- Email: [email protected]
Long-Term Outcome for Clinically Localized Prostate Cancer Treated With Permanent Interstitial Brachytherapy
Int. J. Radiation Oncology Biol. Phys., Vol. 79, No. 5, pp. 1336–1342, 2011 Copyright Ó 2011 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/$–see front matter
doi:10.1016/j.ijrobp.2010.01.005
CLINICAL INVESTIGATION
Prostate
LONG-TERM OUTCOME FOR CLINICALLY LOCALIZED PROSTATE CANCER TREATED WITH PERMANENT INTERSTITIAL BRACHYTHERAPY AL V. TAIRA, M.D.,* GREGORY S. MERRICK, M.D.,y WAYNE M. BUTLER, PH.D.,y ROBERT W. GALBREATH, PH.D.,y JONATHAN LIEF, PH.D.,y EDWARD ADAMOVICH, M.D.,z x AND KENT E. WALLNER, M.D. *Department of Radiation Oncology, University of Washington, Seattle, Washington; ySchiffler Cancer Center, Wheeling Jesuit University, Wheeling, West Virginia; zDepartment of Pathology, Wheeling Hospital, Wheeling, West Virginia; and xPuget Sound Healthcare Corporation, Group Health Cooperative, University of Washington, Seattle, Washington Purpose: To present the largest series of prostate cancer brachytherapy patients treated with modern brachytherapy techniques and postimplant day 0 dosimetric evaluation. Methods and Materials: Between April 1995 and July 2006, 1,656 consecutive patients were treated with permanent interstitial brachytherapy. Risk group stratification was carried out according to the Mt. Sinai guidelines. Median follow-up was 7.0 years. The median day 0 minimum dose covering at least 90% of the target volume was 118.8% of the prescription dose. Cause of death was determined for each deceased patient. Multiple clinical, treatment, and dosimetric parameters were evaluated for impact on the evaluated survival parameters. Results: At 12 years, biochemical progression-free survival (bPFS), cause-specific survival (CSS), and overall survival (OS) for the entire cohort was 95.6%, 98.2%, and 72.6%, respectively. For low-, intermediate-, and high-risk patients, bPFS was 98.6%, 96.5%, and 90.5%; CSS was 99.8%, 99.3%, and 95.2%; and OS was 77.5%, 71.1%, and 69.2%, respectively. For biochemically controlled patients, the median posttreatment prostate-specific antigen (PSA) concentration was 0.02 ng/ml. bPFS was most closely related to percent positive biopsy specimens and risk group, while Gleason score was the strongest predictor of CSS. OS was best predicted by patient age, hypertension, diabetes, and tobacco use. At 12 years, biochemical failure and cause-specific mortality were 1.8% and 0.2%, 5.1% and 2.1%, and 10.4% and 7.1% for Gleason scores 5 to 6 and 7 and $8, respectively. Conclusions: Excellent long-term outcomes are achievable with high-quality brachytherapy for low-, intermediate-, and high-risk patients. These results compare favorably to alternative treatment modalities including radical prostatectomy. Ó 2011 Elsevier Inc. Prostate cancer, Brachytherapy, Biochemical control, Survival.
techniques and documented implant quality to better delineate efficacy and survival outcomes.
INTRODUCTION Men with newly diagnosed prostate cancer are candidates for radical prostatectomy, external beam radiotherapy (XRT), and/or brachytherapy depending on the details of their presentation. Brachytherapy is widely prescribed for men with low-risk disease and is increasingly used for men with higher-risk disease as well, based on recent reports of excellent outcomes in intermediate- and high-risk patients (1–3). However, there has been some reluctance to offer implants to higher-risk men, based primarily on unfavorable results in studies of intermediate- and high-risk brachytherapy patients treated in the 1980s and 1990s without the benefit of modern brachytherapy techniques and dosimetric plan evaluation (4, 5). The purpose of this study was to present long-term results of men treated with modern brachytherapy
METHODS AND MATERIALS Between April 1995 and July 2006, 1,656 consecutive patients with clinically localized prostate cancer, stages T1b to T3c (2002, American Joint Committee on Cancer) were treated with permanent interstitial brachytherapy by a single brachytherapist (GSM). All biopsy slides were reviewed by a single pathologist (EA). Patients were clinically staged by medical history, physical examination, including digital rectal examination and serum prostate-specific antigen (PSA) concentration determination. Bone scans and computed tomography of the abdomen and pelvis were obtained for patients with higher- but not lower-risk disease. Of the 1,656 patients, 575 patients presented with low-risk disease (PSA #10 ng/ml, Gleason score #6, and clinical stage # T2a); 608
Reprint requests to: Gregory S. Merrick, M.D., Schiffler Cancer Center, Wheeling Hospital, 1 Medical Park, Wheeling, WV 26003. Tel: (304) 243-3490; Fax: (304) 243-5047; E-mail: gmerrick@ urologicresearchinstitute.org
Conflict of interest: none. Received Nov 12, 2009, and in revised form Dec 29, 2009. Accepted for publication Jan 6, 2010. 1336
Outcome of permanent interstitial brachytherapy d A. V. TAIRA et al.
patients presented with intermediate-risk disease (1 adverse factor: PSA 10.1-19.9 ng/ml or Gleason score 7 or clinical stage T2b) and 473 presented with high-risk disease (PSA $20 ng/ml or Gleason score $8 or clinical stage $T2c or two to three of the intermediate adverse features). Of these 1,656 men, 622 (37.6%) men received androgen deprivation therapy (ADT). A total of 406 (24.5%) men received #6 months of ADT, while 216 (13.0%) men received >6 months of ADT. In patients receiving ADT, ADT was initiated at 3 months before implantation and consisted of a luteinizing hormone-releasing agonist and an antiandrogen. The maximum duration of ADT was 36 months. A total of 825 (49.8%) patients received supplemental XRT. In general, XRT consisted of 45 Gy delivered in 1.8-Gy fractions with 15 to 18 MV photons delivered via a multifield technique with customized treatment devices. For patients with #10% risk of pelvic lymph node involvement (6), the target volume consisted of the prostate gland and seminal vesicles. For patients with >10% risk of pelvic lymph node involvement, the pelvic lymph nodes were also included in the target volume. Supplemental XRT was delivered prior to implantation. The brachytherapy planning treatment volume (PTV) consisted of the entire prostate gland with a 5-mm periprostatic margin and the proximal 1.0 cm of the seminal vesicles (7, 8). All postimplant dosimetric calculations were based on day 0 evaluation of postimplant coverage of the PTV. For boost brachytherapy, the minimum peripheral dose was 90 to 100 Gy (National Institute of Standards and Technology 1999) for Pd-103 exposure. For monotherapeutic approaches, the minimum peripheral dose was 125 Gy for Pd-103 and 145 Gy (American Association of Physicists in Medicine Task Group 43) for I-125 exposure. Patients were monitored by physical examination, including digital rectal examination and serum PSA determination at 3- and 6-month intervals. The endpoint of the analysis was cause-specific survival (CSS), biochemical progression-free survival (bPFS), and overall survival (OS). Cause of death was determined for each deceased patient. Patients with metastatic prostate cancer or castration-resistant disease without obvious metastases who died of any cause were classified as dead of prostate cancer. All other deaths were attributed to the immediate cause of death. bPFS was defined as a PSA level of #0.40 ng/ml after nadir, which has been shown to be a particularly sensitive definition for identifying patients for whom treatment has failed (9). Patients whose treatment failed to achieve a nadir below 0.40 ng/ml were categorized as a biochemical failure. Multiple clinical, treatment, and dosimetric parameters were evaluated for impact on survival. Clinical and treatment variables that were continuous were compared using an independent t test. Categorical variables were compared using chi-square analysis. Kaplan-Meier analysis and curves were used to determine CSS, OS, and bPFS. Univariate Cox regression analysis was used to determine the variables that predicted bPFS, CSS, and OS. Those variables with p values of <0.10 were entered into a forward conditional, multivariate Cox regression. For all analysis, a p value of <0.05 was considered statistically significant, Statistical analysis was performed with SPSS software version 14.0 (SPSS, Chicago, Ill).
