- Email: [email protected]
Radiotherapy After Prostatectomy: Is the Evidence for Dose Escalation out There?
Int. J. Radiation Oncology Biol. Phys., Vol. 71, No. 2, pp. 346–350, 2008 Copyright Ó 2008 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/08/$–see front matter
doi:10.1016/j.ijrobp.2007.10.008
CLINICAL INVESTIGATION
Prostate
RADIOTHERAPY AFTER PROSTATECTOMY: IS THE EVIDENCE FOR DOSE ESCALATION OUT THERE? CHRISTOPHER R. KING, PH.D., M.D., AND DANIEL S. KAPP, PH.D., M.D. Department of Radiation Oncology, Stanford University School of Medicine, Stanford, CA Purpose: To study the effective doses of radiotherapy (RT) after prostatectomy in search for evidence of a dose– response. Methods and Materials: Original and available data from published studies of adjuvant and salvage RT after prostatectomy were analyzed in the context of biochemical tumor control probability (TCP) dose–response curves. Comparisons were made with dose-escalation studies of radical RT for localized disease. Arguments based on a microscopic vs. macroscopic disease dose-response relationships were used to interpret the clinical data. Results: The tumor control rates after salvage RT were consistent with the TCP dose–response curve of radical RT, suggesting the presence of macroscopic-equivalent disease among salvage patients. For radical RT, the dose to achieve 50% biochemical tumor control was 65.9 Gy (95% confidence interval [CI], 64.8–66.8) and the Slope50 was 2.6%/Gy (95% CI, 2.3–3.0). For salvage RT, the corresponding values were 66.8 Gy (95% CI, 65.1–68.4) and 3.8%/Gy (95% CI, 2.5–7.6). For a comparable TCP, the dose for adjuvant RT was 6 Gy lower, consistent with one-tenth the burden of local disease. The present doses for adjuvant or salvage RT in the range of 60–70 Gy appear to be still on the steep part of the TCP dose–response curve. Conclusions: The effective doses and dose–response relation observed with RT after prostatectomy are consistent with the presence of macroscopic-equivalent disease for salvage patients and about a tenth of the residual disease for adjuvant patients. Greater doses would potentially achieve significantly greater disease-free control rates. A randomized trial with 250 patients comparing 64 vs. 70 Gy for salvage RT or 60 vs. 66 Gy for adjuvant RT would be capable of addressing this issue. Ó 2008 Elsevier Inc. Prostate cancer, Postoperative radiotherapy, Dose response, Dose escalation.
INTRODUCTION
highest radiation dose that can be delivered with acceptable morbidity and suggested a minimum of 64 Gy at conventional dose fractionation (5). Although ample evidence exists in support of dose escalation to $78 Gy for radical RT (RRT) for localized prostate cancer (6–9), very few studies have adequately examined the dose–response relationship in the postoperative setting owing to the relatively narrow range of doses commonly used. Of the multitude of retrospective studies on postoperative RT, only two studies, one for SRT (10) and one for ART (11), each with <90 patients, showed that doses >64.8 Gy provide improved biochemical control compared with lower doses. A recent analysis of patients treated at our institution showed that SRT to 70 Gy achieved superior 5-year bRFS compared with 60 Gy (58% vs. 25%, p = 0.012) and that the high dose was an independent factor for bRFS on multivariate analysis (12). It is commonly presumed, albeit unproven, that the tumor burden in the postoperative setting is microscopic and, therefore, that lower effective doses are necessary compared with the dose needed
The use of adjuvant radiotherapy (ART) (i.e., in the immediate postoperative setting) or salvage RT (SRT) (i.e., after a demonstrated biochemical recurrence) offers potentially curative treatment after unsuccessful radical prostatectomy. Two randomized trials have demonstrated improved biochemical relapse-free survival (bRFS) for ART after prostatectomy for patients with high-risk pathologic features, defined as pT3 or positive surgical margins (1, 2). The role of SRT for biochemical relapse after prostatectomy has also been demonstrated in numerous retrospective studies, most recently in a multi-institutional pooled data analysis (3) and a comprehensive review of published studies (4). Together, these reports form the basis for optimal patient selection and therapeutic guidelines. In the two ART trials, the prostate bed dose was 60 Gy (1) and 60–64 Gy (2). For SRT, the American Society for Therapeutic Radiology and Oncology consensus guidelines have recommended the
Conflict of interest: none. Received Aug 30, 2007, and in revised form Oct 11, 2007. Accepted for publication Oct 12, 2007.
Reprint requests to: Christopher R. King, Ph.D., M.D., Department of Radiation Oncology, Stanford University School of Medicine, Stanford Cancer Center, 875 Blake Wilbur Dr., Stanford, CA 94305. Tel: (650) 736-0698; Fax: (650) 725-8231; E-mail: [email protected] 346
Dose escalation of RT after prostatectomy d C. R. KING AND D. S. KAPP
in the definitive setting. An analysis of the patterns of failure from the Southwestern Oncology Group (SWOG) ART randomized trial demonstrated that treatment failure is predominantly local, and, therefore, an improvement in local therapy will result in improved outcomes (13). The present study compared the data for the dose–response relationship for postoperative treatment and definitive treatment to evaluate the hypothesis that salvage and adjuvant doses should be escalated to improve biochemical diseasefree survival for these patients. METHODS AND MATERIALS We first established the dose–response relationship from the RRT data to serve as a point of reference. We constructed the dose– response curve for RRT from studies of RT alone (i.e., excluding any use of androgen suppression) that provided the 5-year bRFS data. The inclusion criteria were randomized trials or modern series with >200 patients. All studies had consistently used the American Society for Therapeutic Radiology and Oncology definition of biochemical failure of three consecutive increases greater than the nadir with the failure date midway between the first increase and the previous value (14). Only series providing the results for patients predominantly at ‘‘intermediate risk’’ were used (defined as