- Email: [email protected]
Patterns of Local Failure Following Prostate Brachytherapy
Patterns of Local Failure Following Prostate Brachytherapy Nelson N. Stone,* Richard G. Stock, Ida White and Pam Unger From the Departments of Urology (NNS), Radiation Oncology (RGS) and Pathology (IW, PU), Mount Sinai School of Medicine, New York, New York
Purpose: We describe biopsy results in patients with prostate cancer treated with brachytherapy. Materials and Methods: A total of 1,562 men with localized prostate cancer were treated with permanent prostate brachytherapy, of whom 508 agreed to ultrasound guided biopsies 2 years after the completion of all therapy. Median followup was 6.7 years (range 2 to 14.6) and median prostate specific antigen was 7.4 ng/ml (range 0.3 to 300). Disease was categorized as Gleason score less than 7 in 74.8% of patients, stage T2a or less in 64.2%, low risk in 43.1%, intermediate risk in 24.2% and high risk in 32.7%. Of the 508 men 315 (62%) received 125I, 110 (21.7%) received 103Pd and 83 (16.3%) received 103Pd and external beam radiotherapy. A total of 237 men (46.7%) received a short course of hormonal therapy (3 to 9 months). Subsequent biopsies were performed after 2 years if initial biopsy was positive or prostate specific antigen increased. Post-implantation dosimetry results were grouped into low, normal and high dose. Associations were tested by chi-square analysis. Survival functions were calculated with Kaplan-Meier analysis and Cox regression. Results: A total of 643 biopsies were performed in 508 men between 2 and 11 years after implantation. Of the 508 men 39 (7.7%) had a final positive biopsy. Positive biopsy was associated with high prostate specific antigen (p ⫽ 0.035), stage (p ⫽ 0.003), risk (p ⫽ 0.024), no hormonal therapy (p ⫽ 0.002) and low dose (p ⬍0.0001). On multivariate analysis only dose and hormonal therapy were significant (p ⬍0.0001 and p ⫽ 0.004, respectively). Of the patients 80% were free of PSA failure at 10 years if final biopsy was negative compared to 27.3% with a positive biopsy (p ⬍0.0001). Death from prostate cancer was associated with a positive biopsy (OR 18.5, 95% CI 2.3–143, p ⬍0.0001). Of the 52 men with a positive biopsy at year 2, 23 (44.2%) had negative results on subsequent biopsy, while 10 of the 456 (2.2%) with negative 2-year biopsies showed positive results. Positive biopsy occurred in the prostate only in 31 of 39 men (79.5%), in the prostate and seminal vesicles in 3 (7.7%), and in the seminal vesicles only in 5 (12.8%). Conclusions: Patients undergoing prostate brachytherapy must receive an adequate radiation dose to eradicate local disease. Hormonal therapy may benefit local control in patients with intermediate to high risk disease. Extraprostatic biopsies should be performed in patients with local failure who are considering salvage therapy to rule out seminal vesicle involvement. Key Words: prostate; prostatic neoplasms; brachytherapy; biopsy; neoplasm recurrence, local
ermanent seed brachytherapy has become a popular treatment for localized prostate cancer. Brachytherapy can be performed as monotherapy or in combination with HT and/or EBRT.1,2 Different treatments are used for different stages of disease with monotherapy most often reserved for low risk cancer and combined modality therapy used for high risk disease. Treatment outcome is usually measured by biochemical disease control.3 The problem with using PSA as an end point for treatment failure is that it does not separate men with systemic vs local failure. In the past digital rectal examination was used to assess local control. Today we know that posttreatment prostate biopsy is the best method to detect residual or recurrent local disease. There are few reports in the literature that address the long-term results of prostate biopsy following brachytherapy. None evaluated the consequences of sequential biopsy
P
Submitted for publication August 22, 2006. Presented at annual meeting of American Urological Association, Atlanta, Georgia, May 25–26, 2006. * Correspondence: 21 Timber Trail, Suffern, New York 10901 (e-mail: [email protected]).
For another article on a related topic see page 1913.
0022-5347/07/1775-1759/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION
results or the incidence of extraprostatic relapse in these patients. We studied a large cohort of patients with brachytherapy who underwent prostate biopsy a minimum of 2 years after implantation and subsequent biopsy if the initial one was positive or the patient experienced PSA relapse. Biopsy results were correlated with initial patient PSA, Gleason score, HT and implant dose. MATERIALS AND METHODS A total of 1,562 men with biopsy proven prostate cancer were treated with permanent seed implantation at 1 institution. All pathological findings were reviewed by 1 pathologist (PU). Patients were staged using the 1992 American Joint Committee on Cancer system and those with initial PSA more than 10 ng/ml or Gleason score greater than 6 underwent bone and computerized tomography. In addition, most patients with PSA more than 10 ng/ml, Gleason score 7 or greater, or stage T2b or greater underwent SVBs.4 Patients with a positive SVB, Gleason 7 or greater with perineural invasion or PSA more than 30 ng/ml also underwent laparoscopic pelvic lymph node dissection.5 Those with positive pelvic lymph nodes were excluded.