RESULTS Table 1 summarizes the clinical and treatment characteristics of men included in the analysis. Of the 1,656 men, 575 men were low-risk, 608 men were intermediate-risk, and
1337
473 men were high-risk. Median length of follow-up was 7.0 years. The median day 0 minimum dose covering at least 90% of the target volume (D90) and the V100 for all patients was 118.8% of the prescription implant dose and 97.9%, respectively. ADT was used more commonly in higher-risk patients, and long-term ADT was used almost exclusively in patients with high-risk disease. Supplemental radiation therapy also was used more commonly in men with intermediateand high-risk disease. In the men with high-risk cancer, 87.9 % received supplemental XRT, and 58.2% received hormonal therapy. At 12 years, bPFS, CSS, and OS for the group as a whole was 95.6%, 98.2%, and 72.6%, respectively (Fig. 1). Table 2 presents univariate and multivariate analyses of predictors of CSS, bPFS, and OS. Gleason score was by far the strongest predictor of CSS. Although traditional prognostic factors like risk group categorization and percent positive biopsy cores were significant on univariate analysis, these were not significant on multivariate analysis. Pretreatment PSA level, percent positive biopsy samples, Gleason score, and risk group were univariate predictors of bPFS, with percent positive biopsy samples and risk group being the strongest predictors in multivariate analysis. Age, smoking, and diabetes were most strongly associated with OS. bPFS at 12 years was 98.6%, 96.5% and 90.5%, for men with low-, intermediate-, and high-risk disease, respectively (Fig. 2a). CSS was 99.8%, 99.3%, and 95.2%, for men with low-, intermediate-, and high-risk disease (Fig. 2b). OS was 77.5%, 71.1%, and 69.2% for men with low-, intermediate-, and high-risk disease (Fig. 2c). bPFS and CSS rates stratified by Gleason score are presented in Fig. 3. For men with Gleason 5 or 6, actuarial biochemical failure at 12 years was 1.8%, and cancer-specific mortality was 0.2%. For men with Gleason 7 disease, biochemical failure at 12 years was 5.1%, and cancer-specific mortality was 2.1%. For men with Gleason 8 or higher disease, biochemical failure at 12 years was 10.4%, and cancer-specific mortality was 7.1%. For men experiencing biochemical progression, the median time to biochemical failure was 1.9 years vs. 2.4 years for men with a Gleason score of #7 and a Gleason score of $8 (p = 0.231). At the time of this analysis, the longest time spans from implant to biochemical failure was 3.2 years, 8.0 years, and 5.4 years for patients with Gleason scores of 5 to 6, 7, and $8, respectively. When patients were stratified by ADT status, the longest time spans from implant to failure for patients who were hormone-na¨ıve, taking ADT for #6 months, and taking ADT for >6 months were 8.0, 6.8, and 3.1 years, respectively. For the group as a whole, 90% of failures occurred within 3.9 years of implantation. Of the patients who experienced biochemical failure, 33% of men with Gleason scores of 5 to 6, 58% with a Gleason score of 7, and 82% with a Gleason score of $8 eventually developed castrate-resistant disease or metastases by the time of death. bPFS and CSS stratified by percent positive biopsy cores is presented in Fig. 4. Percent positive biopsy samples predicted
I. J. Radiation Oncology d Biology d Physics
1338
Volume 79, Number 5, 2011
Table 1. Clinical, treatment, and dosimetric parameters for brachytherapy patients stratified by risk No. of low-risk patients (n = 575) Parameter Continuous variable
Med.
Mean SD
Age at implant (yrs) 65.0 63.9 7.3 Follow-up (yrs) 6.9 7.1 3.1 PSA 5.7 5.8 1.9 Gleason Score 6.0 5.9 0.3 Percent positive biopsies 16.7 23.4 15.7 BMI 27.7 28.3 4.2 Prostate volume 35.4 35.8 8.5 Planning volume 65.9 65.5 12.1 V100 97.6 96.0 5.1 V150 67.4 64.6 13.8 V200 36.1 34.6 12.0 D90 116.4 116.0 12.9 Most recent PSA 0.02 0.05 0.13 Categorical variable Testosterone Low and Lower 1/3 Normal Middle 1/3 Normal Upper 1/3 normal and high Clinical stage: T1b-T2b T2c-T3b Isotope I-125 Pd-103 ADT None # 6 months > 6 months XRT Yes No Hypertension yes no Diabetes yes no Tobacco use never former current Coronary artery disease yes no Elevated cholesterol yes no Perineural invasion yes no
No. of intermediate-risk patients (n = 608) Med.
Mean SD
66.0 6.7 6.2 7.0 33.3 27.5 34.1 64.1 98.1 72.2 41.4 120.3 0.02
65.7 7.3 6.9 2.9 6.8 2.9 6.8 0.5 38.7 23.0 28.3 4.7 34.3 9.0 63.7 13.1 96.7 4.4 69.0 13.6 39.7 12.3 119.9 13.4 0.04 0.10
No. of high-risk patients (n = 473)
Total (n = 1656)
Mean SD
p value*
67.0 66.7 7.2 7.4 7.7 3.2 10.6 12.6 9.2 7.0 7.6 0.9 50.0 54.1 27.0 27.5 28.4 4.5 30.5 31.2 9.0 56.1 57.2 14.0 98.0 96.1 5.7 71.9 67.5 15.3 40.6 38.6 13.5 120.6 119.3 15.7 0.01 0.04 0.07
<0.001 <0.001 <0.001 <0.001 <0.001 0.898 <0.001 <0.001 0.020 <0.001 <0.001 <0.001 0.037
Med.
Med.
Mean SD
66.0 65.3 7.4 7.0 7.2 3.1 6.5 8.1 6.0 7.0 6.7 0.9 33.3 37.7 25.2 27.5 28.3 4.5 33.6 33.9 9.4 62.9 62.5 13.5 97.9 96.3 5.0 70.4 67.1 14.3 39.4 37.6 12.8 118.8 118.4 14.0 0.02 0.05 0.10
Count n (%)
Count n (%)
Count n (%)
212 (69.3) 71 (23.2) 23 (7.5)
223 (64.5) 86 (24.9) 37 (10.7)
132 (62.9) 52 (24.8) 26 (12.4)
0.354
567 (65.8) 209 (24.2) 86 (10.0)
575 (100) 0 (0)
587 (96.5) 21 (3.5)
250 (52.9) 223 (47.1)
<0.001
1412 (85.3) 244 (14.7)
181 (31.5) 393 (68.5)
54 (8.9) 553 (91.1)
40 (8.5) 433 (91.5)
<0.001
275 (16.6) 1379 (83.4)
392 (68.2) 176 (30.6) 7 (1.2)
444 (73.0) 145 (23.8) 19 (3.1)
198 (41.9) 85 (18.0) 190 (40.2)
<0.001
1034 (62.4) 406 (24.5) 216 (13.0)
15 (2.6) 560 (97.4)
394 (64.8) 214 (35.2)
416 (87.9) 57 (12.1)
<0.001
825 (49.8) 831 (50.2)
231 (43.1) 305 (56.9)
269 (46.5) 309 (53.5)
227 (50.4) 223 (49.6)
0.070
727 (46.5) 837 (53.5)
50 (8.8) 516 (91.2)
70 (11.7) 529 (88.3)
65 (13.8) 405 (86.2)
0.038
185 (11.3) 1450 (88.7)
244 (42.5) 251 (43.7) 79 (13.8)
235 (38.7) 274 (45.1) 99 (16.3)
152 (32.3) 239 (50.7) 80 (17.0)
0.016
631 (38.2) 764 (46.2) 258 (15.6)
91 (15.8) 484 (84.2)
99 (16.3) 509 (83.7)
93 (19.7) 379 (80.3)
0.202
283 (17.1 1372 (82.9)
178 (31.0) 397 (69.0)
187 (30.8) 421 (69.2)
143 (30.3) 329 (69.7)
0.973
508 (30.7) 1147 (69.3)
47 (8.2) 527 (91.8)
211 (35.5) 383 (64.5)
209 (45.4) 251 (54.6)
<0.001
467 (28.7) 1161 (71.3)
p valuey
Count n (%)
Abbreviations: V100, V150, and V200 = percent of the target volume receiving 100%, 150%, and 200% of prescription dose, respectively; D90 = minimum dose, as a percent of the prescription dose, received by at least 90% of target volume; ADT = androgen deprivation therapy; SD = standard deviation; XRT = supplemental external beam radiation therapy; Med. = median. * One-way analysis of variance. Values in boldface are statistically significant. y Chi-square test.
