Distant and local recurrence in patients with biochemical failure after prostate brachytherapy

Brachytherapy 7 (2008) 217e222

Distant and local recurrence in patients with biochemical failure after prostate brachytherapy Richard G. Stock*, Jamie A. Cesaretti, Pamela Unger, Nelson N. Stone Departments of Radiation Oncology, Pathology, and Urology, Mount Sinai School of Medicine, New York, NY

ABSTRACT

PURPOSE: To analyze the patterns of failure after the brachytherapy management of localized prostate cancer. METHODS AND MATERIALS: From 1990 to 2008, 2869 patients underwent prostate brachytherapy and 213 experienced a prostate-specific antigen (PSA) failure by the Phoenix definition. Of these 213 patients, 33.5% were low, 18.5% intermediate, and 58% high risk. RESULTS: Of the 119 patients biopsied, 36 (30%) had a least one positive posttreatment biopsy. In univariate and multivariate analyses, PSA doubling time was the most predictive of a positive biopsy. Patients with doubling times <3, O3e6, >6e10, and O10 months had positive biopsy rates of 9%, 18%, 36%, and 42%, respectively ( p 5 0.01). The actuarial rate of remaining free from distant metastases at 10 years was 73%. Patients with PSA doubling times of <3, O3e6, O6e10, and O10 months had freedom from distant metastases rates of 0%, 74%, 78%, and 94.5% at 10 years, respectively ( p!0.0001). In multivariate analysis, PSA doubling time and time to PSA failure were the most significant predictors of developing distant metastases. CONCLUSIONS: About one third of patients harbor a component of local failure and one fourth demonstrate clinical metastases. PSA doubling time can be used to help predict the source of a rising PSA. Ó 2008 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.

Keywords:

Patterns of failure; Prostate brachytherapy; Prostate cancer; PSA doubling time

Introduction Before the prostate-specific antigen (PSA) era, prostate cancer treatment efficacy was determined by objective clinical relapse and cause-specific survival. With the introduction of PSA, both as a screening and monitoring tool, cancer control rates after definitive therapy for prostate cancer began to be reported using freedom from biochemical failure as the primary endpoint. Although the use of a biochemical marker defined an early recurrence of prostate cancer, it failed to identify the location or type of disease recurrence. Investigators have recently focused on the rate of rise in PSA and time of failure as a method of Received 31 January 2008; received in revised form 28 March 2008; accepted 29 April 2008. Conflict of interest statement: The following authors have relationships that might be considered as potential conflicts of interest: Jamie A. Cesarettidconsultant BARD urologic; Nelson N. Stonedownership Prologics, LLC. * Corresponding author. Department of Radiation Oncology, Mount Sinai School of Medicine, 1184 5th Avenue, New York, NY 10025. Tel.: þ1-212-241-7502; fax: þ1-212-410-7194. E-mail address: [email protected] (R.G. Stock).

determining whether the recurrence is local or distant (1e3). The distinction between these two patterns of failure is critical as subsequent therapies depend on the type of recurrence. The pattern of recurrence after a given treatment can shed light on the efficacy of the therapy and on ways in which the treatment can be improved. For example, a surgical procedure that results in a high local failure rate may need to be changed with the incorporation of wider excisions with greater margins to enhance local control. An analysis of a radiation approach, which results in a high local failure rate, may lead to the conclusion that an escalation of dose is needed to enhance local tumor eradication. To shed light on the patterns of failure after prostate brachytherapy, we elected to examine those patients who had experienced a biochemical failure using the Phoenix definition (4). Among these patients, rates of local failure were calculated using posttreatment prostate biopsy. The proportion of these failures that represented distant recurrence was calculated using actuarial methods over a 15-year period. Finally, those factors which predicted for both local and distant relapse were calculated to help identify the pattern of failure for those patients with increasing PSA levels.

