Radiation Dose Escalation for Localized Prostate Cancer

Transcription

1 Radiation Dose Escalation for Localized Prostate Cancer Intensity-Modulated Radiotherapy Versus Permanent Transperineal Brachytherapy William W. Wong, MD 1 ; Sujay A. Vora, MD 1 ; Steven E. Schild, MD 1 ; Gary A. Ezzell, PhD 1 ; Paul E. Andrews, MD 2 ; Robert G. Ferrigni, MD 2 ; and Scott K. Swanson, MD 2 BACKGROUND: In the current study, the effects of dose escalation for localized prostate cancer treatment with intensity-modulated radiotherapy (IMRT) or permanent transperineal brachytherapy (BRT) in comparison with conventional dose 3-dimensional conformal radiotherapy (3D-CRT) were evaluated. METHODS: This study included 853 patients; 270 received conventional dose 3D-CRT, 314 received high-dose IMRT, 225 received BRT, and 44 received external beam radiotherapy (EBRT) þ BRT boost. The median radiation doses were 68.4 grays (Gy) for 3D-CRT and 75.6 Gy for IMRT. BRT patients received a prescribed dose of 144 Gy with iodine-125 (I-125) or 120 Gy with palladium-103 (Pd-103), respectively. Patients treated with EBRT þ BRT received 45 Gy of EBRT plus a boost of 110 Gy with I-125 or 90 Gy with Pd-103. Risk group categories were low risk (T1-T2 disease, prostate-specific antigen level 10 ng/ml, and a Gleason score 6), intermediate risk (increase in value of 1 of the factors), and high risk (increase in value of 2 factors). RESULTS: With a median follow-up of 58 months, the 5-year biochemical control (bned) rates were 74% for 3D-CRT, 87% for IMRT, 94% for BRT, and 94% for EBRT þ BRT (P <.0001). For the intermediate-risk group, high-dose IMRT, BRT, or EBRT þ BRT achieved significantly better bned rates than 3D-CRT (P <.0001), whereas no improvement was noted for the low-risk group (P ¼.22). There was no increase in gastrointestinal (GI) toxicity from high-dose IMRT compared with conventional dose 3D-CRT, although there was more grade 2 genitourinary (GU) toxicity (toxicities were graded at the time of each follow-up visit using a modified Radiation Therapy Oncology Group [RTOG] scale). BRT caused more GU but less GI toxicity, whereas EBRT þ BRT caused more late GU and GI toxicity than IMRT or 3D-CRT. CONCLUSIONS: The data from the current study indicate that radiation dose escalation improved the bned rate for the intermediate-risk group. IMRT caused less acute and late GU toxicity than BRT or EBRT þ BRT. Cancer 2009; 115: VC 2009 American Cancer Society. KEY WORDS: prostate cancer, radiation dose escalation, 3-dimensional conformal therapy, intensitymodulated radiotherapy, brachytherapy. Prostate cancer is the most common malignancy in men. 1 The use of prostate-specific antigen (PSA) as a screening test for the disease leads to the diagnosis of most new cases while it is localized and amenable to therapy with curative intent. Multiple treatment options are available to patients with localized prostate Corresponding author: William W. Wong, MD, Department of Radiation Oncology, Mayo Clinic Arizona, E. Shea Boulevard, Scottsdale, AZ 85259; Fax: (480) ; wong.william@mayo.edu 1 Department of Radiation Oncology, Mayo Clinic Arizona, Scottsdale, Arizona; 2 Department of Urology, Mayo Clinic Arizona, Scottsdale, Arizona Received: October 27, 2008; Revised: March 5, 2009; Accepted: March 10, 2009 Published online August 10, 2009 in Wiley InterScience ( DOI: /cncr.24558, Cancer December 1, 2009

2 Dose Escalation for Prostate Cancer/Wong et al cancer, including radical prostatectomy, external beam radiotherapy (EBRT), brachytherapy (BRT), and combined EBRT and BRT boost. To the best of our knowledge, there are currently no controlled randomized studies comparing these treatment modalities. Previous reports comparing EBRT with BRT generally included patients who were treated with conventional EBRT doses that were lower than those commonly delivered today. Significant progress has been made in the treatment planning and administration of EBRT as well as BRT. With the advances in technology such as intensity-modulated radiotherapy (IMRT) and image guidance, higher EBRT doses have been administered in recent years, with much better localization of the prostate gland. The biologically equivalent dose (BED) can be calculated for permanent BRT and fractionated EBRT. 2-4 It has been shown that the BED for permanent BRT is higher than that of conventional dose 3-dimensional conformal radiotherapy (3D-CRT). Intuitively, it would be reasonable to assume that dose escalation from 3D-CRT, either by BRT or by IMRT, should achieve better biochemical control (bned) rates. However, the toxicity of each treatment modality must be taken into consideration when making treatment decisions. To our knowledge, only a few reports have been published to date comparing the treatment results and toxicity of dose escalation using high-dose IMRT or BRT with conventional dose 3D-CRT as the baseline. This retrospective study evaluated the effects of dose escalation from conventional dose 3D-CRT (71 grays [Gy]) in the treatment of patients with localized prostate cancer by using high-dose IMRT (75.6 Gy), permanent transperineal BRT or EBRT þ BRT. The bned rates, side effects, as well as potential prognostic factors associated with disease control, were