Outcome of permanent interstitial brachytherapy d A. V. TAIRA et al.
100
Survival (%)
80
60 Cause-Specific Survival, 98.2% Biochemical Progression-Free Survival, 95.6%
40
Overall Survival, 72.6%
20 n= 1656
1575
1013
475
120
0 0
3
6 9 Years Since Implant
12
15
Fig. 1. Kaplan-Meier curves for biochemical progression-free, cause-specific, and overall survival. Each curve represents the same patients.
for both bPFS and CS. Pretreatment PSA level predicted for both bPFS and CSS (Fig. 5). DISCUSSION Studies of men treated with earlier brachytherapy techniques and planning parameters demonstrate what by today’s standards would be considered relatively poor outcomes (4, 5). Based on these reports and nomograms from the same era (10), a perception remains that outcomes in intermediate- and higher-risk brachytherapy patients may not be as good as alternative treatment options. However, over the past 10 to 15 years, brachytherapy techniques and planning have improved significantly. The current study represents the largest series to date of men with long-term follow-up treated with modern brachytherapy techniques and complete postimplant dosimetric evaluation. With current techniques and high-quality implants, excellent long-term outcomes can be achieved in men with low-, intermediate-, and high-risk disease. Specifically, very low rates of prostate cancer-specific mortality are achievable even in high-risk patients. Previously, Zelefsky et al. (5) reported outcomes for a multicenter cohort of 2,693 brachytherapy patients. However, postimplant dosimetric information was available for only 639 patients. With a median follow-up of 63 months, 8-year bPFS for low-, intermediate-, and high-risk patients was 74%, 61%, and 39%, respectively. In the cohort for whom dosimetric data were available, the postimplant D90 was lower than currently recommended. Stone et al. (11) reported a multicenter cohort study of 3,928 brachytherapy patients with a median follow-up of 42 months, all of whom had dosimetric information available. For the cohort as a whole, bPFS for low-, intermediate-, and high-risk patients was 84%, 77%, and 64%. However,
1339
the 1,100 men who received a higher biologically equivalent dose (BED) of >200 Gy via their implant, with or without XRT, had much more favorable outcomes. Among these men, bPFS for low-, intermediate-, and high-risk patients was 88%, 94%, and 90%, respectively. All of the 1,656 men in the current study have complete postimplant dosimetric evaluation with implant quality assessed by standard dosimetric criteria (Table 1). We report excellent long-term bPFS, with 12-year bPFS of 95.6% for the cohort overall. The 12-year bPFS for low-, intermediate-, and high-risk patients was 98.6%, 96.5%, and 90.5%, respectively. Although bPFS is a useful surrogate for treatment efficacy, cancer-specific survival is the most important measure of treatment success. In our study, Gleason score was the strongest predictor of cancer-specific mortality. Few patients with Gleason scores of 5 to 6 experience biochemical failure (1.8% at 12 years), and only a few of those succumbed to prostate cancer (0.2%). On the other hand, even though the biochemical failure rate at 12 years was only 10.4% for men with a Gleason score of $8, prostate cancer-specific mortality was 7.1%, suggesting that men with high-risk disease who experience biochemical failure are much more likely to die of prostate cancer. Our results compare favorably to a recently published report presenting long-term prostate cancer-specific mortality in men treated with radical prostatectomy (RP) at several leading academic medical centers (12). That study consisted of more than 12,500 men treated in the serum PSA level era (median follow-up, 46 months) and represents the largest multiinstitutional study of RP with long-term follow-up. In comparison, prostate cancer-specific mortality for a biopsy sample Gleason score of 5 to 6 at 12 years was 0.2% in brachytherapy patients, compared to 2% and 6% at 10 and 15 years for RP patients. For men with a biopsy Gleason score of 7, the brachytherapy prostate cancer-specific mortality at 12 years was 2% compared to 5% and 17% at 10 and 15 years for RP patients. For patients with biopsy Gleason scores of $8, the risk of cancer-specific death at 12 years for brachytherapy patients was 7% for brachytherapy and 16% and 34% at 10 and 15 years for RP. We expect our results will be durable, particularly in regard to prostate cancer mortality. Over 90% of biochemical failures in our series occurred within 3.9 years of implantation. Among men with Gleason scores of 8 or higher (the group most likely to die of disease), the average time from implantation to biochemical failure was 2.4 years. In our cohort thus far, there have been no biochemical failures in patients with Gleason scores of $8 beyond 5.4 years. In addition, there appears to be a clear plateau on the bPFS curves (Fig. 3a). The timing of biochemical failure in our cohort differs notably from the timing of failure in previously reported series in which implant doses to the prostate were lower. For instance, in the earlier multiinstitutional brachytherapy series reported by Zelefsky et al. (5), approximately half of biochemical failures occurred after 4 years from the time of
I. J. Radiation Oncology d Biology d Physics
1340
Volume 79, Number 5, 2011
Table 2. Univariate and multivariate analyses for CSS, bPFS, and OS Biochemical progression-free survival
Cause-specific survival Univariate analysis Variable
p value
Age at implant (yrs) PSA (ng/mL) Gleason score Body mass index Prostate volume Percent positive biopsy samples D90 XRT (yes/no) ADT (yes/no) ADT duration (0/<6/>6 months) Risk group High vs. low High vs. intermediate Hypertension (yes/no) Diabetes (yes/no) Tobacco use (never/former/current) never vs former never vs current Coronary artery disease (yes/no) Hypercholesterolemia (yes/no) Perineural invasion (yes/no)
0.413 0.089 <0.001 0.898 0.183 0.001 0.905 0.010 0.467 0.308 0.001 0.585 0.008 0.032 0.363 0.186
Multivariate analysis
HR
p value
2.886
0.276 <0.001
1.028
0.138
14.044
0.104
HR
2.804
0.031
p value 0.348 <0.001 <0.001 0.371 0.200 <0.001 0.104 <0.001 0.411 0.165
0.083 15.564 0.256
Univariate analysis
0.254
<0.001 0.189 <0.001 0.093 0.848 0.095
Overall survival
Multivariate analysis
HR
p value
1.045 1.757
0.462 0.781
1.026
<0.001
3.108
0.792
6.6647
<0.001 0.596 0.002
Univariate analysis
Multivariate analysis
HR
p value
HR
p value
HR
1.101
<0.001
1.109
1.138 0.968
0.722 0.937
1.017
<0.001 0.361 0.046 0.036 0.375 0.192 0.053 0.601 0.510 0.506
0.693 1.397 1.433 0.779 1.566
0.327
0.041 0.030 0.021 0.043 0.008 <0.001
0.022
0.005 <0.001 0.002
1.517 2.331 1.584
3.868
0.011 <0.001
1.542
0.017 <0.001 0.129
1.434 2.859
0.319
0.053
0.219
0.039
0.517
0.102
0.357
0.639
0.544
0.015
1.837
0.677
0.697
0.809
0.305
Abbreviations: HR = hazard ratio. Values in boldface are statistically significant.
