- Email: [email protected]
Phase III radiation therapy oncology group (RTOG) trial 86-10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate
Int. J. Radiation Oncology Biol. Phys., Vol. 50, No. 5, pp. 1243–1252, 2001 Copyright © 2001 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/01/$–see front matter
PII S0360-3016(01)01579-6
CLINICAL INVESTIGATION
Prostate
PHASE III RADIATION THERAPY ONCOLOGY GROUP (RTOG) TRIAL 86-10 OF ANDROGEN DEPRIVATION ADJUVANT TO DEFINITIVE RADIOTHERAPY IN LOCALLY ADVANCED CARCINOMA OF THE PROSTATE
JOHN
MILJENKO V. PILEPICH, M.D.,* KATHRYN WINTER, M.S.,† MADHU J. JOHN, M.D.,‡ B. MESIC, M.D.,§ WILLIAM SAUSE, M.D.,㛳 PHILLIP RUBIN, M.D.,¶ COLLEEN LAWTON, M.D.,# MITCHELL MACHTAY, M.D.,** AND DAVID GRIGNON, M.D.††
*Department of Radiation Oncology, Ann Arbor Regional CCOP, Ann Arbor, MI ; †Radiation Therapy Oncology Group, Philadelphia, PA; ‡Department of Radiation Oncology, University of California, San Francisco, CA; §Radiation Oncology Center, Sacramento, CA; 㛳 Department of Radiation Oncology, LDS Hospital, Salt Lake City, UT; ¶Department of Radiation Oncology, University of Rochester, Rochester, NY; #Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI; **Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA; ††Department of Pathology, Wayne State University, Detroit, MI Purpose: To test the hypothesis that androgen ablation before and during radiotherapy for locally advanced carcinoma of the prostate may, by reducing tumor bulk and enhancing tumor cell kill, improve locoregional control and ultimately survival. Methods and Materials: The study was conducted from 1987 to 1991. Eligible patients were those with bulky tumors (T2–T4) with or without pelvic lymph node involvement and without evidence of distant metastases. They were randomized to receive goserelin, 3.6 mg every 4 weeks; and flutamide, 250 mg t.i.d. for 2 months before radiation therapy and during radiation therapy (Arm I), or radiation therapy alone (Arm II). Of 471 randomized patients, 456 were evaluable: 226 on Arm I and 230 on Arm II. Results: As of November 1999, the median follow-up has reached 6.7 years for all patients and 8.6 years for alive patients. At 8 years, androgen ablation has been associated with an improvement in local control (42% vs. 30%, p ⴝ 0.016), reduction in the incidence of distant metastases (34% vs. 45%, p ⴝ 0.04), disease-free survival (33% vs. 21%, p ⴝ 0.004), biochemical disease-free survival ⴝ PSA <1.5 (24% vs. 10%, p < 0.0001), and cause-specific mortality (23% vs. 31%, p ⴝ 0.05). However, subset analysis indicates that the beneficial effect of short-term androgen ablation appears preferentially in patients with Gleason score 2– 6. In that population, there is a highly significant improvement in all endpoints, including survival (70% vs. 52%, p ⴝ 0.015). In patients with Gleason 7–10 tumors, the regimen has not resulted in a significant enhancement in either locoregional control or survival. Conclusion: In patients with Gleason score 2– 6 carcinoma of the prostate, a short course of androgen ablation administered before and during radiotherapy has been associated with a highly significant improvement in local control, reduction in disease progression, and overall survival. © 2001 Elsevier Science Inc. Carcinoma of the prostate, Hormonal management, Radiotherapy.
INTRODUCTION
of enhancing locoregional control and eventually survival. It was postulated that locoregional control would be enhanced if the tumor bulk were reduced before initiation of radiotherapy. Other potential mechanisms of interaction include an enhancement of tumor cell death (apoptosis) and a reduction of tumor cell repopulation during the course of radiotherapy. A preliminary analysis undertaken in 1994 was reported in 1995 (1). The current report represents the long-term update with mature follow-up data.
Radiation Therapy Oncology Group (RTOG) has conducted a number of trials aimed at the evaluation of androgen deprivation adjuvant to definitive radiotherapy. RTOG 86-10 was the first Phase III trial testing the value of androgen deprivation in a neoadjuvant format. In a population of patients with locally advanced tumors, a short course of neoadjuvant hormonal ablation applied for 2 months before as well as during radiotherapy was tested as a means Reprint requests to: Miljenko V. Pilepich, Department of Radiation Oncology, P.O. Box 995, Ann Arbor, MI 48106. Tel: (734) 712-3595; Fax: (734) 712-5344; E-mail: [email protected] Supported by Public Health Service grants CA-21661 and CA-
32115 from the National Cancer Institute, National Institute of Health, Department of Health and Human Services. Received Jan 3, 2001, and in revised form Feb 28, 2001. Accepted for publication Mar 20, 2001. 1243
1244
I. J. Radiation Oncology
●
Biology
●
Physics
METHODS AND MATERIALS Study design Patients were stratified by stage and histologic grade and randomized to receive either radiotherapy alone or goserelin (Zoladex, AstraZeneca Pharmaceuticals, Wilmington, DE) and flutamide (Eulexin, Schering-Plough, Kenilworth, NJ) for 2 months before radiotherapy and during the radiotherapy course. Eligibility criteria Eligible patients were those who had bulky primary tumors, clinical stage T2–T4. In the context of the study, bulky primary lesions were defined as tumors with a product of palpable tumor dimensions (expressed in cm) of 25 or more. The determination was strictly clinical, based on digital rectal examination. The examiner would estimate the transverse and the sagittal dimension of the palpable tumor in centimeters and multiply the two numbers. If several nodularities were palpable, the examiner would measure each of them separately and then add the product of dimensions. Patients with positive lymph nodes were eligible if the involved nodes remained below the common iliac level. Pretreatment evaluation Mandatory studies included physical examination, Karnofsky performance status evaluation, assessment of potency (sexual history), and laboratory studies, including CBC, SGOT, and SGPT in patients assigned to Zoladex and flutamide. Serum acid phosphate determination, serum testosterone levels, and prostatic-specific antigen (PSA) were mandatory for all patients. It should be noted, however, that PSA determination was not widely available early in the study and was made mandatory only in 1990. Mandatory radiographic studies included chest X-ray and a bone scan. Lymph node evaluation was also mandatory and could be performed by either a lymphangiogram, computerized tomography of the pelvis and abdomen, or exploratory laparotomy. Radiotherapy Megavoltage equipment was required with photon energies higher than 1 MV. Any treatment techniques capable of producing the dose distribution specified by the protocol were acceptable, with the exception of the perineal boost. Customized blocks were used in most patients. Patients with no evidence of tumor spread to the pelvic lymphatics were to be treated electively to a target volume that included pelvic nodes to the level of L5–S1 interspace. In patients with evidence of a pelvic lymph node involvement, the regional lymphatics target volume was to include not only pelvic but also lower para-aortic nodes. The superior margin was to extend to L2–L3 interspace. The prostate boost target volume was to include the prostate, with margins sufficiently wide to encompass all of the tumor extensions into the surrounding tissues.
