in 32%, T2c in 16% and T3 in 2% of patients.

BJUI Gleason 7 prostate cancer treated with lowdose-rate brachytherapy: lack of impact of primary Gleason pattern on biochemical failure Richard G. Stock, Joshua Berkowitz, Seth R. Blacksburg and Nelson N. Stone * Departments of Radiation Oncology and * Urology, Mount Sinai School of Medicine, New York, NY, USA Accepted for publication 2 December 211 Study Type Prognosis (case series) Level of Evidence 4 OBJECTIVES To report the biochemical outcomes for Gleason 7 prostate cancer treated with brachytherapy. To analyse the impact of the primary Gleason pattern as well as other diseaseand treatment-related factors on outcome. PATIENTS AND METHODS A total of 56 patients with Gleason 7 prostate cancer were treated between 199 and 28 with brachytherapy, alone or in combination with hormonal therapy and/or external beam radiation therapy. There were 352 patients with Gleason pattern 3 + 4 and 28 with Gleason pattern 4 +3. The mean (range) presenting PSA level was 11.2 (1 3) ng/ml, and the median was 7.8 ng/ml. The presenting clinical stages were T1b in 1%, T1c in 33%, T2a in 16%, T2b What s known on the subject? and What does the study add? There appears to be a clear difference in cancer control outcomes for patients with Gleason scores of 3 + 4 and those with scores of 4 + 3 after radical prostatectomy. It has been documented that patients with Gleason 4 + 3 prostate cancer have higher incidences of non-organ-confined disease than those with primary pattern 3. Higher rates of extracapsular extension, seminal vesicle invasion and positive margins have been found to be associated with primary pattern 4 over 3. These higher rates of non-organ-confined disease can lead to increased biochemical failure, which, in turn, can lead to higher mortality rates. This study provides information on the prognostic significance of primary Gleason pattern in the brachytherapy management of prostate cancer. in 32%, T2c in 16% and T3 in 2% of patients. RESULTS The actuarial freedom from biochemical failure rate at 1 years was 82%. There was no significant difference between 1-year freedom from biochemical failure rates for patients with Gleason scores of 3 + 4 (79%) and those with scores of 4 +3 (82%). Biologically effective dose and presenting PSA level were both significant predictors of biochemical failure in multivariate analysis. CONCLUSIONS The primary Gleason pattern in Gleason 7 prostate cancer shows no significant effect on biochemical failure when treated with brachytherapy. These results are different from those found after radical prostatectomy and are probably attributable to the enhanced local control afforded by a brachytherapy approach to this disease subset. KEYWORDS prostate cancer, Gleason score 7, primary Gleason grade, brachytherapy INTRODUCTION Gleason 7 prostate cancer found on prostate biopsy is used in the National Comprehensive Cancer Network (NCCN) risk group stratification ( http://www.nccn. org ) to denote intermediate-risk prostate cancer. This grade of cancer can behave more like low-risk disease in certain patients or more aggressively in others. The number of cores involved, as well as other presenting disease characteristics, e.g. stage and presenting PSA level, can influence the natural history of Gleason 7 disease. Depending on the primary pattern, it can be read by a pathologist as either Gleason 3 + 4 or 4 + 3. The primary Gleason pattern has been found to be highly prognostic after radical prostatectomy. Most radical prostatectomy reports have found a primary pattern of 4 to be a significant predictor of biochemical failure and mortality [1 6 ]. Different treatment approaches have been recommended for this group of patients with prostate cancer. Radical prostatectomy continues to be the most popular treatment for Gleason 7 disease [1 6 ]. From a radiation approach, various treatments have been used. A Gleason score of 7 has been considered an adverse feature and, historically, patients were selected for trials testing the use of adjuvant hormonal therapy with external beam radiation therapy (EBRT) [7 ]. Brachytherapy alone or in combination with hormonal therapy and/or EBRT have all been used to treat Gleason 7 prostate cancer [8 ]. 212 11, 1257 1261 doi:1.1111/j.1464-41x.212.1157.x 1257