the presence of occult metastatic disease at the time of treatment. We assume that some significant portion of the failures in series with lower prostate doses was due to local disease persistence. When those patients failed, it is not surprising that it could take many years for a local tumor burden to
implant. In contrast, in the current report, nearly all failures occurred within the first 4 years. We postulate that there may be a difference in timing of biochemical failure in patients who fail due to local disease persistence/recurrence versus those patients who fail due to
c.
b. 100
100
100
80 Low Risk, 98.6% Intermediate Risk, 96.5% High Risk, 90.5%
60
40
p < 0.001
20
80
80
Low Risk, 99.8% Intermediate Risk, 99.3% High Risk, 95.2%
Overall Survival (%)
Cause-Specific Survival (%)
Biochemical Progression-Free Survival (%)
a.
60 p < 0.001 40
60
Low Risk, 77.5% Intermediate Risk, 71.1% High Risk, 69.2%
40
20
20
0
p = 0.004
0 0
3
6
9
Years Since Implant
12
15
0 0
3
6
9
Years Since Implant
12
15
0
3
6
9
Years Since Implant
Fig. 2. (a) Kaplan-Meier curves for biochemical progression-free survival, stratified by risk. (b) Kaplan-Meier curves for CSS stratified by risk. (c) Kaplan-Meier curves for OS, stratified by risk. (p values represent single tests for linear trends across all factor levels.)
12
15
Outcome of permanent interstitial brachytherapy d A. V. TAIRA et al.
b.
a.
100
100 Cause-Specific Survival (%)
Biochemical Progression -Free Survival (%)
1341
80 5-6, 98.2% 7, 94.9% 8-10, 89.6%
60
p < 0.001 40
80 5-6, 99.8% 7, 97.9% 8-10, 92.9%
60
p < 0.001 40
20
20
0
0 0
3
6
9
12
15
0
3
Years Since Implant
6
9
12
15
Years Since Implant
Fig. 3. (a) Kaplan-Meier curves for biochemical progression-free survival stratified by Gleason score. (b) Kaplan-Meier curves for CSS stratified by Gleason score. (p values represent single tests for linear trends across all factor levels.)
presents longer-term follow-up based on staging at clinical presentation. Compared with similarly staged patients who undergo RP, the long-term results of brachytherapy when stratified by Gleason score and/or PSA compare favorably.
recover to a volume significant enough to trigger PSA level progression. Higher ablative doses to the prostate, as in this study, should result in improved local disease eradication. Hence, patients who fail despite receiving high ablative doses are probably more likely to have had occult metastatic disease at presentation. When patients fail due to undetected metastatic disease, it is possible that it occurs sooner than when a patient fails from incompletely eradicated local disease. Patients treated with brachytherapy rarely undergo pathology staging. Yet, at the time a patient decides on one definitive treatment option over another, pathology staging is not available. One of the difficulties of comparing results between different treatment modalities is that some publications will present outcomes based upon initial clinical staging (XRT and brachytherapy publications), while others will be based primarily on pathology staging (generally RP series). Also, some publications have relatively short follow-up, begging the question of longer-term outcomes. This report
Excellent long-term prostate cancer outcomes are achievable with high-quality brachytherapy. This work builds on a number of studies demonstrating excellent short- and intermediate-term results with primary brachytherapy, particularly in men with higher-risk disease (1–3, 13). Based on our data and those of others, we believe primary brachytherapy should be considered a viable treatment option for men with low-, intermediate-, and high-risk prostate cancer. If long-term data from other institutions supports our findings, primary brachytherapy may become the preferred treatment option for men with higher-risk disease.
b. 100
100
80
Cause-Specific Survival (%)
Biochemical Progression-Free Survival (%)
a.
CONCLUSIONS
< 34%, 98.0% 34-50%, 93.1% > 50%, 91.5%
60
p < 0.001 40
20
80 < 34, 99.2% 34-50, 98.4% > 50, 95.5%
60
p = 0.030 40
20
0
0 0
3
6
9
Years Since Implant
12
15
0
3
6
9
12
15
Years Since Implant
Fig. 4. (a) Kaplan-Meier curves for biochemical progression-free survival, stratified by percent positive biopsy. (b) Kaplan-Meier curves for CSS stratified by percent positive biopsy. (p values represent single tests for linear trends across all factor levels.)
I. J. Radiation Oncology d Biology d Physics
1342
b. 100
100 Cause-Specific Survival (%)
Biochemical Progression-Free Survival (%)
a.
Volume 79, Number 5, 2011
80 PSA < 10, 96.9% PSA > 20, 92.3% PSA 10-20, 90.2%
60
p < 0.001 40
20
0
80 PSA <10, 98.8% PSA > 20, 97.2% 60
PSA 10-20, 95.3%
40
p = 0.004
20
0 0
3
6
9
12
15
0
Years Since Implant
3
6
9
12
15
Years Since Implant
Fig. 5. (a) Kaplan-Meier curves for biochemical progression-free survival, stratified by PSA level. (b) Kaplan-Meier curves for CSS stratified by PSA level. (p values represent single tests for linear trends across all factor levels.)
REFERENCES 1. Kao J, Stone NN, Lavaf A, et al. (125)I monotherapy using D90 implant doses of 180 Gy or greater. Int J Radiat Oncol Biol Phys 2008;70:96–101. 2. Merrick GS, Butler WM, Wallner KE, et al. Impact of supplemental external beam radiotherapy and/or androgen deprivation therapy on biochemical outcome after permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2005;61:32–43. 3. Stone NN, Potters L, Davis BJ, et al. Multicenter analysis of effect of high biologic effective dose on biochemical failure and survival outcomes in patients with Gleason score 7-10 prostate cancer treated with permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2009;73:341–346. 4. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998;280:969–974. 5. 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;67:327–333. 6. Partin AW, Mangold LA, Lamm DM, et al. Contemporary update of prostate cancer staging nomograms (Partin Tables) for the new millennium. Urology 2001;58:843–848.
7. Merrick GS, Butler WM. Modified uniform seed loading for prostate brachytherapy: Rationale, design, and evaluation. Tech Urol 2000;6:78–84. 8. Merrick GS, Butler WM, Wallner KE, et al. Extracapsular radiation dose distribution after permanent prostate brachytherapy. Am J Clin Oncol 2003;26:e178–189. 9. Kuban DA, Levy LB, Potters L, et al. Comparison of biochemical failure definitions for permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2006;65:1487–1493. 10. Kattan MW, Potters L, Blasko JC, et al. Pretreatment nomogram for predicting freedom from recurrence after permanent prostate brachytherapy in prostate cancer. Urology 2001;58:393–399. 11. Stone NN, Potters L, Davis BJ, et al. Customized dose prescription for permanent prostate brachytherapy: Insights from a multicenter analysis of dosimetry outcomes. Int J Radiat Oncol Biol Phys 2007;69:1472–1477. 12. Stephenson AJ, Kattan MW, Eastham JA, et al. Prostate cancerspecific mortality after radical prostatectomy for patients treated in the prostate-specific antigen era. J Clin Oncol 2009;27:4300–4305. 13. D’Amico AV, Moran BJ, Braccioforte MH, et al. Risk of death from prostate cancer after brachytherapy alone or with radiation, androgen suppression therapy, or both in men with highrisk disease. J Clin Oncol 2009;27:3923–3928.