possessing one or more of the following features: initial prostate-specific antigen [PSA] level of 10–20 ng/mL, biopsy Gleason score of 7, or clinical Stage T2b-T2c). A similar approach was used by Fowler et al. (15) to remove the potential confounding effects of androgen suppression with RT and to produce a reference RT dose–response curve for localized prostate cancer. The series used to construct the current dose–response curve are listed in Table 1 and included The University of Texas M. D. Anderson dose-escalation trial (6), Massachusetts General Hospital dose-escalation trial (7), prospective dose-escalation series from Fox Chase Cancer Center (8) and Memorial Sloan-Kettering Cancer Center (9), and the series from Princess Margaret Hospital reporting on outcomes with low-dose RT (16). We used the standard tumor control probability (TCP) equation to fit these data by the least-squares method. The form of the TCP relationship is given by TCP ¼ expð½d TCD50 =kÞ=½1 þ expð½d TCD50 =kÞ where d is total dose, TCD50 is the dose to achieve 50% biochemical tumor control, and k is a fitting parameter related to the slope at the
347
TCD50 dose point (17). The proportional gain in TCP per additional Gray within the steep part of the TCP curve is given by the parameter Slope50, defined by Slope50 = 25/k (in percentage per Gray). A comprehensive published data search retrieved 10 series on SRT (3, 11, 12, 18–23). Only series that provided 5-year biochemical no evidence of disease results and whose median pre-RT PSA level was <2 ng/mL were included. The details of the SRT series, as well as the two randomized trials of ART, are listed in Table 2. The definition of biochemical failure for postoperative RT is essentially the same as after prostatectomy. In the studies reported, a persistently detectable or increasing PSA >0.2 ng/mL was considered a biochemical failure. Although for SRT all patients actually have disease (measurable by PSA recurrence), of necessity a proportion of ART patients will be without disease. The randomized adjuvant trials provided this critical information from their control (observation) arm, in which the 5-year bRFS rate for the observation arm was 42% in the SWOG trial (2) and 53% in the European Organization for Research and Treatment of Cancer trial (1). Thus, the bRFS for ART must be corrected for the patients without disease to calculate the effectiveness of postoperative RT only among those patients who actually had disease. The actual adjuvant bRFS can then given by the difference in bRFS between the RT and observation arms, divided by the proportion of patients with disease: bRFS (adjuvant) = [bRFS with RT bRFS observation]/[100 bRFS observation]. These two trials differed in their patient selection criteria in the following significant ways. In the EORTC trial, all patients had to have a postoperative PSA level that was <0.4 ng/mL and 10.7% had a postoperative PSA level >0.2 ng/mL. In the SWOG trial, even patients with greater postoperative PSA levels were eligible, and 33.8% had a postoperative PSA level >0.2 ng/mL. Consequently, the median postoperative PSA level was greater in the SWOG trial than in the EORTC trial, and, hence, the absolute bRFS rates in the observation and treatment arms were slightly lower than those in the EORTC trial. The retrospective adjuvant series found were not usable in this context, because they did not estimate the baseline proportion of patients without disease in their patient population.
RESULTS The TCP curve fitted to the RRT series is shown in Fig. 1 (r = 0.990, p <0.0001). The derived value for TCD50 is 65.9 Gy (95% confidence interval [CI], 64.8–66.8) and for the Slope50, it is 2.6%/Gy (95% CI, 2.3–3.0). The SRT series
Table 1. Series with external beam radiotherapy alone for radical radiotherapy of intermediate-risk prostate cancer Series Randomized dose-escalation trial, MDA Randomized dose-escalation trial, MGH Retrospective study, PMH Retrospective study of dose response, FC
Retrospective study of dose response, MSK
Median dose (Gy) 70 78 70.2 79.2 65 (52–67) 69 (<70) 72 (70–74.9) 76 (75–79.9) 82 (>80) 67.5 (64.8–70.2) 78.3 (75.6–81.0)
5-y bNED (%)
Patients (n)
Study
62 75 61 80 45 60 68 76 84 54 79
150 151 197 195 706 43 552 568 367 116 94
Pollack et al. (6) Zietman et al. (7) Catton et al. (17) Eade et al. (8)
Zelefsky et al. (9)
Abbreviations: bNED = biochemical no evidence of disease; MDA = M.D. Anderson Cancer Center; MGH = Massachusetts General Hospital; PMH = Princess Margaret Hospital; FC = Fox Chase Cancer Center; MSK = Memorial Sloan-Kettering Cancer Center. Data in parentheses are ranges.
I. J. Radiation Oncology d Biology d Physics
348
Volume 71, Number 2, 2008
Table 2. Series with external beam radiotherapy for salvage and adjuvant radiotherapy after radical prostatectomy Series Salvage RT retrospective series Catton et al. (19) Tsien et al. (20) Taylor et al. (21) Chawla et al. (22) Buskirk et al. (23) Kalapurakal et al. (24) Anscher et al. (11) Stephenson et al. (3) King and Spiotto (12) Adjuvant RT randomized trials EORTC-22911; Bolla et al. (1) SWOG-8794; Thompson et al. (2)
Median dose (Gy)
5-y bNED (%)
Median pre-RT PSA level (ng/mL)
Patients (n)
60 (54–65) 65 (60–75) 70 (60–78) 64.8 (60.4–64.8) 64.8 (54–72.4) 65 (60–70) 66 (54.6–70) 64.8 (63–66) 60 (60–64) 70 (67.8–70)
29 35 66 35 46 57 50 37 25 58
#2.0 1.2 0.8 1.3 0.7 0.5 1.4 1.1 0.9 0.45
28 57 71 54 368 41 89 1,540 38 84
45* 47*
— —
1,005 425
60 60–64y
Abbreviations: RT = radiotherapy; PSA = prostate-specific antigen; other abbreviations as in Table 1. Data in parentheses are ranges. * Corrected bNED for proportion of patients without disease at baseline as described in text (i.e., [bNED with RT bNED observation]/ [100 bNED observation]). y Median dose not reported.
are shown in Fig. 1 and were remarkably close to the RRT dose–response curve. The TCP curve fit for the SRT series (r = 0.871; p = 0.0004) yielded a value of 66.8 Gy (95% CI, 65.1–68.4) for TCD50 and 3.8%/Gy (95% CI, 2.5–7.6) for the Slope50. Because the 95% CIs overlapped, the difference between the RRT- and SRT-fitted TCP curves was not significant. The ART points from the two randomized trials were also plotted in Fig. 1 and were separated by a lower dose of 6 Gy from the SRT curve.
