Interstitial implant alone or in combination with external beam radiation therapy for intermediate-risk prostate cancer: A survey of practice patterns in the United States

Methods and materials A nomogram-based survey was developed at the Seattle Prostate Institute defining the accepted criteria for intermediate-risk prostate cancer. Patients were defined as having intermediate-risk prostate cancer if they met one of the following criteria: prostate-specific antigen (PSA) >10 ng/dL, Gleason score (GS) ≥7, or cT2b or cT2c disease. Additional potential predictive factors including perineural invasion (PNI), GS 3 + 4 vs. 4 + 3, and high-volume disease were included. Results In the absence of PNI, all of those surveyed would perform monotherapy for intermediate-risk patients, GS 7 (3 + 4) or PSA 10–20, with cT1c and <30% cores +. Up to 80% would perform monotherapy for patients with cT1c, GS 7 (4 + 3), and <30% cores +. Eighty to 90% of physicians would perform an implant alone with cT2a and either a PSA of 10–20 or GS of 7 (3 + 4) and <30% cores +. Fifty to 60% of those surveyed stated that they would treat a patient with cT2b disease, GS 7 (3 + 4), or PSA 11–20, with less than two-thirds of the biopsy cores positive in the absence of PNI. Conclusions This Patterns of Care (POC) study reveals that certain subsets of intermediate-risk localized prostate cancer patients are considered appropriate candidates for an interstitial implant alone. Keywords Prostate Brachytherapy Intermediate risk Introduction Interstitial implants for prostate cancer have evolved over the last century. As early as 1917, radium needles were inserted transperineally into the prostate. In the 1950s, radioactive gold solution was inserted into the prostate, and by the early 1970s, retropubic 125 I seed implants were performed in an open operation. For the men who underwent the surgical retropubic technique, outcomes were generally poor most likely due to poor dosimetric coverage, and the technique rapidly fell from favor (1) . With improved imaging via transrectal ultrasound in the 1980s, template-guided perineal implants were initially performed in Seattle (2) . Techniques were refined over the next decade for men with localized prostate cancer, and the results continued to improve as the techniques were modified to ensure optimal dosimetric coverage of the prostate gland and periprostatic tissue (3) . Tolerances were observed to the rectum, bladder, and urethra providing acceptable side effects to men diagnosed with this potentially life-threatening disease. As outcome data continued to mature in the 1990s and early 2000s, men who were pursuing a nonsurgical approach for their cancer began to choose interstitial implants at an increasing rate. In Patterns of Care study by Zelefsky et al. , brachytherapy increased from 4% in the early 1990s to 36% by 1997 (4) . Factors felt to have contributed to this rise were the favorable early results, increasing concern over the effectiveness of conventional external beam radiation therapy (EBRT), and the duration of recovery and complications associated with radical prostatectomy. The facts that brachytherapy can be performed in an outpatient setting, requires a short recovery time and has manageable side effects (5) also contributed to seed implantation becoming a reasonable option for many patients. Practice patterns of treatment Low-risk localized prostate cancer In retrospective, single-institution studies, interstitial implant monotherapy for men with low-risk prostate cancer has been shown to achieve results equivalent to other local treatments, such as radical prostatectomy and EBRT (6, 7) . The Surgical Prostatectomy vs. Interstitial Radiation Intervention Trial (SPIRIT) was an attempt to verify these data. However, it failed to accrue sufficient patient numbers in the United States, likely due to potential inherent physician biases and/or patient refusal to be randomized to a treatment. Although seed implant monotherapy is considered to be sufficient treatment for most low-risk patients, further risk factors may delineate a high-risk subgroup within this low-risk category (7) . Intermediate-risk localized prostate cancer Historically, patients with intermediate-risk localized prostate cancer ( Table 1 ) have not been considered appropriate candidates for monotherapy because of high risk of extraprostatic disease that has been predicted by nomograms like the Partin tables (8) . Early surgical and implant results suggested that local therapy may not be sufficient in this patient group; therefore, the combination of EBRT and implants was generally recommended. The American Brachytherapy Society (ABS), recognizing that the issue was unresolved, outlined a general recommendation in 1999 suggesting that EBRT plus seeds should be considered for intermediate-risk