1538-4721/08/$ e see front matter Ó 2008 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.brachy.2008.04.002

218

R.G. Stock et al. / Brachytherapy 7 (2008) 217e222

Methods and materials From 1990 to 2008, 2869 patients were treated with lowdose-rate brachytherapy as part of their management for nonmetastatic prostate cancer. In general, staging procedures included bone scans and CT scans of the abdomen/pelvis for intermediate- to high-risk patients. Some low-risk patients were also staged with these tests based on the discretion of the referring urologist. Patients were categorized as low risk if they had PSA <10 ng/ml, stage t2c, Gleason score of 8e10 or two or more intermediate risk features. Of the 2869 patients, 45% were low risk, 22% intermediate risk, and 33% high risk. Of these patients, 213 manifested biochemical failure using the Phoenix definition and are the subject of the following analyses. The disease characteristics for these patients can be found in Table 1. Clinical stage was determined using 1992 American Joint Committee on Cancer (AJCC) staging system (5). These patients with biochemical failure were followed from 1 to 16.7 years (median, 6.8). Treatment All 213 patients were treated with brachytherapy using a real-time ultrasound-guided technique (6). Treatment regimens developed over time so there was overlap in different risk groups being treated by different treatment regimens. Details of the development of these treatment schemas have been previously described (7). Treatments were divided into three main groups: brachytherapy alone (94 patients), brachytherapy and hormonal therapy (41 patients) and trimodality therapy (78 patients) with brachytherapy, hormonal therapy and external beam radiation therapy (EBRT). Table 1 Presenting disease characteristics Factor

No. of patients

Percent

PSA <10 O10e20 O20

114 49 50

53.5 23 23.5

Gleason score 6 7 8e10

116 45 52

54.5 21 24.5

Clinical stage t1b t1c t2a t2b t2c t3

3 57 37 75 27 14

Risk group Low Intermediate High

50 39 124

1 27 17 35 13 7 23.5 18.5 58

Brachytherapy without EBRT (hormonal therapy) was performed using both 125I (prescription dose 160 Gy TG43) (91 patients) and 103Pd (prescription dose 124 Gy NIST 99) (44 patients). In general, 125I was used for patients with Gleason scores of <6 and 103Pd for those with scores >7. Most patients treated with brachytherapy alone were low-risk patients although during the early years of the study period both intermediate- and high-risk patients received implant alone. Hormonal therapy and brachytherapy were used for two main reasons. The first use of hormonal therapy was for downsizing in patients with large prostates (gland sizeO50 cc). It was given for 3 months before implantation and usually 2e3 months postimplant. The second use was as adjuvant therapy with brachytherapy for patients with intermediate- or high-risk features. In this case, the therapy was given for 3 months before and 3 months after implantation (8). Trimodality therapy usually involved 3 months of hormonal therapy followed by a 103Pd brachytherapy implant (76 patients) (prescription dose, 100 Gy NIST 99) or 125I (2 patients) (prescription dose, 120 Gy) and 2 months later EBRT to a dose of 45 Gy. Seminal vesicles were implanted in patients with biopsy positive seminal vesicle disease. The total duration of hormonal therapy was 9 months. In the earlier years of the study, lower implant doses were used with higher external beam doses. EBRT doses ranged from 39.6 to 59.4 Gy (median, 45). Details of this regimen have been previously described (9). EBRT fields were conformal and treated the prostate and seminal vesicles using 1.5e2 cm margins. Overall, when hormonal therapy was used it either involved a lutenizing hormone-releasing hormone analogue alone or in combination with an antiandrogen. Due to changes in disease presentation, treatment philosophy, technique, and experience over time, we created a time-dependant variable that reflected the time sequence of the patients treated. Time course 1 included the first 71 patients treated up to 8/1995. Time course 2 included the second 71 patients treated from 8/1995 through 10/1999. Time course 3 represented the last 71 patients treated from 12/1999 through 2006. To compare delivered doses, all treatment doses were converted into biologically effective dose (BED) values using an alpha/beta ratio of 2. The BED values for the 184 PSA failures ranged from 15 to 278 Gy2 (median, 182). Details of these calculations have been previously described (10). Overall, 197 patients underwent postimplant dosimetry and had BED values calculated. Reasons for no dosimetry were bilateral hip prostheses causing severe metallic interference or patient noncompliance in returning for the CT scan. Endpoints The determination of PSA failure was made using the Phoenix definition (4). PSA doubling times were calculated using first order kinetics. Posttreatment trans-rectal ultrasound-guided prostate biopsies were offered 2 or more years after completion of all therapy (including hormones).