Volume 50, Number 5, 2001
The dose was prescribed on the central axis at the projected center of the target volumes. Regional lymphatics were to receive a total of 44 to 46 Gy, although doses up to 50 Gy were acceptable. The prostatic target volume was to receive a boost to a total prescribed dose of 65 to 70 Gy. The minimum target doses to the regional lymphatics and the prostatic target volume were 44 Gy and 65 Gy, respectively. The maximum target dose, defined as the greatest dose in target volume delivered to an area greater than 2 cm2, was 50 Gy for the regional lymphatics target volume and 72 Gy for the prostatic boost target volume. The daily doses were 1.8 –2 Gy given 4 –5 times a week. Drug therapy Goserelin (Zoladex) was administered s.c. every 4 weeks, starting 2 months before initiation of radiotherapy and continuing during radiotherapy course, the total overall treatment time being 112 days. Flutamide (Eulexin) was administered by mouth, 250 mg, 3 times per day over the same time period (112 days). Patient assessments After completion of treatment, the patients were to be seen every 3 months during the first year, every 4 months during the second and third years, every 6 months through the fifth year, and then annually for the remainder of the patient’s life. Follow-up evaluation included a clinical examination and a PSA determination. A central review of the radiation therapy information was performed for each case by the study chair after completion of treatment. Radiation therapy equipment used in treatment was calibrated by the Radiologic Physics Center at the University of Texas M. D. Anderson Hospital. Compliance with the drug administration was reviewed by RTOG headquarters staff and the study chair. Central review of the pathology material was undertaken to achieve uniformity of histologic grade (Gleason score) determination. End points and statistical analysis The primary end point of the study was locoregional control; secondary end points were disease-free survival (freedom from progression) and survival. For the purposes of the current analysis, local failure is defined as progression of palpable tumor at any time, persistence of a palpable tumor beyond the second year, and a positive biopsy 2 years or more after study entry. Regional metastasis is defined as clinical or radiographic evidence of involvement of the pelvic lymphatics by the tumor. Distant metastases are defined as clinical or radiographic evidence of disease beyond the pelvis. Survival was measured from the date of randomization to the date of death, or the most recent follow-up. Estimates of survival and progression-free survival were
Androgen deprivation and radiotherapy in carcinoma of the prostate
● M. PILEPICH et al.
1245
Table 1. All patients (n ⫽ 456)
Local failure Distant metastases NED bNED (w/PSA ⬍4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of Failures
5 years
8 years
p value
226 230 226 230 226 230 201 203 201 203 226 230 226 230
72 98 82 104 160 183 158 180 170 194 112 136 52 74
22% 35% 29% 39% 49% 34% 39% 20% 28% 10% 72% 68% 15% 20%
30% 42% 34% 45% 33% 21% 24% 10% 16% 3% 53% 44% 23% 31%
0.016 0.04 0.004 ⬍0.0001 ⬍0.0001 0.10 0.05
Follow-up for all patients
Median follow-up (Yr)
derived by the Kaplan-Meier method (2). Statistical comparisons for survival and progression-free survival were made by the log-rank statistic in the case of censored data or by the proportional hazards analysis to control for prognostic factors. Statistical comparisons for the cumulative incidence of local progression, distant metastases, and cause-specific failure were performed using Gray’s test (3).
All patients (n ⫽ 456)
Alive patients (n ⫽ 208)
6.7
8.6
RESULTS Patient characteristics A total of 471 patients were accessioned from April 15, 1987 through June 1, 1991. Of these, 15 were excluded for the following reasons: no follow-up (4), tumor too small (5), refused all treatment and follow-up (3), lung primary (1), bone metastases (1), and benign disease (1).
Fig. 1. Local progression, all patients.
1246
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 50, Number 5, 2001
Fig. 2. Disease-free survival, all patients.
There were 456 analyzable patients: 226 on the treatment arm (Arm I) and 230 on the control arm (Arm II). As of the end of 1999, the median follow-up reached 6.7 years for all patients and 8.6 years for alive patients. The pretreatment prognostic factors were well balanced be-
tween the treatment groups. The details have been published previously (1). Of the 225 Arm I patients with hormone compliance information, 211 (94%) completed goserelin treatment. Of these, 188 (84%) completed flutamide treatment as planned.
Fig. 3. bNED (PSA ⬍1.5), all patients.
Androgen deprivation and radiotherapy in carcinoma of the prostate
Table 2. Causes of death Cause of death Treated cancer Other* Unknown Total Other causes of death Second primary Heart related Pneumonia Pulmonary Mets. Other Not specified Total
RT ⫹ hormones
RT alone
Total
52 56 4 112
74 56 6 136
126 112 10 248
RT ⫹ hormones
RT alone
Total
11 21 2 3 1 10 8 56
7 21 3 3 3 11 8 56
18 42 5 6 4 21 16 112
* Breakdown of other causes of death.
The reasons for terminating flutamide in 23 patients included diarrhea in 11 patients,1 hot flashes in 3 patients, and miscellaneous other problems, including liver function abnormalities, rash, nausea, etc., in the remaining patients. Two patients refused goserelin, and one of these also re-
● M. PILEPICH et al.