patients. The ABS stopped short of making specific recommendations and recognized the need for individualization of treatment recommendations. Therefore, patients may receive either implant alone or combination therapy at the discretion of the brachytherapy team (9) . Patients deemed at an arbitrarily high risk for extracapsular penetration, seminal vesicle involvement, or lymph node spread were considered appropriate candidates for the combination of EBRT and seeds (10) . The rationale for considering EBRT plus seeds was based on the surgical and early implant outcomes that supported the hypothesis that additional treatment was required for extraprostatic disease beyond the implant volume. Although some early implant series results seemed to confirm this hypothesis, recent single-institution retrospective data have emerged that appears to contradict this notion. For example, two recent studies comparing the outcomes of monotherapy and EBRT plus implant with or without hormonal therapy have resulted in equivalent if not superior results with monotherapy treatment (11, 12) . These data suggest that monotherapy may be an appropriate option for appropriately selected intermediate-risk patients. Therefore, the question remains, “What therapy is appropriate for intermediate-risk localized prostate cancer?” Whereas randomized trials are the ideal method of defining the role of new treatment modalities to the current standard of care, patient accrual to a study comparing prostate brachytherapy to surgery was unsuccessful (SPIRIT). In a current Phase III randomized trial, the Radiation Treatment Oncology Group is currently investigating whether EBRT should be added to an interstitial implant for select patients with intermediate-risk localized prostate cancer. On the basis of the current rate of accrual and the potential subgroups, it may take over a decade before enough data are generated to find an answer to this important question (personal communication DAK, 2006). Furthermore, it is unlikely that another brachytherapy trial will be initiated to answer this critical question. In the meantime, the field of brachytherapy continues to move forward as more and more men are choosing this minimally invasive procedure. Therefore, single-institution experiences in both prospective and retrospective protocols will continue to guide the field of prostate brachytherapy by providing the data that will help both radiation oncologists and urologists make appropriate treatment recommendations. This study is aimed at understanding and defining the current patterns of care among experts in prostate brachytherapy both in the academic and community setting. Methods and materials A nomogram-based survey was developed at the Seattle Prostate Institute defining the accepted criteria for intermediate-risk prostate cancer, including the serum prostate-specific antigen (PSA) level, Gleason score (GS), and clinical stage. Patients were defined as having intermediate-risk prostate cancer if they met one of the following criteria: PSA >10 ng/dL, GS ≥7, or clinical stage T2b or T2c disease. Additional potential predictive factors, including perineural invasion (PNI), GS 3 + 4 vs. 4 + 3, and extent of positive biopsies, were also added to the nomogram to help better elucidate the clinician's decision to offer additional EBRT to the interstitial implant for potentially more aggressive disease. Twenty brachytherapy experts, chosen based on their clinical and academic productivity, were surveyed and asked to select either monotherapy (seed implant alone) or EBRT plus seed implantation on the basis of stage, grade, serum PSA level, percentage of biopsy-positive cores, and PNI. The data from the survey serve as the basis for this report. The percentages reported in this pattern of care study are based on the number of clinicians choosing that modality of therapy with respect to the total number of surveys returned. Results Of the 20 surveys sent out, 18 of 20 (90%) were completed and returned. Table 2 illustrates the clinical situations considered suitable for monotherapy by over 50% of those brachytherapy clinicians surveyed. Table 3 illustrates clinical situations that over 50% of the clinicians surveyed would add EBRT to a seed implant for patients with intermediate-risk, localized prostate cancer. The following conclusions may be drawn from the patterns of care survey: 1. In the absence of PNI, all of those surveyed would perform monotherapy for intermediate-risk patients with nonpalpable (cT1c), low-volume (<30%) disease with a GS 7 (3 + 4) or a serum PSA level between 10 and 20 ng/mL. In the presence of PNI, up to 90% of those surveyed would perform monotherapy on intermediate-risk patients. 