analyzed. MATERIALS AND METHODS Patient Population This study included 853 consecutive patients who were treated with radiotherapy for localized prostate cancer (T1c-T3N0M0 disease) between May 1993 and July 2004 at Mayo Clinic Arizona. This project was approved by the Institutional Review Board at the Mayo Clinic. Informed consent was obtained from all patients before any therapy was administered. Patient characteristics are listed in Table 1. The median pretreatment PSA level was 7 ng/ml (range, ng/ml). All prostate biopsy slides were reviewed by pathologists at the Mayo Clinic. Patients were categorized into 3 risk groups: low risk (stage T1-T2 disease, PSA level 10 ng/ml, and a Gleason score 6), intermediate risk (increase in value of 1 of the risk factors), and high risk (increase in value of 2 or more of the risk factors). 5 Treatment Between 1993 and 2000, 270 patients were treated with EBRT using 3D-CRT. The techniques generally included a 4-field box technique, delivering 45 Gy to the prostate and seminal vesicles, while the prostate was boosted to a median dose of 68.4 Gy (range, Gy). Treatment was administered in daily fractions of 1.8 to 2 Gy. The dose was prescribed to encompass the clinical target volume. Pelvic lymph nodes were not treated. Customized blocking with a 1.0-cm to 2.0-cm margin from the planning target volume to block edge was used for sparing the bladder and rectum. Since November of 2000, high-dose IMRT has been used for the delivery of EBRT in our department. Three hundred fourteen patients were treated with IMRT and were included in this analysis. The treatment volume included the prostate and seminal vesicles, with a 6-mm to 10-mm margin. The median dose to the prostate gland was 75.6 Gy (range, Gy), whereas the seminal vesicles received 50.4 Gy. Daily transabdominal ultrasonography was performed to localize the prostate gland at the time of treatment, thus minimizing the risk of a geographic miss. Adjuvant androgen deprivation therapy (ADT) was administered to 161 patients (28%) who received 3D- CRT or IMRT. The median duration of ADT was 9 months (range, 1-72 months), depending on the risk category of the disease. Patients with high-grade cancer or a PSA level >20 ng/ml usually received a longer course of ADT that ranged from 2 to 3 years. Transperineal BRT was performed in 225 patients using iodine-125 (I-125) or palladium-103 (Pd-103) seeds. The prescribed minimal peripheral dose was 144 Gy for I-125 and 120 Gy for Pd-103, respectively. Shortterm ADT (2-14 months) was used in 72 patients to Cancer December 1,

3 Table 1. Patient Characteristics Characteristics 3D-CRT IMRT BRT EBRT1BRT Total no T classification T1c 42 (16%) 109 (35%) 114 (51%) 13 (30%) T2a 78 (29%) 122 (39%) 83 (37%) 10 (23%) T2b 59 (21%) 36 (11%) 24 (11%) 16 (36%) T2c 64 (24%) 33 (11%) 4 (2%) 4 (9%) T3 27 (10%) 14 (4%) 0 1 (2%) PSA, ng/ml (71%) 238 (76%) 193 (86%) 29 (65%) (19%) 54 (17%) 28 (12%) 13 (30%) (10%) 22 (7%) 4 (2%) 2 (5%) Gleason score (65%) 138 (44%) 173 (77%) 20 (45%) 7 95 (35%) 176 (56%) 52 (23%) 24 (55%) Perineural invasion No 246 (91%) 263 (84%) 216 (96%) 41 (93%) Yes 24 (9%) 51 (16%) 9 (4%) 3 (7%) Adjuvant hormonal treatment No 223 (83%) 200 (64%) 153 (68%) 32 (73%) Yes 47 (17%) 114 (36%) 72 (32%) 12 (27%) Risk group Low 119 (44%) 109 (35%) 158 (70%) 14 (32%) Intermediate 111 (41%) 151 (48%) 58 (26%) 23 (52%) High 40 (15%) 54 (17%) 9 (4%) 7 (16%) 3D-CRT indicates conventional dose 3-dimensional conformal radiotherapy; IMRT, intensity-modulated radiotherapy; BRT, permanent transperineal brachytherapy; XRT, radiotherapy; PSA, prostate-specific antigen. downsize the prostate gland if the prostate gland size was significantly enlarged, or if there was significant pubic arch interference noted on pelvic computed tomography (CT) scans. The median duration of ADT was 3 months. Forty-four patients were treated with combined EBRT þ BRT boost. These patients usually had 1 of these features: PSA level 10 ng/ml, Gleason score 7, and a clinical disease stage 2B. The treatment regimen was comprised of 45 Gy of EBRT to the prostate and seminal vesicles using 3D-CRT, followed by a BRT boost of 110 Gy using I-125 seeds, or 90 Gy using Pd-103 seeds. Twelve patients also received short-term ADT. All BRT planning was performed using preoperative ultrasonography, and preloaded needles were used. Before 2000, CT-based postimplant dosimetry was not available in our department. For the 130 patients treated since 2000, the dose covering 90% of the prostate volume (D90) and the fractional volume of prostate covered by 100% of the prescribed dose (V100) were calculated. Follow-Up Patients were seen for follow-up visits every 3 to 6 months initially for the first 2 years, and then every 6 to 12 months thereafter. PSA measurement and digital rectal examination were performed routinely at each visit. Urinary (genitourinary [GU]) and rectal (gastrointestinal [GI]) side effects were graded at the time of each follow-up visit using a modified Radiation Therapy Oncology Group (RTOG) scale (Tables 2 and 3). Acute side effects were defined as those that occurred within 3 months of treatment, and late side effects were those that occurred after 3 months. Statistical Analysis Treatment outcomes were measured in terms of bned rate, overall survival, local control, and distant control of disease. Biochemical failure was defined as an increase in the PSA level of 2 ng/ml above the nadir with no backdating (ASTRO Phoenix definition). Local failure was 5598 Cancer December 1, 2009