1247
fused flutamide. A total of 186 patients completed both goserelin and flutamide as planned per the study guidelines. Table 1 summarizes the treatment results2 for the entire population and gives estimated 5- and 8-year rates for each endpoint. There is a highly significant improvement in local control and reduction in disease progression for Arm I (Figs. 1, 2, and 3). In addition, cause-specific failure (mortality due to prostate cancer) is reduced significantly for Arm I within the entire population (p ⫽ 0.05). Table 2 lists the causes of death for the entire population. Figures 4 and 5 show disease-specific mortality and absolute survival for all patients. Subset analyses indicate that the beneficial effect of hormonal management appears preferentially in patients with lower Gleason score (Gleason 2– 6). Table 3 summarizes the end-point analysis in Gleason 2– 6 patients. In this population, experimental treatment has been associated with a highly significant improvement in all end points, including survival (p ⫽ 0.015). Figures 6 –9 show the pattern of local progression (failure), no biochemical evidence of disease (bNED) (PSA ⬍1.5), and absolute survival for Gleason 2– 6 patients. Gleason 7 patients did not benefit significantly from the experimental treatment in terms of either local failure, sur-
Fig. 4. Disease-specific mortality, all patients.
1
Diarrhea was graded as Grade 2 in 7 patients and Grade 4 in 3 patients. Grading system is defined as follows: Grade 1, increase in frequency of bowel movements (no medication required), stools semisolid; Grade 2, watery stools, 3– 6/day; Grade 3, watery
stools, ⬎6/day; Grade 4, diarrhea requiring hospitalization. 2 Regional metastases were recorded in 6 patients. All of these patients had either local failure or distant metastases as the first site of failure and are accounted for elsewhere.
1248
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 50, Number 5, 2001
Fig. 5. Absolute survival, all patients.
vival, or cause-specific failure. The results in this population are summarized in Table 4. Table 5 summarizes the results in Gleason 8 –10 patients. These patients did not benefit significantly from the experimental treatment. DISCUSSION The mechanisms of interaction between radiation therapy and androgen deprivation in carcinoma of the prostate re-
main largely unknown. It is likely that the modes of interaction vary, depending on the timing (initiation) of the androgen deprivation and the duration of its application. If applied for a period of time before radiotherapy, androgen deprivation is likely to result in a significant reduction of tumor volume (a cytoreduction). If applied in close proximity to radiation therapy and/or during radiation therapy, androgen deprivation may interact with radiation on a cellular level and also affect cell kinetics.
Table 3. Centrally reviewed Gleason 2– 6 (n ⫽ 129)
Local failure Distant metastases NED bNED (w/PSA ⬍4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of failures
5 years
8 years
p value
71 58 71 58 71 58 64 52 64 52 71 58 71 58
17 29 11 21 39 44 39 44 47 47 24 34 1 13
14% 35% 12% 28% 67% 48% 61% 24% 45% 13% 84% 79% 0% 9%
21% 46% 13% 34% 50% 32% 47% – 30% – 70% 52% 2% 17%
0.005 0.006 0.004 ⬍0.0001 ⬍0.0001 0.015 0.0002
Follow-up for centrally reviewed Gleason 2– 6 patients
Median follow-up (Yr)
All patients (n ⫽ 129)
Alive patients (n ⫽ 71)
7.9
8.6
Androgen deprivation and radiotherapy in carcinoma of the prostate
● M. PILEPICH et al.
1249
Fig. 6. Local progression, Gleason 2– 6.
Cytoreduction refers to debulking of tumor before initiation of radiotherapy. It is well recognized that reduced tumor volume generally correlates with an increased statistical probability of tumor control with radiotherapy. Tumor size is a wellrecognized predictor of local control in carcinoma of the prostate. In RTOG experience (4), patients with bulky prima-
ries, with the product of tumor dimensions 25 or more, exhibited a particularly high recurrence rate, with more than 50% of the patients exhibiting evidence of local recurrence at 6 years after radiotherapy. This observation was the basis for defining selection criteria for several RTOG studies of hormonal cytoreduction in locally advanced carcinoma of the prostate.
Fig. 7. bNED (PSA ⬍1.5), Gleason 2– 6.
1250
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 50, Number 5, 2001
Fig. 8. Disease-specific mortality, Gleason 2– 6.
There is now considerable evidence in surgical and radiation oncology literature that several months of androgen deprivation predictably results in significant downsizing of prostate cancer volume in the majority of patients (5–11). However, clinical data addressing the potential value of hormonal cytoreduction before radiotherapy have been quite limited. Green et
al. (12) reported on a nonrandomized study suggesting that the local control could be significantly improved. The results were corroborated by the findings of Laverdiere et al. (13). There is some experimental evidence that neoadjuvant androgen suppression may enhance the effects of radiotherapy. Zietman et al. (14, 15) tested the concept in a Shionogi
Fig. 9. Absolute survival, Gleason 2– 6.
Androgen deprivation and radiotherapy in carcinoma of the prostate
● M. PILEPICH et al.