2. Up to 30% of those surveyed indicated that the variation in pathology of GS 7 from 3 + 4 to 4 + 3 would alter their decision concerning the use of EBRT vs. a monotherapy approach. Additionally, an increase in volume of tumor in the gland from 30% to 50% of biopsy cores positive would influence the decision to use EBRT. 3. For patients who present with a palpable nodule (cT2a), low-volume disease (<30% biopsy cores positive), and no PNI, up to 90% of those surveyed would perform monotherapy with a GS 7 (3 + 4) tumor or a serum PSA level between 10 and 20 ng/mL. 4. In the absence of PNI, over 50% of those surveyed would perform monotherapy on patients presenting with disease palpable on half of the prostate gland (cT2b) but low-volume disease (<30% biopsy cores positive), with a GS 7 (3 + 4) or a serum PSA level between 10 and 20 ng/mL. 5. Up to 80% of those surveyed would perform monotherapy for patients with a GS 7 (4 + 3), if the patient had nonpalpable disease and <30% of the biopsy cores were positive. In the presence of PNI, the percentage of clinicians who would choose monotherapy decreased to 60%. 6. Up to 60% of those surveyed stated that they would treat with monotherapy if a patient presented with clinically palpable disease on half of the gland (cT2b), ≤30% biopsy cores positive, no PNI, and GS 7 (3 + 4), or serum PSA values between 10 and 20 ng/mL. 7. Up to 90% of those surveyed would add EBRT to a seed implant for patients who presenting with high-volume disease (50% biopsy cores positive) and either GS 7 (3 + 4) or serum PSA level between 10 and 20 ng/mL, even in the presentation of nonpalpable disease (cT1c). 8. For patients who present with bilateral palpable disease (cT2c), over 50% of the clinicians would administer EBRT before the seed implant. Additionally, the presence of GS 8–10, even with low-volume, nonpalpable disease and a serum PSA level <4 ng/mL would guide over 50% of the physicians to add EBRT during treatment. 9. The presence or absence of PNI became less of a factor in the decision to add EBRT as patients presented with higher volume disease (>50% biopsy cores positive). 10. Less than 30% of those surveyed would perform an implant alone on a patient presenting with either GS 7 or PSA between 10 and 29 and with >50% of the biopsy cores positive. Treatment observations from the patterns of care survey on individual clinical presentations may be derived from Tables 2 and 3 . Discussion The outcomes of this survey reflect an evolving trend in the brachytherapy community in the treatment of intermediate-risk localized prostate cancer with a seed implant alone. For patients with intermediate-risk localized prostate cancer, the role of EBRT combined with a permanent implant is theoretically to treat microscopic disease beyond the reach of the brachytherapy implant. The pretreatment challenge with permanent interstitial implants is to predict the likelihood of disease beyond the margin of local therapy (10) . No current technology can predict this microscopic spread with accuracy. Monotherapy, with permanent prostate brachytherapy alone, if deemed equally effective as combined therapy, has the advantage of decreased patient morbidity, decreased cost, and greater convenience for the patient. The EBRT treatment issues concerning daily setup variability, rectal filling, and intrafraction motion do not exist with permanent implants. Brachytherapy's advantage over local treatment with EBRT lies in the decreased integral dose to the patient while providing a larger effective dose to the tumor in the prostate gland with a margin for extraprostatic extension of disease. Within the intermediate-risk group, several factors may be involved in accounting for the differences in results in some studies vs. equivalent results in others between those patients treated with monotherapy and those treated with EBRT plus implant: patient selection, percent positive cores, implant technique and philosophy, and PNI. Patient selection Patients with intermediate-risk localized prostate cancer represent a heterogeneous group, with a wide range of risks for extraprostatic disease as predicted by the Partin tables (8) . Analysis of the intermediate-risk group comparing treatment schemes could easily be affected by the weighting of “high-risk patients” within this cohort group. In the future, subgroup analysis may be able to sort out these specific patient risk groups from the rudimentary risk groupings of the past. Percent positive cores Butler and Merrick argue against altering treatment decision with respect to the number of biopsy cores positive, GS 3 + 4 vs. 4 + 3, and the presence or absence of PNI (13–15) . With the lack of randomized trials addressing these issues, they effectively point out that there are little data in the brachytherapy literature to support the conclusions drawn in the radical prostatectomy or EBRT series. They also conclude that these “limitations” are obviated in brachytherapy by implanting with consistent extracapsular dosimetric margins (16) . The role of EBRT in patients with a larger number of positive cores needs further clarification. D'Amico et al. acknowledged that dose escalation, the addition of hormonal therapy, or both should be considered in patients with greater than 50% biopsy cores positive (17) . Kestin et al. evaluated the predictive power of the percentage of positive biopsy cores in men with locally advanced prostate cancer treated with EBRT and high dose-rate brachytherapy (18) . On multivariate analysis, percentage of positive biopsy cores appeared to be more reliable than clinical T stage in predicting biochemical and clinical failure. Patients with less than 33% positive cores had 5-year biochemical control and clinical failure rates of 83% and 7%, respectively. Patients who had more than 67% positive cores had 5-year biochemical control and clinical failure rates of 57% and 25%, respectively. Rossi et al. recently reported that the 5-year estimate of freedom of biochemical relapse was 95% for patients with less than 50% biopsy core positive vs. 63% in patients with greater than 50% biopsy cores positive (19) . Merrick et al. disputes this notion and found no clinical significance in the percentage of biopsy cores predicting worse 5-year biochemical serum PSA outcome. Their data suggest that the lack of significance may be because dose escalation with brachytherapy exceeds other radiation modalities, and the seed distribution in the periprostatic region provides superior margins compared to radical prostatectomy (14) . Treatment volume Surgical margins are typically within 3 mm of the prostate edge and closer if nerve sparing is performed. Current implant philosophy and techniques have allowed margins of 5–10 mm or greater. The treatment volume of brachytherapy may therefore theoretically have an advantage over that of surgery due to the ability to treat a wider margin around the prostate laterally, particularly at the base and at the apex of the gland. Because adding EBRT to the treatment may add only the seminal vesicles and possibly a slightly greater margin to an implant, this option may not provide significant advantage for many patients. Butler and Merrick argue against the “myths and fallacies” that have been created concerning the appropriateness of an implant alone in the presence of these poor predictive factors (16) . They contend that much of the data that have generated these notions come from the surgical and external beam literature and are not supported in the brachytherapy data. A high-quality implant with modified peripheral loading and seed placement in the periprostatic tissue should provide appropriate dosimetric coverage of the prostate and extracapsular disease in the periprostatic tissues. With appropriate dosimetric coverage, their data suggest that none of these factors are predictive of biochemical failure and therefore, an implant alone may be appropriate. Perineural invasion Although several physicians in this survey seem to alter their treatment decisions due to the presence or absence of PNI, its predictive value in the literature is controversial. Whereas most data that define the prognostic value of PNI in predicting biochemical outcome arise from radical prostatectomy (20) and EBRT series (21) , its role in predicting biochemical failure in the brachytherapy literature is not well defined. In a study by Merrick et al. , there was no significant difference in biochemical control after prostate brachytherapy using either 103 Pd or 125 I seeds in a series of 425 patients as a function of the presence or absence of PNI (93.1% vs. 94.8%, respectively) (15) . The current ABS recommendations for treating prostate cancer with an interstitial permanent implant alone are for patients with low-risk localized disease (cT1c–cT2a, serum PSA level <10 ng/mL, and GS <7). The presence of either clinical stage ≥T2c, PSA >10, or GS ≥7 have been considered indications for the addition of EBRT due to the increased risk of extraprostatic disease. Nevertheless, 100% of those surveyed indicated that they would use an implant alone for patients with low-volume, nonpalpable disease, no PNI, and either a serum PSA level between 10 and 20 ng/mL or a GS of 7 (3 + 4). In this survey, a GS 7 (4 + 3), greater than one-third of the biopsy cores positive, and the presence of PNI altered the treatment decision of whether or not to treat with an implant alone. Interestingly, there are little data in the brachytherapy literature that would suggest that the presence of any one of these factors predicts for worse outcome. These factors may be surrogates for the presence of extracapsular extension of disease, implying that the additional margin with EBRT may provide the appropriate