roughly one-tenth the disease burden of SRT patients. This estimate of disease burden differential between the ART and SRT patients is consistent with the observation that, for the same bRFS, ART required about 6-Gy lower doses than did the SRT patients (Fig. 1). If one considers that a dose of 2 Gy yields a surviving fraction of about 0.5, three fractions (i.e., 6 Gy total) will correspond to a surviving fraction of 0.12 (i.e., about one-tenth), roughly the same estimated
100
DISCUSSION 80
5-year % bNED
By examining the dose–response relationship of SRT, we have shown the interesting result that it is essentially the same as the dose–response relationship for RRT for localized prostate cancer. The conclusion is that SRT patients harbor disease that effectively follows nearly the same macroscopic disease dose–response curve as that for RRT patients. This is not altogether surprising when one considers that SRT patients have clinically measurable disease by PSA that is of the same order of magnitude as for patients with localized disease. The pre-RT PSA level for the salvage patients among the studies used was 1 ng/mL (Table 2). A 1-cm3 tumor contains 109 cells and produces a PSA level of about 3.5 ng/mL (24); therefore, a tumor producing a PSA level of 1 ng/mL will still contain about 108–109 cells and should be considered macroscopic. The TCP curve for the salvage patients is slightly steeper than that for radical patients (i.e., the Slope50 is slightly greater). This observation is consistent with the fact that salvage patients, on the basis of their pre-RT PSA level spanning a very narrow range compared with that of radical patients, form a more homogenous group. In contrast, the ART patients in the two randomized trials were defined as having an undetectable postoperative PSA level if it was <0.2 ng/mL. Therefore, ART patients possess
TCP fit to Radical RT TCP fit to Salvage RT SRT ART (EORTC)
60
ART (SWOG) RRT (MSK)
40
RRT (MDA) RRT (PMH)
20
RRT (FC) RRT (MGH)
50
60
70
80
90
Total Dose (Gy)
Fig. 1. Dose–response graph plotting 5-year biochemical relapsefree survival vs. dose. Individual radical radiotherapy (RRT) series identified by separate symbol (see also Table 1). Dashed line derived from fitting tumor control probability (TCP) relationship only with RRT series. Dotted line indicates TCP fit to salvage RT (SRT) series only. bNED = biochemically no evidence of disease; ART = adjuvant RT; EORTC = European Organization for Research and Treatment of Cancer; SWOG = Southwestern Oncology Group; MSK = Memorial Sloan-Kettering Cancer Center; MDA = M.D. Anderson Cancer Center; PMH = Princess Margaret Hospital; MGH = Massachusetts General Hospital; FC = For Chase Cancer Center.
Dose escalation of RT after prostatectomy d C. R. KING AND D. S. KAPP
fractional difference in disease burden between the salvage and ART patients on the basis of the pre-RT PSA level. However, in each trial, a significant proportion of patients had a pre-RT PSA level >0.2 ng/mL; thus, the practical distinction between the ART and SRT patients was somewhat blurred. The availability of today’s ultrasensitive PSA assays with a threshold of 0.02 ng/mL will allow for the selection of ART patients who possess an even lower burden of disease. Our fitted values of 65.9 Gy for TCD50 and 2.6%/Gy for Slope50 for RRT is in agreement with those (67.5 Gy and 3.2%/Gy, respectively) from M.D. Anderson’s analysis of intermediate-risk patients (25). A radiation dose–response study by Okunieff et al. (17) demonstrated that for most human cancers, the TCD50 for macroscopic disease was about 52 Gy. In the spectrum of human tumors, prostate cancer is considered to be somewhat radioresistant. In agreement with that, we have found the TCD50 for prostate cancer to be 66 Gy. The aforementioned study also showed that for microscopic disease with a 10-fold lower tumor burden than macroscopic disease, the TCD50 was about 6 Gy lower, also in agreement with our comparison between the effective doses for ART and SRT. Perhaps the strongest direct evidence in support of more aggressive local treatment is the findings from both randomized adjuvant trials that local failure rates were significantly greater than the metastatic failure rates (local failure was four times greater than the metastatic failure in the EORTC trial and 30% greater in the SWOG trial). Furthermore, improvement in local control in the treatment arms resulted in a decrease in metastatic rates compared with the rates in the observation arm (13). A similar association of the reduction of distant metastatic disease with improved local control was also described after RRT for prostate cancer (14). The reluctance to deliver greater doses in the postoperative setting has historically been because of the belief that the toxicity would be excessive and represent an unwarranted risk because no evidence for the effectiveness of greater doses existed. Similarly, until the demonstration of improved outcomes with dose escalation for localized prostate cancer, the same beliefs prevailed and the RRT doses until the late 1980s were nearly universally well below 70 Gy. In Table 3, we have summarized the toxicity data of RT in the postoperative setting (1, 26). For adjuvant or salvage
349
doses of 60 Gy, the rate of high-grade genitourinary and gastrointestinal toxicities was quite low (<4%). In the large multi-institutional analysis of postoperative RT, the radiation dose (which ranged mostly from 60 to 70 Gy) was not predictive of either late genitourinary or gastrointestinal toxicity on multivariate analysis (26). We propose that greater RT doses in the postoperative setting, in the range of 70–74 Gy, are practical and likely to be safe, if good techniques are used. A recent study using fiducial-based image-guided RT to the prostate bed demonstrated the feasibility of such a technique (27). The use of fiducial markers or an alternate means of image guidance would be recommended for any such dose-escalation study to ensure that toxicity is minimized. Finally, we need to briefly consider the potential role that androgen suppression might have in the context of postoperative RT. A recent retrospective study showed a significant improvement in bRFS when 4 months of total androgen suppression was added to RT compared with RT alone for SRT or ART (28). Currently, one randomized trial, the Radiation Therapy Oncology Group 96-01 trial, is comparing 64.8 Gy postoperative RT with and without 2 years of bicalutamide for patients with a detectable PSA level and either Stage pT3 or pT2 with positive margins. This trial will likely be interpreted, however, in terms of the effect of long-term hormonal therapy instead of the enhancement of local RT. The hypothesis that hormonal therapy could itself be a substitute for dose escalation remains to be studied. With an expected proportional gain in the bRFS rate of 3% per incremental Gray, a randomized trial testing a SRT dose of 64 vs. 70 Gy or an adjuvant trial testing 60 vs. 66 Gy would require a total of approximately 250 patients and would be expected to detect a 20% difference in the 5-year bRFS rate between the two treatment arms (assuming a power of 0.8, a two-sided a of 0.05, 4 years of accrual, and 3 years of follow-up). Should such trials confirm our hypothesis, then even further dose escalation would be warranted. CONCLUSIONS We suggest that for prostate cancer: (1) the dose–response curve for SRT is very similar to that of RRT; (2) a clinically significant gain in bRFS would be achieved with doses
Table 3. Late toxicity after adjuvant or salvage radiotherapy reported as 5-year rates on RTOG grade scale Series
Median dose (Gy)
Patients (n)
GU (%)
GI (%)
Toxicity grade
10 1 0.2 4.2* 0*
4 0.4 0.3
2 3 4 3 4
Retrospective pooled data on adjuvant and salvage RT (26)
64 (50–78)
959
Randomized adjuvant RT trial, EORTC (1)
60
503
Abbreviations: RTOG = Radiation Therapy Oncology Group; GU = genitourinary; GI = gastrointestinal; RT = radiotherapy; EORTC = European Organization for Research and Treatment of Cancer. * Reported GU and GI rates were combined. Data in parentheses are ranges.