doi:10.1016/j.ijrobp.2010.01.005
CLINICAL INVESTIGATION
Prostate
LONG-TERM OUTCOME FOR CLINICALLY LOCALIZED PROSTATE CANCER TREATED WITH PERMANENT INTERSTITIAL BRACHYTHERAPY AL V. TAIRA, M.D.,* GREGORY S. MERRICK, M.D.,y WAYNE M. BUTLER, PH.D.,y ROBERT W. GALBREATH, PH.D.,y JONATHAN LIEF, PH.D.,y EDWARD ADAMOVICH, M.D.,z x AND KENT E. WALLNER, M.D. *Department of Radiation Oncology, University of Washington, Seattle, Washington; ySchiffler Cancer Center, Wheeling Jesuit University, Wheeling, West Virginia; zDepartment of Pathology, Wheeling Hospital, Wheeling, West Virginia; and xPuget Sound Healthcare Corporation, Group Health Cooperative, University of Washington, Seattle, Washington Purpose: To present the largest series of prostate cancer brachytherapy patients treated with modern brachytherapy techniques and postimplant day 0 dosimetric evaluation. Methods and Materials: Between April 1995 and July 2006, 1,656 consecutive patients were treated with permanent interstitial brachytherapy. Risk group stratification was carried out according to the Mt. Sinai guidelines. Median follow-up was 7.0 years. The median day 0 minimum dose covering at least 90% of the target volume was 118.8% of the prescription dose. Cause of death was determined for each deceased patient. Multiple clinical, treatment, and dosimetric parameters were evaluated for impact on the evaluated survival parameters. Results: At 12 years, biochemical progression-free survival (bPFS), cause-specific survival (CSS), and overall survival (OS) for the entire cohort was 95.6%, 98.2%, and 72.6%, respectively. For low-, intermediate-, and high-risk patients, bPFS was 98.6%, 96.5%, and 90.5%; CSS was 99.8%, 99.3%, and 95.2%; and OS was 77.5%, 71.1%, and 69.2%, respectively. For biochemically controlled patients, the median posttreatment prostate-specific antigen (PSA) concentration was 0.02 ng/ml. bPFS was most closely related to percent positive biopsy specimens and risk group, while Gleason score was the strongest predictor of CSS. OS was best predicted by patient age, hypertension, diabetes, and tobacco use. At 12 years, biochemical failure and cause-specific mortality were 1.8% and 0.2%, 5.1% and 2.1%, and 10.4% and 7.1% for Gleason scores 5 to 6 and 7 and $8, respectively. Conclusions: Excellent long-term outcomes are achievable with high-quality brachytherapy for low-, intermediate-, and high-risk patients. These results compare favorably to alternative treatment modalities including radical prostatectomy. Ó 2011 Elsevier Inc. Prostate cancer, Brachytherapy, Biochemical control, Survival.
techniques and documented implant quality to better delineate efficacy and survival outcomes.
INTRODUCTION Men with newly diagnosed prostate cancer are candidates for radical prostatectomy, external beam radiotherapy (XRT), and/or brachytherapy depending on the details of their presentation. Brachytherapy is widely prescribed for men with low-risk disease and is increasingly used for men with higher-risk disease as well, based on recent reports of excellent outcomes in intermediate- and high-risk patients (1–3). However, there has been some reluctance to offer implants to higher-risk men, based primarily on unfavorable results in studies of intermediate- and high-risk brachytherapy patients treated in the 1980s and 1990s without the benefit of modern brachytherapy techniques and dosimetric plan evaluation (4, 5). The purpose of this study was to present long-term results of men treated with modern brachytherapy
METHODS AND MATERIALS Between April 1995 and July 2006, 1,656 consecutive patients with clinically localized prostate cancer, stages T1b to T3c (2002, American Joint Committee on Cancer) were treated with permanent interstitial brachytherapy by a single brachytherapist (GSM). All biopsy slides were reviewed by a single pathologist (EA). Patients were clinically staged by medical history, physical examination, including digital rectal examination and serum prostate-specific antigen (PSA) concentration determination. Bone scans and computed tomography of the abdomen and pelvis were obtained for patients with higher- but not lower-risk disease. Of the 1,656 patients, 575 patients presented with low-risk disease (PSA #10 ng/ml, Gleason score #6, and clinical stage # T2a); 608
Reprint requests to: Gregory S. Merrick, M.D., Schiffler Cancer Center, Wheeling Hospital, 1 Medical Park, Wheeling, WV 26003. Tel: (304) 243-3490; Fax: (304) 243-5047; E-mail: gmerrick@ urologicresearchinstitute.org
Conflict of interest: none. Received Nov 12, 2009, and in revised form Dec 29, 2009. Accepted for publication Jan 6, 2010. 1336
Outcome of permanent interstitial brachytherapy d A. V. TAIRA et al.
patients presented with intermediate-risk disease (1 adverse factor: PSA 10.1-19.9 ng/ml or Gleason score 7 or clinical stage T2b) and 473 presented with high-risk disease (PSA $20 ng/ml or Gleason score $8 or clinical stage $T2c or two to three of the intermediate adverse features). Of these 1,656 men, 622 (37.6%) men received androgen deprivation therapy (ADT). A total of 406 (24.5%) men received #6 months of ADT, while 216 (13.0%) men received >6 months of ADT. In patients receiving ADT, ADT was initiated at 3 months before implantation and consisted of a luteinizing hormone-releasing agonist and an antiandrogen. The maximum duration of ADT was 36 months. A total of 825 (49.8%) patients received supplemental XRT. In general, XRT consisted of 45 Gy delivered in 1.8-Gy fractions with 15 to 18 MV photons delivered via a multifield technique with customized treatment devices. For patients with #10% risk of pelvic lymph node involvement (6), the target volume consisted of the prostate gland and seminal vesicles. For patients with >10% risk of pelvic lymph node involvement, the pelvic lymph nodes were also included in the target volume. Supplemental XRT was delivered prior to implantation. The brachytherapy planning treatment volume (PTV) consisted of the entire prostate gland with a 5-mm periprostatic margin and the proximal 1.0 cm of the seminal vesicles (7, 8). All postimplant dosimetric calculations were based on day 0 evaluation of postimplant coverage of the PTV. For boost brachytherapy, the minimum peripheral dose was 90 to 100 Gy (National Institute of Standards and Technology 1999) for Pd-103 exposure. For monotherapeutic approaches, the minimum peripheral dose was 125 Gy for Pd-103 and 145 Gy (American Association of Physicists in Medicine Task Group 43) for I-125 exposure. Patients were monitored by physical examination, including digital rectal examination and serum PSA determination at 3- and 6-month intervals. The endpoint of the analysis was cause-specific survival (CSS), biochemical progression-free survival (bPFS), and overall survival (OS). Cause of death was determined for each deceased patient. Patients with metastatic prostate cancer or castration-resistant disease without obvious metastases who died of any cause were classified as dead of prostate cancer. All other deaths were attributed to the immediate cause of death. bPFS was defined as a PSA level of #0.40 ng/ml after nadir, which has been shown to be a particularly sensitive definition for identifying patients for whom treatment has failed (9). Patients whose treatment failed to achieve a nadir below 0.40 ng/ml were categorized as a biochemical failure. Multiple clinical, treatment, and dosimetric parameters were evaluated for impact on survival. Clinical and treatment variables that were continuous were compared using an independent t test. Categorical variables were compared using chi-square analysis. Kaplan-Meier analysis and curves were used to determine CSS, OS, and bPFS. Univariate Cox regression analysis was used to determine the variables that predicted bPFS, CSS, and OS. Those variables with p values of <0.10 were entered into a forward conditional, multivariate Cox regression. For all analysis, a p value of <0.05 was considered statistically significant, Statistical analysis was performed with SPSS software version 14.0 (SPSS, Chicago, Ill).