coverage in the periprostatic tissues to sterilize microscopic disease. The predictive nature of Gleason grade 7 (3 + 4) vs. (4 + 3) in outcomes and the subsequent addition of EBRT with an implant is controversial. Differences in biochemical outcomes have been noted in patients who present with GS 7 (3 + 4) and (4 + 3). Potters et al. looked retrospectively at 208 patients with GS 7 (3 + 4) and 150 patients with GS 7 (4 + 3) (22) . The 7-year biochemical freedom from relapse was 78.4% and 56.7% for the (3 + 4) and (4 + 3) groups, respectively. Whereas GS was significant in predicting biochemical freedom from recurrence, the addition of EBRT was not. They concluded that the clinical outcome of a GS 7 (3 + 4) tumor was similar to a GS 6 (3 + 3) tumor, and that the decision between monotherapy or combined therapy could be “liberalized” toward monotherapy in patients with GS 7 (3 + 4) tumor (22) . Similar conclusions concerning the predictive nature of 3 + 4 vs. 4 + 3 have been noted in the radical prostatectomy literature (23, 24) . In contrast, Merrick et al. reporting on 273 consecutive patients from 1995 to 2001 found that the 8-year biochemical progression-free survival rate in patients with GS 7 tumor was 94% with no statistical difference noted when the group was stratified by the dominant histologic pattern (13) . He concluded that in hormone-naïve patients with a GS 7 tumor, biochemical progression-free survival is independent of GS 3 + 4 vs. 4 + 3 histology. Treatment decisions by those surveyed were altered depending on the volume of disease in the prostate as determined by the number of biopsy cores positive. Even in patients with cT1c disease, GS 7 (3 + 4), and serum PSA level <10 ng/mL, if a patient had 30–50% or greater than 50% of his cores positive, 70–90% of the clinicians surveyed would add EBRT during treatment. D'Amico et al. evaluated the clinical utility of the percentage of positive biopsies in predicting PSA outcome after radical prostatectomy (17) . Of the 269 patients with intermediate-risk disease, 80% (214 patients) could be classified as having <34% or >50% biopsy cores positive. The 4-year rate of serum PSA control using the preoperative prostate biopsy was 86%, 48%, and 11% for <34%, 34–50%, and >50% cores positive, respectively. The study concluded that this information may be helpful in selecting men for adjuvant therapy trials after radical prostatectomy. Although the addition of EBRT should have little affect in improving outcomes in patients treated with a high-quality implant who do not have extracapsular disease beyond the reach of the implant, it is important to issue a cautionary note with regard to permanent interstitial implants for patients with intermediate-risk disease. The key to achieving the outcomes published with monotherapy is to ensure a high-quality brachytherapy program. Potters et al. , in an updated report on combined therapy and monotherapy, reported that EBRT becomes insignificant when considering the D 90 in multivariate analysis (25) In the brachytherapy community, even though the technique, prescription dose, and seed placement philosophies may vary widely, the quality assurance with postimplant dosimetry must be consistent. Data in terms of outcomes and morbidity should only be quoted to patients if the treatment is consistent with the philosophies of the published institutions. According to Merrick and Butler, “without standardization of planning and postimplant dosimetry techniques, there can be no meaningful multi-institutional comparisons of biochemical outcomes and morbidity” (26) . The data presented in this study regarding the survey results suggest that in the brachytherapy community, there are certain subsets of intermediate-risk patients that are considered by many clinicians appropriate candidates for implant alone. The presence of GS 7 (4 + 3) tumor, PNI, and greater than 30% biopsy cores positive alter the decision patterns with respect to addition of EBRT to the implant. In the absence of randomized data, the best data comes from single-institution experiences evaluated in a prospective fashion. Although these data are subject to inherent bias, they provide a framework for hypothesis generation that has culminated in the current Radiation Treatment Oncology Group randomized trial comparing the outcomes in patients with intermediate-risk disease who are treated with an implant alone vs. those treated with an implant in combination with EBRT. Conclusions This pattern of care survey has identified the changing trend in the treatment of prostate cancer with permanent seed implantation. The survey has identified specific clinical conditions in the intermediate-risk patient group, which brachytherapy experts consider monotherapy with a permanent prostate implant to be an adequate treatment. Similarly, the