4 Dose Escalation for Prostate Cancer/Wong et al Table 2. MCA Modification of the RTOG GU Toxicity Scale Grade 1 Grade 2 Grade 3 Grade 4 Grade 5 Minimal symptoms that require no medications Symptoms that require medications such as alpha blocking agents, urinary anesthetic agents, NSAIDs, or mild narcotics Incontinence requiring 2 pads per d Mild hematuria Nocturia every h or more frequently (after alpha blocking agents) (this must be >2x more nocturia than baseline state) Painful passage of macroscopic blood or clots Pain requiring stronger narcotics (eg, morphine, oxycodone, etc) Obstruction requiring prolonged (>1 wk) catheterization (obstruction was not present prior to radiotherapy) Toxicity requiring minor surgical intervention Hematuria requiring coagulation or hyperbaric oxygen Incontinence requiring >2 pads per d Requires major surgical intervention and/or hospitalization (such as cystectomy) Fatal complication MCA indicates Mayo Clinic Arizona; RTOG, Radiation Therapy Oncology Group; GU, genitourinary; NSAIDS, nonsteroidal anti-inflammatory agents. defined as clinically palpable disease recurrence. Distant failure was defined as the development of metastasis documented by radiologic studies. Estimates of rates of bned, overall survival, local control, and distant control of disease were calculated using the Kaplan-Meier product limit method. To evaluate the effects of potential prognostic factors on treatment, analysis was performed using the log-rank test and multivariate analysis with the Cox proportional hazards model. The Pearson chi-square test was performed to compare the side effects of each treatment modality. Calculation of BED To determine the BED for fractionated EBRT, the equation based on the linear-quadratic model was used: BED ¼ ndð1 þðd=a=bþþ in which n ¼ number of fractions; d ¼ dose per fraction; and a/b ¼ a tissue-specific and effect-specific parameter associated with the linear-quadratic model. To calculate the BED for I-125 and Pd-103 low dose-rate permanent decaying implants, the equation proposed by Dale was used 2 : BED ¼ðR 0 =kþf1 þðr 0 =ðl þ kþða=bþþg Table 3. MCA Modification of the RTOG GI Toxicity Scale Grade 1 Grade2 Grade 3 Grade 4 Grade 5 Minimal symptoms that require no medications except fiber supplements Bleeding requiring no therapy or fiber alone, not severe enough to induce anemia Symptoms that require medications such as loperamide, pramoxine and hydrocortisone, and hydrocortisone Mucous production or discharge requiring 2 pads/d Pain requiring mild narcotics Bleeding requiring 2 coagulation procedures, or anemia Diarrhea requiring hospitalization Pain requiring stronger narcotics (eg, morphine, oxycodone, etc) Toxicity requiring minor surgical intervention Bleeding requiring coagulation >2 times, transfusion, or hyperbaric oxygen Incontinence requiring >2 pads per d Requires major surgical intervention and/or hospitalization (such as colostomy or repair of a fistula) Fatal complication MCA indicates Mayo Clinic Arizona; RTOG, Radiation Therapy Oncology Group; GI, gastrointestinal. in which R 0 ¼ initial dose rate of implant; k ¼ radioactive decay constant ¼ 0.693/T 1/2 ¼ radioactive half-life of isotope; l ¼ repair rate constant ¼ 0.693/t 1/2 ; and t 1/2 ¼ tissue repair half-time. The a/b ratio for prostate cancer was believed to be low, and a value of 2 Gy was used here. Other values used for BED calculations were t 1/2 ¼ 1 hour and T 1/2 ¼ 60 days for I-125 and 17 days for Pd ,4 The BED values for combined EBRT and BRT were obtained by adding the BED for each individual modality. 6 RESULTS The median follow-up was 58 months (range, 3 to 121 months) for the entire cohort of 853 patients. The median follow-up was 62 months, 56 months, 49 months, and 63 months, for patients treated with 3D-CRT, IMRT, BRT alone, and EBRT þ BRT, respectively. The distribution of risk groups among the different treatment modalities revealed that patients who received BRT alone constituted a select group of generally low-risk patients (70% at low risk, 26% at intermediate risk, and 4% at high risk). There was a significant association noted between the presence of perineural invasion and disease risk group (P <.0001). The percentages of biopsy demonstrating perineural invasion were 5%, 14%, and 21%, respectively, for the low-risk, intermediate-risk, and high-risk groups. CT-based postimplant dosimetry was available for 130 patients. All these cases used I-125 seeds. For the 106 Cancer December 1,