RESULTS Table 1 summarizes the clinical and treatment characteristics of men included in the analysis. Of the 1,656 men, 575 men were low-risk, 608 men were intermediate-risk, and
1337
473 men were high-risk. Median length of follow-up was 7.0 years. The median day 0 minimum dose covering at least 90% of the target volume (D90) and the V100 for all patients was 118.8% of the prescription implant dose and 97.9%, respectively. ADT was used more commonly in higher-risk patients, and long-term ADT was used almost exclusively in patients with high-risk disease. Supplemental radiation therapy also was used more commonly in men with intermediateand high-risk disease. In the men with high-risk cancer, 87.9 % received supplemental XRT, and 58.2% received hormonal therapy. At 12 years, bPFS, CSS, and OS for the group as a whole was 95.6%, 98.2%, and 72.6%, respectively (Fig. 1). Table 2 presents univariate and multivariate analyses of predictors of CSS, bPFS, and OS. Gleason score was by far the strongest predictor of CSS. Although traditional prognostic factors like risk group categorization and percent positive biopsy cores were significant on univariate analysis, these were not significant on multivariate analysis. Pretreatment PSA level, percent positive biopsy samples, Gleason score, and risk group were univariate predictors of bPFS, with percent positive biopsy samples and risk group being the strongest predictors in multivariate analysis. Age, smoking, and diabetes were most strongly associated with OS. bPFS at 12 years was 98.6%, 96.5% and 90.5%, for men with low-, intermediate-, and high-risk disease, respectively (Fig. 2a). CSS was 99.8%, 99.3%, and 95.2%, for men with low-, intermediate-, and high-risk disease (Fig. 2b). OS was 77.5%, 71.1%, and 69.2% for men with low-, intermediate-, and high-risk disease (Fig. 2c). bPFS and CSS rates stratified by Gleason score are presented in Fig. 3. For men with Gleason 5 or 6, actuarial biochemical failure at 12 years was 1.8%, and cancer-specific mortality was 0.2%. For men with Gleason 7 disease, biochemical failure at 12 years was 5.1%, and cancer-specific mortality was 2.1%. For men with Gleason 8 or higher disease, biochemical failure at 12 years was 10.4%, and cancer-specific mortality was 7.1%. For men experiencing biochemical progression, the median time to biochemical failure was 1.9 years vs. 2.4 years for men with a Gleason score of #7 and a Gleason score of $8 (p = 0.231). At the time of this analysis, the longest time spans from implant to biochemical failure was 3.2 years, 8.0 years, and 5.4 years for patients with Gleason scores of 5 to 6, 7, and $8, respectively. When patients were stratified by ADT status, the longest time spans from implant to failure for patients who were hormone-na¨ıve, taking ADT for #6 months, and taking ADT for >6 months were 8.0, 6.8, and 3.1 years, respectively. For the group as a whole, 90% of failures occurred within 3.9 years of implantation. Of the patients who experienced biochemical failure, 33% of men with Gleason scores of 5 to 6, 58% with a Gleason score of 7, and 82% with a Gleason score of $8 eventually developed castrate-resistant disease or metastases by the time of death. bPFS and CSS stratified by percent positive biopsy cores is presented in Fig. 4. Percent positive biopsy samples predicted
I. J. Radiation Oncology d Biology d Physics
1338
Volume 79, Number 5, 2011
Table 1. Clinical, treatment, and dosimetric parameters for brachytherapy patients stratified by risk No. of low-risk patients (n = 575) Parameter Continuous variable
Med.
Mean SD
Age at implant (yrs) 65.0 63.9 7.3 Follow-up (yrs) 6.9 7.1 3.1 PSA 5.7 5.8 1.9 Gleason Score 6.0 5.9 0.3 Percent positive biopsies 16.7 23.4 15.7 BMI 27.7 28.3 4.2 Prostate volume 35.4 35.8 8.5 Planning volume 65.9 65.5 12.1 V100 97.6 96.0 5.1 V150 67.4 64.6 13.8 V200 36.1 34.6 12.0 D90 116.4 116.0 12.9 Most recent PSA 0.02 0.05 0.13 Categorical variable Testosterone Low and Lower 1/3 Normal Middle 1/3 Normal Upper 1/3 normal and high Clinical stage: T1b-T2b T2c-T3b Isotope I-125 Pd-103 ADT None # 6 months > 6 months XRT Yes No Hypertension yes no Diabetes yes no Tobacco use never former current Coronary artery disease yes no Elevated cholesterol yes no Perineural invasion yes no
No. of intermediate-risk patients (n = 608) Med.
Mean SD
66.0 6.7 6.2 7.0 33.3 27.5 34.1 64.1 98.1 72.2 41.4 120.3 0.02
65.7 7.3 6.9 2.9 6.8 2.9 6.8 0.5 38.7 23.0 28.3 4.7 34.3 9.0 63.7 13.1 96.7 4.4 69.0 13.6 39.7 12.3 119.9 13.4 0.04 0.10
No. of high-risk patients (n = 473)
Total (n = 1656)
Mean SD
p value*
67.0 66.7 7.2 7.4 7.7 3.2 10.6 12.6 9.2 7.0 7.6 0.9 50.0 54.1 27.0 27.5 28.4 4.5 30.5 31.2 9.0 56.1 57.2 14.0 98.0 96.1 5.7 71.9 67.5 15.3 40.6 38.6 13.5 120.6 119.3 15.7 0.01 0.04 0.07
<0.001 <0.001 <0.001 <0.001 <0.001 0.898 <0.001 <0.001 0.020 <0.001 <0.001 <0.001 0.037
Med.
Med.
Mean SD
66.0 65.3 7.4 7.0 7.2 3.1 6.5 8.1 6.0 7.0 6.7 0.9 33.3 37.7 25.2 27.5 28.3 4.5 33.6 33.9 9.4 62.9 62.5 13.5 97.9 96.3 5.0 70.4 67.1 14.3 39.4 37.6 12.8 118.8 118.4 14.0 0.02 0.05 0.10
Count n (%)
Count n (%)
Count n (%)
212 (69.3) 71 (23.2) 23 (7.5)
223 (64.5) 86 (24.9) 37 (10.7)
132 (62.9) 52 (24.8) 26 (12.4)
0.354
567 (65.8) 209 (24.2) 86 (10.0)
575 (100) 0 (0)
587 (96.5) 21 (3.5)
250 (52.9) 223 (47.1)
<0.001
1412 (85.3) 244 (14.7)
181 (31.5) 393 (68.5)
54 (8.9) 553 (91.1)
40 (8.5) 433 (91.5)
<0.001
275 (16.6) 1379 (83.4)
392 (68.2) 176 (30.6) 7 (1.2)
444 (73.0) 145 (23.8) 19 (3.1)
198 (41.9) 85 (18.0) 190 (40.2)
<0.001
1034 (62.4) 406 (24.5) 216 (13.0)
15 (2.6) 560 (97.4)
394 (64.8) 214 (35.2)
416 (87.9) 57 (12.1)
<0.001
825 (49.8) 831 (50.2)
231 (43.1) 305 (56.9)
269 (46.5) 309 (53.5)
227 (50.4) 223 (49.6)
0.070
727 (46.5) 837 (53.5)
50 (8.8) 516 (91.2)
70 (11.7) 529 (88.3)
65 (13.8) 405 (86.2)
0.038
185 (11.3) 1450 (88.7)
244 (42.5) 251 (43.7) 79 (13.8)
235 (38.7) 274 (45.1) 99 (16.3)
152 (32.3) 239 (50.7) 80 (17.0)
0.016
631 (38.2) 764 (46.2) 258 (15.6)
91 (15.8) 484 (84.2)
99 (16.3) 509 (83.7)
93 (19.7) 379 (80.3)
0.202
283 (17.1 1372 (82.9)
178 (31.0) 397 (69.0)
187 (30.8) 421 (69.2)
143 (30.3) 329 (69.7)
0.973
508 (30.7) 1147 (69.3)
47 (8.2) 527 (91.8)
211 (35.5) 383 (64.5)
209 (45.4) 251 (54.6)
<0.001
467 (28.7) 1161 (71.3)
p valuey
Count n (%)
Abbreviations: V100, V150, and V200 = percent of the target volume receiving 100%, 150%, and 200% of prescription dose, respectively; D90 = minimum dose, as a percent of the prescription dose, received by at least 90% of target volume; ADT = androgen deprivation therapy; SD = standard deviation; XRT = supplemental external beam radiation therapy; Med. = median. * One-way analysis of variance. Values in boldface are statistically significant. y Chi-square test.
