VOLUME 26 䡠 NUMBER 15 䡠 MAY 20 2008

JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T

Ten-Year Follow-Up of Radiation Therapy Oncology Group

From the Department of Radiation
Oncology, Fox Chase Cancer Center;
Department of Biostatistics, Radiation
Therapy Oncology Group, Philadelphia;
Northeast Radiation Oncology Center,
Scranton, PA; Department of Radiation
Oncology, McGill University, Montréal,
Québec; Department of Radiation
Oncology, University of Western
Ontario, London, Ontario, Canada;
Department of Pathology, Indiana
University, Indianapolis, IN; Department
of Radiation Oncology, Medical College
of Wisconsin, Milwaukee, WI; Depart-
ment of Radiation Oncology, Radiologi-
cal Associates of Sacramento,
Sacramento, CA; Department of Radia-
tion Oncology, University of Michigan,
Ann Arbor, MI; and the Department of
Radiation Oncology, Massachusetts
General Hospital, Boston, MA.
Submitted October 19, 2007; accepted
February 4, 2008; published online
ahead of print at www.jco.org on April
14, 2008.
Supported by Grants No. RTOG U10
CA21661, CCOP U10 CA37422, and
STAT U10 CA32115 from the National
Cancer Institute.
Presented at the 48th Annual Meeting
of the American Society for Therapeutic
Radiology and Oncology, November
5-9, 2006, Philadelphia, PA.
Protocol 92-02: A Phase III Trial of the Duration of Elective
Androgen Deprivation in Locally Advanced Prostate Cancer
Eric M. Horwitz, Kyounghwa Bae, Gerald E. Hanks, Arthur Porter, David J. Grignon, Harmar D. Brereton,
Varagur Venkatesan, Colleen A. Lawton, Seth A. Rosenthal, Howard M. Sandler, and William U. Shipley
A B S T R A C T
Purpose
To determine whether adding 2 years of androgen-deprivation therapy (ADT) improved outcome
for patients electively treated with ADT before and during radiation therapy (RT).
Patients and Methods
Prostate cancer patients with T2c-T4 prostate cancer with no extra pelvic lymph node involvement
and prostate-specific antigen (PSA) less than 150 ng/mL were included. All patients received 4
months of goserelin and flutamide before and during RT. They were randomized to no further ADT
(short-term ADT [STAD] ⫹ RT) or 24 months of goserelin (long-term ADT [LTAD] ⫹ RT). A total of
1,554 patients were entered. RT was 45 Gy to the pelvic nodes and 65 to 70 Gy to the prostate.
Median follow-up of all survival patients is 11.31 and 11.27 years for the two arms.
Results
At 10 years, the LTAD ⫹ RT group showed significant improvement over the STAD ⫹ RT group
for all end points except overall survival: disease-free survival (13.2% v 22.5%; P ⬍ .0001),
disease-specific survival (83.9% v 88.7%; P ⫽ .0042), local progression (22.2% v 12.3%;
P ⬍ .0001), distant metastasis (22.8% v 14.8%; P ⬍ .0001), biochemical failure (68.1% v 51.9%;
P ⱕ .0001), and overall survival (51.6% v 53.9%, P ⫽ .36). One subgroup analyzed consisted of all
cancers with a Gleason score of 8 to 10 cancers. An overall survival difference was observed
(31.9% v 45.1%; P ⫽ .0061), as well as in all other end points herein.
Conclusion
LTAD as delivered in this study for the treatment of locally advanced prostate cancer is superior
to STAD for all end points except survival. A survival advantage for LTAD ⫹ RT in the treatment
of locally advanced tumors with a Gleason score of 8 to 10 suggests that this should be the
standard of treatment for these high-risk patients.
J Clin Oncol 26:2497-2504. © 2008 by American Society of Clinical Oncology