R.G. Stock et al. / Brachytherapy 7 (2008) 217e222

Of the 213 patients, 119 underwent posttreatment biopsy. Reasons for not undergoing biopsy were patient refusal or followup care given by a physician outside of our institution. The biopsies (total 6e12, 3e6 from each side) were repeated yearly if the 2-year biopsy was positive until it reverted to negative or there was clear evidence of biochemical progression. In addition, if biochemical progression occurred after an initial negative 2-year biopsy, repeat biopsies were performed to determine if the patient was experiencing a late local recurrence. All prostate needle biopsies were evaluated by a single genitourinary pathologist (PU). The biopsies were interpreted as negative for tumor, presence of prostate carcinoma without radiation effect on the tumor, or prostatic carcinoma with radiation effect. Radiation effect on the tumor meant the presence of the following histologic features: infiltrative pattern to the malignant glands, malignant glands with either pyknotic or slightly enlarged nuclei, and frequently incomplete gland formation. High molecular weight cytokeratin immunohistochemistry (k903) was performed on all the cases of carcinoma with radiation effect, which displayed the absence of basal cells supporting the carcinoma diagnosis (11, 12). No Gleason score was assigned to the carcinomas with radiation effect. The cases showing prostate carcinoma with no radiation effect on the tumor were given Gleason scores. A positive biopsy was one that had the presence of cancer with or without the presence of radiation effect. A local failure was scored for at least one positive biopsy posttreatment. Distant metastases included disease spread to bones, visceral organs, or lymph nodes outside of the pelvis as documented by imaging studies. Statistics The effect of factors on biopsy was tested using the Pearson chi-square test. Freedom from distant metastases (FFDM) rates was calculated using actuarial methods. Difference in survival rates was tested using the log rank test. The effect of multiple variables on biopsy and distant metastases rates was tested using logistic and Cox regression, respectively (13).

Results

219

PSA failure, time from completion of treatment to biopsy, PSA doubling time, BED, treatment group, treatment time group, risk group, PSA, Gleason score, and clinical stage. The results of these analyses can be found in Table 2. Using univariate analyses, PSA doubling time, treatment group, treatment time group, and time to biopsy significantly predicted for a positive biopsy. A logistic regression analysis was performed using PSA doubling time as a categorical variable separated into 3-month intervals, treatment group as a categorical variable (three groups), treatment time group as a categorical variable (Groups 1, 2, 3), and time to biopsy as a categorical variable (three groups) (see Table 3). This analysis demonstrated that only doubling time significantly predicted for a positive biopsy ( p 5 0.05). In the overall database (2869 patients), 82 patients had at least one posttreatment positive biopsy. Of these patients, 37 had experienced a Phoenix-defined biochemical failure. Distant metastases Distant metastatic disease developed in 51 of the 213 patients with PSA failure patients during the followup period. The overall FFDM at 10 years was 73% (see Fig. 1). A FFDM curve from the date of biochemical failure to the time of distant metastases revealed a rate of 76% at 10 years. Patients with distant metastases had a decreased overall survival rate at 10 years of 33% compared to 80% for those without clinical evidence of metastatic disease ( p! 0.0001). The following factors were analyzed for their effect on FFDM rates: pretreatment PSA, clinical stage, Gleason score, risk group, treatment group, BED, time to PSA failure, and PSA doubling time (see Table 4). Using univariate analysis, risk group, Gleason score, treatment group, BED (<, O150), time to PSA failure, and PSA doubling time all predicted for developing distant metastases. A multivariate analysis was performed using Cox regression of these significant factors (see Table 5). PSA doubling time and time to PSA failure were the most significant predictors. The effect of PSA doubling time on developing distant metastases can be seen in Fig. 2. Of the 51 patients with distant metastases, 3 also had a positive biopsy. Among patients with biochemical failure but no evidence of distant disease, 49% were placed on hormonal therapy and 51% were just observed.