350
I. J. Radiation Oncology d Biology d Physics
greater than currently recommended for ART or SRT; (3) the dose required for ART is about 6 Gy lower than that for SRT
Volume 71, Number 2, 2008
for a similar level of bRFS; and (4) a randomized dose-escalation trial for SRT or ART would be feasible and safe.
REFERENCES 1. Bolla M, van Poppel H, Collette L, et al. Postoperative radiotherapy after radical prostatectomy: A randomized controlled trial (EORTC trial 22911). Lancet 2005;366:572–578. 2. Thompson IM, Tangen CM, Paradelo J, et al. Adjuvant radiotherapy for pathologically advanced prostate cancer: A randomized trial. JAMA 2006;296:2329–2335. 3. Stephenson AJ, Scardino PT, Kattan MW, et al. Predicting the outcome of salvage radiation therapy for recurrent prostate cancer after radical prostatectomy. J Clin Oncol 2007;25: 2035–2041. 4. Hayes SB, Pollack A. Parameters for treatment decisions for salvage radiation therapy. J Clin Oncol 2005;23:8204–8211. 5. Cox JD, Gallagher MJ, Hammond EH, et al. for the ASTRO Consensus Panel. Consensus statement on radiation therapy of prostate cancer: Guidelines for prostate re-biopsy after radiation and for radiation therapy with rising PSA levels after radical prostatectomy. J Clin Oncol 1999;17:1155–1163. 6. 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: 1097–1105. 7. 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;294:1233–1239. 8. Eade TN, Hanlon AL, Horwitz EM, et al. What dose of external beam radiation is high enough for prostate cancer? Int J Radiat Oncol Biol Phys 2007;68:682–689. 9. Zelefsky MJ, Leibel SA, Gaudin PB, et al. Dose-escalation with three-dimensional conformal radiation therapy affects the outcome in prostate cancer. Int J Radiat Oncol Biol Phys 1998; 41:491–500. 10. Valicenti RK, Gomella LG, Ismail M, et al. Effect of higher radiation dose on biochemical control after radical prostatectomy for pT3N0 prostate cancer. Int J Radiat Oncol Biol Phys 1998; 42:501–506. 11. Anscher MS, Clough R, Dodge R. Radiotherapy for a rising PSA after radical prostatectomy: The first 10 years. Int J Radiat Oncol Biol Phys 2000;48:369–375. 12. King CR, Spiotto MT. Improved outcomes with higher doses for salvage radiotherapy after prostatectomy. Int J Radiat Oncol Biol Phys 2008;71:23–27. 13. Swanson GP, Hussey MA, Tangen CM, et al. Predominant treatment failure in post-prostatectomy patients is local: Analysis of patterns of treatment failure in SWOG 8794. J Clin Oncol 2007;25:2225–2229. 14. Coen JJ, Zietman AL, Thakral H, et al. Radical radiation for localized prostate cancer: Local persistence of disease results in a late wave of metastases. J Clin Oncol 2002;20:3199–3205.
15. Cox JD, Grignon D, Kaplan R, et al. Consensus statement: Guidelines for PSA following radiation therapy. Int J Radiat Oncol Biol Phys 1997;37:1035–1041. 16. Fowler JF, Ritter MA, Chappell RJ, et al. What hypofractionated protocols should be tested for prostate cancer? Int J Radiat Oncol Biol Phys 2003;56:1093–1104. 17. Catton C, Gospodarowicz M, Mui J, et al. Clinical and biochemical outcome of conventional dose radiotherapy for localized prostate cancer. Can J Urol 2002;9:1444–1452. 18. Okunieff P, Morgan D, Niemierko A, et al. Radiation doseresponse of human tumors. Int J Radiat Oncol Biol Phys 1995; 32:1227–1237. 19. Catton C, Gospodarowicz M, Warde P, et al. Adjuvant and salvage radiation therapy after radical prostatectomy for adenocarcinoma of the prostate. Radiother Oncol 2001;59:51–60. 20. Tsien C, Griffith KA, Sandler HA, et al. Long-term results of three-dimensional conformal adjuvant and salvage radiotherapy after radical prostatectomy. Urology 2003;62:93–98. 21. Taylor N, Kelly JF, Kuban DA, et al. Adjuvant and salvage radiotherapy after radical prostatectomy for prostate cancer. Int J Radiat Oncol Biol Phys 2003;56:755–763. 22. Chawla AK, Thakral HK, Zietman AL, et al. Salvage radiotherapy after radical prostatectomy for prostate adenocarcinoma: Analysis of efficacy and prognostic factors. Urology 2002;59: 726–731. 23. Buskirk SJ, Pisansky TM, Schild SE, et al. Salvage radiotherapy for isolated prostate specific antigen increase after radical prostatectomy: Evaluation of prognostic factors and creation of a prognostic scoring system. J Urol 2006;176:985–990. 24. Kalapurakal JA, Huang C-F, Neriamparampil MM, et al. Biochemical disease-free survival following adjuvant and salvage irradiation after radical prostatectomy. Int J Radiat Oncol Biol Phys 2002;54:1047–1054. 25. Stamey T, Kabalin J, McNeal J, et al. Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. II. Radical prostatectomy treated patients. J Urol 1998; 141:1076–1083. 26. Feng M, Hanlon AL, Pisansky TM, et al. Predictive factors for late genitourinary and gastrointestinal toxicity in patients with prostate cancer treated with adjuvant or salvage radiotherapy. Int J Radiat Oncol Biol Phys 2007;68:1417–1423. 27. Schiffner DC, Gottschalk AR, Lometti M, et al. Daily electronic portal imaging of implanted gold seed fiducials in patients undergoing radiotherapy after radical prostatectomy. Int J Radiat Oncol Biol Phys 2007;67:610–619. 28. King CR, Presti JC Jr., Gill H, et al. Radiotherapy after radical prostatectomy: Does transient androgen suppression improve outcomes? Int J Radiat Oncol Biol Phys 2004;59:341–347.