The contents of this article are solely
the responsibility of the authors and do
not necessarily represent the official
views of the National Cancer Institute.
Authors’ disclosures of potential con-
flicts of interest and author contribu-
tions are found at the end of this
article.
Corresponding author: Eric M. Horwitz,
MD, Fox Chase Cancer Center, Depart-
ment of Radiation Oncology, 333 Cott-
man Ave, Philadelphia, PA 19111-2497;
e-mail: eric.horwitz@fccc.edu.
© 2008 by American Society of Clinical
Oncology
0732-183X/08/2615-2497/$20.00
DOI: 10.1200/JCO.2007.14.9021
INTRODUCTION
The role of androgen-deprivation therapy (ADT)
used in conjunction with external-beam radiation
therapy (RT) to treat locally advanced prostate can-
cer has been investigated since the 1960s.1 Early
studies demonstrated a benefit in clinical outcome
when ADT was used, but many questions remained
unanswered, including the duration and timing of
treatment. Substantial toxicity was also identified
when hormones were used. Beginning in the 1980s,
the Radiation Therapy Oncology Group (RTOG)
developed trials based on information from these
early studies, which incorporated new hormonal
agents with less cardiovascular toxicity. RTOG
85-31 randomly assigned patients with locally ad-
vanced prostate cancer between RT and goserelin
indefinitely versus RT alone. A statistically signifi-
cant improvement in 10-year biochemical no evi-
dence of disease (bNED) control, distant metastasis
survival (DMS), local control (LC), and overall sur-
vival (OS) was observed for patients treated with
long-term ADT (LTAD).2 The role of short-term
ADT (STAD) was examined in RTOG 86-10. Pa-
tients were randomly assigned between ADT before
and during RT or RT alone. Significant improve-
ment in LC was seen for patients treated with STAD.3
The difference between the two significant
study arms from RTOG 85-31 and 86-10 was de-
finitively tested in the largest phase III random-
ized prospective trial of prostate cancer patients in
North America. The first analysis of RTOG 92-02,
demonstrated statistically significant improve-
ments in outcome in the long-term hormone arm

© 2008 by American Society of Clinical Oncology 2497

Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042

Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
 Horwitz et al

for disease-free survival (DFS), cause-specific survival (CSS), bio-
chemical failure (BF), distant metastasis (DM), and local failure
(LF).4 This study updates the results of RTOG 92-02 and reports
the clinical and biochemical results as well as toxicity at 10 years.
PATIENTS AND METHODS
Patient Population
Patients eligible for inclusion on RTOG 92-02 included those men with
histologically confirmed adenocarcinoma of the prostate (T2c-T4N0-X). A
pretreatment PSA less than 150 ng/mL was mandatory. The 1992 American
Joint Committee on Cancer Staging Manual was used for clinical staging.5
There could be no documented, DM and the participants’ Karnofsky perfor-
mance status had to be at least 70%. No prior ADT, RT, or chemotherapy was
allowed. Institutional review board approval was obtained at each participat-
ing center before patient enrollment and data was transferred to a single
statistical center per Health Insurance Portability and Accountability Act
(HIPAA) guidelines. Informed consent was obtained from each participant
before enrollment, random assignment, and treatment. The details of pretreat-
ment patient evaluations have been summarized in the previous report.4 After
RT was completed, follow-up with prostate-specific antigen (PSA) was every 3
to 6 months during the first 5 years, and then yearly.
LP, DM, BF, and DS death9 in the presence of competing risks. The log-rank
test was used to test the significance between two treatment arms for end point
of survival curves (OS and DFS).10 Gray’s test was used to test the significance
between two treatment arms for the end point of failure curves (LP, DM and
BF, and DS death).11 The testing was performed at a significance level of .05
with two-sided tests.
RESULTS
Table 1 shows pretreatment characteristics for 1,521 eligible patients.
There was no statistically significant difference between treatment
arms with respect to the stratification factors and other characteristics.
As reported by the institutions, RT was completed as assigned in 96%
of the cases in the STAD ⫹ RT arm and 95% of the cases in the LTAD
⫹ RT arm. Four percent and 3% of the reviewed cases were judged
unacceptable major variations in the STAD ⫹ RT and LTAD ⫹ RT
Table 1. Pretreatment Characteristics for All Eligible Patients

