recently, concerns have been raised regarding the potential under-sampling of larger glands by the traditional sextant biopsy protocol [3 5]. One of the aims of the Vienna nomogram is to increase prostate cancer detection by increasing the number of cores taken at biopsy of larger prostates (Table 1). In our study, in the subgroup of patients with a prostate volume >50 ml, prostate cancer detection was higher in the eight-core group (25.8%) compared with the Vienna nomogram group (22%; mean number of cores taken 14.4), but this difference was not significant (Table 3). Given the high prevalence of histological prostate cancer [21], the increased detection due to (i) structured screening studies, (ii) a greater awareness of prostate cancer among the general public, (iii) a reduction in PSA thresholds for biopsy, (iv) an increase in the number of biopsy cores [22] and (v) the long lead-time of prostate cancer, the risk of overdiagnosis and overtreatment is substantial [23]. Another aim of the Vienna nomogram is to reduce the likelihood of detecting insignificant prostate cancer, by reducing the number of biopsies taken from older patients. Criteria which could indicate clinically insignificant prostate cancer, and which could be used in active surveillance of patients, include: a Gleason score 6 (no pattern 4 or 5 disease), PSA level 10 ng/ml, clinical stage T1 or T2a disease, PSA density 0.15, percentage positive biopsy cores <33%, <50% involvement of any biopsy core and stable PSA kinetics [24 26]. Using the above definition, only 14 (4.6%) of the 303 patients in our study had insignificant prostate cancer, 5/152 (3.3%) in the Vienna nomogram group and 9/151(6.0%) in the control group. Unfortunately, no conclusions can be drawn from so few patients, except that insignificant prostate cancer is extremely rare in our referral population. In our study, more patients were diagnosed with prostate cancer in the age group >70 years, using the eight-core biopsy protocol compared with the Vienna nomogram (46.9% vs 32.6%; NS). The mean number of cores taken with the Vienna nomogram in patients older than 70 years was 8.5. Increasing the number of prostate biopsy cores raises the concern of increased morbidity. TRUS-guided biopsies of the prostate are associated with several minor but self-limiting complications and are usually well tolerated by patients [27]. In our study the complication rates in the two groups overall were almost identical (Table 2). Patients in the subgroup with prostate volume >50 ml had a higher complication rate in the Vienna nomogram group (average of 14.4 cores) than in the eight-core group (59.3% vs 37.8, NS) (Table 3). These complications were minor, consisting mostly of self-limiting macroscopic haematuria. Serious complications were rare: four patients in the Vienna nomogram group and five patients in the control group developed post-biopsy fever requiring parenteral antibiotic treatment at their local hospital or general practitioner. All patients had an uneventful recovery. One patient in the Vienna nomogram group (16 cores) developed urinary retention secondary to gross haematuria that required catheterization and irrigation for 2 days (Table 2). The main limitation of this study is that stratification before randomization was performed according to serum PSA level alone and did not include DRE, with the result that, by chance, a greater proportion of subjects with DRE findings suspicious of malignancy were included in group B, but controlling for this discrepancy did not reveal a significant difference in cancer detection. The main objective of this prospective, randomized study was to determine whether the Vienna nomogram (which was originally developed in a screening setting) would be useful, when applied in a urology referral unit. Overall, when comparing the results from biopsies taken using the Vienna nomogram versus an eight-core biopsy protocol, no significant differences were noted in the cancer detection or complication rates. CONFLICT OF INTEREST None declared. 2010 207