5 Table 4. 5-Year Biochemical Disease Control Rates 5-Year Univariate Multivariate Biochemical analysis analysis Control, % P P Treatment modality 3D-CRT 74% <.0001 <.0001 IMRT 87% BRT 94% XRT1BRT 94% FIGURE 1. Biochemical control (bned) rate by treatment modality is shown. BRT indicates permanent transperineal brachytherapy; EB, external beam; IMRT, intensity-modulated radiotherapy; 3D-CRT, conventional dose 3-dimensional conformal radiotherapy. patients who were treated with I-125 seeds alone, the median D90 was Gy (range, Gy), and the median V100 was 0.92 (range, ). For 24 patients treated with EBRT þ BRT boost, the median D90 was Gy (range, Gy), and the median V100 was 0.95 (range, ). To assess the extent of dose escalation from IMRT and BRT compared with conventional dose 3D-CRT, the BED for each treatment modality was calculated. For 3D- CRT doses of 66 to 71 Gy, the corresponding BEDs were to Gy. For IMRT doses of 75.6 to 77.4 Gy, the corresponding BEDs were to 147 Gy. For BRT alone, the BEDs were Gy for I-125 and Gy for Pd-103. For EBRT þ BRT boost, the BEDs were Gy for I-125 and Gy for Pd-103, respectively. The 5-year overall survival for the entire group was 97%. The 5-year bned rates, local control rates, and distant control rates were 74%, 93%, and 96%, respectively, for 3D-CRT; 87%, 99%, and 97%, respectively, for IMRT; 94%, 100%, and 99%, respectively, for BRT alone; and 94%, 100%, and 97%, respectively, for EBRT þ BRT. The bned rates for 3D-CRT were significantly less than those of the other higher dose modalities (P <.0001) (Fig. 1). Table 4 summarized the 5-year bned rates for these patients according to treatment modality, T classification, PSA level, Gleason score, presence of perineural invasion, use of adjuvant hormonal treatment, and risk T classification T1c 92% < T2a 89% T2b 82% T2c 74% T3 48% PSA, ng/ml 10 89% < % >20 66% Gleason score 6 88% < % Perineural invasion No 86% < Yes 73% Adjuvant hormonal treatment No 85%.95 Yes 83% Risk group Low 92% <.0001 Intermediate 82% High 64% 3D-CRT indicates conventional dose 3-dimensional conformal radiotherapy; IMRT, intensity-modulated radiotherapy; BRT, permanent transperineal brachytherapy; XRT, radiotherapy; PSA, prostate-specific antigen. group. For the low-risk group, the 5-year bned rates for patients treated with conventional dose 3D-CRT were similar to those of patients treated with other modalities (P ¼.22) (Fig. 2). Dose escalation with IMRT or BRT did not achieve better bned rates. For the intermediaterisk group, bned rates were significantly improved with high-dose IMRT, BRT alone, or EBRT þ BRT, compared with conventional dose 3D-CRT (88% vs 94% vs 94% vs 74%, respectively; P <.0001) (Fig. 3). However, comparison between EBRT þ BRT with other treatment modalities would have limited power because of the relatively small number of patients. For the high-risk group, high-dose IMRT achieved significantly better bned rates compared with conventional dose 3D-CRT (74% vs 49%; P ¼.027). Patients receiving BRT alone or EBRT 5600 Cancer December 1, 2009

6 Dose Escalation for Prostate Cancer/Wong et al FIGURE 2. Biochemical control (bned) rate for the low-risk group by treatment modality is shown. EB indicates external beam; BRT, permanent transperineal brachytherapy; IMRT, intensity-modulated radiotherapy; 3D-CRT, conventional dose 3-dimensional conformal radiotherapy. FIGURE 3. Biochemical control (bned) rate for the intermediate-risk group by treatment modality is shown. 3D-CRT indicates conventional dose 3-dimensional conformal radiotherapy; EB, external beam; BRT, permanent transperineal brachytherapy; IMRT, intensity-modulated radiotherapy. þ BRT were combined as a group for analysis of treatment outcome. Although BRT or EBRT þ BRT achieved a bned rate of 72% for the high-risk group, the small number of patients and events did not allow for a meaningful comparison of treatment outcome with other modalities. On univariate analysis, the following factors were found to be significantly associated with bned rate: treatment modality (P <.0001), T classification (P <.0001), FIGURE 4. Biochemical control (bned) rate for the intermediate-risk group by treatment modality excluding those patients who received androgen deprivation therapy is shown. EB indicates external beam; BRT, permanent transperineal brachytherapy; IMRT, intensity-modulated radiotherapy; 3D-CRT, conventional dose 3-dimensional conformal radiotherapy. Gleason score (P <.0001), PSA level (P <.0001), presence of perineural invasion (P <.0001), and risk group (P <.0001) (Table 4). The use of adjuvant hormonal treatment was not found to be associated with treatment outcome (P ¼.95).Anattemptwasmadetoevaluatethe significance of percentage positive biopsy as a prognostic factor. However, many biopsy reports did not provide the total number of cores submitted as well as the number of positive cores to allow the calculation of the percentage positive biopsy, making it infeasible to conduct this analysis. Multivariate analysis using the Cox proportional hazards model was performed to evaluate the effect of treatment modality, clinical T classification, PSA level, Gleason score, and presence of perineural invasion on treatment outcome (Table 4). All these factors remained significantly associated with the bned rate. Risk grouping was not included in multivariate analysis because it contained elements of T classification, Gleason score, and PSA level and would confound the analysis. Use of ADT was also not included in multivariate analysis because it was not found to be significant in univariate analysis. Although the use of ADT was not significantly associated with treatment outcome on univariate analysis, a separate analysis in each risk group was performed for patients who did not receive any ADT. This would eliminate the effect of ADT as a confounding factor on bned rate (Fig. 4) (Table 5). For the low-risk group, the 5-year bned rate by 3D-CRT was not improved by dose Cancer December 1,