STOCK ET AL. The present study examines the biochemical outcomes for Gleason 7 prostate cancer treated with low-dose-rate brachytherapy. The prognostic importance of the primary Gleason pattern and other disease- and treatment-related factors, with regard to biochemical failure, were analysed. The purpose was to determine the relative importance of primary Gleason pattern in the brachytherapy management of Gleason 7 prostate cancer as well as to identify unique features that portend a worse prognosis. PATIENTS AND METHODS A total of 56 patients with Gleason score 7 adenocarcinoma on prostate biopsy were treated with brachytherapy between 199 and 28 at Mount Sinai Hospital in New York. These patients had a minimum of 2 years follow-up. All pathology was reviewed by a single pathologist. There were 352 patients with Gleason pattern 3 + 4 and 28 with Gleason pattern 4 + 3. The number of cores and percentage involvement by core were not available on a consistent basis and, therefore, were not analysed. The mean (range) presenting PSA level was 11.2 (1 3) ng/ml with a median of 7.8 ng/ml. The presenting clinical stages were T1b in 1%, T1c in 33%, T2a in 16%, T2b in 32%, T2c in 16% and T3 in 2% of patients. Using the NCCN risk group stratification, there were 429 patients in the intermediate-risk and 131 in the high-risk group. Treatment included four main groups: implant alone (palladium [pd ]-13 or iodine [ I ] -125; 24 patients), implant plus neoadjuvant (3 months) and adjuvant hormonal therapy (3 months; 89 patients), implant (pd-13 or I-125) plus EBRT (39.6 5.4 Gy; median: 59.4 Gy; 79 patients) and 9 months of hormonal therapy in addition to combined implant and EBRT (368 patients). At our institution, the treatment approach for patients with intermediate disease has changed over time from treatment with implant alone in the early years to a more combined approach with hormonal therapy or brachytherapy in the latter years of the study period [8 ]. The primary Gleason pattern was never considered a criterion for selecting a particular treatment regimen. All implants were performed using a real-time ultrasonography-guided system, which has been described previously [9 ]. Pd-13 was used in 544 patients and I-125 in 16 patients. One month after seed implant, CT-based dosimetry was performed. The implant dose was calculated using the dose-volume histogram parameter of D9. Biologically effective dose (BED) values were calculated for all treatments. The formula used has been reported previously [1 ]. The median (range) BED for all treatments was 22 (91 278) Gy2. All patients were followed up by PSA assessment and DREs every 6 months. The median (range) follow-up was 65 (24 2) months. Biochemical failure was defined using the Phoenix definition [11 ]. Actuarial survival curves were calculated using Kaplan Meier analysis. Comparisons between survival curves were calculated using the log-rank test. Multivariate analysis of survival was performed using Cox regression [12 ]. A P value of <5 was considered to indicate statistical significance. RESULTS There were a total of 55 PSA failures detected over the follow-up period (a 9.8% crude failure rate). The actuarial freedom from biochemical failure (FFBF) at 1 years was 82% ( Fig. 1 ). The effect of disease- and treatment-related factors on biochemical failure is shown in Table 1. Using univariate analysis, there was no significant difference between patients with Gleason scores of 3 + 4 (79%) and those with scores of 4 +3 (82% [Fig. 2 ] ). BED was a significant predictor of PSA failure. Both thresholds of 16Gy 2 and 18Gy 2 yielded significant results, although the smallest P value was found using a dose threshold of 16 Gy 2. ( Fig. 3 ). This threshold also yielded a significant result in Cox regression ( Table 2 ). The most significant predictor in both univariate and multivariate analysis was PSA level. There was little difference between groups of 1 ng/ml and >1 2 ng/ml, but patients with PSA levels >2 ng/ml had significantly worse outcomes, with a 1-year FFBF rate of 54% (Fig. 4 ). In univariate analysis, treatment with implant alone yielded significantly worse outcomes than were found in the other treatment groups but effect was nonsignificant in multivariate analysis. In addition, NCCN risk group was a significant predictor (favouring FIG. 1. Overall freedom from biochemical failure. 