Biopsy outcomes Of the 119 patients, 36 (30%) had a least one positive posttreatment biopsy. Of note, 10 of the 36 patients with at least one positive biopsy had followup biopsies, the last of which was read as negative. Among patients with a positive biopsy, 19% were read as having radiation effect, 16% were Gleason score 6, 26% were Gleason score 7, 23% were Gleason score 8, 6% were Gleason score 9, and 10% were Gleason score 10. The following analyses were done using an endpoint of at least one positive biopsy . Factors examined for their potential impact on biopsy outcomes were as follows: time to

Discussion The optimal management of a patient with a biochemical failure after brachytherapy is often unclear. Options include observation, hormonal therapy, and local salvage therapies such as radical prostatectomy and cryotherapy. Choosing an appropriate strategy requires insight into the probable source of a biochemical recurrence. Local salvage therapies should not be done unless definite evidence of local recurrence has been documented by prostate biopsy. The delivery of a systemic therapy usually assumes distant disease.

220

R.G. Stock et al. / Brachytherapy 7 (2008) 217e222

Table 2 Effect of prognostic factors on the outcome (positive or negative) of the posttreatment prostate biopsy No. of patients

95% Confidence interval Exp (B)

p-Value (log rank test)

Factor

Negative

Positive

PSA (ng/ml) <10 O10e20 O20

44 20 19

19 9 8

0.99

Gleason score <6 7 8e10

48 19 16

27 4 5

0.18

Clinical stage t2b

27 56

7 29

0.15

Risk group Low Intermediate High

21 17 45

9 9 18

0.85

Time of PSA failure (yr) <1 2 O1e2 18 O2e3 14 O3 49

1 6 5 24

0.87

Time of biopsy posttreatment (yr) <3 59 O3e5 15 O5 9

19 5 13

0.006

PSA doubling time (mo) <3 21 O3e6 18 O6e10 14 O10 30

2 4 8 22

0.01

Treatment Implant alone Implant þ HRM Trimodality

39 18 26

27 6 3

0.01

Time of treatment Course 1 Course 2 Course 3

31 32 20

24 6 6

0.01

16 16

0.24

BED (111/119 patients had dosimetry) <150 30 O150 49

Table 3 Logistic regression analysis of factors potentially affecting biopsy outcomes

Factor

Wald

p-Value

Exp (B)

Lower

Upper

Doubling time Time to biopsy Treatment group Time of treatment

3.74 2.79 2.53 0.161

0.05 0.09 0.11 0.69

1.63 1.60 0.574 1.16

0.993 0.922 0.290 0.559

2.75 2.77 1.14 2.4

most common problem is sampling error, in that the biopsy cores can simply miss recurrent cancer. For this reason, certain investigators have cautioned against using it as an early surrogate for outcome (14). In fact, the ASTRO consensus statement on posttreatment biopsy cautions against using it in routine clinical practice as a treatment endpoint (15). Despite these limitations, it is still the best method currently available to detect a local recurrence short of performing a prostatectomy on every patient with a PSA failure. Although the number of patients with at least one positive biopsy was small (36), in univariate analysis, BED, treatment group, treatment time group, and PSA doubling time were all shown to significantly impact the detection of a local failure by biopsy. A doseeresponse relationship between implant dose and biochemical failure has been well documented (10, 16e19). In addition, clear relationships between dose and biopsy results have also been documented for both brachytherapy and external beam (10, 12, 20). A patient with a PSA failure and a low BED (!150) should be considered to be at high risk for a local failure and should undergo posttreatment prostate biopsy. Positive biopsy rates after treatment with hormonal therapy and implant or trimodality therapy were low at 9% and 0%, respectively. The use of hormonal therapy with both implant and EBRT has been shown to result in low positive biopsy rates (12, 20, 21). In addition, the trimodality regimen has also been shown to result in extremely low positive local failure rates (9, 22). Patients experiencing a PSA

BED 5 biologically effective dose; HRM 5 hormonal therapy.

A staging workup including CT imaging and bone scan, which reveals radiographic evidence of metastatic disease provides an obvious source for a rising PSA. The difficult situation is one in which both imaging and posttreatment biopsy fail to reveal the source of biochemical progression. Identifying those factors that predict for a positive biopsy and the development of metastatic disease can help delineate a particular pattern of failure in patients before they have been clinically manifested. The use of posttreatment biopsy as an endpoint for prostate cancer outcomes is fraught with uncertainties. The

Fig. 1. Freedom from developing distant metastases.