Outcome of permanent interstitial brachytherapy d A. V. TAIRA et al.
100
Survival (%)
80
60 Cause-Specific Survival, 98.2% Biochemical Progression-Free Survival, 95.6%
40
Overall Survival, 72.6%
20 n= 1656
1575
1013
475
120
0 0
3
6 9 Years Since Implant
12
15
Fig. 1. Kaplan-Meier curves for biochemical progression-free, cause-specific, and overall survival. Each curve represents the same patients.
for both bPFS and CS. Pretreatment PSA level predicted for both bPFS and CSS (Fig. 5). DISCUSSION Studies of men treated with earlier brachytherapy techniques and planning parameters demonstrate what by today’s standards would be considered relatively poor outcomes (4, 5). Based on these reports and nomograms from the same era (10), a perception remains that outcomes in intermediate- and higher-risk brachytherapy patients may not be as good as alternative treatment options. However, over the past 10 to 15 years, brachytherapy techniques and planning have improved significantly. The current study represents the largest series to date of men with long-term follow-up treated with modern brachytherapy techniques and complete postimplant dosimetric evaluation. With current techniques and high-quality implants, excellent long-term outcomes can be achieved in men with low-, intermediate-, and high-risk disease. Specifically, very low rates of prostate cancer-specific mortality are achievable even in high-risk patients. Previously, Zelefsky et al. (5) reported outcomes for a multicenter cohort of 2,693 brachytherapy patients. However, postimplant dosimetric information was available for only 639 patients. With a median follow-up of 63 months, 8-year bPFS for low-, intermediate-, and high-risk patients was 74%, 61%, and 39%, respectively. In the cohort for whom dosimetric data were available, the postimplant D90 was lower than currently recommended. Stone et al. (11) reported a multicenter cohort study of 3,928 brachytherapy patients with a median follow-up of 42 months, all of whom had dosimetric information available. For the cohort as a whole, bPFS for low-, intermediate-, and high-risk patients was 84%, 77%, and 64%. However,
1339
the 1,100 men who received a higher biologically equivalent dose (BED) of >200 Gy via their implant, with or without XRT, had much more favorable outcomes. Among these men, bPFS for low-, intermediate-, and high-risk patients was 88%, 94%, and 90%, respectively. All of the 1,656 men in the current study have complete postimplant dosimetric evaluation with implant quality assessed by standard dosimetric criteria (Table 1). We report excellent long-term bPFS, with 12-year bPFS of 95.6% for the cohort overall. The 12-year bPFS for low-, intermediate-, and high-risk patients was 98.6%, 96.5%, and 90.5%, respectively. Although bPFS is a useful surrogate for treatment efficacy, cancer-specific survival is the most important measure of treatment success. In our study, Gleason score was the strongest predictor of cancer-specific mortality. Few patients with Gleason scores of 5 to 6 experience biochemical failure (1.8% at 12 years), and only a few of those succumbed to prostate cancer (0.2%). On the other hand, even though the biochemical failure rate at 12 years was only 10.4% for men with a Gleason score of $8, prostate cancer-specific mortality was 7.1%, suggesting that men with high-risk disease who experience biochemical failure are much more likely to die of prostate cancer. Our results compare favorably to a recently published report presenting long-term prostate cancer-specific mortality in men treated with radical prostatectomy (RP) at several leading academic medical centers (12). That study consisted of more than 12,500 men treated in the serum PSA level era (median follow-up, 46 months) and represents the largest multiinstitutional study of RP with long-term follow-up. In comparison, prostate cancer-specific mortality for a biopsy sample Gleason score of 5 to 6 at 12 years was 0.2% in brachytherapy patients, compared to 2% and 6% at 10 and 15 years for RP patients. For men with a biopsy Gleason score of 7, the brachytherapy prostate cancer-specific mortality at 12 years was 2% compared to 5% and 17% at 10 and 15 years for RP patients. For patients with biopsy Gleason scores of $8, the risk of cancer-specific death at 12 years for brachytherapy patients was 7% for brachytherapy and 16% and 34% at 10 and 15 years for RP. We expect our results will be durable, particularly in regard to prostate cancer mortality. Over 90% of biochemical failures in our series occurred within 3.9 years of implantation. Among men with Gleason scores of 8 or higher (the group most likely to die of disease), the average time from implantation to biochemical failure was 2.4 years. In our cohort thus far, there have been no biochemical failures in patients with Gleason scores of $8 beyond 5.4 years. In addition, there appears to be a clear plateau on the bPFS curves (Fig. 3a). The timing of biochemical failure in our cohort differs notably from the timing of failure in previously reported series in which implant doses to the prostate were lower. For instance, in the earlier multiinstitutional brachytherapy series reported by Zelefsky et al. (5), approximately half of biochemical failures occurred after 4 years from the time of
I. J. Radiation Oncology d Biology d Physics
1340
Volume 79, Number 5, 2011
Table 2. Univariate and multivariate analyses for CSS, bPFS, and OS Biochemical progression-free survival
Cause-specific survival Univariate analysis Variable
p value
Age at implant (yrs) PSA (ng/mL) Gleason score Body mass index Prostate volume Percent positive biopsy samples D90 XRT (yes/no) ADT (yes/no) ADT duration (0/<6/>6 months) Risk group High vs. low High vs. intermediate Hypertension (yes/no) Diabetes (yes/no) Tobacco use (never/former/current) never vs former never vs current Coronary artery disease (yes/no) Hypercholesterolemia (yes/no) Perineural invasion (yes/no)
0.413 0.089 <0.001 0.898 0.183 0.001 0.905 0.010 0.467 0.308 0.001 0.585 0.008 0.032 0.363 0.186
Multivariate analysis
HR
p value
2.886
0.276 <0.001
1.028
0.138
14.044
0.104
HR
2.804
0.031
p value 0.348 <0.001 <0.001 0.371 0.200 <0.001 0.104 <0.001 0.411 0.165
0.083 15.564 0.256
Univariate analysis
0.254
<0.001 0.189 <0.001 0.093 0.848 0.095
Overall survival
Multivariate analysis
HR
p value
1.045 1.757
0.462 0.781
1.026
<0.001
3.108
0.792
6.6647
<0.001 0.596 0.002
Univariate analysis
Multivariate analysis
HR
p value
HR
p value
HR
1.101
<0.001
1.109
1.138 0.968
0.722 0.937
1.017
<0.001 0.361 0.046 0.036 0.375 0.192 0.053 0.601 0.510 0.506
0.693 1.397 1.433 0.779 1.566
0.327
0.041 0.030 0.021 0.043 0.008 <0.001
0.022
0.005 <0.001 0.002
1.517 2.331 1.584
3.868
0.011 <0.001
1.542
0.017 <0.001 0.129
1.434 2.859
0.319
0.053
0.219
0.039
0.517
0.102
0.357
0.639
0.544
0.015
1.837
0.677
0.697
0.809
0.305
Abbreviations: HR = hazard ratio. Values in boldface are statistically significant.