1251
Table 4. Centrally reviewed Gleason 7 (n ⫽ 176)
Local failure Distant metastases NED bNED (with PSA ⬍ 4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of failures
5 years
8 years
88 88 88 88 88 88 79 80 79 80 88 88 88 88
30 37 40 39 70 72 68 74 71 78 48 48 24 23
22% 38% 34% 36% 44% 28% 30% 18% 20% 9% 71% 67% 17% 16%
33% 43% 44% 44% 22% 19% 14% 9% 9% 3% 49% 46% 27% 26%
p value 0.26 0.88 0.40 ⬍0.016 ⬍0.0001 0.91 0.68
Follow-up for centrally reviewed Gleason 7 patients All patients (n ⫽ 176)
Alive patients (n ⫽ 80)
Median follow-up (Yr)
6.7
8.5
tumor in mice. Using tumor control as an end point, a dose–response curve was constructed. When orchiectomy was performed 24 – 48 h before radiation, the tumor control dose for 50% of tumors (TCD 50) fell from 89 Gy to 60.3 Gy. When the tumors were allowed to regress maximally before irradiation, the TCD 50 fell to 42.1 Gy. The degree to which this effect is related to cytoreduction vs. other mechanisms of action on a cellular level and possible changes in tumor bed and blood supply is not clear. RTOG 86-10 tested the potential value of androgen deprivation in conjunction with definitive radiotherapy in a
select population of patients with locally very advanced tumors. Moreover, approximately 40% of the patients had abnormal elevation of serum acid phosphates, reflecting the high potential for subclinical metastatic disease in a large proportion of patients. If observed as a whole, the study population benefited significantly from the experimental treatment. This refers to the incidence of local failures, disease progression, and reduction in the disease-specific mortality. The subset analyses reveal that the experimental treatment has proven to be much more effective in patients with
Table 5. Centrally reviewed Gleason 8 –10 (n ⫽ 124)
Local Failure Distant metastases NED bNED (with PSA ⬍4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of failures
5 years
8 years
57 67 57 67 57 67 49 57 49 57 57 67 57 67
23 28 29 40 46 56 43 53 44 57 37 46 25 36
36% 37% 45% 58% 32% 27% 24% 14% 14% 5% 54% 59% 33% 36%
40% 43% 47% 60% 22% 13% 12% 5% 10% — 38% 31% 44% 54%
p value 0.69 0.35 0.51 0.07 ⬍0.0001 0.98 0.36
Follow-up for centrally reviewed Gleason 8 –10 patients
Median follow-up (Yr)
All patients (n ⫽ 124)
Alive patients (n ⫽ 41)
5.3
8.7
1252
I. J. Radiation Oncology
●
Biology
●
Physics
low Gleason scores (Gleason 2– 6). In this population, the impact on local control, disease progression, and survival is profound and unequivocal. In patients with higher Gleason scores (Gleason 7–10), experimental treatment has not produced a significant improvement in local control and survival. Impact on disease progression as judged by bNED is of limited clinical significance. It is of interest that a parallel RTOG study (RTOG 85-31), which tested the potential value of long-term adjuvant deprivation after radiotherapy, demonstrated a beneficial effect of hormonal treatment in high Gleason score
Volume 50, Number 5, 2001
patients but not in the low Gleason score patients (16). This indicates that in Gleason 7–10 patients, alternative methods of hormonal management need to be used. These observations stress a need to consider different strategies for adjuvant hormonal management for different subsets of patients. In high Gleason score patients, 4 months of androgen deprivation before and during radiotherapy course are not adequate. In Gleason 2– 6 patients with locally advanced disease, such treatment exerts a major beneficial effect on all end points, including survival, and should be considered a standard of care.
REFERENCES 1. Pilepich MV, Krall JM, Al-Sarraf M, et al. Androgen deprivation with radiation therapy compared with radiation therapy alone for locally advanced prostatic carcinoma: A randomized comparative trial of the Radiation Therapy Oncology Group. Urology 1995;45:616 – 623. 2. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;52:457– 481. 3. Gray RL. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat 1988;16:1141– 1154. 4. Pilepich MV, Krall JM, Sause WT, et al. Prognostic factors in carcinoma of the prostate—Analysis of RTOG study 75-06. Int J Radiat Oncol Biol Phys 1987;13:339 –349. 5. Aus G, Abrahamsson P-A, Ahlgren F, et al. Hormonal treatment before radical prostatectomy: A 3-year follow-up. J Urol 1998;159:2013–2017. 6. Goldenberg SL, Klotz LH, Srigley J, et al. Randomized, prospective, controlled study comparing radical prostatectomy alone and neoadjuvant androgen withdrawal in treatment of localized prostate cancer. J Urol 1996;156:873– 877. 7. Soloway MS, Sharifi R, Wajsman Z, et al. Randomized prospective study comparing radical prostatectomy alone versus radical prostatectomy preceded by androgen blockade in clinical stage B2 (T2bNxM0) prostate cancer. J Urol 1995;154: 424 – 428. 8. Witjes WPJ, Schulman CC, Debruyne FMMJ. Preliminary results of a prospective randomized study comparing radical prostatectomy versus radical prostatectomy associated with neoadjuvant hormonal combination therapy in T2–2N0M0 prostatic carcinoma. Urology 1997;49:65– 69. 9. Dalkin BL, Ahmann FR, Nagle R, et al. Randomized study of
10.
11.
12. 13.
14.
15. 16.
neoadjuvant testicular androgen ablation therapy before radical prostatectomy in men with clinically localized prostate cancer. J Urol 1996;155:1357–1360. Bonney WW, Schned AR, Timberlake DS. Neoadjuvant androgen ablation for localized prostatic cancer. Pathology methods, surgical end points and meta-analysis of randomized trials. J Urol 1998;160:1754 –1760. Zelefsky MJ, Leibel SA, Burman CM, et al. Neoadjuvant hormonal therapy improves the therapeutic ratio in patients with bulky prostatic cancer treated with three-dimensional conformal radiation therapy. Int J Radiat Oncol Biol Phys 1994;29:755–761. Green N, Bodner H, Broth E, et al. Improved control of bulky prostate carcinoma with sequential estrogen and radiation therapy. Int J Radiat Oncol Biol Phys 1984;10:971–976. Laverdiere J, Gomez JL, Cusan L, et al. Beneficial effect of combination therapy administered prior and following external beam radiation therapy in localized prostate cancer. Int J Radiat Oncol Biol Phys 1995;37:247–252. Zietman AL, Prince EA, Nakfoor BM, Shipley WU. Neoadjuvant androgen suppression with radiation in the management of locally advanced adenocarcinoma of the prostate: Experimental and clinical results. Urology 1997;49:74 – 83. Zietman AL, Nakfoor BN, Prince EA, Gerweck LE. The effect of androgen deprivation on an androgen sensitive tumor; an in vivo and in vitro study. Cancer J Sci Am 1997;3:31–36. Pilepich MV, Caplan R, Byhardt R, et al. Phase III trial of androgen suppression using goserelin in unfavorable prognosis carcinoma of the prostate treated with definitive radiotherapy—report of RTOG protocol. J Clin Oncol 1997;15:1013– 1021.