survey has also identified those patients who many believe would likely benefit from combined therapy. References [1] H. Ragde J.C. Blasko P.D. Grimm Brachytherapy for clinically localized prostate cancer: Results at 7- and 8-year follow-up Semin Surg Oncol 13 1997 438 443 [2] H.H. Holm The history of interstitial brachytherapy of prostatic cancer Semin Surg Oncol 13 1997 431 437 [3] J.C. Blasko K. Wallner P.D. Grimm Prostate specific antigen based disease control following ultrasound guided 125iodine implantation for stage T1/T2 prostatic carcinoma J Urol 154 1995 1096 1099 [4] M.J. Zelefsky J. Moughan J. Owen Changing trends in national practice for external beam radiotherapy for clinically localized prostate cancer: 1999 Patterns of Care survey for prostate cancer Int J Radiat Oncol Biol Phys 59 2004 1053 1061 [5] N.N. Stone R.G. Stock Permanent seed implantation for localized adenocarcinoma of the prostate Curr Urol Rep 3 2002 201 206 [6] H. Ragde J.C. Blasko P.D. Grimm Interstitial iodine-125 radiation without adjuvant therapy in the treatment of clinically localized prostate carcinoma Cancer 80 1997 442 453 [7] A.V. D'Amico R. Whittington S.B. Malkowicz Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer JAMA 280 1998 969 974 [8] A.W. Partin M.W. Kattan E.N. Subong Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update JAMA 277 1997 1445 1451 [9] S. Nag D. Beyer J. Friedland American Brachytherapy Society (ABS) recommendations for transperineal permanent brachytherapy of prostate cancer Int J Radiat Oncol Biol Phys 44 1999 789 799 [10] P.B.J. Grimm J. Sylvester Clinical selection issues for permanent seed prostate brachytherapy J Brachyther Int 17 2001 143 152 [11] L. Potters C. Morgenstern E. Calugaru 12-year outcomes following permanent prostate brachytherapy in patients with clinically localized prostate cancer J Urol 173 2005 1562 1566 [12] G.S. Merrick W.M. Butler K.E. Wallner Impact of supplemental external beam radiotherapy and/or androgen deprivation therapy on biochemical outcome after permanent prostate brachytherapy Int J Radiat Oncol Biol Phys 161 2005 32 43 [13] G.S. Merrick W.M. Butler R.W. Galbreath Biochemical outcome for hormone-naive patients with Gleason score 3 + 4 versus 4 + 3 prostate cancer undergoing permanent prostate brachytherapy Urology 60 2002 98 103 [14] G.S. Merrick W.M. Butler R.W. Galbreath Relationship between percent positive biopsies and biochemical outcome after permanent interstitial brachytherapy for clinically organ-confined carcinoma of the prostate gland Int J Radiat Oncol Biol Phys 52 2002 664 673 [15] G.S. Merrick W.M. Butler R.W. Galbreath Perineural invasion is not predictive of biochemical outcome following prostate brachytherapy Cancer J 8 2002 79 80 [16] W.M. Butler G.S. Merrick Myths and fallacies in permanent prostate brachytherapy Med Dosim 28 2003 155 165 [17] A.V. D'Amico A.A. Renshaw K. Cote Impact of the percentage of positive prostate cores on prostate cancer-specific mortality for patients with low or favorable intermediate-risk disease Am J Clin Oncol 22 2004 3726 3732 [18] L.L. Kestin N.S. Goldstein F.A. Vicini Percentage of positive biopsy cores as predictor of clinical outcome in prostate cancer treated with radiotherapy J Urol 168 2002 1994 1999 [19] P.J. Rossi P.E. Clark M.A. Papagikos Percentage of positive biopsies associated with freedom from biochemical recurrence after low-dose-rate prostate brachytherapy alone for clinically localized prostate cancer Urology 67 2006 349 353 [20] J.I. Epstein The role of perineural invasion and other biopsy characteristics as prognostic markers for localized prostate cancer Semin Surg Oncol 16 1998 124 128 [21] A.V. D'Amico Perineural invasion as a predictor of PSA outcome following local therapy for patients with clinically localized prostate cancer Cancer J 7 2001 375 376 [22] L. Potters R. Purrazzella S. Brustein The prognostic significance of Gleason Grade in patients treated with permanent prostate brachytherapy Int J Radiat Oncol Biol Phys 56 2003 749 754 [23] A.V. D'Amico A.A. Renshaw D. Schultz The impact of the biopsy Gleason score on PSA outcome for prostate cancer patients with PSA < or = 10 ng/ml and T1c, 2a: Implications for patient selection for prostate-only therapy Int J Radiat Oncol Biol Phys 45 1999 847 857 [24] C.M. Herman M.W. Kattan M. Ohori Primary Gleason pattern as a predictor of disease progression in Gleason score 7 prostate cancer: A multivariate analysis of 823 men treated with radical prostatectomy Am J Surg Pathol 25 2001 657 660 [25] L. Potters P. Fearn M.W. Kattan External radiotherapy and permanent prostate brachytherapy in patients with localized prostate cancer Brachytherapy 1 2002 36 41 [26] B.R. Thomadsen M.J. Rivard W.M. Butler Brachytherapy physics second edition 2nd ed. 2005 Medical Physics Publishing Madison, WI