the presence of occult metastatic disease at the time of treatment. We assume that some significant portion of the failures in series with lower prostate doses was due to local disease persistence. When those patients failed, it is not surprising that it could take many years for a local tumor burden to
implant. In contrast, in the current report, nearly all failures occurred within the first 4 years. We postulate that there may be a difference in timing of biochemical failure in patients who fail due to local disease persistence/recurrence versus those patients who fail due to
c.
b. 100
100
100
80 Low Risk, 98.6% Intermediate Risk, 96.5% High Risk, 90.5%
60
40
p < 0.001
20
80
80
Low Risk, 99.8% Intermediate Risk, 99.3% High Risk, 95.2%
Overall Survival (%)
Cause-Specific Survival (%)
Biochemical Progression-Free Survival (%)
a.
60 p < 0.001 40
60
Low Risk, 77.5% Intermediate Risk, 71.1% High Risk, 69.2%
40
20
20
0
p = 0.004
0 0
3
6
9
Years Since Implant
12
15
0 0
3
6
9
Years Since Implant
12
15
0
3
6
9
Years Since Implant
Fig. 2. (a) Kaplan-Meier curves for biochemical progression-free survival, stratified by risk. (b) Kaplan-Meier curves for CSS stratified by risk. (c) Kaplan-Meier curves for OS, stratified by risk. (p values represent single tests for linear trends across all factor levels.)
12
15
Outcome of permanent interstitial brachytherapy d A. V. TAIRA et al.
b.
a.
100
100 Cause-Specific Survival (%)
Biochemical Progression -Free Survival (%)
1341
80 5-6, 98.2% 7, 94.9% 8-10, 89.6%
60
p < 0.001 40
80 5-6, 99.8% 7, 97.9% 8-10, 92.9%
60
p < 0.001 40
20
20
0
0 0
3
6
9
12
15
0
3
Years Since Implant
6
9
12
15
Years Since Implant
Fig. 3. (a) Kaplan-Meier curves for biochemical progression-free survival stratified by Gleason score. (b) Kaplan-Meier curves for CSS stratified by Gleason score. (p values represent single tests for linear trends across all factor levels.)
presents longer-term follow-up based on staging at clinical presentation. Compared with similarly staged patients who undergo RP, the long-term results of brachytherapy when stratified by Gleason score and/or PSA compare favorably.
recover to a volume significant enough to trigger PSA level progression. Higher ablative doses to the prostate, as in this study, should result in improved local disease eradication. Hence, patients who fail despite receiving high ablative doses are probably more likely to have had occult metastatic disease at presentation. When patients fail due to undetected metastatic disease, it is possible that it occurs sooner than when a patient fails from incompletely eradicated local disease. Patients treated with brachytherapy rarely undergo pathology staging. Yet, at the time a patient decides on one definitive treatment option over another, pathology staging is not available. One of the difficulties of comparing results between different treatment modalities is that some publications will present outcomes based upon initial clinical staging (XRT and brachytherapy publications), while others will be based primarily on pathology staging (generally RP series). Also, some publications have relatively short follow-up, begging the question of longer-term outcomes. This report
Excellent long-term prostate cancer outcomes are achievable with high-quality brachytherapy. This work builds on a number of studies demonstrating excellent short- and intermediate-term results with primary brachytherapy, particularly in men with higher-risk disease (1–3, 13). Based on our data and those of others, we believe primary brachytherapy should be considered a viable treatment option for men with low-, intermediate-, and high-risk prostate cancer. If long-term data from other institutions supports our findings, primary brachytherapy may become the preferred treatment option for men with higher-risk disease.
b. 100
100
80
Cause-Specific Survival (%)
Biochemical Progression-Free Survival (%)
a.
CONCLUSIONS
< 34%, 98.0% 34-50%, 93.1% > 50%, 91.5%
60
p < 0.001 40
20
80 < 34, 99.2% 34-50, 98.4% > 50, 95.5%
60
p = 0.030 40
20
0
0 0
3
6
9
Years Since Implant
12
15
0
3
6
9
12
15
Years Since Implant
Fig. 4. (a) Kaplan-Meier curves for biochemical progression-free survival, stratified by percent positive biopsy. (b) Kaplan-Meier curves for CSS stratified by percent positive biopsy. (p values represent single tests for linear trends across all factor levels.)
I. J. Radiation Oncology d Biology d Physics
1342
b. 100
100 Cause-Specific Survival (%)
Biochemical Progression-Free Survival (%)
a.
Volume 79, Number 5, 2011
80 PSA < 10, 96.9% PSA > 20, 92.3% PSA 10-20, 90.2%
60
p < 0.001 40
20
0
80 PSA <10, 98.8% PSA > 20, 97.2% 60
PSA 10-20, 95.3%
40
p = 0.004
20
0 0
3
6
9
12
15
0
Years Since Implant
3
6
9
12
15
Years Since Implant
Fig. 5. (a) Kaplan-Meier curves for biochemical progression-free survival, stratified by PSA level. (b) Kaplan-Meier curves for CSS stratified by PSA level. (p values represent single tests for linear trends across all factor levels.)
REFERENCES 1. Kao J, Stone NN, Lavaf A, et al. (125)I monotherapy using D90 implant doses of 180 Gy or greater. Int J Radiat Oncol Biol Phys 2008;70:96–101. 2. Merrick GS, Butler WM, Wallner KE, et al. Impact of supplemental external beam radiotherapy and/or androgen deprivation therapy on biochemical outcome after permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2005;61:32–43. 3. Stone NN, Potters L, Davis BJ, et al. Multicenter analysis of effect of high biologic effective dose on biochemical failure and survival outcomes in patients with Gleason score 7-10 prostate cancer treated with permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2009;73:341–346. 4. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998;280:969–974. 5. 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;67:327–333. 6. Partin AW, Mangold LA, Lamm DM, et al. Contemporary update of prostate cancer staging nomograms (Partin Tables) for the new millennium. Urology 2001;58:843–848.
7. Merrick GS, Butler WM. Modified uniform seed loading for prostate brachytherapy: Rationale, design, and evaluation. Tech Urol 2000;6:78–84. 8. Merrick GS, Butler WM, Wallner KE, et al. Extracapsular radiation dose distribution after permanent prostate brachytherapy. Am J Clin Oncol 2003;26:e178–189. 9. Kuban DA, Levy LB, Potters L, et al. Comparison of biochemical failure definitions for permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2006;65:1487–1493. 10. Kattan MW, Potters L, Blasko JC, et al. Pretreatment nomogram for predicting freedom from recurrence after permanent prostate brachytherapy in prostate cancer. Urology 2001;58:393–399. 11. Stone NN, Potters L, Davis BJ, et al. Customized dose prescription for permanent prostate brachytherapy: Insights from a multicenter analysis of dosimetry outcomes. Int J Radiat Oncol Biol Phys 2007;69:1472–1477. 12. Stephenson AJ, Kattan MW, Eastham JA, et al. Prostate cancerspecific mortality after radical prostatectomy for patients treated in the prostate-specific antigen era. J Clin Oncol 2009;27:4300–4305. 13. D’Amico AV, Moran BJ, Braccioforte MH, et al. Risk of death from prostate cancer after brachytherapy alone or with radiation, androgen suppression therapy, or both in men with highrisk disease. J Clin Oncol 2009;27:3923–3928.