7 Table 5. 5-Year Biochemical Disease Control Rates Based on Risk Group Category and Treatment Modality for Patients Who Did Not Receive ADT Treatment No. of Patients Low Risk Intermediate Risk High Risk PSA Control No. of Patients PSA Control No. of Patients PSA Control 3D-CRT % 93 65% 21 55% IMRT 94 93% 88 85% 15 76% BRT % 32 83% 3 50% EBRTþBRT 8 100% % 5 100% P ADT indicates androgen deprivation therapy; PSA, prostate-specific antigen; 3D-CRT, conventional dose 3-dimensional conformal radiotherapy; IMRT, intensity-modulated radiotherapy; BRT, permanent transperineal brachytherapy; EB, external beam radiotherapy. Table 6. Maximum Toxicity Scores (MCA Modifications of RTOG Toxicity Scales) Grade 3D-CRT IMRT BRT EBRT1BRT Chi-Square P Acute GI 0 26% 28% 77% 75% < % 26% 14% 14% 2 54% 45% 8% 11% 3 0 1% 0 0 Late GI 0 57% 63% 72% 39% % 23% 15% 34% 2 15% 14% 12% 23% 3 2% 1% 1% 5% Acute GU 0 38% 27% 11% 9% < % 22% 14% 16% 2 39% 49% 68% 73% 3 1% 3% 6% 2% Late GU 0 66% 47% 23% 11% < % 22% 14% 18% 2 16% 27% 45% 52% 3 5% 5% 18% 18% MCA indicates Mayo Clinic Arizona; RTOG, Radiation Therapy Oncology Group; 3D-CRT, conventional dose 3-dimensional conformal radiotherapy; IMRT, intensity-modulated radiotherapy; BRT, permanent transperineal brachytherapy; EB, external beam radiotherapy; GI, gastrointestinal; GU, genitourinary. escalation with IMRT, BRT or EBRT þ BRT. For the intermediate-risk group, higher dose by IMRT, BRT or EBRT þ BRT achieved a significantly better outcome compared with 3D-CRT (P ¼.0003). The 5-year bned rates were 65%, 85%, 83%, and 100% for 3D-CRT, IMRT, BRT, and EBRT þ BRT, respectively. However, the number of patients in the high-risk group was too small to draw a meaningful conclusion. The maximum acute and late GI and GU toxicity grades of each treatment modality are listed in Table 6. No grade 4 GI or GU toxicities were reported to occur. Grade 2 toxicity could usually be managed by conservative measures, whereas grade 3 toxicity had more detrimental impact on the quality of life of patients. Despite the delivery of higher doses of EBRT, there was no increase in acute or late GI toxicity from IMRT compared with 3D-CRT. There were more grade 2 acute (49% vs 39%) and late (27% vs 16%) GU toxicities, but no increase in grade 3 toxicity, from high-dose IMRT compared with conventional dose 3D-CRT. The toxicity profile of BRT was very different from IMRT or 3D-CRT. BRT alone caused a much lower incidence of acute or late GI toxicity than EBRT. Late grade 3 GI toxicity occurred in 2% of the patients treated with 3D-CRT. Four patients developed proctitis and hematochezia that required multiple argon plasma coagulation (APC) and/or blood transfusion. No patient in the IMRT group developed significant hematochezia, although Cancer December 1, 2009

8 Dose Escalation for Prostate Cancer/Wong et al patient developed a grade 3 late GI toxicity in the form of fecal incontinence. For the group of patients treated with BRT alone, 2 patients developed grade 3 proctitis (1%), 1 of whom required hyperbaric oxygen treatment. On the contrary, EBRT þ BRT resulted in more late grade 2 and 3 GI side effects than other modalities, despite a lower incidence of acute GI toxicities. Of the group of patients treated with EBRT þ BRT, 2 patients (5%) had grade 3 late GI toxicities, 1 of whom was treated with APC, whereas the patient other eventually received hyperbaric oxygen. Both BRT alone and EBRT þ BRT caused significantly more grade 2 and 3 acute and late GU toxicity compared with 3D-CRT or IMRT. Although grade 2 GU toxicity could be managed with medications such as a-blockers and urinary anesthetics, grade 3 toxicity often require surgical interventions. The incidence of late grade 3 GU side effects was 5% for both 3D-CRT and IMRT, and was 18% for both BRT and EBRT þ BRT. Fourteen patients in the 3D-CRT group had grade 3 late GU toxicity. Six patients underwent dilatation and 3 underwent direct vision incisional urethrotomy (DVIU) for urethral stricture. Five patients had bladder outlet obstruction, 3 of which ultimately underwent transurethral resection of prostate (TURP), and 1 required prolonged self-intermittent catheterization (SIC). Twelve patients in the IMRT group had a grade 3 late GU toxicity. Urethral dilatation was performed in 6 patients and DVIU in 1 patient. Three patients underwent TURP, and 1 patient was treated with prolonged SIC. One patient developed hematuria that required treatment with formalin. In the group of patients treated with BRT alone, 25 patients underwent urethral dilatation and 6 underwent DVIU for stricture. Five patients had bladder outlet obstruction, 4 of whom were managed with prolonged SIC, whereas 1 patient underwent TURP and subsequently developed incontinence. Four patients had moderate incontinence that required >2 pads per day. In the group of patients treated with EBRT þ BRT group, 8 patients underwent urethral dilatation and 1 underwent DVIU. One patient had