PII S0360-3016(01)01579-6
CLINICAL INVESTIGATION
Prostate
PHASE III RADIATION THERAPY ONCOLOGY GROUP (RTOG) TRIAL 86-10 OF ANDROGEN DEPRIVATION ADJUVANT TO DEFINITIVE RADIOTHERAPY IN LOCALLY ADVANCED CARCINOMA OF THE PROSTATE
JOHN
MILJENKO V. PILEPICH, M.D.,* KATHRYN WINTER, M.S.,† MADHU J. JOHN, M.D.,‡ B. MESIC, M.D.,§ WILLIAM SAUSE, M.D.,㛳 PHILLIP RUBIN, M.D.,¶ COLLEEN LAWTON, M.D.,# MITCHELL MACHTAY, M.D.,** AND DAVID GRIGNON, M.D.††
*Department of Radiation Oncology, Ann Arbor Regional CCOP, Ann Arbor, MI ; †Radiation Therapy Oncology Group, Philadelphia, PA; ‡Department of Radiation Oncology, University of California, San Francisco, CA; §Radiation Oncology Center, Sacramento, CA; 㛳 Department of Radiation Oncology, LDS Hospital, Salt Lake City, UT; ¶Department of Radiation Oncology, University of Rochester, Rochester, NY; #Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI; **Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA; ††Department of Pathology, Wayne State University, Detroit, MI Purpose: To test the hypothesis that androgen ablation before and during radiotherapy for locally advanced carcinoma of the prostate may, by reducing tumor bulk and enhancing tumor cell kill, improve locoregional control and ultimately survival. Methods and Materials: The study was conducted from 1987 to 1991. Eligible patients were those with bulky tumors (T2–T4) with or without pelvic lymph node involvement and without evidence of distant metastases. They were randomized to receive goserelin, 3.6 mg every 4 weeks; and flutamide, 250 mg t.i.d. for 2 months before radiation therapy and during radiation therapy (Arm I), or radiation therapy alone (Arm II). Of 471 randomized patients, 456 were evaluable: 226 on Arm I and 230 on Arm II. Results: As of November 1999, the median follow-up has reached 6.7 years for all patients and 8.6 years for alive patients. At 8 years, androgen ablation has been associated with an improvement in local control (42% vs. 30%, p ⴝ 0.016), reduction in the incidence of distant metastases (34% vs. 45%, p ⴝ 0.04), disease-free survival (33% vs. 21%, p ⴝ 0.004), biochemical disease-free survival ⴝ PSA <1.5 (24% vs. 10%, p < 0.0001), and cause-specific mortality (23% vs. 31%, p ⴝ 0.05). However, subset analysis indicates that the beneficial effect of short-term androgen ablation appears preferentially in patients with Gleason score 2– 6. In that population, there is a highly significant improvement in all endpoints, including survival (70% vs. 52%, p ⴝ 0.015). In patients with Gleason 7–10 tumors, the regimen has not resulted in a significant enhancement in either locoregional control or survival. Conclusion: In patients with Gleason score 2– 6 carcinoma of the prostate, a short course of androgen ablation administered before and during radiotherapy has been associated with a highly significant improvement in local control, reduction in disease progression, and overall survival. © 2001 Elsevier Science Inc. Carcinoma of the prostate, Hormonal management, Radiotherapy.
INTRODUCTION
of enhancing locoregional control and eventually survival. It was postulated that locoregional control would be enhanced if the tumor bulk were reduced before initiation of radiotherapy. Other potential mechanisms of interaction include an enhancement of tumor cell death (apoptosis) and a reduction of tumor cell repopulation during the course of radiotherapy. A preliminary analysis undertaken in 1994 was reported in 1995 (1). The current report represents the long-term update with mature follow-up data.
Radiation Therapy Oncology Group (RTOG) has conducted a number of trials aimed at the evaluation of androgen deprivation adjuvant to definitive radiotherapy. RTOG 86-10 was the first Phase III trial testing the value of androgen deprivation in a neoadjuvant format. In a population of patients with locally advanced tumors, a short course of neoadjuvant hormonal ablation applied for 2 months before as well as during radiotherapy was tested as a means Reprint requests to: Miljenko V. Pilepich, Department of Radiation Oncology, P.O. Box 995, Ann Arbor, MI 48106. Tel: (734) 712-3595; Fax: (734) 712-5344; E-mail: [email protected] Supported by Public Health Service grants CA-21661 and CA-
32115 from the National Cancer Institute, National Institute of Health, Department of Health and Human Services. Received Jan 3, 2001, and in revised form Feb 28, 2001. Accepted for publication Mar 20, 2001. 1243
1244
I. J. Radiation Oncology
●
Biology
●
Physics
METHODS AND MATERIALS Study design Patients were stratified by stage and histologic grade and randomized to receive either radiotherapy alone or goserelin (Zoladex, AstraZeneca Pharmaceuticals, Wilmington, DE) and flutamide (Eulexin, Schering-Plough, Kenilworth, NJ) for 2 months before radiotherapy and during the radiotherapy course. Eligibility criteria Eligible patients were those who had bulky primary tumors, clinical stage T2–T4. In the context of the study, bulky primary lesions were defined as tumors with a product of palpable tumor dimensions (expressed in cm) of 25 or more. The determination was strictly clinical, based on digital rectal examination. The examiner would estimate the transverse and the sagittal dimension of the palpable tumor in centimeters and multiply the two numbers. If several nodularities were palpable, the examiner would measure each of them separately and then add the product of dimensions. Patients with positive lymph nodes were eligible if the involved nodes remained below the common iliac level. Pretreatment evaluation Mandatory studies included physical examination, Karnofsky performance status evaluation, assessment of potency (sexual history), and laboratory studies, including CBC, SGOT, and SGPT in patients assigned to Zoladex and flutamide. Serum acid phosphate determination, serum testosterone levels, and prostatic-specific antigen (PSA) were mandatory for all patients. It should be noted, however, that PSA determination was not widely available early in the study and was made mandatory only in 1990. Mandatory radiographic studies included chest X-ray and a bone scan. Lymph node evaluation was also mandatory and could be performed by either a lymphangiogram, computerized tomography of the pelvis and abdomen, or exploratory laparotomy. Radiotherapy Megavoltage equipment was required with photon energies higher than 1 MV. Any treatment techniques capable of producing the dose distribution specified by the protocol were acceptable, with the exception of the perineal boost. Customized blocks were used in most patients. Patients with no evidence of tumor spread to the pelvic lymphatics were to be treated electively to a target volume that included pelvic nodes to the level of L5–S1 interspace. In patients with evidence of a pelvic lymph node involvement, the regional lymphatics target volume was to include not only pelvic but also lower para-aortic nodes. The superior margin was to extend to L2–L3 interspace. The prostate boost target volume was to include the prostate, with margins sufficiently wide to encompass all of the tumor extensions into the surrounding tissues.