RT STAD ⫹ RT LTAD ⫹ RT
(n ⫽ 763) (n ⫽ 758)
All patients received conventional RT to the pelvis using a four-field
technique with megavoltage x-rays (ⱖ 4 MV) to 44 to 46 Gy. This was followed Characteristic No. % No. % ␹2 Test P
by a conedown to the prostate for a total dose of 65 to 70 Gy for T2c tumors and Age, years
67.5 to 70 Gy for T3 to T4 tumors. The prescribed dose was recorded at the Median 70 70
center of the prostate target volume. Range 43-87 43-88
ADT PSA, ng/mL
ADT was begun 2 months before the start of RT and continued until the ⱕ 30 511 67 510 67
RT was completed. All patients in this study received flutamide (250 mg three ⬎ 30 252 33 248 33 .8977
times a day) with goserelin (3.6 mg subcutaneously monthly) until the RT was Karnofsky status, %
completed. Patients were randomly assigned to no further treatment (STAD ⫹ 70 11 2 2 ⬍1
RT) or 24 additional months of monthly goserelin (LTAD ⫹ RT). 80 54 7 48 6
90 376 49 358 47
Study Design and End Points 100 322 42 350 46 .0425
This study was designed to test for an absolute 10% improvement in Intercurrent disease
DFS, from 40% to 50%, at 5 years. Randomization was performed before any No 254 33 246 33
treatment was started using the treatment allocation scheme described by Yes 503 66 508 67
Zelen.6 Patients were stratified according to stage, pretreatment PSA, grade, Unknown 6 1 4 ⬍1 .8881
and nodal status although the study was not powered to assess the primary end Clinical stage
point on the basis of these strata. RTOG 92-02 opened for accrual on June 26, T2 348 45 344 45
1992, and the study closed on April 15, 1995, with 1,554 registered cases. T3 394 52 376 50
Study end points included DFS, OS, local progression (LP), DM, BF, and T4 21 3 38 5 .0697
disease-specific survival (DSS). LP was defined as clinical evidence of local Pathological nodal status
recurrence (by any method) or persistent disease. DM was defined as the NX 658 87 648 86
clinical evidence of distant disease (by any method). Patients without an event N0 70 9 81 11
for either LP or DM end point were censored at the date of last follow-up or N1 24 3 18 2
death. BF was defined as the earliest of the following: three consecutive rises N2 11 1 11 1 .6328
after a post-treatment PSA nadir (the 1997 American Society for Therapeutic Institutional Gleason score
Radiology and Oncology [ASTRO] definition7), receiving additional ADT or 2-5 143 19 137 18
PSA greater than 4 ng/mL. Patients were censored at the date of last PSA. DFS 6 149 19 154 20
was defined as the first occurrence of LP, DF, BF, or death. DSS was defined as 7 226 30 251 33
a death resulting from to prostate cancer, treatment toxicity or unknown 8-10 187 24 174 23
causes with DM. All event times were measured from the date of random Unknown 58 8 42 6 .3386
assignment. Acute RT toxicities were defined as those occurring within 90 days Central Gleason score
from the start of RT. Any toxicity continuing or developing after 90 days was ⱕ6 118 15 116 15
considered a late RT toxicity. Results from a hypothesis-generating postran- 7 271 36 295 39
domization subset analysis of patients with a Gleason score of 8 to 10 was 8-10 108 14 107 14
performed in the first report and, on the basis of results from similar random- Unknown 266 35 240 32 .5013
ized prospective studies, was included in this update.
Abbreviations: STAD ⫹ RT, short-term androgen deprivation with external-
Statistical Methods beam radiation therapy; LTAD ⫹ RT, long-term androgen deprivation with
The Kaplan-Meier method was used to estimate the survival rate for OS external-beam radiation therapy followed by goserelin.
and DFS.8 The cumulative incidence approach was used to estimate the rate for

2498 © 2008 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042

Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
 Long-Term Results of RTOG 92-02

arm, respectively. The median follow-up for all living patients in this
study is 11.3 years for both arms.
Toxicity
Toxicity from treatment was scored using the previously re-
ported RTOG criteria.12 Acute and late RT and hormone toxicities are
listed in Table 2. Nine percent and 1% of the cases have reported grade
3 and grade 4 acute toxicity, respectively, caused by RT in the STAD ⫹
RT arm. Seven percent and 1% of the cases have reported grade 3 and
grade 4 acute toxicity, respectively, caused by RT in the LTAD ⫹ RT
arm. Hormone grade 3 and grade 4 toxicities are 4% and less than 1%,
respectively, at any time in both arms. One patient in the LTAD ⫹ RT
arm died as a result of chemical hepatitis during the hormone treat-
ment. As seen in Table 2, LTAD increased late toxicity. Grade 3 and
grade 4 late toxicity caused by RT have been reported in 6% and 1% of
the cases in the STAD ⫹ RT arm and 7% and 3% of the cases in the
LTAD ⫹ RT arm, respectively. One patient died as a result of proctitis
and one as a result of sigmoid perforation in the STAD ⫹ RT arm, and
one patient died as a result of colostomy obstruction in the LTAD ⫹
RT arm. Three patients experienced grade 3 cardiac toxicities, one in
the STAD ⫹ RT group (congestive heart failure [CHF]) and two in the
LTAD ⫹ RT arm (one with CHF and one with ischemia). There was
no statistically significant difference in cardiac toxicity between the
two groups, and there were no grade 4 or 5 cardiac events, although
this may be an underestimation.
Treatment Outcomes
Table 3 shows the 10-year treatment outcomes for the entire
study population. The estimated DFS (the primary end point) at 10
years was 13.2% and 22.5% for the STAD ⫹ RT arm and LTAD ⫹ RT

Table 2. Acute and Chronic Toxicity for All Eligible Patients

STAD ⫹ RT (n ⫽ 763) LTAD ⫹ RT (n ⫽ 758)