1759
Vol. 177, 1759-1764, May 2007 Printed in U.S.A. DOI:10.1016/j.juro.2007.01.069
1760
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
Patients were offered a treatment protocol based on staging results. Low risk (PSA less than 10 ng/ml, Gleason score 6 or less, stage T2a or less and prostate volume less than 50 cc) were treated with an 125I implant alone to a prescription dose of 160 Gy (Task Group 43). Patients at low risk with a prostate of 50 cc or greater were also placed on 3 months of neoadjuvant HT before implantation. About 50% of these patients also continued on HT for an additional 2 months. Those at intermediate risk (PSA 10 to 20 ng/ml, Gleason 7 or stage T2b and negative SVB) were treated with 6 months of HT, including 3 before implantation, and 103Pd to a dose of 124 Gy (National Institute of Standards and Technology 1999). Patients at high risk (PSA more than 20 ng/ml, Gleason 8 –10, stage T2c–T3 or positive SVB with negative laparoscopic peritoneal lymph node dissection) were treated with 9 months of HT, including 3 before implantation, and 103Pd to a dose of 100 Gy (National Institute of Standards and Technology 1999), followed 2 months later by 45 Gy EBRT in 25 fractions.2 Patients with a positive SVB received seed placement in the SVs with the intent of providing a radiation boost to the proximal 50% of the SVs2. Median age of the 508 patients with biopsy was 66 years (range 41 to 88) and median PSA was 7.4 ng/ml (range 0.3 to 300). Patients were categorized with Gleason score less than 7 in 74.8%, stage T2a or less in 64.2%, low risk in 43.1%, intermediate risk in 24.2% and high risk in 32.7%. Median implant prostate volume was 36.5 cc (range 2.4 to 114). Of the 508 men 315 (62%) received 125I, 110 (21.7%) received 103 Pd and 83 (16.3%) received 103Pd with EBRT (table 1). A total of 237 patients (46.7%) received a short course of HT (3 to 9 months). One month after implantation computerized tomography based dosimetry was performed using ADAC Pinnacle or Varian software. These data were separated into 3 groups based on post-implantation dosimetry D90 results. Patients receiving a dose of less than 116 Gy 125I or less than 100 Gy 103Pd were considered low dose. A delivered dose of more than 116 to 160 Gy 125I and more than 100 to 124 Gy was considered normal dose. Anything above these doses or when combination therapy was used were considered high dose. The rationale for using these dose cutoff points was based on reports in the literature suggesting that an adequate implant should deliver at least 80% of the typical prescription dose (145 Gy for 125I and 124 Gy for 103Pd).6
TABLE 1. Patient characteristics No. (%) PSA (ng/ml): 4 or Less 4.1–10 10.1–20 Greater than 20 Stage: T1b–T2a T2b–c T3 Gleason score: 2–6 7 8–10 HT: Yes No Risk: Low Intermediate High
50 (9.8) 302 (59.5) 112 (22.1) 44 (8.6) 317 (62.4) 180 (35.5) 11 (2.1) 380 (74.8) 78 (15.4) 50 (9.8) 272 (53.3) 236 (46.7) 219 (43.1) 123 (24.2) 166 (32.7)
Prostate biopsy was offered 2 years after completion of all therapy. It was performed under ultrasound guidance with a minimum of 6 cores taken. All biopsies were read by 1 pathologist (PU) experienced with evaluating post-irradiation specimens. Biopsies were scored as negative or positive only. Any tissue with remaining cancer, even with radiation effect, was considered positive. Patients agreeing to biopsy underwent repeat biopsy yearly if 2-year biopsy was positive until biopsy became negative or there was clear evidence of PSA progression (minimum of 3 increases above a nadir). In addition, patients with evidence of PSA progression in whom initial posttreatment biopsies were negative were offered repeat biopsies. They had a minimum of 12 core samples taken (range 12 to 30) with at least 6 from the prostate and 6 from the SVs. Patients with 2 sets of negative biopsies (12 cores) and increasing PSA underwent saturation biopsies with 30 core samples taken. This procedure was done by a mapping technique using anesthesia. Associations were tested by chi-square analysis and ANOVA. Survival functions were computed with KaplanMeier analysis and Cox regression. Results of the last biopsy were used for association analyses with p ⬍0.05 considered significant. RESULTS Median followup was 6.7 years (range 2 to 14.6). A total of 643 biopsies were performed in 508 men, including 1 to 5 biopsies in 416, 61, 22, 6 and 3, respectively. Of the 508 men 39 (7.7%) had PB and 99 (19.5%) experienced PSA failure. PB was associated with PSA greater than 10 (p ⫽ 0.035), stage greater than T2a (p ⫽ 0.003), risk greater than low (p ⫽ 0.024), no HT (p ⫽ 0.002) and low radiation dose (p ⬍0.0001, table 2). Gleason score was not associated with a PB. On multivariate analysis only dose and HT were significant (p ⬍0.0001 and p ⫽ 0.004, respectively, table 3). Patients were grouped into risk groups and variables associated with a PB were determined (table 4). Dose was the only significant variable in patients at low risk (p ⬍0.001), while HT (p ⫽ 0.008) and dose (p ⫽ 0.039) were significant in those at intermediate risk, and HT (p ⬍0.001), dose (p ⫽ 0.035) and combination therapy (p ⫽ 0.026) were significant in those at high risk. In patients at intermediate risk treated with a high dose HT decreased the positive biopsy rate to 2% (1 of 50) from 10.8% (4 of 37) in those not treated with HT (p ⫽ 0.08). In patients at high risk treated with a high dose HT decreased the positive biopsy rate to 6.5% (8 of 124) from 28.6% (12 of 42) (OR 4.4, 95% CI 1.9 –10.1, p ⬍0.0001). When these patients were further analyzed by implant type (103Pd vs partial 103Pd plus EBRT), those treated with combination therapy with HT had a decrease in PBs to 3.3% (2 of 60) compared to 11.8% (6 of 51) in those receiving 103Pd plus HT (p ⫽ 0.087). The 10-year biological failure-free rate in the entire cohort was 84%. Of the 508 men with biopsies 80% were free of PSA failure at 10 years if final biopsy was negative compared to 27.3% for a PB (p ⬍0.0001, see figure). Of the patients 102 (6.5%) died during the study period, including 14 (0.9%) who died of prostate cancer. Biopsy data were available on 43 deceased patients (41.2%). Of these 43 patients 1 of 37 (2.5%) had a PB and died of other causes, while 3 of 6 (50%) had a PB and died of prostate cancer (OR 18.5,
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY TABLE 2. Univariate analysis of positive prostate biopsy Variable PSA (ng/ml): Less than 10 10.1–20 Greater than 20 Stage: T1–T2a T2b T2c–T3 Risk: Low Intermediate High Prostate vol (cc): 30 or Less 30.1–50 Greater than 50 HT: Yes No Implant: ⫹ EBRT Alone Dose: Low Normal High
No. Pos Biopsy/ Total No. (%) 21/347 (6.1) 10/112 (8.9) 8/49 (16.3) 15/347 (4.6) 19/135 (7.3) 5/56 (8.9)
p Value
0.035
0.003
10/219 (4.6) 9/123 (7.3) 20/166 (12)
0.024
21/196 (10.7) 12/225 (5.3) 4/78 (5.1)
0.07
9/237 (3.8) 30/271 (11.1)
0.002
2/83 (2.4) 37/425 (8.7)
0.03
15/51 (29.4) 4/66 (6.1) 16/367 (4.4)
⬍0.001
95% CI 2.3–143, p ⬍0.0001). A Cox regression proportional hazard model was performed to assess competing causes of death in the subset of 508 patients evaluated by biopsy (table 5). Higher 1-year post-implantation PSA predicted a subsequent final PB vs a negative biopsy (mean 2.1 vs 0.99 ng/ml, p ⬍0.0001). However, lower 1-year PSA was also associated with HT, that is 0.5 ng/ml with HT vs 1.57 ng/ml without HT (p ⬍0.0001). Linear regression analysis with HT plus 1-year PSA demonstrated that each was significant (p ⫽ 0.048 and 0.001, respectively). Of the 52 men with a PB at year 2, 23 (44.2%) had negative results on subsequent biopsy. The likelihood of positive or negative results in these patients at subsequent year biopsies was the same for all 5 years. In contrast, 10 of the 456 patients (2.2%) with negative 2-year biopsies showed positive results on subsequent biopsy, which was performed because of increasing PSA. Mean time to local failure was 4.5 years (range 2 to 11). Mean time to PSA failure in patients with a negative last biopsy was 3.3 compared to 5.2 years in those with final PB (p ⫽ 0.036). A PB occurred in the prostate only in 31 of 39 men (79.5%), in the prostate and SVs in 3 (7.7%), and in the SVs only in 5 (12.8%).