doi:10.1016/j.ijrobp.2007.10.008
CLINICAL INVESTIGATION
Prostate
RADIOTHERAPY AFTER PROSTATECTOMY: IS THE EVIDENCE FOR DOSE ESCALATION OUT THERE? CHRISTOPHER R. KING, PH.D., M.D., AND DANIEL S. KAPP, PH.D., M.D. Department of Radiation Oncology, Stanford University School of Medicine, Stanford, CA Purpose: To study the effective doses of radiotherapy (RT) after prostatectomy in search for evidence of a dose– response. Methods and Materials: Original and available data from published studies of adjuvant and salvage RT after prostatectomy were analyzed in the context of biochemical tumor control probability (TCP) dose–response curves. Comparisons were made with dose-escalation studies of radical RT for localized disease. Arguments based on a microscopic vs. macroscopic disease dose-response relationships were used to interpret the clinical data. Results: The tumor control rates after salvage RT were consistent with the TCP dose–response curve of radical RT, suggesting the presence of macroscopic-equivalent disease among salvage patients. For radical RT, the dose to achieve 50% biochemical tumor control was 65.9 Gy (95% confidence interval [CI], 64.8–66.8) and the Slope50 was 2.6%/Gy (95% CI, 2.3–3.0). For salvage RT, the corresponding values were 66.8 Gy (95% CI, 65.1–68.4) and 3.8%/Gy (95% CI, 2.5–7.6). For a comparable TCP, the dose for adjuvant RT was 6 Gy lower, consistent with one-tenth the burden of local disease. The present doses for adjuvant or salvage RT in the range of 60–70 Gy appear to be still on the steep part of the TCP dose–response curve. Conclusions: The effective doses and dose–response relation observed with RT after prostatectomy are consistent with the presence of macroscopic-equivalent disease for salvage patients and about a tenth of the residual disease for adjuvant patients. Greater doses would potentially achieve significantly greater disease-free control rates. A randomized trial with 250 patients comparing 64 vs. 70 Gy for salvage RT or 60 vs. 66 Gy for adjuvant RT would be capable of addressing this issue. Ó 2008 Elsevier Inc. Prostate cancer, Postoperative radiotherapy, Dose response, Dose escalation.
INTRODUCTION
highest radiation dose that can be delivered with acceptable morbidity and suggested a minimum of 64 Gy at conventional dose fractionation (5). Although ample evidence exists in support of dose escalation to $78 Gy for radical RT (RRT) for localized prostate cancer (6–9), very few studies have adequately examined the dose–response relationship in the postoperative setting owing to the relatively narrow range of doses commonly used. Of the multitude of retrospective studies on postoperative RT, only two studies, one for SRT (10) and one for ART (11), each with <90 patients, showed that doses >64.8 Gy provide improved biochemical control compared with lower doses. A recent analysis of patients treated at our institution showed that SRT to 70 Gy achieved superior 5-year bRFS compared with 60 Gy (58% vs. 25%, p = 0.012) and that the high dose was an independent factor for bRFS on multivariate analysis (12). It is commonly presumed, albeit unproven, that the tumor burden in the postoperative setting is microscopic and, therefore, that lower effective doses are necessary compared with the dose needed
The use of adjuvant radiotherapy (ART) (i.e., in the immediate postoperative setting) or salvage RT (SRT) (i.e., after a demonstrated biochemical recurrence) offers potentially curative treatment after unsuccessful radical prostatectomy. Two randomized trials have demonstrated improved biochemical relapse-free survival (bRFS) for ART after prostatectomy for patients with high-risk pathologic features, defined as pT3 or positive surgical margins (1, 2). The role of SRT for biochemical relapse after prostatectomy has also been demonstrated in numerous retrospective studies, most recently in a multi-institutional pooled data analysis (3) and a comprehensive review of published studies (4). Together, these reports form the basis for optimal patient selection and therapeutic guidelines. In the two ART trials, the prostate bed dose was 60 Gy (1) and 60–64 Gy (2). For SRT, the American Society for Therapeutic Radiology and Oncology consensus guidelines have recommended the
Conflict of interest: none. Received Aug 30, 2007, and in revised form Oct 11, 2007. Accepted for publication Oct 12, 2007.
Reprint requests to: Christopher R. King, Ph.D., M.D., Department of Radiation Oncology, Stanford University School of Medicine, Stanford Cancer Center, 875 Blake Wilbur Dr., Stanford, CA 94305. Tel: (650) 736-0698; Fax: (650) 725-8231; E-mail: [email protected] 346
Dose escalation of RT after prostatectomy d C. R. KING AND D. S. KAPP
in the definitive setting. An analysis of the patterns of failure from the Southwestern Oncology Group (SWOG) ART randomized trial demonstrated that treatment failure is predominantly local, and, therefore, an improvement in local therapy will result in improved outcomes (13). The present study compared the data for the dose–response relationship for postoperative treatment and definitive treatment to evaluate the hypothesis that salvage and adjuvant doses should be escalated to improve biochemical diseasefree survival for these patients. METHODS AND MATERIALS We first established the dose–response relationship from the RRT data to serve as a point of reference. We constructed the dose– response curve for RRT from studies of RT alone (i.e., excluding any use of androgen suppression) that provided the 5-year bRFS data. The inclusion criteria were randomized trials or modern series with >200 patients. All studies had consistently used the American Society for Therapeutic Radiology and Oncology definition of biochemical failure of three consecutive increases greater than the nadir with the failure date midway between the first increase and the previous value (14). Only series providing the results for patients predominantly at ‘‘intermediate risk’’ were used (defined as possessing one or more of the following features: initial prostate-specific antigen [PSA] level of 10–20 ng/mL, biopsy Gleason score of 7, or