chronic pain that required narcotic pain medication. DISCUSSION To the best of our knowledge, the current study is among the few reports to provide an updated comparison of dose escalation from conventional dose 3D-CRT, using either high-dose IMRT, BRT or EBRT þ BRT, in the treatment of localized prostate cancer. To evaluate the extent of dose escalation, the BEDs of high-dose IMRT, BRT, and EBRT þ BRT were calculated and compared with that of conventional dose 3D-CRT. The differences in BED between 66 Gy of 3D-CRT and 77.4 Gy of IMRT, I-125 BRT, Pd-103 BRT, 45 Gy of EBRT þ I-125 BRT boost, and 45 Gy of EBRT þ Pd-103 boost were 21.6 Gy, 33 Gy, 30.9 Gy, 81.1 Gy, and 77.4 Gy, respectively. Data from the current study demonstrated that radiation dose escalation, either with IMRT or with BRT or EBRT þ BRT, leads to improved bned rates compared with conventional dose 3D-CRT for patients in the intermediate-risk group. On the contrary, for patients with low-risk disease, dose escalation does not appear to achieve better a bned rate. The current study data also demonstrated better bned rates with IMRT compared with 3D-CRT for patients in the high-risk group. However, the small number of events and patients who were treated with BRT or EBRT þ BRT in this group limited our power to compare them with other radiation modalities, and therefore studies with a larger number of patients would be needed. To eliminate any confounding effects of ADT on the treatment outcome, a separate analysis was performed excluding patients who received ADT. The study data again demonstrated that dose escalation with IMRT, BRT or EBRT þ BRT achieved significantly better bned rates for patients in the intermediate-risk group compared with patients treated with conventional dose 3D-CRT. For the patients in the low-risk group, there was no significant differenceintreatmentoutcomewithdoseescalationasused in our study observed. The small number of patients in the high-risk group limited the power to draw a conclusion. There is clearly a dose response for prostate cancer irradiation. Several randomized studies have demonstrated that dose escalation by EBRT leads to improved bned. 7-9 To our knowledge, there are no randomized trials performed to date comparing the various RT modalities for prostate cancer. Although several retrospective studies have compared EBRT with BRT, many of these reports included EBRT with doses <72 Gy, and the conclusions were different from those noted in the current study. In a study by D Amico et al 766 patients were treated with EBRT with a median dose of 66 Gy (median Cancer December 1,

9 follow-up, 38 months), 66 patients were treated with BRT with Pd-103, and 152 patients received Pd-103 implant and neoadjuvant ADT (median follow-up, 41 months). 10 Using 3 consecutively rising PSA values with backdating as the definition of biochemical disease failure, disease control for low-risk patients was similar with EBRT, BRT, or BRT þ neoadjuvant ADT. For patients in the intermediate-risk and high-risk groups, EBRT achieved better results compared with BRT or BRT þ neoadjuvant ADT. The follow-up duration of this study was relatively short, and the results of BRT ADT appeared to be inferior to those of other published studies. Brachman et al reported their results for 1527 patients treated with BRT or EBRT. 11 ThemedianEBRTdosewas 66 Gy, and the median follow-up was 41 months for EBRT and 51 months for BRT. Patients with a Gleason score of 8 to 10, or a PSA level of 10 to 20 ng/ml were found to have significantly better 5-year bned rates when treatedwithebrtcomparedwithbrt(5-yearbned rates, 52% vs 28% for patients with a Gleason score of 8-10, and 70% vs 53% for patients with a PSA level of ng/ml). No difference in disease control was found between EBRT or BRT for patients with a PSA level <10 ng/ml or apsalevel>20 ng/ml, a Gleason score of 4 to 6, or T1 and T2 disease. Neither EBRT nor BRT was found to be particularly effective for patients with a PSA level >20 ng/ ml, with 5-year bned rate of approximately 50%. In another study, Zelefsky et al reported the results for 282 patients with favorable-risk disease (a PSA level 10 ng/ml, a Gleason score 6 and T classification 2b). 12 The 5-year PSA control rate was 88% for the patients treated with 3D-CRT (median dose, 70.2 Gy) and 82% for the patients treated with BRT. To our knowledge, there are only a few reports in the literature that compare high-dose EBRT with BRT. The results of the current study are in agreement with the published reports. Potters et al reported a multi-institutional study for patients clinical T1 to T2 prostate cancer that included 733 patients treated with BRT alone and 340 patients treated with EBRT alone. 