R.G. Stock et al. / Brachytherapy 7 (2008) 217e222

221

Table 4 Predictors of developing distant metastases Factor

No. of patients 10-Yr FFDM (%) p-Value

Pretreatment PSA (ng/ml) <10 114 O10e20 49 O20 50

73 73 70

Gleason score <6 7 8e10

116 45 52

88.5 70 38.5

Clinical stage t2b

60 153

81 70

0.35

Risk group Low Intermediate High

50 39 124

90.5 87 69

0.002

Treatment Implant Implant Implant

group alone and HRM RX and EBRT

0.71

!0.0001

Fig. 2. The effect of PSA doubling time on the development of metastatic disease.

94 41 78

94 90 29

!0.0001

BED <150 O150

58 139

95 60

!0.0001

Time of PSA failure (yr) <1 O1e2 O2e3 O3

12 45 39 117

16 57.5 47 87

!0.0001

PSA doubling time (mo) <3 O3e6 O6e10 O10

54 53 30 76

0 74 78 94.5

!0.0001

BED 5 biologically effective dose; HRM 5 hormonal therapy; FFDM 5 freedom from distant metastases; EBRT 5 external beam radiation therapy.

failure after these regimens are far less likely to harbor local disease. In addition, high-risk patients, treated with trimodality, who experience a biochemical recurrence, exhibit PSA profiles and eventual outcomes that support a pattern of systemic over local failure (22). Table 5 Cox regression analysis of factors affecting the development of distant metastases 95% Confidence interval Exp (B) Factor

p-Value

Exp (B)

Lower

Upper

Treatment group Risk group Doubling time Gleason score Time to failure BED (<150, 150)

0.005 0.15 0.003 0.035 0.000 0.94

3.03 0.577 0.586 1.62 0.486 0.959

1.39 0.275 0.412 1.03 0.36 0.294

6.58 1.21 0.833 2.53 0.657 3.12

BED 5 biologically effective dose.

The most significant delineator between local and system progression in patients with biochemical failure was PSA doubling time. Patients with a doubling time O10 months had a positive biopsy rate of 49% vs. 0% for those with doubling time <6 months. Patients with long doubling times should undergo a prostate biopsy and if negative should be considered for repeat biopsies because they are likely to harbor residual or recurrent local disease. In the setting of negative biopsies and a rising PSA, it is important to identify those at high risk of having systemic disease. The identification of such patients can help select out those for immediate hormonal therapy vs. observation. In multivariate analysis, time to PSA failure, Gleason score, and PSA doubling time were all predictive of developing distant metastases. Time to PSA failure has been shown in other studies to impact the development of distant disease. In a study by Pound et al., patients who develop a PSA failure <2 years after radical prostatectomy had a 10-year FFDDM rate of 25% vs. 58% for those failing O2 years (!0.001) (23). In addition, PSA doubling time has been shown to be highly predicting in selecting those patients at risk for developing metastatic disease and dying of prostate cancer (23, 24). In the current study, patients with !3 months PSA doubling time, those failing !1 year after treatment, and those with Gleason scores 8e10 had low FFDM rates of 22%, 58%, and 50%, respectively. Patients with PSA failure and any of these factors are at high risk of developing distant metastases despite negative systemic workup and should be considered for early hormonal therapy or other systemic approaches. Conclusion Known treatment and disease factors can be used to predict for both local and systemic failure in patients with biochemical failure. In addition to staging procedures such as

222

R.G. Stock et al. / Brachytherapy 7 (2008) 217e222

posttreatment prostate biopsy and radiologic imaging, these factors can aid in identifying the source a rising PSA and designing subsequent management.

[13] [14]