LECUONA and HEYNS REFERENCES 1 Eskew LA, Bare RL, McCullough DL. Systematic 5 region prostate biopsy is superior to sextant method for diagnosing carcinoma of the prostate. J Urol 1997; 157: 199 203 2 Ravery V, Goldblatt L, Royer B, Blanc E, Toublanc M, Boccon-Gibod L. Extensive biopsy protocol improves the detection rate of prostate cancer. J Urol 2000; 164: 393 6 3 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 9 4 Uzzo RG, Wei JT, Waldbaum RS, Perlmutter AP, Byrne JC, Vaughn ED Jr. The influence of prostate size on cancer detection. Urology 1995; 46: 831 6 5 Remzi M, Djavan B, Wammack R et al. Can total and transition zone volume of the prostate determine whether to perform a repeat biopsy? Urology 2003; 61: 161 6 6 Borboroglu PG, Comer SW, Riffenburgh RH, Amling CL. Extensive repeat transrectal ultrasound guided prostate biopsy in patients with previous benign sextant biopsies. J Urol 2000; 163: 158 62 7 Stewart CS, Leibovich BC, Weaver AL, Lieber MM. Prostate cancer diagnosis using a saturation needle biopsy technique after previous negative sextant biopsies. J Urol 2001; 166: 86 91; discussion 91 2 8 Remzi M, Fong YK, Dobrovits M et al. The Vienna nomogram: validation of a novel biopsy strategy defining the optimal number of cores based on patient age and total prostate volume. J Urol 2005; 174: 1256 61 9 McNeal JE. Origin and development of carcinoma of the prostate. Cancer 1969; 23: 24 34 10 Hodge KK, McNeal JE, Terris MK, Stamey TA. Random systematic versus directed ultrasound guided transrectal core biopsies of the prostate. J Urol 1989; 142: 71 5 11 Stamey TA. Making the most out of six systematic sextant biopsies. Urology 1995; 45: 2 12 12 Pelzer AE, Bektic J, Berger AP et al. Are transition zone biopsies still necessary to improve prostate cancer detection? Results from the Tyrol screening project. Eur Urol 2005; 48: 916 21 13 Philip J, Ragavan N, Desouza J, Foster CS, Javle P. Effect of peripheral biopsies in maximising early prostate cancer detection in 8-, 10- or 12-core biopsy regimens. BJU Int 2004; 93: 1218 20 14 Presti JC, Chang JJ, Bhargava V, Shinohara K. The optimal systematic prostate biopsy scheme should include 8 rather than 6 biopsies: results of a prospective clinical trial. J Urol 2000; 163: 163 7 15 Inahara M, Suzuki H, Kojima S et al. Improved prostate cancer detection using systematic 14-core biopsy for large glands with normal digital rectal examination findings. Urology 2006; 68: 815 19 16 Jones JS, Patel A, Schoenfield L, Rabets JC, Zippe CD, Magi-Galluzzi C. Saturation technique does not improve cancer detection as an initial prostate biopsy strategy. J Urol 2006; 175: 485 8 17 Guichard G, Larré S, Gallina A et al. Extended 21-sample needle biopsy protocol for diagnosis of prostate cancer in 1000 consecutive patients. Eur Urol 2007; 52: 430 5 18 Descazeaud A, Rubin M, Chemama S et al. Saturation biopsy protocol enhances prediction of pt3 and surgical margin status on prostatectomy specimen. World J Urol 2006; 24: 676 80 19 Walz J, Graefen M, Chun FK et al. High incidence of prostate cancer detected by saturation biopsy after previous negative biopsy series. Eur Urol 2006; 50: 498 505 20 Scattoni V, Zlotta A, Montironi R, Schulman C, Rigatti P, Montorsi F. Extended and saturation prostatic biopsy in the diagnosis and characterisation of prostate cancer: a critical analysis of the literature. Eur Urol 2007; 52: 1309 22 21 Sakr WA, Grignon DJ, Crissman JD et al. High grade prostatic intraepithelial neoplasia (HGPIN) and prostatic adenocarcinoma between the ages of 20 69: an autopsy study of 249 cases. In Vivo 1994; 8: 439 43 22 Bangma CH, Roemeling S, Schröder FH. Overdiagnosis and overtreatment of early detected prostate cancer. World J Urol 2007; 25: 3 9 23 Draisma G, Boer R, Otto SJ et al. Lead times and overdetection due to prostatespecific antigen screening: estimates from the European Randomized Study of Screening for Prostate Cancer. J Natl Cancer Inst 2003; 95: 838 9 24 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 25 Dall Era MA, Cooperberg MR, Chan JM et al. Active surveillance for early-stage prostate cancer. Cancer 2008; 112: 1650 9 26 O Donnell H, Parker C. What is low-risk cancer and what is its natural history? World J Urol 2008; 26: 415 22 27 Paul R, Schöler S, van Randenborgh H, Kübler H, Alschibaja M, Busch R. Morbidity of prostatic biopsy for different biopsy strategies: is there a relation to core number and sampling region? Eur Urol 2004; 45: 450 6 Correspondence: Angus Lecuona, Tygerberg Hospital, University of Stellenbosch, Department of Urology, Room 4078 4th Floor, Clinical Building Faculty of Health Sciences Tygerberg Western Cape 7505, South Africa. e-mail: anguslecuona@yahoo.com Abbreviation: NS, not significant. 208 2010