clinical Stage T2b-T2c). A similar approach was used by Fowler et al. (15) to remove the potential confounding effects of androgen suppression with RT and to produce a reference RT dose–response curve for localized prostate cancer. The series used to construct the current dose–response curve are listed in Table 1 and included The University of Texas M. D. Anderson dose-escalation trial (6), Massachusetts General Hospital dose-escalation trial (7), prospective dose-escalation series from Fox Chase Cancer Center (8) and Memorial Sloan-Kettering Cancer Center (9), and the series from Princess Margaret Hospital reporting on outcomes with low-dose RT (16). We used the standard tumor control probability (TCP) equation to fit these data by the least-squares method. The form of the TCP relationship is given by TCP ¼ expð½d TCD50 =kÞ=½1 þ expð½d TCD50 =kÞ where d is total dose, TCD50 is the dose to achieve 50% biochemical tumor control, and k is a fitting parameter related to the slope at the
347
TCD50 dose point (17). The proportional gain in TCP per additional Gray within the steep part of the TCP curve is given by the parameter Slope50, defined by Slope50 = 25/k (in percentage per Gray). A comprehensive published data search retrieved 10 series on SRT (3, 11, 12, 18–23). Only series that provided 5-year biochemical no evidence of disease results and whose median pre-RT PSA level was <2 ng/mL were included. The details of the SRT series, as well as the two randomized trials of ART, are listed in Table 2. The definition of biochemical failure for postoperative RT is essentially the same as after prostatectomy. In the studies reported, a persistently detectable or increasing PSA >0.2 ng/mL was considered a biochemical failure. Although for SRT all patients actually have disease (measurable by PSA recurrence), of necessity a proportion of ART patients will be without disease. The randomized adjuvant trials provided this critical information from their control (observation) arm, in which the 5-year bRFS rate for the observation arm was 42% in the SWOG trial (2) and 53% in the European Organization for Research and Treatment of Cancer trial (1). Thus, the bRFS for ART must be corrected for the patients without disease to calculate the effectiveness of postoperative RT only among those patients who actually had disease. The actual adjuvant bRFS can then given by the difference in bRFS between the RT and observation arms, divided by the proportion of patients with disease: bRFS (adjuvant) = [bRFS with RT bRFS observation]/[100 bRFS observation]. These two trials differed in their patient selection criteria in the following significant ways. In the EORTC trial, all patients had to have a postoperative PSA level that was <0.4 ng/mL and 10.7% had a postoperative PSA level >0.2 ng/mL. In the SWOG trial, even patients with greater postoperative PSA levels were eligible, and 33.8% had a postoperative PSA level >0.2 ng/mL. Consequently, the median postoperative PSA level was greater in the SWOG trial than in the EORTC trial, and, hence, the absolute bRFS rates in the observation and treatment arms were slightly lower than those in the EORTC trial. The retrospective adjuvant series found were not usable in this context, because they did not estimate the baseline proportion of patients without disease in their patient population.
RESULTS The TCP curve fitted to the RRT series is shown in Fig. 1 (r = 0.990, p <0.0001). The derived value for TCD50 is 65.9 Gy (95% confidence interval [CI], 64.8–66.8) and for the Slope50, it is 2.6%/Gy (95% CI, 2.3–3.0). The SRT series
Table 1. Series with external beam radiotherapy alone for radical radiotherapy of intermediate-risk prostate cancer Series Randomized dose-escalation trial, MDA Randomized dose-escalation trial, MGH Retrospective study, PMH Retrospective study of dose response, FC
Retrospective study of dose response, MSK
Median dose (Gy) 70 78 70.2 79.2 65 (52–67) 69 (<70) 72 (70–74.9) 76 (75–79.9) 82 (>80) 67.5 (64.8–70.2) 78.3 (75.6–81.0)
5-y bNED (%)
Patients (n)
Study
62 75 61 80 45 60 68 76 84 54 79
150 151 197 195 706 43 552 568 367 116 94
Pollack et al. (6) Zietman et al. (7) Catton et al. (17) Eade et al. (8)
Zelefsky et al. (9)
Abbreviations: bNED = biochemical no evidence of disease; MDA = M.D. Anderson Cancer Center; MGH = Massachusetts General Hospital; PMH = Princess Margaret Hospital; FC = Fox Chase Cancer Center; MSK = Memorial Sloan-Kettering Cancer Center. Data in parentheses are ranges.
I. J. Radiation Oncology d Biology d Physics
348
Volume 71, Number 2, 2008
Table 2. Series with external beam radiotherapy for salvage and adjuvant radiotherapy after radical prostatectomy Series Salvage RT retrospective series Catton et al. (19) Tsien et al. (20) Taylor et al. (21) Chawla et al. (22) Buskirk et al. (23) Kalapurakal et al. (24) Anscher et al. (11) Stephenson et al. (3) King and Spiotto (12) Adjuvant RT randomized trials EORTC-22911; Bolla et al. (1) SWOG-8794; Thompson et al. (2)
Median dose (Gy)
5-y bNED (%)
Median pre-RT PSA level (ng/mL)
Patients (n)
60 (54–65) 65 (60–75) 70 (60–78) 64.8 (60.4–64.8) 64.8 (54–72.4) 65 (60–70) 66 (54.6–70) 64.8 (63–66) 60 (60–64) 70 (67.8–70)
29 35 66 35 46 57 50 37 25 58
#2.0 1.2 0.8 1.3 0.7 0.5 1.4 1.1 0.9 0.45
28 57 71 54 368 41 89 1,540 38 84
45* 47*
— —
1,005 425
60 60–64y
Abbreviations: RT = radiotherapy; PSA = prostate-specific antigen; other abbreviations as in Table 1. Data in parentheses are ranges. * Corrected bNED for proportion of patients without disease at baseline as described in text (i.e., [bNED with RT bNED observation]/ [100 bNED observation]). y Median dose not reported.
are shown in Fig. 1 and were remarkably close to the RRT dose–response curve. The TCP curve fit for the SRT series (r = 0.871; p = 0.0004) yielded a value of 66.8 Gy (95% CI, 65.1–68.4) for TCD50 and 3.8%/Gy (95% CI, 2.5–7.6) for the Slope50. Because the 95% CIs overlapped, the difference between the RRT- and SRT-fitted TCP curves was not significant. The ART points from the two randomized trials were also plotted in Fig. 1 and were separated by a lower dose of 6 Gy from the SRT curve.