1761
TABLE 4. Positive biopsy by risk group Variable Low risk: Low dose Normal dose High dose Intermediate risk: No HT HT Low dose Normal dose High dose High risk: No HT HT Low dose Normal dose High dose 125 I 103 Pd 103 Pd/EBRT
No. Pos Biopsy/ Total No. (%)
p Value
2/4 (50) 1/20 (5) 7/192 (3.6)
⬍0.001
8/57 (14) 1/66 (1.5) 3/11 (27.3) 1/18 (5.6) 5/87 (5.7) 12/42 (28.6) 8/125 (6.5) 10/36 (27.8) 2/28 (7.1) 4/88 (4.5) 6/37 (16.2) 12/67 (17.9) 2/62 (3.2)
0.008 0.039 ⬍0.001 0.035 0.026
option for prostate cancer. Data in this study suggest that it can but only if certain conditions are met. As determined by biopsy in this study, the overall local failure rate was 7.7%. The primary requirement for disease eradication was the delivery of higher doses (table 4). Patients receiving the highest doses had a local failure rate of 3.6% for low risk (p ⬍0.001), 5.7% for intermediate risk (p ⫽ 0.039) and
DISCUSSION Brachytherapy should be able to eradicate local disease in most patients if it is to be considered a reasonable treatment TABLE 3. Multivariate analysis of positive prostate biopsy after brachytherapy Variable
p Value
Risk group PSA Stage HT Additional EBRT Dose
0.804 0.550 0.653 0.004 0.650 ⬍0.001
Biochemical (PSA) freedom from failure by biopsy results. Cum, cumulative.
1762
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
TABLE 5. Cox regression proportional hazard model of competing causes of death in 508 patients evaluated by biopsy Variable
Wald
p Value
Stage Gleason PSA HT Risk group Dose Biopsy result Supplemental EBRT
4.1 20.5 1.5 18.9 21.4 23.1 1.0 8.5
0.043 ⬍0.001 0.221 ⬍0.001 ⬍0.001 ⬍0.001 0.307 0.004
4.5% for high risk (p ⫽ 0.035) disease. For individual isotopes this required a D90 of at least 160 Gy for 125I and 124 Gy for 103Pd or combination therapy. In addition, HT appeared to further improve biopsy results in patients at intermediate and high risk who received the higher doses. While the difference was not quite significant (p ⫽ 0.087), combination therapy in patients at high risk treated with HT appeared to further decrease PB rates from 11.8% to 3.3%. The need to deliver higher doses of radiation was shown by a number of investigators to improve biochemical relapse-free survival in patients treated with prostate brachytherapy.7,8 In a small cohort of patients treated with 125I Stone et al also previously observed that those treated to a dose of at least 140 Gy had a 2-year postimplantation PB rate of 4.8 % vs 20.5% in those not achieving that dose (p ⫽ 0.002).1 The current study with longer followup in a larger cohort of patients further substantiates these results. Regardless of patient risk category higher doses are required to eradicate disease. It is interesting that Gleason score was not a significant variable for predicting local failure. Patients with high Gleason scores have a substantially greater rate of biochemical relapse than those with lower scores. While high grade lesions may be more likely to metastasize, they are not more difficult to eradicate locally if high enough radiation doses are delivered. Zelefsky et al studied the impact of higher radiation doses in patients receiving conformal EBRT.9 Of 480 patients who underwent post-EBRT biopsies 121 (25.2%) had positive findings. The positive biopsy rate decreased from 90 of 320 patients (28.1%) for doses 70 Gy or less to 27 of 119 (22.7%) for 75.6 Gy and to 4 of 41 (9.8%) for 81 Gy doses. Recently Stock et al described the effect of BED on biopsy outocmes.10 D90s were converted into a BED for each isotope, which was summed with the BED of EBRT for patients with combination therapy. This allowed the different treatment regimens to be assigned a BED value and the effect of different BED levels on biopsy outcome to be determined. Multivariate analysis revealed that only HT and BED significantly affected biopsy results (p ⫽ 0.05 and 0.006, respectively). The 10-year biological failure-free rate in the entire cohort was 84%. Of patients undergoing prostate biopsies 80% were free of failure if biopsy was negative vs 27.3% if the final biopsy was positive (p ⬍0.0001). The significance of a positive post-irradiation prostate was controversial in the past. The current study with long-term followup and repeat biopsies confirms that a PB leads to disease progressions. In fact, a positive biopsy was 18.5 times more likely to result in prostate cancer death than a negative one (p ⬍0.0001). How-
ever, in the Cox proportional hazards model the last biopsy result was not significant (table 5). This result most likely reflects the low number of deaths (43) in the 508 patients at this time. Longer followup is needed to adequately address this issue. In this study routine prostate biopsies were offered 2 years after therapy regardless of disease status and 10% of the patients (52 of 508) had a positive biopsy. Of these cases 44% eventually reverted to negative results. This is in contrast to the study reported by Prestidge et al, in which 22% of posttreatment biopsies were positive or indeterminate after 1 year.11 Most indeterminate pathological findings in this report demonstrated prostate cancer with a radiation treatment effect. Almost all positive 2-year biopsies in our patients demonstrated the same results. While Prestidge et al believed that most of these results would clear with time, although this was not actually shown in their study, our data suggest that this happens less than half of the time. These data can present a conundrum to the urologist who elects to perform biopsy in patients 2 years after brachytherapy. While this is not routinely performed in clinical practice, the decision to perform biopsy after brachytherapy may be made if the patient experiences an increase in PSA after attaining a nadir. Unfortunately up to 31% of patients can experience a temporary increase in PSA after brachytherapy that is not associated with local or systemic progression.12 Our suggestion would be to observe these patients and not perform biopsy if the delivered D90 dose was high. If the patient experiences several consecutive PSA increases, biopsy may become necessary. The PSA nadir at 1 year appears to be somewhat predictive of local control. Patients with 1-year PSA less than 1 ng/ml were far less likely to experience a PB than patients with PSA greater than 2.0 ng/ml (p ⬍0.0001). However, these results are also confounded by HT use, which also contributed to a lower 1-year PSA and to the final biopsy results on multivariate analysis. Late increasing PSA may also suggest local failure. In fact, local failure was identified in 1 of our patients 11 years after implantation. Of patients experiencing biochemical failure those with a negative biopsy were more likely to experience relapse earlier than patients with a PB (3.3 vs. 5.2 years, p ⫽ 0.036). This relationship between the failure pattern and time to PSA failure is similar to that seen with radical prostatectomy. The type of local failure was also noteworthy. Of the cases 20.5% failed in the prostate and SVs or the SVs alone. If a salvage procedure is considered in a patient experiencing local failure after brachytherapy, biopsy of the SVs is necessary to rule out extraprostatic disease.
CONCLUSIONS Patients undergoing prostate brachytherapy must receive an adequate radiation dose to eradicate local disease. Local failure can occur as long as 11 years after implantation. HT may benefit local control in patients with intermediate to high risk disease. Extraprostatic biopsies should be performed in patients with local failure who are considering salvage therapy to rule out SV involvement.