Toxicity Grade 1 Grade 2 Grade 3 Grade 4 Grade 5 Grade 1 Grade 2 Grade 3 Grade 4 Grade 5
Acute radiotherapy toxicityⴱ
Skin 205 97 2 1 0 192 111 3 1 0
Bladder 284 188 36 3 0 289 177 22 1 0
Bowel 229 358 10 0 0 212 354 14 0 0
Other GU 45 19 4 0 0 41 22 1 1 0
Other GI 63 38 1 0 0 45 55 1 0 0
Other 82 54 20 6 0 68 62 16 7 0
Maximum toxicity per patient 198 418 65 10 0 213 406 50 10 0
% 26 55 9 1 0 28 54 7 1 0
Late radiotherapy toxicity†
Skin 47 19 0 0 0 55 17 0 0 0
Bladder 200 98 31 2 0 169 121 25 13 0
Bowel 228 112 5 2 1 216 107 15 2 2
Other GU 115 34 7 2 0 115 45 4 3 0
Other GI 76 24 2 0 0 93 31 4 2 0
Other 94 28 5 0 0 78 36 10 1 0
Maximum toxicity per patient 282 210 45 5 1 236 210 50 19 2
% 37 28 6 1 ⬍1 31 28 7 3 ⬍1
Hormone toxicity‡
Nausea 0 3 0 0 0 0 0 2 0 0
Vomiting 1 2 0 0 0 0 1 1 0 0
Diarrhea 12 32 7 1 0 0 21 7 1 0
Headache 0 0 1 0 0 0 0 0 0 0
Fluid retention 0 1 0 0 0 0 0 0 0 0
Gynecomastia 0 0 0 0 0 1 0 0 0 0
Skin Rash 1 1 0 0 0 1 1 1 0 0
Infection 1 1 0 0 0 0 1 0 0 0
Elevated AST 51 20 8 1 0 95 22 11 1 1
Thrombophi/emboli 1 0 0 0 0 0 0 0 0 0
Cardiac 0 0 1 0 0 0 0 2 0 0
Hot Flashes 0 0 0 0 0 1 0 0 0 0
Impotence 0 0 1 0 0 0 0 0 0 0
Other 40 16 15 1 0 64 14 13 3 1
Maximum toxicity per patient 72 56 29 2 0 119 45 34 5 1
% 9 7 4 ⬍1 0 16 6 4 1 ⬍1

NOTE. P values calculated from statistics ⱕ grade 2 versus ⱖ grade 3 between the two arms.
Abbreviations: STAD ⫹ RT, short-term androgen deprivation with external-beam radiation therapy; LTAD ⫹ RT, long-term androgen deprivation with external-beam
radiation therapy followed by goserelin; GU, genitourinary.
ⴱ 2
␹ test P ⫽ .2104; n ⫽ 763 for STAD ⫹ RT and 758 for LTAD ⫹ RT.
†␹2 test P ⫽ .0269 (statistically significant at ␣ ⫽ .05); n ⫽ 762 for STAD ⫹ RT and 755 for LTAD ⫹ RT.
‡␹2 test P ⫽ .9789; n ⫽ 763 for STAD ⫹ RT and 758 for LTAD ⫹ RT.

www.jco.org © 2008 by American Society of Clinical Oncology 2499

Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042

Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
 Horwitz et al

Table 3. 10-Year Treatment Outcomes for All Eligible Patients

STAD ⫹ RT (n ⫽ 763) LTAD ⫹ RT (n ⫽ 758)

Outcome No. of Failures Estimated Rate 95% CI No. of Failures Estimated Rate 95% CI Log-Rank ␹2 Test P
Disease-free survival 653 13.2 11 to 16 571 22.5 19 to 26 ⬍ .0001ⴱ
Overall survival 351 51.6 48 to 55 330 53.9 50 to 58 .3590
Disease-specific survival 116 83.9 81 to 87 80 88.7 86 to 91 .0042ⴱ
Local progression 166 22.2 19 to 25 90 12.3 10 to 15 ⬍ .0001ⴱ
Distant metastasis 167 22.8 20 to 26 107 14.8 12 to 17 ⬍ .0001ⴱ
Biochemical failure 513 68.1 65 to 71 384 51.9 48 to 55 ⬍ .0001ⴱ

Abbreviations: STAD ⫹ RT, short-term androgen deprivation with external-beam radiation therapy; LTAD ⫹ RT, long-term androgen deprivation with external-beam
radiation therapy followed by goserelin.
ⴱ
Statistically significant at .05.