Volume 50, Number 5, 2001
The dose was prescribed on the central axis at the projected center of the target volumes. Regional lymphatics were to receive a total of 44 to 46 Gy, although doses up to 50 Gy were acceptable. The prostatic target volume was to receive a boost to a total prescribed dose of 65 to 70 Gy. The minimum target doses to the regional lymphatics and the prostatic target volume were 44 Gy and 65 Gy, respectively. The maximum target dose, defined as the greatest dose in target volume delivered to an area greater than 2 cm2, was 50 Gy for the regional lymphatics target volume and 72 Gy for the prostatic boost target volume. The daily doses were 1.8 –2 Gy given 4 –5 times a week. Drug therapy Goserelin (Zoladex) was administered s.c. every 4 weeks, starting 2 months before initiation of radiotherapy and continuing during radiotherapy course, the total overall treatment time being 112 days. Flutamide (Eulexin) was administered by mouth, 250 mg, 3 times per day over the same time period (112 days). Patient assessments After completion of treatment, the patients were to be seen every 3 months during the first year, every 4 months during the second and third years, every 6 months through the fifth year, and then annually for the remainder of the patient’s life. Follow-up evaluation included a clinical examination and a PSA determination. A central review of the radiation therapy information was performed for each case by the study chair after completion of treatment. Radiation therapy equipment used in treatment was calibrated by the Radiologic Physics Center at the University of Texas M. D. Anderson Hospital. Compliance with the drug administration was reviewed by RTOG headquarters staff and the study chair. Central review of the pathology material was undertaken to achieve uniformity of histologic grade (Gleason score) determination. End points and statistical analysis The primary end point of the study was locoregional control; secondary end points were disease-free survival (freedom from progression) and survival. For the purposes of the current analysis, local failure is defined as progression of palpable tumor at any time, persistence of a palpable tumor beyond the second year, and a positive biopsy 2 years or more after study entry. Regional metastasis is defined as clinical or radiographic evidence of involvement of the pelvic lymphatics by the tumor. Distant metastases are defined as clinical or radiographic evidence of disease beyond the pelvis. Survival was measured from the date of randomization to the date of death, or the most recent follow-up. Estimates of survival and progression-free survival were
Androgen deprivation and radiotherapy in carcinoma of the prostate
● M. PILEPICH et al.
1245
Table 1. All patients (n ⫽ 456)
Local failure Distant metastases NED bNED (w/PSA ⬍4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of Failures
5 years
8 years
p value
226 230 226 230 226 230 201 203 201 203 226 230 226 230
72 98 82 104 160 183 158 180 170 194 112 136 52 74
22% 35% 29% 39% 49% 34% 39% 20% 28% 10% 72% 68% 15% 20%
30% 42% 34% 45% 33% 21% 24% 10% 16% 3% 53% 44% 23% 31%
0.016 0.04 0.004 ⬍0.0001 ⬍0.0001 0.10 0.05
Follow-up for all patients
Median follow-up (Yr)
derived by the Kaplan-Meier method (2). Statistical comparisons for survival and progression-free survival were made by the log-rank statistic in the case of censored data or by the proportional hazards analysis to control for prognostic factors. Statistical comparisons for the cumulative incidence of local progression, distant metastases, and cause-specific failure were performed using Gray’s test (3).
All patients (n ⫽ 456)
Alive patients (n ⫽ 208)
6.7
8.6
RESULTS Patient characteristics A total of 471 patients were accessioned from April 15, 1987 through June 1, 1991. Of these, 15 were excluded for the following reasons: no follow-up (4), tumor too small (5), refused all treatment and follow-up (3), lung primary (1), bone metastases (1), and benign disease (1).
Fig. 1. Local progression, all patients.
1246
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 50, Number 5, 2001
Fig. 2. Disease-free survival, all patients.
There were 456 analyzable patients: 226 on the treatment arm (Arm I) and 230 on the control arm (Arm II). As of the end of 1999, the median follow-up reached 6.7 years for all patients and 8.6 years for alive patients. The pretreatment prognostic factors were well balanced be-
tween the treatment groups. The details have been published previously (1). Of the 225 Arm I patients with hormone compliance information, 211 (94%) completed goserelin treatment. Of these, 188 (84%) completed flutamide treatment as planned.
Fig. 3. bNED (PSA ⬍1.5), all patients.
Androgen deprivation and radiotherapy in carcinoma of the prostate
Table 2. Causes of death Cause of death Treated cancer Other* Unknown Total Other causes of death Second primary Heart related Pneumonia Pulmonary Mets. Other Not specified Total
RT ⫹ hormones
RT alone
Total
52 56 4 112
74 56 6 136
126 112 10 248
RT ⫹ hormones
RT alone
Total
11 21 2 3 1 10 8 56
7 21 3 3 3 11 8 56
18 42 5 6 4 21 16 112
* Breakdown of other causes of death.
The reasons for terminating flutamide in 23 patients included diarrhea in 11 patients,1 hot flashes in 3 patients, and miscellaneous other problems, including liver function abnormalities, rash, nausea, etc., in the remaining patients. Two patients refused goserelin, and one of these also re-
● M. PILEPICH et al.