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
Abbreviations and Acronyms BED D90 EBRT HT PB PSA SV SVB
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
biological equivalent dose dose to 90% of the area external beam radiotherapy hormonal therapy last positive biopsy prostate specific antigen seminal vesicle SV biopsy
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Stone NN, Stock RG and Unger P: Intermediate-term biochemical and local control following I-125 brachytherapy for prostate cancer. J Urol 2005; 173: 803. Stock RG, Cahlon O, Cesaretti J, Kolhmeier MA and Stone NN: Combined modality treatment in the management of high risk prostate cancer. Int J Rad Oncol Biol Phys 2004; 59: 1352. Cox JD, Grignon DJ and Kaplan RS: Consensus statement: guidelines for PSA following radiation therapy. Int J Radiat Oncol Biol Phys 1997; 37: 1035. Linzer DG, Stock RG, Stone NN, Ianuzzi C, Ratnow R and Unger P: Seminal vesicle biopsy: accuracy and implications for staging of prostate cancer. Urology 1996; 48: 757. Stone NN, Stock RG, Parikh D, Yeghiayan P and Unger P: Perineural invasion and seminal vesicle involvement predict pelvic lymph node metastasis in men with localized carcinoma of the prostate. J Urol 1998; 160: 1722. Willins J and Wallner K: CT-based dosimetry for transperineal I-125 prostate brachytherapy. Int J Rad Oncol Biol Phys 1997; 39: 347. Stock RG, Cesaretti J and Stone NN: Disease specific survival following brachytherapy management of prostate cancer. Int J Rad Oncol Biol Phys 2006; 64: 810. Potters L, Cao Y, Calugaru E, Torre T, Fearn and Wang XH: A comprehensive review of CT-based dosimetry parameters and biochemical control in patients treated with permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2001; 50: 605. Zelefsky MJ, Fuks Z, Hunt M, Lombardi D, Ling CC, Rueter VE et al: High dose radiation delivered by intensity modulated conformal radiotherapy improves the outcome of localized prostate cancer. J Urol 2001; 166: 876. Stock RG, Stone NN, Cesaretti J and Rosenstein BS: Biologically effective dose values for prostate brachytherapy: effects on PSA failure and posttreatment biopsies. Int J Rad Oncol Biol Phys 2006; 64: 527. Prestidge BR, Hoak DC, Grimm PD, Ragde H, Cavanagh W and Blasko JC: Post-treatment biopsy results following interstitial brachytherapy in early-stage prostate cancer. Int J Radiat Oncol Biol Phys 1997; 37: 31. Stock RG, Stone NN and Cessaretti J: Prostate specific antigen bounce following prostate seed implantation for localized prostate cancer: descriptions and implications. Int J Rad Oncol Biol Phys 2003; 56: 448.
EDITORIAL COMMENTS These authors report extensive experience with postbrachytherapy prostate biopsies offered 2 years after therapy. Of the cohort of 1,562 men 508 accepted biopsy, of whom 61 underwent a second biopsy and 31 underwent 3 or more biopsies. The population consisted of 43% of men with low risk, 24% with intermediate risk and 33% with high risk disease. HT (3 to 9 months) was done in 37% of the patients
1763
and 16% received supplemental EBRT. Of the 52 men who had PB after 2 years 44% subsequently converted to negative results and only 7.7% had positive results at final biopsy. Only dose was predictive of PB in the low risk group. For those at intermediate risk dose and HT were predictive and in those at high risk dose, HT and the addition of EBRT were predictive. These results emphasize the importance of quality in brachytherapy outcome. Although the authors did not specifically address the question of the timing of post-brachytherapy biopsies,1 44% of biopsies that were positive at 2 years subsequently reverted to negative. This creates a clinical dilemma for urologists faced with increasing PSA early after brachytherapy. Of men 30% to 40% experience a transient increase in PSA within the first 2 to 2½ years.2,3 Biopsy at this time cannot reliably prove local treatment failure because insufficient time has passed for histological tumor resolution.4 In the scenario of low risk disease and a high quality implant observation with continued PSA monitoring is recommended. Juanita Crook Department of Radiation Oncology University of Toronto/Princess Margaret Hospital Toronto, Ontario Canada 1.
Crook J, Malone S, Perry G, Bahadur Y, Robertson S and Abdolell M: Postradiotherapy prostate biopsies: what do they really mean? Results for 498 patients. Int J Radiat Oncol Biol Phys 2000; 48: 355. 2. Cavanagh W, Blasko JC, Grimm PD and Sylvester JE: Transient elevation of serum prostate-specific antigen following (125)I/(103)Pd brachytherapy for localized prostate cancer. Semin Urol Oncol 2000; 18: 160. 3. Merrick GS, Butler WM, Wallner KE, Galbreath RW and Anderson RL: Prostate-specific antigen spikes after permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2002; 54: 450. 4. Smathers S, Wallner K, Sprouse J and True L: Temporary PSA rises and repeat prostate biopsies after brachytherapy. Int J Radiat Oncol Biol Phys 2001; 50: 1207.
These authors describe a 7.7% rate of prostate cancer PBs in 508 men who underwent permanent seed brachytherapy combined with EBRT 2 to 11 years after implantation. They conclude that the adequate radiation dose to eradicate local disease is substantial. Even extraprostatic biopsies of the SVs should be performed to rule out involvement. The 7.7% incidence of prostate cancer PBs is only the minimum of therapy failures after seed implantation because prostate biopsy represents only a small part of the prostate and many cancers might have been missed by biopsy, that is up to 17%.1 Patients with T3b, Gleason score greater than 7 and PSA greater than 30 ng/ml underwent laparoscopic lymphadenectomy. This may have been insufficient due to the low number of lymph nodes retrieved by this procedure since patients with node positive disease were excluded from study. The question of the PSA nadir of 0.2 to 0.5 ng/ml after seed implantation remains.2 Furthermore, the question also remains how good seed therapy is for prostate cancer treat-
1764
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
ment at long-term followup. Relying on prognosis alone is not enough. Long-term results with 15 and 20 years of followup might provide an answer. Prospective, interdisciplinary, multicenter studies with longer followup are required. Stefan Hautmann Peter Martin Braun Klaus-Peter Jünemann Department of Urology and Pediatric Urology University of Kiel Kiel, Germany 1.
2.
Stock RG, Ho A, Cesaretti JA and Stone NN: Changing the patterns of failure for high-risk prostate cancer patients by optimizing local control. Int J Radiat Oncol Biol Phys 2006; 66: 389. Critz FA and Levinson K.: 10-Year disease-free survival rates after simultaneous irradiation for prostate cancer with a focus on calculation methodology. J Urol 2004; 172: 2232.
REPLY BY AUTHORS We agree with the need to be cautious when evaluating 2-year post-implant PBs. Radiation oncologists should routinely communicate dosimetry results to urology colleagues, who can then incorporate this information in patient management. We believe that our biopsy technique (6 to 30 cores) adequately assesses the likelihood of local failure. A well done laparoscopic pelvic lymph node dissection is just as accurate as an open one when performed by an experienced surgeon (reference 5 in article).1 We agree that prospective, randomized trials are the best means to assess competitive treatment strategies but until such results become available, large single or multi-institution experiences remain the most persuasive data from which to make management decisions. 1.