arm, respectively (P ⬍ .0001). The 10-year OS rates are 51.6% and
53.9% for the STAD ⫹ RT and LTAD ⫹ RT arms, respectively
(P ⫽ .359). The LTAD ⫹ RT arm statistically significantly improved
patients’ 10-year DFS (P ⬍ .0001) and DSS (P ⫽ .004), and reduced LP
(P ⬍ .0001), DM (P ⬍ .0001) and BF (P ⬍ .0001) compared with the
STAD ⫹ RT arm. These differences remain when adjusted for PSA
and T stage (Table 4). The LTAD ⫹ RT arm significantly improved
patients’ DFS (P ⬍ .0001), DSS (P ⫽ .002), LP (P ⬍ .0001), DM
(⬍ 0.0001), and BF (P ⬍ .0001) at 10 years compared with the STAD
⫹ RT arm (Fig 1).
Treatment Outcomes of the Subgroup With a Gleason
Score of 8 to 10
For patients with a Gleason score of 8 to 10, there are statistically
significant improvements in DFS (P ⬍ .0001), OS (P ⫽ 0061), and
DSS (P ⫽ .0072), and reductions in LP (P ⫽ .0338), DM (P ⫽ .0019),
and BF (P ⬍ .0001) for the patients in the LTAD ⫹ RT arm compared
with the STAD ⫹ RT arm (Table 5; Fig 2).
DISCUSSION
In this study, we report the long-term clinical, biochemical, and tox-
icity results from RTOG 92-02. At 10 years, statistically significant
differences in DFS, DSS, LP, DM, and BF were observed within the
entire population, with patients randomly assigned to the LTAD ⫹ RT
arm demonstrating superior outcome. The primary end point testing
for an absolute 10% improvement in DFS, from 40% to 50%, was
positive at 5 and 10 years. This trial was not designed to detect a
survival difference, yet a difference in cause-specific mortality was
seen. The subset of patients with a Gleason score of 8 to 10 treated with
LTAD ⫹ RT had statistically significant improvements in OS, DFS,
and DSS, and statistically significant reductions in LP, DM, and BF.
Toxicity remained low, although patients who received LTAD had
increased rates of GI toxicity.
The RTOG began investigating the role of ADT combined with
RT in the treatment of locally advanced prostate cancer in the 1980s.
The long-term results of RTOG 85-31, which randomly assigned 977
patients to receive between 65 and 70 Gy external beam RT and
goserelin started during the last week of RT (arm 1) and continuing
indefinitely versus RT alone (arm 2) were reported by Pilepich et al2
and demonstrated significant differences between the two treatment
arms at 10 years for bNED control, DMF, LF, and OS (P ⬍ .0001) after
long follow-up.
RTOG 86-10 opened after RTOG 85-31 began accruing patients.
Prostate cancer patients with T2b-4N0-1M0 and tumors larger than
25 cm2 were randomly assigned between goserelin and flutamide 2

Table 4. 10-Year Multivariate Analysis for All Eligible Patients

Outcome Comparison HR 95% CI ␹2 Test P
Disease-free survival STAD⫹RT RL 0.582 to 0.726 ⬍ .0001ⴱ
LTAD⫹RT 0.65
Overall survival STAD⫹RT RL 0.804 to 1.057 .2447
LTAD⫹RT 0.922
Disease-specific survival STAD⫹RT RL 0.504 to 0.857 .0019ⴱ
LTAD⫹RT 0.657
Local progression STAD⫹RT RL 0.399 to 0.663 ⬍ .0001ⴱ
LTAD⫹RT 0.514
Distant metastasis STAD⫹RT RL 0.456 to 0.732 ⬍ .0001ⴱ
LTAD⫹RT 0.578
Biochemical failure STAD⫹RT RL 0.476 to 0.621 ⬍ .0001ⴱ
LTAD⫹RT 0.544

NOTE. Multivariate Cox proportional hazard models include the following covariates: prostate-specific antigen (ⱕ 30 关RL兴, ⬎ 30 ng/mL), T stage (T2 关RL兴, T3, T4)
and treatment (STAD ⫹ RT 关RL兴, LTAD⫹RT).
Abbreviations: STAD ⫹ RT, short-term androgen deprivation with external-beam radiation therapy; LTAD ⫹ RT, long-term androgen deprivation with external-beam
radiation therapy followed by goserelin; HR, hazard ratio; RL, reference level.
ⴱ
Statistically significant at .05.