2 years 1 2 3 4 5 6 7 8 9 1 No.@ risk 559 546 442 352 278 26 159 115 76 49 1 intermediate- over high-risk) in univariate but not in Cox regression analysis. DISCUSSION 1 2 3 4 5 6 7 8 9 1 There has been much discussion in the urological literature on the relative importance of the primary Gleason pattern in Gleason score 7 prostate cancer. The vast majority of reports on outcomes after radical prostatectomy show inferior biochemical control for patients with Gleason score 7 with primary pattern 4 vs primary pattern 3 (Table 3 [1 6 ] ). Various explanations have been given to explain these findings. The original hypothesis was that patients with primary pattern 4 did worse than those with pattern 3 because they had biologically more aggressive disease and were more likely to harbour microscopic systemic disease [3 ]. This conclusion worked well in explaining the differences seen in progression-free survival between the primary patterns after radical prostatectomy. The theory of microscopic spread is supported by a study by Stark et al. [13 ] which analysed radical prostatectomy outcomes from patients in two prospective studies, the Physicians Health study and the Health Professional follow-up study. The results of their analysis found a threefold increase in mortality for patients treated with radical prostatectomy with Gleason score 4 +3 vs those with 3 + 4 (HR 2.4; 95% CI, 1. to 5.6). The data from the present study, as well as those from the study by Merrick et al. [14 ], which also examines this issue in patients 1258 212

GLEASON 7 PROSTATE CANCER TABLE 1 Effect of disease- and treatment-related factors on FFBF rates Factor No. of patients No. of failures 1-year FFBF, % P Gleason score (primary + secondary pattern) 3 + 4 352 35 79 4 + 3 28 2 82.659 NCCN risk group Intermediate 429 35 83 High 131 2 73 43 PSA, ng/ml 1 39 27 85 >1 2 116 14 82.5 >2 54 14 54 <1 Stage t2a 28 22 84 t2b 28 33 77.19 Treatment Implant alone 24 7 48 Implant + EBRT 79 6 83 Implant + HRM 89 11 84 Trimodality 368 31 81 41 Supplemental EBRT No 113 18 77 Yes 447 37 81.35 Adjuvant hormonal therapy No 13 13 66 Yes 457 42 83.148 BED 16 Gy2 5 13 68 >16 Gy2 488 37 85 5 <18 Gy2 17 19 75 >18 Gy2 431 31 85 33 FIG. 2. Effect of primary Gleason grade. Percent Free From PSA Failure 2 1 Effect of Primary Gleason Grade P =.66 1 2 3 4 5 6 7 8 9 1 FIG. 3. Effect of BED on biochemical failure. 2 1 Effect of Biologically Effective dose P = 5 1 2 3 4 5 6 7 8 9 1 4+3 3+4 >16 Gy2 16 Gy2 FIG. 4. Effect of initial PSA on biochemical failure. TABLE 2 Cox regression analysis of biochemical failure 95% CI for Exp(B) Factor P Exp(B) Lower Upper Risk group.846.91.353 2.35 Treatment groups.17 2.7.653 11.29 Hormonal treatment 8 71 4 1.38 PSA groups ( 1, >1 2, >2 ng/ml) 9 1.93 1.18 3.18 Stage ( T2a, T2b).49 1.1.828 1.48 BED ( 16, >16 Gy2 ) 11.154 36.651 Effect of initial PSA PSA 1 ng/ml PSA > 1 2 ng/ml PSA > 2 ng/ml 2 1 P < 1 1 2 3 4 5 67891 Exp(B), exponential (B). undergoing brachytherapy, suggest a different theory to explain the results from the radical prostatectomy studies: patients with primary pattern 4 have a higher local failure rate than those with primary pattern 3. It has been documented that patients with Gleason 4 + 3 prostate cancer have higher incidences of non-organ-confined disease than those with primary pattern 3. Higher rates of extracapsular extension, seminal vesicle invasion and positive margins have been correlated with primary pattern 4 over 3. These higher rates of non-organ-confined disease can lead to increased biochemical failure, which, in turn, can lead to higher mortality rates. The idea that disease left behind after radical prostatectomy and margin status are somehow related to the primary pattern has been examined in a study by Brimo et al. [15 ]. In looking at patients with extracapsular extension and positive margins, those authors found that the primary pattern at the positive margin site 212 1259