13 The median follow-up was 58 months. The median EBRT dose was 74 Gy (range, Gy). The 7-year bned rates were 77% and 74%, respectively, for EBRT and BRT. There was no difference in the bned rate noted between high-dose EBRT and BRT. PSA level and Gleason score were found to be independent predictors for disease control. In a report by Kupelian et al 484 patients were treated with an EBRT dose <72 Gy, 301 patients received an EBRT dose 72 Gy, 950 patients received BRT, and 222 patients received EBRT þ BRT. 14 The median follow-up was 56 months. The 5-year bned rates were 51% for an EBRT dose <72 Gy, 81% for an EBRT dose 72 Gy, 83% for BRT, and 77% for EBRT þ BRT. When the EBRT dose <72 Gy was excluded, there was no significant difference in the bned rate noted by treatment modality, suggesting that the intrinsic tumor characteristics would determine the final outcome if a high dose of radiation is delivered, whether by IMRT or BRT. The results of the current study have demonstrated that treatment modality, clinical T classification, Gleason score, PSA level, risk group, and presence of perineural invasion are significant prognostic factors on univariate analysis. The use of ADT was not found to be associated with treatment outcome. A possible explanation for this finding was that ADT was generally used for patients with disease that had unfavorable features, and not for those who had more favorable disease. ADT may have improved the bned rate for patients with unfavorable disease, making the result closer to that of those with more favorable disease. Another indication for ADT was to downsize the prostate gland, which should not have an impact on the treatment outcome. On multivariate analysis, treatment modality, T classification, Gleason score, PSA level, and presence of perineural invasion were found to remain significant. Although T classification, Gleason score, and PSA level are generally accepted as important prognostic factors for prostate cancer, the presence of perineural invasion is not commonly evaluated in many published reports, and there is controversy regarding its significance. 15,16 Although some studies do not find the presence of perineural invasion to adversely affect treatment outcome in patients treated with EBRT or BRT, the data from the current study indicate that the presence of perineural invasion is associated with risk category and is an unfavorable prognostic factor. Harnden et al published a literature review on the prognostic significance of perineural invasion, concluding that the weight of evidence supports the significance of perineural invasion as a prognostic factor. 17 The main concern with dose escalation is an increase in normal tissue toxicity. Both IMRT and BRT allow 5604 Cancer December 1, 2009

10 Dose Escalation for Prostate Cancer/Wong et al dose escalation while limiting the doses to the surrounding normal tissues. In the current study, patients treated with high-dose IMRT experienced more grades 1 and 2, but not grade 3, acute and late GU side effects compared with those treated with conventional dose 3D-CRT. There was no increase in acute or late GI toxicity noted for patients treated with high-dose IMRT compared with those treated with conventional dose 3D-CRT. BRT can conceptually be thought of as the ultimate example of CRT, with the delivery of a high dose of radiation to the prostate and rapid dose drop-off in the surrounding tissues. Many studies have reported acceptable side effects and quality of life for patients treated with BRT However, the published data also suggest that BRT or EBRT þ BRT cause more urinary side effects compared with EBRT alone. Litwin et al reported that BRT caused more obstructive and irritative urinary symptoms than EBRT. 20 A recent report from Fox Chase Cancer Center comparing IMRT with I-125 permanent BRT reported grade 2 or higher GI and GU toxicity rates of 2.4% and 3.5%, respectively, with IMRT using 74 to 78 Gy at 3 years versus 7.7% and 19.2%, respectively, for I- 125 BRT. In an analysis of 5621 Medicare patients who received BRT alone or EBRT þ BRT with at least 2 years of follow-up, 14.1% required an invasive procedure for GI or GU complications. 21 In this report, EBRT þ BRT were found to result in significantly more late GI and GU side effects that required invasive procedures compared with BRT alone. The urinary and rectal side effects of treatment have significant impacts on the health-related quality of life (HR-QOL) of patients. In a multi-institutional prospective study assessing the HR-QOL of 1201 patients and 625 spouses before and after treatment for prostate cancer, 18% of patients in the BRT group and 11% of patients in the EBRT group reported moderate or worse distress from overall urinary symptoms at 1 year. 22 Bowel symptoms causing moderate or worse distress were reported in 11% of patients who had been treated with EBRT and 8% of patients who had been treated with BRT at 2 years. Data from the current study indicate that BRT or EBRT þ BRT result in significantly more grade 2 and 3 acute and late GU toxicities than 3D-CRT or IMRT. Proper patient selection for BRT or EBRT þ BRT is especially important. Patients who have significant obstructive urinary symptoms preoperatively should be counseled regarding the increased risk of urinary side effects. The current study data also demonstrate that BRT alone results in less GI toxicity than EBRT, whereas the addition of EBRT to BRT increases the risk of grade 2 or higher late GI toxicities. The retrospective nature of the current study introduces biases and should be kept in mind when interpreting the results. The decision regarding which treatment modality to administer to each individual patient was subject to both physician and patient preferences. Randomized studies will be needed to evaluate the relative efficacy of high-dose IMRT and BRT. Although CT-based postimplant dosimetry was not available for all patients, the D90 and V100 values for the 130 cases with such dosimetry suggested that the quality of these implants was good. We believed that the implant quality for those cases without CT-based postimplant dosimetry should be similar to that of patients who had CT-based postimplant dosimetry, because all the implants were performed in a consistent manner by the same team of radiation oncologists and urologists. In addition, the bned rates of our BRT cases were comparable to large published series from experienced centers, suggesting that the quality of our implants should be acceptable. 