2500 © 2008 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042

Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
 Long-Term Results of RTOG 92-02

A 1.0 B 1.0
0.9 0.9

0.8 0.8
DSS Non Failure Rate

STAD + RT STAD + RT
LTAD + RT LTAD + RT

Non Failure Rate

0.7 P = .0042 0.7 P < .0001

0.6 0.6

0.5 0.5

0.4 0.4
0.3 0.3

0.2 0.2

0.1 763 722 631 552 441 322 0.1 763 690 578 506 393 279
758 714 640 552 453 336 758 698 611 518 418 314

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10

Time Since Random Assignment (years) Time Since Random Assignment (years)
C 1.0 D 1.0
0.9 0.9

0.8 0.8
Non Failure Rate

0.7 0.7

Survival Rate
0.6 0.6

0.5 0.5

0.4 0.4
STAD + RT STAD + RT
0.3 LTAD + RT 0.3 LTAD + RT
P < .0001 P = .3590
0.2 0.2

0.1 763 398 243 182 138 98 0.1 763 722 631 552 441 322
758 607 366 282 207 152 758 714 640 552 453 336

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10

Time Since Random Assignment (years) Time Since Random Assignment (years)
Fig 1. (A) Disease-specific survival, (B) distant metastasis failure, (C) biochemical failure, and (D) overall survival for all eligible patients. STAD, short-term
androgen-deprivation therapy; LTAD, long-term androgen-deprivation therapy; RT, radiotherapy.

months before and during RT (arm 1, 226 patients) and RT alone
(arm 2, 232 patients). At 8 years, a significant difference in LC was seen
between the two arms (P ⫽ .004) as well as an improvement in PFS for
arm I (P ⫽ .0019). With longer follow-up, a difference in the rate of
DM was observed between the two groups (P ⫽ .04). All patients in
RTOG 85-31 and 86-10, as well as those in the current study were
treated with conventional RT doses (65 to 70 Gy).3 Both of these
studies showed statistically significant improvements in outcome for
patients treated with RT and ADT compared with patients treated
with RT alone; however, the ideal duration and timing of hormones
remained unknown. One of the primary goals of RTOG 92-02 was to
determine the duration of the ADT. The initial report with 5-year data

Table 5. 10-Year Treatment Outcomes for Patients With a Gleason Score of 8 to 10 and N0/NX
STAD ⫹ RT (n ⫽ 171) LTAD ⫹ RT (n ⫽ 166)

No. of Failures 10-Year No. of Failures 10-Year

Outcome by 10 Years Estimate (%) 95% CI (%) by 10 Years Estimate (%) 95% CI (%) Log-Rank ␹2 Test P
Overall survival 111 31.91 24.62 to 39.203 85 45.05 37.07 to 53.03 .0061
Disease-free survival 154 9.35 4.90 to 13.80 127 20.81 14.30 to 27.32 ⬍ .0001
Disease-specific survival 53 66.91 59.60 to 74.21 30 79.81 73.38 to 86.24 .0072
Local progression 46 27.26 20.49 to 34.02 29 17.77 11.87 to 23.67 .0338
Distant metastasis 66 39.69 32.16 to 47.21 39 25.56 18.64 to 32.48 .0019
Biochemical failure 126 73.87 67.20 to 80.55 89 55.95 48.06 to 63.83 ⬍ .0001

Abbreviations: STAD ⫹ RT, short-term androgen deprivation with external-beam radiation therapy; LTAD ⫹ RT, long-term androgen deprivation with external-beam
radiation therapy followed by goserelin.

www.jco.org © 2008 by American Society of Clinical Oncology 2501

Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042

Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
 Horwitz et al

A 1.0 B 1.0
0.9 0.9

0.8 0.8
DSS Non Failure Rate

STAD + RT STAD + RT
LTAD + RT LTAD + RT

Non Failure Rate

0.7 P = .0061 0.7 P = .0019

0.6 0.6

0.5 0.5

0.4 0.4
0.3 0.3

0.2 0.2

0.1 171 160 133 106 75 45 0.1 171 160 133 106 75 45
166 161 142 119 98 58 166 161 142 119 98 58

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10

Time Since Random Assignment (years) Time Since Random Assignment (years)
C 1.0 D 1.0
0.9 0.9

0.8 0.8
Non Failure Rate

0.7 0.7

Survival Rate
0.6 0.6

0.5 0.5

0.4 0.4
STAD + RT STAD + RT
0.3 LTAD + RT 0.3 LTAD + RT
P < .0001 P = .7159
0.2 0.2

0.1 171 77 45 30 22 15 0.1 499 475 418 377 308 227

166 140 76 58 45 28 522 484 439 385 316 250

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10

Time Since Random Assignment (years) Time Since Random Assignment (years)
Fig 2. (A) Overall survival, (B) distant metastasis failure, and (C) biochemical failure) for the subgroup of patients with a Gleason score of 8 to 10 and N0/NX. (D) Overall
survival for the subgroup of patients with a Gleason score of 2 to 7 and N0/NX subgroup. STAD, short-term androgen-deprivation therapy; LTAD, long-term
androgen-deprivation therapy; RT, radiotherapy.