Stone NN, Stock RG and Unger P: Laparoscopic pelvic lymph node dissection for prostate cancer: comparison of the extended and modified techniques. J Urol 1997; 158: 1891.
Purpose: We describe biopsy results in patients with prostate cancer treated with brachytherapy. Materials and Methods: A total of 1,562 men with localized prostate cancer were treated with permanent prostate brachytherapy, of whom 508 agreed to ultrasound guided biopsies 2 years after the completion of all therapy. Median followup was 6.7 years (range 2 to 14.6) and median prostate specific antigen was 7.4 ng/ml (range 0.3 to 300). Disease was categorized as Gleason score less than 7 in 74.8% of patients, stage T2a or less in 64.2%, low risk in 43.1%, intermediate risk in 24.2% and high risk in 32.7%. Of the 508 men 315 (62%) received 125I, 110 (21.7%) received 103Pd and 83 (16.3%) received 103Pd and external beam radiotherapy. A total of 237 men (46.7%) received a short course of hormonal therapy (3 to 9 months). Subsequent biopsies were performed after 2 years if initial biopsy was positive or prostate specific antigen increased. Post-implantation dosimetry results were grouped into low, normal and high dose. Associations were tested by chi-square analysis. Survival functions were calculated with Kaplan-Meier analysis and Cox regression. Results: A total of 643 biopsies were performed in 508 men between 2 and 11 years after implantation. Of the 508 men 39 (7.7%) had a final positive biopsy. Positive biopsy was associated with high prostate specific antigen (p ⫽ 0.035), stage (p ⫽ 0.003), risk (p ⫽ 0.024), no hormonal therapy (p ⫽ 0.002) and low dose (p ⬍0.0001). On multivariate analysis only dose and hormonal therapy were significant (p ⬍0.0001 and p ⫽ 0.004, respectively). Of the patients 80% were free of PSA failure at 10 years if final biopsy was negative compared to 27.3% with a positive biopsy (p ⬍0.0001). Death from prostate cancer was associated with a positive biopsy (OR 18.5, 95% CI 2.3–143, p ⬍0.0001). Of the 52 men with a positive biopsy at year 2, 23 (44.2%) had negative results on subsequent biopsy, while 10 of the 456 (2.2%) with negative 2-year biopsies showed positive results. Positive biopsy occurred in the prostate only in 31 of 39 men (79.5%), in the prostate and seminal vesicles in 3 (7.7%), and in the seminal vesicles only in 5 (12.8%). Conclusions: Patients undergoing prostate brachytherapy must receive an adequate radiation dose to eradicate local disease. Hormonal therapy may benefit local control in patients with intermediate to high risk disease. Extraprostatic biopsies should be performed in patients with local failure who are considering salvage therapy to rule out seminal vesicle involvement. Key Words: prostate; prostatic neoplasms; brachytherapy; biopsy; neoplasm recurrence, local
ermanent seed brachytherapy has become a popular treatment for localized prostate cancer. Brachytherapy can be performed as monotherapy or in combination with HT and/or EBRT.1,2 Different treatments are used for different stages of disease with monotherapy most often reserved for low risk cancer and combined modality therapy used for high risk disease. Treatment outcome is usually measured by biochemical disease control.3 The problem with using PSA as an end point for treatment failure is that it does not separate men with systemic vs local failure. In the past digital rectal examination was used to assess local control. Today we know that posttreatment prostate biopsy is the best method to detect residual or recurrent local disease. There are few reports in the literature that address the long-term results of prostate biopsy following brachytherapy. None evaluated the consequences of sequential biopsy
P
Submitted for publication August 22, 2006. Presented at annual meeting of American Urological Association, Atlanta, Georgia, May 25–26, 2006. * Correspondence: 21 Timber Trail, Suffern, New York 10901 (e-mail: [email protected]).
For another article on a related topic see page 1913.
0022-5347/07/1775-1759/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION
results or the incidence of extraprostatic relapse in these patients. We studied a large cohort of patients with brachytherapy who underwent prostate biopsy a minimum of 2 years after implantation and subsequent biopsy if the initial one was positive or the patient experienced PSA relapse. Biopsy results were correlated with initial patient PSA, Gleason score, HT and implant dose. MATERIALS AND METHODS A total of 1,562 men with biopsy proven prostate cancer were treated with permanent seed implantation at 1 institution. All pathological findings were reviewed by 1 pathologist (PU). Patients were staged using the 1992 American Joint Committee on Cancer system and those with initial PSA more than 10 ng/ml or Gleason score greater than 6 underwent bone and computerized tomography. In addition, most patients with PSA more than 10 ng/ml, Gleason score 7 or greater, or stage T2b or greater underwent SVBs.4 Patients with a positive SVB, Gleason 7 or greater with perineural invasion or PSA more than 30 ng/ml also underwent laparoscopic pelvic lymph node dissection.5 Those with positive pelvic lymph nodes were excluded.
1759
Vol. 177, 1759-1764, May 2007 Printed in U.S.A. DOI:10.1016/j.juro.2007.01.069
1760
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
Patients were offered a treatment protocol based on staging results. Low risk (PSA less than 10 ng/ml, Gleason score 6 or less, stage T2a or less and prostate volume less than 50 cc) were treated with an 125I implant alone to a prescription dose of 160 Gy (Task Group 43). Patients at low risk with a prostate of 50 cc or greater were also placed on 3 months of neoadjuvant HT before implantation. About 50% of these patients also continued on HT for an additional 2 months. Those at intermediate risk (PSA 10 to 20 ng/ml, Gleason 7 or stage T2b and negative SVB) were treated with 6 months of HT, including 3 before implantation, and 103Pd to a dose of 124 Gy (National Institute of Standards and Technology 1999). Patients at high risk (PSA more than 20 ng/ml, Gleason 8 –10, stage T2c–T3 or positive SVB with negative laparoscopic peritoneal lymph node dissection) were treated with 9 months of HT, including 3 before implantation, and 103Pd to a dose of 100 Gy (National Institute of Standards and Technology 1999), followed 2 months later by 45 Gy EBRT in 25 fractions.2 Patients with a positive SVB received seed placement in the SVs with the intent of providing a radiation boost to the proximal 50% of the SVs2. Median age of the 508 patients with biopsy was 66 years (range 41 to 88) and median PSA was 7.4 ng/ml (range 0.3 to 300). Patients were