roughly one-tenth the disease burden of SRT patients. This estimate of disease burden differential between the ART and SRT patients is consistent with the observation that, for the same bRFS, ART required about 6-Gy lower doses than did the SRT patients (Fig. 1). If one considers that a dose of 2 Gy yields a surviving fraction of about 0.5, three fractions (i.e., 6 Gy total) will correspond to a surviving fraction of 0.12 (i.e., about one-tenth), roughly the same estimated
100
DISCUSSION 80
5-year % bNED
By examining the dose–response relationship of SRT, we have shown the interesting result that it is essentially the same as the dose–response relationship for RRT for localized prostate cancer. The conclusion is that SRT patients harbor disease that effectively follows nearly the same macroscopic disease dose–response curve as that for RRT patients. This is not altogether surprising when one considers that SRT patients have clinically measurable disease by PSA that is of the same order of magnitude as for patients with localized disease. The pre-RT PSA level for the salvage patients among the studies used was 1 ng/mL (Table 2). A 1-cm3 tumor contains 109 cells and produces a PSA level of about 3.5 ng/mL (24); therefore, a tumor producing a PSA level of 1 ng/mL will still contain about 108–109 cells and should be considered macroscopic. The TCP curve for the salvage patients is slightly steeper than that for radical patients (i.e., the Slope50 is slightly greater). This observation is consistent with the fact that salvage patients, on the basis of their pre-RT PSA level spanning a very narrow range compared with that of radical patients, form a more homogenous group. In contrast, the ART patients in the two randomized trials were defined as having an undetectable postoperative PSA level if it was <0.2 ng/mL. Therefore, ART patients possess
TCP fit to Radical RT TCP fit to Salvage RT SRT ART (EORTC)
60
ART (SWOG) RRT (MSK)
40
RRT (MDA) RRT (PMH)
20
RRT (FC) RRT (MGH)
50
60
70
80
90
Total Dose (Gy)
Fig. 1. Dose–response graph plotting 5-year biochemical relapsefree survival vs. dose. Individual radical radiotherapy (RRT) series identified by separate symbol (see also Table 1). Dashed line derived from fitting tumor control probability (TCP) relationship only with RRT series. Dotted line indicates TCP fit to salvage RT (SRT) series only. bNED = biochemically no evidence of disease; ART = adjuvant RT; EORTC = European Organization for Research and Treatment of Cancer; SWOG = Southwestern Oncology Group; MSK = Memorial Sloan-Kettering Cancer Center; MDA = M.D. Anderson Cancer Center; PMH = Princess Margaret Hospital; MGH = Massachusetts General Hospital; FC = For Chase Cancer Center.
Dose escalation of RT after prostatectomy d C. R. KING AND D. S. KAPP
fractional difference in disease burden between the salvage and ART patients on the basis of the pre-RT PSA level. However, in each trial, a significant proportion of patients had a pre-RT PSA level >0.2 ng/mL; thus, the practical distinction between the ART and SRT patients was somewhat blurred. The availability of today’s ultrasensitive PSA assays with a threshold of 0.02 ng/mL will allow for the selection of ART patients who possess an even lower burden of disease. Our fitted values of 65.9 Gy for TCD50 and 2.6%/Gy for Slope50 for RRT is in agreement with those (67.5 Gy and 3.2%/Gy, respectively) from M.D. Anderson’s analysis of intermediate-risk patients (25). A radiation dose–response study by Okunieff et al. (17) demonstrated that for most human cancers, the TCD50 for macroscopic disease was about 52 Gy. In the spectrum of human tumors, prostate cancer is considered to be somewhat radioresistant. In agreement with that, we have found the TCD50 for prostate cancer to be 66 Gy. The aforementioned study also showed that for microscopic disease with a 10-fold lower tumor burden than macroscopic disease, the TCD50 was about 6 Gy lower, also in agreement with our comparison between the effective doses for ART and SRT. Perhaps the strongest direct evidence in support of more aggressive local treatment is the findings from both randomized adjuvant trials that local failure rates were significantly greater than the metastatic failure rates (local failure was four times greater than the metastatic failure in the EORTC trial and 30% greater in the SWOG trial). Furthermore, improvement in local control in the treatment arms resulted in a decrease in metastatic rates compared with the rates in the observation arm (13). A similar association of the reduction of distant metastatic disease with improved local control was also described after RRT for prostate cancer (14). The reluctance to deliver greater doses in the postoperative setting has historically been because of the belief that the toxicity would be excessive and represent an unwarranted risk because no evidence for the effectiveness of greater doses existed. Similarly, until the demonstration of improved outcomes with dose escalation for localized prostate cancer, the same beliefs prevailed and the RRT doses until the late 1980s were nearly universally well below 70 Gy. In Table 3, we have summarized the toxicity data of RT in the postoperative setting (1, 26). For adjuvant or salvage
349
doses of 60 Gy, the rate of high-grade genitourinary and gastrointestinal toxicities was quite low (<4%). In the large multi-institutional analysis of postoperative RT, the radiation dose (which ranged mostly from 60 to 70 Gy) was not predictive of either late genitourinary or gastrointestinal toxicity on multivariate analysis (26). We propose that greater RT doses in the postoperative setting, in the range of 70–74 Gy, are practical and likely to be safe, if good techniques are used. A recent study using fiducial-based image-guided RT to the prostate bed demonstrated the feasibility of such a technique (27). The use of fiducial markers or an alternate means of image guidance would be recommended for any such dose-escalation study to ensure that toxicity is minimized. Finally, we need to briefly consider the potential role that androgen suppression might have in the context of postoperative RT. A recent retrospective study showed a significant improvement in bRFS when 4 months of total androgen suppression was added to RT compared with RT alone for SRT or ART (28). Currently, one randomized trial, the Radiation Therapy Oncology Group 96-01 trial, is comparing 64.8 Gy postoperative RT with and without 2 years of bicalutamide for patients with a detectable PSA level and either Stage pT3 or pT2 with positive margins. This trial will likely be interpreted, however, in terms of the effect of long-term hormonal therapy instead of the enhancement of local RT. The hypothesis that hormonal therapy could itself be a substitute for dose escalation remains to be studied. With an expected proportional gain in the bRFS rate of 3% per incremental Gray, a randomized trial testing a SRT dose of 64 vs. 70 Gy or an adjuvant trial testing 60 vs. 66 Gy would require a total of approximately 250 patients and would be expected to detect a 20% difference in the 5-year bRFS rate between the two treatment arms (assuming a power of 0.8, a two-sided a of 0.05, 4 years of accrual, and 3 years of follow-up). Should such trials confirm our hypothesis, then even further dose escalation would be warranted. CONCLUSIONS We suggest that for prostate cancer: (1) the dose–response curve for SRT is very similar to that of RRT; (2) a clinically significant gain in bRFS would be achieved with doses
Table 3. Late toxicity after adjuvant or salvage radiotherapy reported as 5-year rates on RTOG grade scale Series
Median dose (Gy)
Patients (n)
GU (%)
GI (%)
Toxicity grade
10 1 0.2 4.2* 0*
4 0.4 0.3
2 3 4 3 4
Retrospective pooled data on adjuvant and salvage RT (26)
64 (50–78)
959
Randomized adjuvant RT trial, EORTC (1)
60
503
Abbreviations: RTOG = Radiation Therapy Oncology Group; GU = genitourinary; GI = gastrointestinal; RT = radiotherapy; EORTC = European Organization for Research and Treatment of Cancer. * Reported GU and GI rates were combined. Data in parentheses are ranges.