STOCK ET AL. TABLE 3 Effect of primary Gleason grade on biochemical failure after prostatectomy Study Gleason score No. of patients Reporting year FFBF P Khoddami et al. [1] 3 + 4 215 5 83% 16 4 + 3 9 5 65% Tollefson et al. [2] 3 + 4 1256 1 48% <1 4 + 3 432 1 38% Chan et al. [3] 3 + 4 456 1 63% <1 4 + 3 114 1 44% Sim et al. [4] 3 + 4 355 5 8% <1 4 + 3 88 5 42% Trock et al. [5] 3 + 4 626 6 8% <1 4 + 3 168 6 37% Alenda et al. [6] 3 + 4 721 5 76% <1 4 + 3 527 5 61% the patient s presentation, including PSA level, clinical stage and volume of cancer on biopsy. In conclusion, we found that the primary Gleason pattern in Gleason 7 prostate cancer had no significant effect on biochemical failure when treated with brachytherapy. These results are different from those found after radical prostatectomy and are probably attributable to the enhanced local control afforded by a brachytherapy approach to this disease subset. CONFLICT OF INTEREST predicted biochemical failure. Those with 4 + 3 had only a 12% FFBF rate at 6 years compared with 25% for those with 3 +4. Results from SWOG 8794 [16 ], which tested the efficacy of adjuvant postoperative radiation therapy after radical prostatectomy for high-risk features, found that radiation therapy increased local control and subsequently survival and concluded that the predominant mode of failure of patients in the high-risk group was local [16,17 ]. Brachytherapy results in the highest possible delivered radiation dose and also treats extracapsular disease [1 ]. The results from the present study, as well as those from Merrick et al. [14 ], suggest that brachytherapy, with or without supplemental EBRT, improves local control. This is supported by the high negative posttreatment biopsy results after brachytherapy and brachytherapy plus EBRT [18 ]. This enhanced local control seems to result in similar biochemical control rates between primary pattern 4 vs pattern 3. The poorer biochemical control rates seen with primary pattern 4 after radical prostatectomy are probably attributable to local failure which is not as much of an issue when a brachytherapy approach is used. In the present study, the FFBF at 1 years was 79% for 3 + 4 patients and 82% for 4 +3 patients ( P =.66). Similarly, Merrick et al. [14 ] found no difference in biochemical failure at 1 years between patients with 4 + 3 and those with 3 +4. The two most significant predictors of biochemical failure were PSA level and BED. A big decrease in the cancer control rate occurred at a PSA level of 2 ng/ml. This supports the NCCN risk group stratification, which places a patient with Gleason 7 disease into a high-risk group as the PSA level increases to > 2 ng/ml. The FFBF rate at 1 years of 54% in patients with PSA > 2 ng/ml suggests that these patients carry a high risk of having microscopic disease in the lymph nodes or occult systemic spread. Better systemic approaches need to be explored in these patients. The BED showed a highly significant effect on biochemical control. This highlights the need for high-quality, high-dose treatment to eradicate Gleason 7 prostate cancer. Although the type of treatment did not significantly affect biochemical failure in multivariate analysis, in univariate analysis, treatment with implant alone resulted in significantly worse biochemical control than with other treatment approaches. We reported a similar finding in a previous publication on patients in the intermediaterisk group [19 ]. This emphasizes the importance of the current Radiation Therapy Oncology Group trial (232), which randomizes patients with intermediate-risk prostate cancer to implant alone vs combined therapy with implant and EBRT ( www.rtog.org ). It is hoped that the RTOG trial will help to identify the optimum brachytherapy treatment for patients with intermediate-risk prostate cancer. Currently, we treat patients with Gleason 7 prostate cancer with either combined hormonal therapy and implant or implant and