categorized with Gleason score less than 7 in 74.8%, stage T2a or less in 64.2%, low risk in 43.1%, intermediate risk in 24.2% and high risk in 32.7%. Median implant prostate volume was 36.5 cc (range 2.4 to 114). Of the 508 men 315 (62%) received 125I, 110 (21.7%) received 103 Pd and 83 (16.3%) received 103Pd with EBRT (table 1). A total of 237 patients (46.7%) received a short course of HT (3 to 9 months). One month after implantation computerized tomography based dosimetry was performed using ADAC Pinnacle or Varian software. These data were separated into 3 groups based on post-implantation dosimetry D90 results. Patients receiving a dose of less than 116 Gy 125I or less than 100 Gy 103Pd were considered low dose. A delivered dose of more than 116 to 160 Gy 125I and more than 100 to 124 Gy was considered normal dose. Anything above these doses or when combination therapy was used were considered high dose. The rationale for using these dose cutoff points was based on reports in the literature suggesting that an adequate implant should deliver at least 80% of the typical prescription dose (145 Gy for 125I and 124 Gy for 103Pd).6
TABLE 1. Patient characteristics No. (%) PSA (ng/ml): 4 or Less 4.1–10 10.1–20 Greater than 20 Stage: T1b–T2a T2b–c T3 Gleason score: 2–6 7 8–10 HT: Yes No Risk: Low Intermediate High
50 (9.8) 302 (59.5) 112 (22.1) 44 (8.6) 317 (62.4) 180 (35.5) 11 (2.1) 380 (74.8) 78 (15.4) 50 (9.8) 272 (53.3) 236 (46.7) 219 (43.1) 123 (24.2) 166 (32.7)
Prostate biopsy was offered 2 years after completion of all therapy. It was performed under ultrasound guidance with a minimum of 6 cores taken. All biopsies were read by 1 pathologist (PU) experienced with evaluating post-irradiation specimens. Biopsies were scored as negative or positive only. Any tissue with remaining cancer, even with radiation effect, was considered positive. Patients agreeing to biopsy underwent repeat biopsy yearly if 2-year biopsy was positive until biopsy became negative or there was clear evidence of PSA progression (minimum of 3 increases above a nadir). In addition, patients with evidence of PSA progression in whom initial posttreatment biopsies were negative were offered repeat biopsies. They had a minimum of 12 core samples taken (range 12 to 30) with at least 6 from the prostate and 6 from the SVs. Patients with 2 sets of negative biopsies (12 cores) and increasing PSA underwent saturation biopsies with 30 core samples taken. This procedure was done by a mapping technique using anesthesia. Associations were tested by chi-square analysis and ANOVA. Survival functions were computed with KaplanMeier analysis and Cox regression. Results of the last biopsy were used for association analyses with p ⬍0.05 considered significant. RESULTS Median followup was 6.7 years (range 2 to 14.6). A total of 643 biopsies were performed in 508 men, including 1 to 5 biopsies in 416, 61, 22, 6 and 3, respectively. Of the 508 men 39 (7.7%) had PB and 99 (19.5%) experienced PSA failure. PB was associated with PSA greater than 10 (p ⫽ 0.035), stage greater than T2a (p ⫽ 0.003), risk greater than low (p ⫽ 0.024), no HT (p ⫽ 0.002) and low radiation dose (p ⬍0.0001, table 2). Gleason score was not associated with a PB. On multivariate analysis only dose and HT were significant (p ⬍0.0001 and p ⫽ 0.004, respectively, table 3). Patients were grouped into risk groups and variables associated with a PB were determined (table 4). Dose was the only significant variable in patients at low risk (p ⬍0.001), while HT (p ⫽ 0.008) and dose (p ⫽ 0.039) were significant in those at intermediate risk, and HT (p ⬍0.001), dose (p ⫽ 0.035) and combination therapy (p ⫽ 0.026) were significant in those at high risk. In patients at intermediate risk treated with a high dose HT decreased the positive biopsy rate to 2% (1 of 50) from 10.8% (4 of 37) in those not treated with HT (p ⫽ 0.08). In patients at high risk treated with a high dose HT decreased the positive biopsy rate to 6.5% (8 of 124) from 28.6% (12 of 42) (OR 4.4, 95% CI 1.9 –10.1, p ⬍0.0001). When these patients were further analyzed by implant type (103Pd vs partial 103Pd plus EBRT), those treated with combination therapy with HT had a decrease in PBs to 3.3% (2 of 60) compared to 11.8% (6 of 51) in those receiving 103Pd plus HT (p ⫽ 0.087). The 10-year biological failure-free rate in the entire cohort was 84%. Of the 508 men with biopsies 80% were free of PSA failure at 10 years if final biopsy was negative compared to 27.3% for a PB (p ⬍0.0001, see figure). Of the patients 102 (6.5%) died during the study period, including 14 (0.9%) who died of prostate cancer. Biopsy data were available on 43 deceased patients (41.2%). Of these 43 patients 1 of 37 (2.5%) had a PB and died of other causes, while 3 of 6 (50%) had a PB and died of prostate cancer (OR 18.5,
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY TABLE 2. Univariate analysis of positive prostate biopsy Variable PSA (ng/ml): Less than 10 10.1–20 Greater than 20 Stage: T1–T2a T2b T2c–T3 Risk: Low Intermediate High Prostate vol (cc): 30 or Less 30.1–50 Greater than 50 HT: Yes No Implant: ⫹ EBRT Alone Dose: Low Normal High
No. Pos Biopsy/ Total No. (%) 21/347 (6.1) 10/112 (8.9) 8/49 (16.3) 15/347 (4.6) 19/135 (7.3) 5/56 (8.9)
p Value
0.035
0.003
10/219 (4.6) 9/123 (7.3) 20/166 (12)
0.024
21/196 (10.7) 12/225 (5.3) 4/78 (5.1)
0.07
9/237 (3.8) 30/271 (11.1)
0.002
2/83 (2.4) 37/425 (8.7)
0.03
15/51 (29.4) 4/66 (6.1) 16/367 (4.4)
⬍0.001
95% CI 2.3–143, p ⬍0.0001). A Cox regression proportional hazard model was performed to assess competing causes of death in the subset of 508 patients evaluated by biopsy (table 5). Higher 1-year post-implantation PSA predicted a subsequent final PB vs a negative biopsy (mean 2.1 vs 0.99 ng/ml, p ⬍0.0001). However, lower 1-year PSA was also associated with HT, that is 0.5 ng/ml with HT vs 1.57 ng/ml without HT (p ⬍0.0001). Linear regression analysis with HT plus 1-year PSA demonstrated that each was significant (p ⫽ 0.048 and 0.001, respectively). Of the 52 men with a PB at year 2, 23 (44.2%) had negative results on subsequent biopsy. The likelihood of positive or negative results in these patients at subsequent year biopsies was the same for all 5 years. In contrast, 10 of the 456 patients (2.2%) with negative 2-year biopsies showed positive results on subsequent biopsy, which was performed because of increasing PSA. Mean time to local failure was 4.5 years (range 2 to 11). Mean time to PSA failure in patients with a negative last biopsy was 3.3 compared to 5.2 years in those with final PB (p ⫽ 0.036). A PB occurred in the prostate only in 31 of 39 men (79.5%), in the prostate and SVs in 3 (7.7%), and in the SVs only in 5 (12.8%).