13,14 Conclusions Dose escalation from conventional dose 3D-CRT can be achieved with IMRT, BRT, or EBRT þ BRT. Data from the current study indicate that there is a dose-response effect in the bned rate for patients with intermediaterisk prostate cancer. IMRT causes fewer grade 2 and 3 acute and late GU toxicities than BRT or EBRT þ BRT. For patients with low-risk disease, dose escalation from conventional dose 3D-CRT does not provide an improvement in bned rate. Conflict of Interest Disclosures The authors made no disclosures. References 1. Jemal A,Ward E, Hao Y, Xu J, Murray T, Thun MJ. Cancer statistics, CA Cancer J Clin. 2008;58: Dale RG. The application of the linear-quadratic doseeffect equation to fractionated and protracted radiotherapy. Br J Radiol. 1985;58: Cancer December 1,

11 3. Fowler J, Chappell R, Ritter M. Is alpha/beta for prostate tumors really low? Int J Radiat Oncol Biol Phys. 2001; 50: Stock RG, Stone NN, Cesaretti JA, Rosenstein BS. Biologically effective dose values for prostate brachytherapy: effects on PSA failure and posttreatment biopsy results. Int J Radiat Oncol Biol Phys. 2006;64: Zelefsky MJ, Leibel SA, Gaudin PB, et al. Dose escalation with 3-dimensional conformal radiation therapy affects the outcome in prostate cancer. Int J Radiat Oncol Biol Phys. 1998;41: Jani AB, Hand CM, Lujan AE, et al. Biological effective dose for comparison and combination of external beam and low-dose rate interstitial brachytherapy prostate cancer treatment plans. Med Dosim. 2004;29: Pollack A, Zagars GK, Starkschall G, et al. Prostate cancer radiation dose response: results of the M. D. Anderson phase III randomized trial. Int J Radiat Oncol Biol Phys. 2002;53: Zietman AL, DeSilvio ML, Slater JD, et al. Comparison of conventional-dose vs high-dose conformal radiation therapy in clinically localized adenocarcinoma of the prostate: a randomized controlled trial. JAMA. 2005;294: Peeters ST, Heemsbergen WD, Koper PC, et al. Doseresponse in radiotherapy for localized prostate cancer: results of the Dutch multicenter randomized phase III trial comparing 68 Gy of radiotherapy with 78 Gy. J Clin Oncol. 2006;24: 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: Brachman DG, Thomas T, Hilbe J, Beyer DC. Failure-free survival following brachytherapy alone or external beam irradiation alone for T1-2 prostate tumors in 2222 patients: results from a single practice. Int J Radiat Oncol Biol Phys. 2000;48: Zelefsky MJ, Wallner KE, Ling CC, et al. Comparison of the 5-year outcome and morbidity of 3-dimensional conformal radiotherapy versus transperineal permanent iodine-125 implantation for early-stage prostatic cancer. J Clin Oncol. 1999;17: Potters L, Klein EA, Kattan MW, et al. Monotherapy for stage T1-T2 prostate cancer: radical prostatectomy, external beam radiotherapy, or permanent seed implantation. Radiother Oncol. 2004;71: Kupelian PA, Potters L, Khuntia D, et al. Radical prostatectomy, external beam radiotherapy <72 Gy, external beam radiotherapy > or ¼72 Gy, permanent seed implantation, or combined seeds/external beam radiotherapy for stage T1-T2 prostate cancer. Int J Radiat Oncol Biol Phys. 2004;58: Merrick GS, Butler WM, Galbreath RW, Lief JH, Adamovich E. Perineural invasion is not predictive of biochemical outcome following prostate brachytherapy. Cancer J. 2001; 7: Pollack A, Hanlon AL, Horwitz EM, Feigenberg SJ, Uzzo RG, Hanks GE. Prostate cancer radiotherapy dose response: an update of the fox chase experience. J Urol. 2004; 171: Harnden P, Shelley MD, Clements H, et al. The prognostic significance of perineural invasion in prostatic cancer biopsies: a systematic review. Cancer. 2007;109: Ash D, Bottomley D, Al-Qaisieh B, Carey B, Gould K, Henry A. A prospective analysis of long-term quality of life after permanent I-125 brachytherapy for localised prostate cancer. Radiother Oncol. 2007;84: Crook J, Fleshner N, Roberts C, Pond G. Long-term urinary sequelae following 125iodine prostate brachytherapy. J Urol. 2008;179: ; discussion Litwin MS, Gore JL, Kwan L, et al. Quality of life after surgery, external beam irradiation, or brachytherapy for early-stage prostate cancer. Cancer. 2007;109: Chen AB, D Amico AV, Neville BA, Earle CC. Patient and treatment factors associated with complications after prostate brachytherapy. J Clin Oncol. 2006;24: Sanda MG, Dunn RL, Michalski J, et al. Quality of life and satisfaction with outcome among prostate-cancer survivors. N Engl J Med. 2008;358: Cancer December 1, 2009