demonstrated the superiority of LTAD, and one of the strengths of the
present analysis is that these results remain valid at 10 years.
RTOG 92-02 showed improved results with LTAD for all end
points with additional follow-up, except OS. Were these benefits ob-
served elsewhere? Bolla et al13 reported results from European Organi-
sation for Research and Treatment of Cancer (EORTC) 22863, where
415 men were randomized between RT alone (70 Gy), and RT with
goserelin starting the first day of treatment and continuing for 3 years.
The differences in clinical outcome persist between the two arms is
similar to the results observed in the present study. LC, PSA DFS, and
OS were 98%, 50%, and 70%, respectively for the combined modality
arm compared with 84%, 20%, and 40%, respectively for the RT alone
arm. These differences were all statistically significant (P ⬍ .0001).
Patients in the control arm in the EORTC 22863 study received no
adjuvant ADT, whereas the patients in the control arm of RTOG 92-02
received 4 months of neoadjuvant and adjuvant ADT. The difference
in the trial design may be one explanation for the failure of RTOG
92-02 to demonstrate an OS benefit for the entire trial cohort. Another
reason may include the age of the patients in this study and the issue of
competing medical comorbidities. It is encouraging and significant
that the difference in CSS persisted with longer follow-up in this study.
The subset analysis performed on patients with a Gleason score of 8 to
10 demonstrated statistically significant improvements in OS and
CSS. Bolla et al14 reported the early results from the successor trial to
EORTC 22863 in 2007. EORTC 22961 was designed to test whether
similar prostate cancer patients would benefit from STAD therapy
instead of 3 years of LTAD therapy. An interim analysis demonstrated
decreased PFS in the STAD therapy arm (68.9% v 81.8% at 5 years).
Biochemical PFS was also improved with LTAD (78.3% v 58.9%).
Preliminary results of this trial support the conclusion observed in the
10-year results of RTOG 92-02, although the survival advantage in the
EORTC study may be explained by the inclusion of patients with more
advanced disease (⫹N2).
As the benefits of ADT have become known, so has the potential
toxicity of this treatment.15,16 Patients treated with LTAD had overall
increased late toxicity. This has been observed in the Fox Chase expe-
rience reported by Feigenberg et al. The authors reported the results
for 1,204 patients were treated for prostate cancer with three-
dimensional conformal RT (3DCRT). STAD therapy was adminis-
tered to 140 patients, 119 patients received LTAD therapy and 945