1247
fused flutamide. A total of 186 patients completed both goserelin and flutamide as planned per the study guidelines. Table 1 summarizes the treatment results2 for the entire population and gives estimated 5- and 8-year rates for each endpoint. There is a highly significant improvement in local control and reduction in disease progression for Arm I (Figs. 1, 2, and 3). In addition, cause-specific failure (mortality due to prostate cancer) is reduced significantly for Arm I within the entire population (p ⫽ 0.05). Table 2 lists the causes of death for the entire population. Figures 4 and 5 show disease-specific mortality and absolute survival for all patients. Subset analyses indicate that the beneficial effect of hormonal management appears preferentially in patients with lower Gleason score (Gleason 2– 6). Table 3 summarizes the end-point analysis in Gleason 2– 6 patients. In this population, experimental treatment has been associated with a highly significant improvement in all end points, including survival (p ⫽ 0.015). Figures 6 –9 show the pattern of local progression (failure), no biochemical evidence of disease (bNED) (PSA ⬍1.5), and absolute survival for Gleason 2– 6 patients. Gleason 7 patients did not benefit significantly from the experimental treatment in terms of either local failure, sur-
Fig. 4. Disease-specific mortality, all patients.
1
Diarrhea was graded as Grade 2 in 7 patients and Grade 4 in 3 patients. Grading system is defined as follows: Grade 1, increase in frequency of bowel movements (no medication required), stools semisolid; Grade 2, watery stools, 3– 6/day; Grade 3, watery
stools, ⬎6/day; Grade 4, diarrhea requiring hospitalization. 2 Regional metastases were recorded in 6 patients. All of these patients had either local failure or distant metastases as the first site of failure and are accounted for elsewhere.
1248
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 50, Number 5, 2001
Fig. 5. Absolute survival, all patients.
vival, or cause-specific failure. The results in this population are summarized in Table 4. Table 5 summarizes the results in Gleason 8 –10 patients. These patients did not benefit significantly from the experimental treatment. DISCUSSION The mechanisms of interaction between radiation therapy and androgen deprivation in carcinoma of the prostate re-
main largely unknown. It is likely that the modes of interaction vary, depending on the timing (initiation) of the androgen deprivation and the duration of its application. If applied for a period of time before radiotherapy, androgen deprivation is likely to result in a significant reduction of tumor volume (a cytoreduction). If applied in close proximity to radiation therapy and/or during radiation therapy, androgen deprivation may interact with radiation on a cellular level and also affect cell kinetics.
Table 3. Centrally reviewed Gleason 2– 6 (n ⫽ 129)
Local failure Distant metastases NED bNED (w/PSA ⬍4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of failures
5 years
8 years
p value
71 58 71 58 71 58 64 52 64 52 71 58 71 58
17 29 11 21 39 44 39 44 47 47 24 34 1 13
14% 35% 12% 28% 67% 48% 61% 24% 45% 13% 84% 79% 0% 9%
21% 46% 13% 34% 50% 32% 47% – 30% – 70% 52% 2% 17%
0.005 0.006 0.004 ⬍0.0001 ⬍0.0001 0.015 0.0002
Follow-up for centrally reviewed Gleason 2– 6 patients
Median follow-up (Yr)
All patients (n ⫽ 129)
Alive patients (n ⫽ 71)
7.9
8.6
Androgen deprivation and radiotherapy in carcinoma of the prostate
● M. PILEPICH et al.
1249
Fig. 6. Local progression, Gleason 2– 6.
Cytoreduction refers to debulking of tumor before initiation of radiotherapy. It is well recognized that reduced tumor volume generally correlates with an increased statistical probability of tumor control with radiotherapy. Tumor size is a wellrecognized predictor of local control in carcinoma of the prostate. In RTOG experience (4), patients with bulky prima-
ries, with the product of tumor dimensions 25 or more, exhibited a particularly high recurrence rate, with more than 50% of the patients exhibiting evidence of local recurrence at 6 years after radiotherapy. This observation was the basis for defining selection criteria for several RTOG studies of hormonal cytoreduction in locally advanced carcinoma of the prostate.
Fig. 7. bNED (PSA ⬍1.5), Gleason 2– 6.
1250
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 50, Number 5, 2001
Fig. 8. Disease-specific mortality, Gleason 2– 6.
There is now considerable evidence in surgical and radiation oncology literature that several months of androgen deprivation predictably results in significant downsizing of prostate cancer volume in the majority of patients (5–11). However, clinical data addressing the potential value of hormonal cytoreduction before radiotherapy have been quite limited. Green et
al. (12) reported on a nonrandomized study suggesting that the local control could be significantly improved. The results were corroborated by the findings of Laverdiere et al. (13). There is some experimental evidence that neoadjuvant androgen suppression may enhance the effects of radiotherapy. Zietman et al. (14, 15) tested the concept in a Shionogi
Fig. 9. Absolute survival, Gleason 2– 6.
Androgen deprivation and radiotherapy in carcinoma of the prostate
● M. PILEPICH et al.