EBRT (with or without hormonal therapy). The exact treatment is individualized based on None declared. REFERENCES 1 Khoddami SM, Shariat SF, Lotan Y et al. Predictive value of primary Gleason pattern 4 in patients with Gleason score 7 tumours treated with radical prostatectomy. BJU Int 24 ; 94 : 42 6 2 Tollefson MK, Leibovich BC, Slezak JM, Zincke H, Blute ML. Long-term prognostic significance of primary Gleason pattern in patients with Gleason score 7 prostate cancer: impact on prostate cancer specific survival. J Urol 26 ; 175 : 547 51 3 Chan TY, Partin AW, Walsh PC, Epstein JI. Prognostic significance of Gleason score 3 + 4 versus Gleason score 4 + 3 tumor at radical prostatectomy. Urology 2 ; 56 : 823 7 4 Sim HG, Telesca D, Culp SH et al. Tertiary Gleason pattern 5 in Gleason 7 prostate cancer predicts pathological stage and biochemical recurrence. J Urol 28 ; 179 : 1775 9 5 Trock BJ, Guo CC, Gonzalgo ML, Magheli A, Loeb S, Epstein JI. Tertiary Gleason patterns and biochemical recurrence after prostatectomy: proposal for a modified Gleason scoring system. J Urol 29 ; 182 : 1364 7 6 Alenda O, Ploussard G, Mouracade P et al. Impact of the primary Gleason pattern on biochemical recurrence-free survival after radical prostatectomy: a single-center cohort of 1,248 patients with Gleason 7 tumors. World J Urol 2 11 ; 5 : 671 6 126 212

GLEASON 7 PROSTATE CANCER 7 D Amico AV, Manola J, Loffredo M, Renshaw AA, DellaCroce A, Kantoff PW. 6-Month androgen suppression plus radiation therapy vs radiation therapy alone for patients with clinically localized prostate cancer: a randomized controlled trial. JAMA 24 ; 292 : 821 7 8 Stone NN, Stock RG. Prostate brachytherapy: treatment strategies. J Urol 1999 ; 162 : 421 6 9 Stock RG, Stone NN, Wesson MF, DeWyngaert JK. A modified technique allowing interactive ultrasound-guided three-dimensional transperineal prostate implantation. Int J Radiat Oncol Biol Phys 1995 ; 32 : 219 25 1 Stock RG, Stone NN, Cesaretti JA, Rosenstein BS. Biologically effective dose values for prostate brachytherapy: effects on PSA failure and posttreatment biopsy results. Int J Radiat Oncol Biol Phys 26 ; 64 : 527 33 11 Roach M 3rd, Hanks G, Thames H et al. D efining 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 26 ; 65 : 965 74 12 Norusis MJ. SPSS 1 Guide to Data Analysis. Upper Saddle River : Prentice Hall, 2 13 Stark JR, Perner S, Stampfer MJ et al. Gleason score and lethal prostate cancer: does 3 + 4 = 4 + 3? J Clin Oncol 29 ; 27 : 3459 64 14 Merrick GS, Galbreath RW, Butler WM et al. Primary Gleason pattern does not impact survival after permanent interstitial brachytherapy for Gleason score 7 prostate cancer. Cancer 27 ; 11 : 289 96 15 Brimo F, Partin AW, Epstein JI. Tumor grade at margins of resection in radical prostatectomy specimens is an independent predictor of prognosis. Urology 21 ; 76 : 126 9 16 Swanson GP, Hussey MA, Tangen CM et al. Predominant treatment failure in postprostatectomy patients is local: analysis of patterns of treatment failure in SWOG 8794. J Clin Oncol 27 ; 25 : 2225 9 17 Thompson IM, Tangen CM, Paradelo J et al. Adjuvant radiotherapy for pathological T3NM prostate cancer significantly reduces risk of metastases and improves survival: long-term followup of a randomized clinical trial. J Urol 29 ; 181 : 956 62 18 Stone NN, Stone MM, Rosenstein BS, Unger P, Stock RG. Influence of pretreatment and treatment factors on intermediate to long-term outcome after prostate brachytherapy. J Urol 211 ; 185 : 495 5 19 Ho AY, Burri RJ, Cesaretti JA, Stone NN, Stock RG. Radiation dose predicts for biochemical control in intermediaterisk prostate cancer patients treated with low-dose-rate brachytherapy. Int J Radiat Oncol Biol Phys 29 ; 75 : 16 22 Correspondence: Richard G. Stock, Director of Prostate Programs, Department of Radiation Oncology Mount Sinai School of Medicine, New York, NY 129, USA. e-mail: Richard.stock@mountsinai.org Abbreviations : NCCN, National Comprehensive Cancer Network ; pd-13, palladium-13 ; I-125, iodine-125 ; FFBF, freedom from biochemical failure ; EBRT, external beam radiation therapy ; BED, biologically effective dose. 212 1261