1761
TABLE 4. Positive biopsy by risk group Variable Low risk: Low dose Normal dose High dose Intermediate risk: No HT HT Low dose Normal dose High dose High risk: No HT HT Low dose Normal dose High dose 125 I 103 Pd 103 Pd/EBRT
No. Pos Biopsy/ Total No. (%)
p Value
2/4 (50) 1/20 (5) 7/192 (3.6)
⬍0.001
8/57 (14) 1/66 (1.5) 3/11 (27.3) 1/18 (5.6) 5/87 (5.7) 12/42 (28.6) 8/125 (6.5) 10/36 (27.8) 2/28 (7.1) 4/88 (4.5) 6/37 (16.2) 12/67 (17.9) 2/62 (3.2)
0.008 0.039 ⬍0.001 0.035 0.026
option for prostate cancer. Data in this study suggest that it can but only if certain conditions are met. As determined by biopsy in this study, the overall local failure rate was 7.7%. The primary requirement for disease eradication was the delivery of higher doses (table 4). Patients receiving the highest doses had a local failure rate of 3.6% for low risk (p ⬍0.001), 5.7% for intermediate risk (p ⫽ 0.039) and
DISCUSSION Brachytherapy should be able to eradicate local disease in most patients if it is to be considered a reasonable treatment TABLE 3. Multivariate analysis of positive prostate biopsy after brachytherapy Variable
p Value
Risk group PSA Stage HT Additional EBRT Dose
0.804 0.550 0.653 0.004 0.650 ⬍0.001
Biochemical (PSA) freedom from failure by biopsy results. Cum, cumulative.
1762
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
TABLE 5. Cox regression proportional hazard model of competing causes of death in 508 patients evaluated by biopsy Variable
Wald
p Value
Stage Gleason PSA HT Risk group Dose Biopsy result Supplemental EBRT
4.1 20.5 1.5 18.9 21.4 23.1 1.0 8.5
0.043 ⬍0.001 0.221 ⬍0.001 ⬍0.001 ⬍0.001 0.307 0.004
4.5% for high risk (p ⫽ 0.035) disease. For individual isotopes this required a D90 of at least 160 Gy for 125I and 124 Gy for 103Pd or combination therapy. In addition, HT appeared to further improve biopsy results in patients at intermediate and high risk who received the higher doses. While the difference was not quite significant (p ⫽ 0.087), combination therapy in patients at high risk treated with HT appeared to further decrease PB rates from 11.8% to 3.3%. The need to deliver higher doses of radiation was shown by a number of investigators to improve biochemical relapse-free survival in patients treated with prostate brachytherapy.7,8 In a small cohort of patients treated with 125I Stone et al also previously observed that those treated to a dose of at least 140 Gy had a 2-year postimplantation PB rate of 4.8 % vs 20.5% in those not achieving that dose (p ⫽ 0.002).1 The current study with longer followup in a larger cohort of patients further substantiates these results. Regardless of patient risk category higher doses are required to eradicate disease. It is interesting that Gleason score was not a significant variable for predicting local failure. Patients with high Gleason scores have a substantially greater rate of biochemical relapse than those with lower scores. While high grade lesions may be more likely to metastasize, they are not more difficult to eradicate locally if high enough radiation doses are delivered. Zelefsky et al studied the impact of higher radiation doses in patients receiving conformal EBRT.9 Of 480 patients who underwent post-EBRT biopsies 121 (25.2%) had positive findings. The positive biopsy rate decreased from 90 of 320 patients (28.1%) for doses 70 Gy or less to 27 of 119 (22.7%) for 75.6 Gy and to 4 of 41 (9.8%) for 81 Gy doses. Recently Stock et al described the effect of BED on biopsy outocmes.10 D90s were converted into a BED for each isotope, which was summed with the BED of EBRT for patients with combination therapy. This allowed the different treatment regimens to be assigned a BED value and the effect of different BED levels on biopsy outcome to be determined. Multivariate analysis revealed that only HT and BED significantly affected biopsy results (p ⫽ 0.05 and 0.006, respectively). The 10-year biological failure-free rate in the entire cohort was 84%. Of patients undergoing prostate biopsies 80% were free of failure if biopsy was negative vs 27.3% if the final biopsy was positive (p ⬍0.0001). The significance of a positive post-irradiation prostate was controversial in the past. The current study with long-term followup and repeat biopsies confirms that a PB leads to disease progressions. In fact, a positive biopsy was 18.5 times more likely to result in prostate cancer death than a negative one (p ⬍0.0001). How-
ever, in the Cox proportional hazards model the last biopsy result was not significant (table 5). This result most likely reflects the low number of deaths (43) in the 508 patients at this time. Longer followup is needed to adequately address this issue. In this study routine prostate biopsies were offered 2 years after therapy regardless of disease status and 10% of the patients (52 of 508) had a positive biopsy. Of these cases 44% eventually reverted to negative results. This is in contrast to the study reported by Prestidge et al, in which 22% of posttreatment biopsies were positive or indeterminate after 1 year.11 Most indeterminate pathological findings in this report demonstrated prostate cancer with a radiation treatment effect. Almost all positive 2-year biopsies in our patients demonstrated the same results. While Prestidge et al believed that most of these results would clear with time, although this was not actually shown in their study, our data suggest that this happens less than half of the time. These data can present a conundrum to the urologist who elects to perform biopsy in patients 2 years after brachytherapy. While this is not routinely performed in clinical practice, the decision to perform biopsy after brachytherapy may be made if the patient experiences an increase in PSA after attaining a nadir. Unfortunately up to 31% of patients can experience a temporary increase in PSA after brachytherapy that is not associated with local or systemic progression.12 Our suggestion would be to observe these patients and not perform biopsy if the delivered D90 dose was high. If the patient experiences several consecutive PSA increases, biopsy may become necessary. The PSA nadir at 1 year appears to be somewhat predictive of local control. Patients with 1-year PSA less than 1 ng/ml were far less likely to experience a PB than patients with PSA greater than 2.0 ng/ml (p ⬍0.0001). However, these results are also confounded by HT use, which also contributed to a lower 1-year PSA and to the final biopsy results on multivariate analysis. Late increasing PSA may also suggest local failure. In fact, local failure was identified in 1 of our patients 11 years after implantation. Of patients experiencing biochemical failure those with a negative biopsy were more likely to experience relapse earlier than patients with a PB (3.3 vs. 5.2 years, p ⫽ 0.036). This relationship between the failure pattern and time to PSA failure is similar to that seen with radical prostatectomy. The type of local failure was also noteworthy. Of the cases 20.5% failed in the prostate and SVs or the SVs alone. If a salvage procedure is considered in a patient experiencing local failure after brachytherapy, biopsy of the SVs is necessary to rule out extraprostatic disease.
CONCLUSIONS Patients undergoing prostate brachytherapy must receive an adequate radiation dose to eradicate local disease. Local failure can occur as long as 11 years after implantation. HT may benefit local control in patients with intermediate to high risk disease. Extraprostatic biopsies should be performed in patients with local failure who are considering salvage therapy to rule out SV involvement.