350
I. J. Radiation Oncology d Biology d Physics
greater than currently recommended for ART or SRT; (3) the dose required for ART is about 6 Gy lower than that for SRT
Volume 71, Number 2, 2008
for a similar level of bRFS; and (4) a randomized dose-escalation trial for SRT or ART would be feasible and safe.
REFERENCES 1. Bolla M, van Poppel H, Collette L, et al. Postoperative radiotherapy after radical prostatectomy: A randomized controlled trial (EORTC trial 22911). Lancet 2005;366:572–578. 2. Thompson IM, Tangen CM, Paradelo J, et al. Adjuvant radiotherapy for pathologically advanced prostate cancer: A randomized trial. JAMA 2006;296:2329–2335. 3. Stephenson AJ, Scardino PT, Kattan MW, et al. Predicting the outcome of salvage radiation therapy for recurrent prostate cancer after radical prostatectomy. J Clin Oncol 2007;25: 2035–2041. 4. Hayes SB, Pollack A. Parameters for treatment decisions for salvage radiation therapy. J Clin Oncol 2005;23:8204–8211. 5. Cox JD, Gallagher MJ, Hammond EH, et al. for the ASTRO Consensus Panel. Consensus statement on radiation therapy of prostate cancer: Guidelines for prostate re-biopsy after radiation and for radiation therapy with rising PSA levels after radical prostatectomy. J Clin Oncol 1999;17:1155–1163. 6. 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: 1097–1105. 7. 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;294:1233–1239. 8. Eade TN, Hanlon AL, Horwitz EM, et al. What dose of external beam radiation is high enough for prostate cancer? Int J Radiat Oncol Biol Phys 2007;68:682–689. 9. Zelefsky MJ, Leibel SA, Gaudin PB, et al. Dose-escalation with three-dimensional conformal radiation therapy affects the outcome in prostate cancer. Int J Radiat Oncol Biol Phys 1998; 41:491–500. 10. Valicenti RK, Gomella LG, Ismail M, et al. Effect of higher radiation dose on biochemical control after radical prostatectomy for pT3N0 prostate cancer. Int J Radiat Oncol Biol Phys 1998; 42:501–506. 11. Anscher MS, Clough R, Dodge R. Radiotherapy for a rising PSA after radical prostatectomy: The first 10 years. Int J Radiat Oncol Biol Phys 2000;48:369–375. 12. King CR, Spiotto MT. Improved outcomes with higher doses for salvage radiotherapy after prostatectomy. Int J Radiat Oncol Biol Phys 2008;71:23–27. 13. Swanson GP, Hussey MA, Tangen CM, et al. Predominant treatment failure in post-prostatectomy patients is local: Analysis of patterns of treatment failure in SWOG 8794. J Clin Oncol 2007;25:2225–2229. 14. Coen JJ, Zietman AL, Thakral H, et al. Radical radiation for localized prostate cancer: Local persistence of disease results in a late wave of metastases. J Clin Oncol 2002;20:3199–3205.
15. Cox JD, Grignon D, Kaplan R, et al. Consensus statement: Guidelines for PSA following radiation therapy. Int J Radiat Oncol Biol Phys 1997;37:1035–1041. 16. Fowler JF, Ritter MA, Chappell RJ, et al. What hypofractionated protocols should be tested for prostate cancer? Int J Radiat Oncol Biol Phys 2003;56:1093–1104. 17. Catton C, Gospodarowicz M, Mui J, et al. Clinical and biochemical outcome of conventional dose radiotherapy for localized prostate cancer. Can J Urol 2002;9:1444–1452. 18. Okunieff P, Morgan D, Niemierko A, et al. Radiation doseresponse of human tumors. Int J Radiat Oncol Biol Phys 1995; 32:1227–1237. 19. Catton C, Gospodarowicz M, Warde P, et al. Adjuvant and salvage radiation therapy after radical prostatectomy for adenocarcinoma of the prostate. Radiother Oncol 2001;59:51–60. 20. Tsien C, Griffith KA, Sandler HA, et al. Long-term results of three-dimensional conformal adjuvant and salvage radiotherapy after radical prostatectomy. Urology 2003;62:93–98. 21. Taylor N, Kelly JF, Kuban DA, et al. Adjuvant and salvage radiotherapy after radical prostatectomy for prostate cancer. Int J Radiat Oncol Biol Phys 2003;56:755–763. 22. Chawla AK, Thakral HK, Zietman AL, et al. Salvage radiotherapy after radical prostatectomy for prostate adenocarcinoma: Analysis of efficacy and prognostic factors. Urology 2002;59: 726–731. 23. Buskirk SJ, Pisansky TM, Schild SE, et al. Salvage radiotherapy for isolated prostate specific antigen increase after radical prostatectomy: Evaluation of prognostic factors and creation of a prognostic scoring system. J Urol 2006;176:985–990. 24. Kalapurakal JA, Huang C-F, Neriamparampil MM, et al. Biochemical disease-free survival following adjuvant and salvage irradiation after radical prostatectomy. Int J Radiat Oncol Biol Phys 2002;54:1047–1054. 25. Stamey T, Kabalin J, McNeal J, et al. Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. II. Radical prostatectomy treated patients. J Urol 1998; 141:1076–1083. 26. Feng M, Hanlon AL, Pisansky TM, et al. Predictive factors for late genitourinary and gastrointestinal toxicity in patients with prostate cancer treated with adjuvant or salvage radiotherapy. Int J Radiat Oncol Biol Phys 2007;68:1417–1423. 27. Schiffner DC, Gottschalk AR, Lometti M, et al. Daily electronic portal imaging of implanted gold seed fiducials in patients undergoing radiotherapy after radical prostatectomy. Int J Radiat Oncol Biol Phys 2007;67:610–619. 28. King CR, Presti JC Jr., Gill H, et al. Radiotherapy after radical prostatectomy: Does transient androgen suppression improve outcomes? Int J Radiat Oncol Biol Phys 2004;59:341–347.