References [1] Partin AW, Pearson JD, Landis PK, et al. Evaluation of serum prostate-specific antigen velocity after radical prostatectomy to distinguish local recurrence from distant metastases. Urology 1994;43: 649e659. [2] D’Amico AV, Hanks GE. Linear regressive analysis using prostatespecific antigen doubling time for predicting tumor biology and clinical outcome in prostate cancer. Cancer 1993;72:2638e2643. [3] Pollack A, Zagars GK, Kavadi VS. Prostate specific antigen doubling time and disease relapse after radiotherapy for prostate cancer. Cancer 1994;74:670e678. [4] Roach M 3rd, Hanks G, Thames H, et al. Defining biochemical failure following radiotherapy with or without hormonal therapy in men with clinically localized prostate cancer; recommendations of the RTOG-ASTRO Phoenix Consensus Conference. Int J Radiat Oncol Biol Phys 2006;65:965e974. [5] Fleming ID, Cooper JS. AJCC 1992 staging system. Philadelphia, PA: Lippincott-Raven; 1997. [6] Stock RG, Stone NN, Wesson MF, et al. A modified technique allowing interactive ultrasound-guided three-dimensional transperineal prostate implantation. Int J Radiat Oncol Biol Phys 1995;32:219e225. [7] Stone NN, Stock RG. Prostate brachytherapy: Treatment strategies. J Urol 1999;162:421e426. [8] Lee LN, Stock RG, Stone NN. Role of hormonal therapy in the management of intermediate- to high-risk prostate cancer treated with permanent radioactive seed implantation. Int J Radiat Oncol Biol Phys 2002;52:444e452. [9] Stock RG, Cahlon O, Cesaretti JA, et al. Combined modality treatment in the management of high-risk prostate cancer. Int J Radiat Oncol Biol Phys 2004;59:1352e1359. [10] Stock RG, Stone NN, Cesaretti JA, et al. Biologically effective dose values for prostate brachytherapy: Effects on PSA failure and posttreatment biopsy results. Int J Radiat Oncol Biol Phys 2006;64: 527e533. [11] Petraki CD, Sfikas CP. Histopathological changes induced by therapies in the benign prostate and prostate adenocarcinoma. Histol Histopathol 2007;1:107e118. [12] Stock RG, Stone NN, Kao J, et al. The effect of disease and treatment-related factors on biopsy results after prostate brachytherapy:

[15]

[16] [17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

Implications for treatment optimization. Cancer 2000;89:1829e 1834. Norusis MJ. SPSS 10.0 guide to data analysis. Upper Saddle River, NJ: Prentice-Hall; 2000. Pollack A, Zagars GK, Antolak JA, et al. Prostate biopsy status and PSA nadir level as early surrogates for treatment failure: Analysis of a prostate cancer randomized radiation dose escalation trial. Int J Radiat Oncol Biol Phys 2002;54:677e685. Cox JD, Gallagher MJ, Hammond EH, et al. Consensus statements on radiation therapy of prostate cancer: Guidelines for prostate rebiopsy after radiation and for radiation therapy with rising prostatespecific antigen levels after radical prostatectomy. American Society for Therapeutic Radiology and Oncology Consensus Panel. J Clin Oncol 1999;17:1155e1163. Stock RG, Stone NN, Tabert A, et al. A dose-response study for I-125 prostate implants. Int J Radiat Oncol Biol Phys 1998;41:101e108. Potters L, Cao Y, Calugaru E, et al. A comprehensive review of CTbased dosimetry parameters and biochemical control in patients treated with permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2001;50:605e614. Wallner K, Merrick G, True L, et al. 125I versus 103Pd for low-risk prostate cancer: Preliminary PSA outcomes from a prospective randomized multicenter trial. Int J Radiat Oncol Biol Phys 2003;57: 1297e1303. Zelefsky MJ, Kuban DA, Levy LB, et al. Multi-institutional analysis of long-term outcome for stages T1-T2 prostate cancer treated with permanent seed implantation. Int J Radiat Oncol Biol Phys 2007; 67:327e333. Zelefsky MJ, Fuks Z, Hunt M, et al. High dose radiation delivered by intensity modulated conformal radiotherapy improves the outcome of localized prostate cancer. J Urol 2001;166:876e881. Laverdiere J, Gomez JL, Suburu ER, et al. Beneficial effect of combination hormonal therapy administered prior and following external beam radiation therapy in localized prostate cancer. Int J Radiat Oncol Biol Phys 1997;37:247e252. Stock RG, Cesaretti JA, Stone NN. Changing the patterns of failure for high-risk prostate cancer patients by optimizing local control. Int J Radiat Oncol Biol Phys 2006;66:389e394. Pound CR, Partin AW, Eisenberger MA, et al. Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999;281:1591e1597. D’Amico AV, Moul J, Carroll PR, et al. Prostate specific antigen doubling time as a surrogate end point for prostate cancer specific mortality following radical prostatectomy or radiation therapy. J Urol 2004;172:42e46.