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
Abbreviations and Acronyms BED D90 EBRT HT PB PSA SV SVB
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
biological equivalent dose dose to 90% of the area external beam radiotherapy hormonal therapy last positive biopsy prostate specific antigen seminal vesicle SV biopsy
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Stone NN, Stock RG and Unger P: Intermediate-term biochemical and local control following I-125 brachytherapy for prostate cancer. J Urol 2005; 173: 803. Stock RG, Cahlon O, Cesaretti J, Kolhmeier MA and Stone NN: Combined modality treatment in the management of high risk prostate cancer. Int J Rad Oncol Biol Phys 2004; 59: 1352. Cox JD, Grignon DJ and Kaplan RS: Consensus statement: guidelines for PSA following radiation therapy. Int J Radiat Oncol Biol Phys 1997; 37: 1035. Linzer DG, Stock RG, Stone NN, Ianuzzi C, Ratnow R and Unger P: Seminal vesicle biopsy: accuracy and implications for staging of prostate cancer. Urology 1996; 48: 757. Stone NN, Stock RG, Parikh D, Yeghiayan P and Unger P: Perineural invasion and seminal vesicle involvement predict pelvic lymph node metastasis in men with localized carcinoma of the prostate. J Urol 1998; 160: 1722. Willins J and Wallner K: CT-based dosimetry for transperineal I-125 prostate brachytherapy. Int J Rad Oncol Biol Phys 1997; 39: 347. Stock RG, Cesaretti J and Stone NN: Disease specific survival following brachytherapy management of prostate cancer. Int J Rad Oncol Biol Phys 2006; 64: 810. Potters L, Cao Y, Calugaru E, Torre T, Fearn and Wang XH: A comprehensive review of CT-based dosimetry parameters and biochemical control in patients treated with permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2001; 50: 605. Zelefsky MJ, Fuks Z, Hunt M, Lombardi D, Ling CC, Rueter VE et al: High dose radiation delivered by intensity modulated conformal radiotherapy improves the outcome of localized prostate cancer. J Urol 2001; 166: 876. Stock RG, Stone NN, Cesaretti J and Rosenstein BS: Biologically effective dose values for prostate brachytherapy: effects on PSA failure and posttreatment biopsies. Int J Rad Oncol Biol Phys 2006; 64: 527. Prestidge BR, Hoak DC, Grimm PD, Ragde H, Cavanagh W and Blasko JC: Post-treatment biopsy results following interstitial brachytherapy in early-stage prostate cancer. Int J Radiat Oncol Biol Phys 1997; 37: 31. Stock RG, Stone NN and Cessaretti J: Prostate specific antigen bounce following prostate seed implantation for localized prostate cancer: descriptions and implications. Int J Rad Oncol Biol Phys 2003; 56: 448.
EDITORIAL COMMENTS These authors report extensive experience with postbrachytherapy prostate biopsies offered 2 years after therapy. Of the cohort of 1,562 men 508 accepted biopsy, of whom 61 underwent a second biopsy and 31 underwent 3 or more biopsies. The population consisted of 43% of men with low risk, 24% with intermediate risk and 33% with high risk disease. HT (3 to 9 months) was done in 37% of the patients
1763
and 16% received supplemental EBRT. Of the 52 men who had PB after 2 years 44% subsequently converted to negative results and only 7.7% had positive results at final biopsy. Only dose was predictive of PB in the low risk group. For those at intermediate risk dose and HT were predictive and in those at high risk dose, HT and the addition of EBRT were predictive. These results emphasize the importance of quality in brachytherapy outcome. Although the authors did not specifically address the question of the timing of post-brachytherapy biopsies,1 44% of biopsies that were positive at 2 years subsequently reverted to negative. This creates a clinical dilemma for urologists faced with increasing PSA early after brachytherapy. Of men 30% to 40% experience a transient increase in PSA within the first 2 to 2½ years.2,3 Biopsy at this time cannot reliably prove local treatment failure because insufficient time has passed for histological tumor resolution.4 In the scenario of low risk disease and a high quality implant observation with continued PSA monitoring is recommended. Juanita Crook Department of Radiation Oncology University of Toronto/Princess Margaret Hospital Toronto, Ontario Canada 1.
Crook J, Malone S, Perry G, Bahadur Y, Robertson S and Abdolell M: Postradiotherapy prostate biopsies: what do they really mean? Results for 498 patients. Int J Radiat Oncol Biol Phys 2000; 48: 355. 2. Cavanagh W, Blasko JC, Grimm PD and Sylvester JE: Transient elevation of serum prostate-specific antigen following (125)I/(103)Pd brachytherapy for localized prostate cancer. Semin Urol Oncol 2000; 18: 160. 3. Merrick GS, Butler WM, Wallner KE, Galbreath RW and Anderson RL: Prostate-specific antigen spikes after permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2002; 54: 450. 4. Smathers S, Wallner K, Sprouse J and True L: Temporary PSA rises and repeat prostate biopsies after brachytherapy. Int J Radiat Oncol Biol Phys 2001; 50: 1207.
These authors describe a 7.7% rate of prostate cancer PBs in 508 men who underwent permanent seed brachytherapy combined with EBRT 2 to 11 years after implantation. They conclude that the adequate radiation dose to eradicate local disease is substantial. Even extraprostatic biopsies of the SVs should be performed to rule out involvement. The 7.7% incidence of prostate cancer PBs is only the minimum of therapy failures after seed implantation because prostate biopsy represents only a small part of the prostate and many cancers might have been missed by biopsy, that is up to 17%.1 Patients with T3b, Gleason score greater than 7 and PSA greater than 30 ng/ml underwent laparoscopic lymphadenectomy. This may have been insufficient due to the low number of lymph nodes retrieved by this procedure since patients with node positive disease were excluded from study. The question of the PSA nadir of 0.2 to 0.5 ng/ml after seed implantation remains.2 Furthermore, the question also remains how good seed therapy is for prostate cancer treat-
1764
LOCAL FAILURE PATTERNS AFTER PROSTATE BRACHYTHERAPY
ment at long-term followup. Relying on prognosis alone is not enough. Long-term results with 15 and 20 years of followup might provide an answer. Prospective, interdisciplinary, multicenter studies with longer followup are required. Stefan Hautmann Peter Martin Braun Klaus-Peter Jünemann Department of Urology and Pediatric Urology University of Kiel Kiel, Germany 1.
2.
Stock RG, Ho A, Cesaretti JA and Stone NN: Changing the patterns of failure for high-risk prostate cancer patients by optimizing local control. Int J Radiat Oncol Biol Phys 2006; 66: 389. Critz FA and Levinson K.: 10-Year disease-free survival rates after simultaneous irradiation for prostate cancer with a focus on calculation methodology. J Urol 2004; 172: 2232.
REPLY BY AUTHORS We agree with the need to be cautious when evaluating 2-year post-implant PBs. Radiation oncologists should routinely communicate dosimetry results to urology colleagues, who can then incorporate this information in patient management. We believe that our biopsy technique (6 to 30 cores) adequately assesses the likelihood of local failure. A well done laparoscopic pelvic lymph node dissection is just as accurate as an open one when performed by an experienced surgeon (reference 5 in article).1 We agree that prospective, randomized trials are the best means to assess competitive treatment strategies but until such results become available, large single or multi-institution experiences remain the most persuasive data from which to make management decisions. 1.
Stone NN, Stock RG and Unger P: Laparoscopic pelvic lymph node dissection for prostate cancer: comparison of the extended and modified techniques. J Urol 1997; 158: 1891.