1251
Table 4. Centrally reviewed Gleason 7 (n ⫽ 176)
Local failure Distant metastases NED bNED (with PSA ⬍ 4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of failures
5 years
8 years
88 88 88 88 88 88 79 80 79 80 88 88 88 88
30 37 40 39 70 72 68 74 71 78 48 48 24 23
22% 38% 34% 36% 44% 28% 30% 18% 20% 9% 71% 67% 17% 16%
33% 43% 44% 44% 22% 19% 14% 9% 9% 3% 49% 46% 27% 26%
p value 0.26 0.88 0.40 ⬍0.016 ⬍0.0001 0.91 0.68
Follow-up for centrally reviewed Gleason 7 patients All patients (n ⫽ 176)
Alive patients (n ⫽ 80)
Median follow-up (Yr)
6.7
8.5
tumor in mice. Using tumor control as an end point, a dose–response curve was constructed. When orchiectomy was performed 24 – 48 h before radiation, the tumor control dose for 50% of tumors (TCD 50) fell from 89 Gy to 60.3 Gy. When the tumors were allowed to regress maximally before irradiation, the TCD 50 fell to 42.1 Gy. The degree to which this effect is related to cytoreduction vs. other mechanisms of action on a cellular level and possible changes in tumor bed and blood supply is not clear. RTOG 86-10 tested the potential value of androgen deprivation in conjunction with definitive radiotherapy in a
select population of patients with locally very advanced tumors. Moreover, approximately 40% of the patients had abnormal elevation of serum acid phosphates, reflecting the high potential for subclinical metastatic disease in a large proportion of patients. If observed as a whole, the study population benefited significantly from the experimental treatment. This refers to the incidence of local failures, disease progression, and reduction in the disease-specific mortality. The subset analyses reveal that the experimental treatment has proven to be much more effective in patients with
Table 5. Centrally reviewed Gleason 8 –10 (n ⫽ 124)
Local Failure Distant metastases NED bNED (with PSA ⬍4) bNED (w/PSA ⬍1.5) Survival Cause-specific failure
Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm Arm
I II I II I II I II I II I II I II
Total
No. of failures
5 years
8 years
57 67 57 67 57 67 49 57 49 57 57 67 57 67
23 28 29 40 46 56 43 53 44 57 37 46 25 36
36% 37% 45% 58% 32% 27% 24% 14% 14% 5% 54% 59% 33% 36%
40% 43% 47% 60% 22% 13% 12% 5% 10% — 38% 31% 44% 54%
p value 0.69 0.35 0.51 0.07 ⬍0.0001 0.98 0.36
Follow-up for centrally reviewed Gleason 8 –10 patients
Median follow-up (Yr)
All patients (n ⫽ 124)
Alive patients (n ⫽ 41)
5.3
8.7
1252
I. J. Radiation Oncology
●
Biology
●
Physics
low Gleason scores (Gleason 2– 6). In this population, the impact on local control, disease progression, and survival is profound and unequivocal. In patients with higher Gleason scores (Gleason 7–10), experimental treatment has not produced a significant improvement in local control and survival. Impact on disease progression as judged by bNED is of limited clinical significance. It is of interest that a parallel RTOG study (RTOG 85-31), which tested the potential value of long-term adjuvant deprivation after radiotherapy, demonstrated a beneficial effect of hormonal treatment in high Gleason score
Volume 50, Number 5, 2001
patients but not in the low Gleason score patients (16). This indicates that in Gleason 7–10 patients, alternative methods of hormonal management need to be used. These observations stress a need to consider different strategies for adjuvant hormonal management for different subsets of patients. In high Gleason score patients, 4 months of androgen deprivation before and during radiotherapy course are not adequate. In Gleason 2– 6 patients with locally advanced disease, such treatment exerts a major beneficial effect on all end points, including survival, and should be considered a standard of care.
REFERENCES 1. Pilepich MV, Krall JM, Al-Sarraf M, et al. Androgen deprivation with radiation therapy compared with radiation therapy alone for locally advanced prostatic carcinoma: A randomized comparative trial of the Radiation Therapy Oncology Group. Urology 1995;45:616 – 623. 2. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;52:457– 481. 3. Gray RL. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat 1988;16:1141– 1154. 4. Pilepich MV, Krall JM, Sause WT, et al. Prognostic factors in carcinoma of the prostate—Analysis of RTOG study 75-06. Int J Radiat Oncol Biol Phys 1987;13:339 –349. 5. Aus G, Abrahamsson P-A, Ahlgren F, et al. Hormonal treatment before radical prostatectomy: A 3-year follow-up. J Urol 1998;159:2013–2017. 6. Goldenberg SL, Klotz LH, Srigley J, et al. Randomized, prospective, controlled study comparing radical prostatectomy alone and neoadjuvant androgen withdrawal in treatment of localized prostate cancer. J Urol 1996;156:873– 877. 7. Soloway MS, Sharifi R, Wajsman Z, et al. Randomized prospective study comparing radical prostatectomy alone versus radical prostatectomy preceded by androgen blockade in clinical stage B2 (T2bNxM0) prostate cancer. J Urol 1995;154: 424 – 428. 8. Witjes WPJ, Schulman CC, Debruyne FMMJ. Preliminary results of a prospective randomized study comparing radical prostatectomy versus radical prostatectomy associated with neoadjuvant hormonal combination therapy in T2–2N0M0 prostatic carcinoma. Urology 1997;49:65– 69. 9. Dalkin BL, Ahmann FR, Nagle R, et al. Randomized study of
10.
11.
12. 13.
14.
15. 16.
neoadjuvant testicular androgen ablation therapy before radical prostatectomy in men with clinically localized prostate cancer. J Urol 1996;155:1357–1360. Bonney WW, Schned AR, Timberlake DS. Neoadjuvant androgen ablation for localized prostatic cancer. Pathology methods, surgical end points and meta-analysis of randomized trials. J Urol 1998;160:1754 –1760. Zelefsky MJ, Leibel SA, Burman CM, et al. Neoadjuvant hormonal therapy improves the therapeutic ratio in patients with bulky prostatic cancer treated with three-dimensional conformal radiation therapy. Int J Radiat Oncol Biol Phys 1994;29:755–761. Green N, Bodner H, Broth E, et al. Improved control of bulky prostate carcinoma with sequential estrogen and radiation therapy. Int J Radiat Oncol Biol Phys 1984;10:971–976. Laverdiere J, Gomez JL, Cusan L, et al. Beneficial effect of combination therapy administered prior and following external beam radiation therapy in localized prostate cancer. Int J Radiat Oncol Biol Phys 1995;37:247–252. Zietman AL, Prince EA, Nakfoor BM, Shipley WU. Neoadjuvant androgen suppression with radiation in the management of locally advanced adenocarcinoma of the prostate: Experimental and clinical results. Urology 1997;49:74 – 83. Zietman AL, Nakfoor BN, Prince EA, Gerweck LE. The effect of androgen deprivation on an androgen sensitive tumor; an in vivo and in vitro study. Cancer J Sci Am 1997;3:31–36. Pilepich MV, Caplan R, Byhardt R, et al. Phase III trial of androgen suppression using goserelin in unfavorable prognosis carcinoma of the prostate treated with definitive radiotherapy—report of RTOG protocol. J Clin Oncol 1997;15:1013– 1021.