2502 © 2008 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042

Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
 Long-Term Results of RTOG 92-02
patients received no ADT. Independent predictors of grade 2 or higher
GI morbidity in the multivariable analysis were the use of AD (P ⫽
.0079), higher total radiation dose (P ⬍ .0001), the lack of a rectal
shield (P ⫽ .0003), and older age (P ⫽ .0009). The 5-year actuarial risk
of grade 2 or higher GI morbidity was 17% for no AD versus 18% for
STAD and 26% for LTAD (P ⫽ .017).17 Recent studies have docu-
mented the potential long-term cardiovascular risks associated with
ADT.16 In RTOG 92-02, only three grade 3 cardiovascular events were
documented, one in the STAD ⫹ RT arm and two in the LTAD ⫹
RT arm.
Although answering many questions, the results in this study
leave several questions unanswered. These long-term results from
RTOG 92-02 strongly support the recommendation that LTAD ther-
apy used in conjunction with RT for locally advanced nonmetastatic
prostate cancer is superior to STAD ⫹ RT. However, an OS advantage
was not observed for the entire study group. Is it possible the excess
toxicity for the LTAD arm negated a positive effect on CSS? Was the
study underpowered to detect a survival difference? The arms were
evenly balanced, and the accrual goals were exceeded. We do not
believe that excess toxicity is a probable reason. One potential expla-
nation for the lack of an OS advantage is that the follow-up interval is
not yet long enough for the patients with lower Gleason scores. Pa-
tients with a Gleason score of 7 and lower represented more than 70%
of the patient population, and so despite the locally advanced nature of
the patient sample, the number of patients with a Gleason score of 8 to
10 at highest risk for DS mortality is relatively small. RTOG 92-02 has
enough follow-up to observe a survival advantage in the high-grade
patients because their life expectancy would be less, but there may not
be enough follow-up time to see a difference in the low-grade patients.
The significance of RT dose in this setting is also unclear. Other
randomized, prospective studies have demonstrated the importance
of dose, but this has not been formally tested with STAD or LTAD
therapy in North America. The University of Texas M.D. Anderson
Cancer Center trial reported by Pollack et al18 is the most mature
randomized prospective prostate cancer dose escalation trial. In this
study, 301 men were randomly assigned to between 70 Gy and 78 Gy.
With 60 months of median follow-up, patients with an initial pretreat-
ment PSA greater than 10 ng/mL experienced the greatest benefit,
whereas the additional 8 Gy had no effect on the more favorable
patients (pretreatment PSA ⱕ 10 ng/mL). In the second randomized
prospective trial reported by Zietman et al,19 393 patients with stage
T1b/T2b prostate cancer and PSA levels less than 15 ng/mL were
randomly assigned to a total dose of either 70.2 Gy (conventional
dose) or 79.2 Gy (high dose) with a combination of conformal photon
and proton beams. The 5-year rates of biochemical control were
61.4% for conventional-dose and 80.4% for high-dose therapy (P ⬍
.001). The advantage to high-dose therapy was observed in both the
low-risk and the higher-risk subgroups.
The long-term results for RTOG 92-02 presented in this report
demonstrated an OS benefit for prostate cancer patients with a Glea-
son score of 8 to 10 treated with prolonged LTAD ⫹ RT. LTAD ⫹ RT
www.jco.org
Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
should be regarded as the standard treatment for patients with pros-
tate cancer patients with a Gleason score of 8 to 10. Subsequent RTOG
trials have adopted this regimen of LTAD ⫹ RT as the control arm in
randomized clinical trials evaluating new therapies for patients with
high-risk prostate cancer. Two successor trials introduced cytotoxic
chemotherapy after the RT has been completed, with the strategy of
intensifying systemic therapy. RTOG 99-02 utilized a three-drug reg-
imen of paclitaxel, etoposide, and estramustine, in addition to LTAD
⫹ RT.20 RTOG 05-21 utilizes a similar schema, but with docetaxel-
based chemotherapy. Updated RT techniques, such as intensity-
modulated radiation therapy, and doses are utilized in contemporary
trials. The fact that the experimental arm of RTOG 92-02 is still being
utilized as the basis of the control arm in ongoing clinical trials for
high-risk, poorly differentiated prostate cancer provides additional
testimony to the significance of the results of this trial.
The long-term results of RTOG 92-02 demonstrate that 24
months of ADT after total androgen suppression and RT is superior to
total androgen suppression and RT alone in all end points except
survival for patients with locally advanced nonmetastatic prostate
cancer. In a hypothesis-generating postrandomization analysis, an
overall survival benefit was observed in patients with tumors hav-
ing a Gleason score of 8 to 10. No change in significant results from
5-year to 10-year follow-up was observed, proving the durability of
these results.
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS
OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Eric M. Horwitz, Gerald E. Hanks,
William U. Shipley
Administrative support: Kyounghwa Bae, David J. Grignon,
Howard M. Sandler
Provision of study materials or patients: Gerald E. Hanks, Arthur
Porter, Varagur Venkatesan, Colleen A. Lawton, Seth A. Rosenthal,
Howard M. Sandler, William U. Shipley
Collection and assembly of data: Eric M. Horwitz, Kyounghwa Bae,
Gerald E. Hanks, David J. Grignon, Harmar D. Brereton, Varagur
Venkatesan, Seth A. Rosenthal, Howard M. Sandler, William U. Shipley
Data analysis and interpretation: Eric M. Horwitz, Kyounghwa Bae,
Gerald E. Hanks, Arthur Porter, David J. Grignon, Harmar D. Brereton,
Varagur Venkatesan, Colleen A. Lawton, Seth A. Rosenthal, Howard M.
Sandler, William U. Shipley
Manuscript writing: Eric M. Horwitz, Kyounghwa Bae, Gerald E. Hanks,
Seth A. Rosenthal, Howard M. Sandler
Final approval of manuscript: Eric M. Horwitz, Kyounghwa Bae, Gerald
E. Hanks, Arthur Porter, David J. Grignon, Harmar D. Brereton, Varagur
Venkatesan, Colleen A. Lawton, Seth A. Rosenthal, Howard M. Sandler,
William U. Shipley
© 2008 by American Society of Clinical Oncology 2503

REFERENCES
1. Zagars GK, Johnson DE, von Eschenbach AC,
et al: Adjuvant estrogen following radiation therapy
for stage C adenocarcinoma of the prostate: Long-
term results of a prospective randomized study. Red
Journal 14:1085-1091, 1988
2. Pilepich MV, Winter K, Lawton CA, et al:
Androgen suppression adjuvant to definitive radio-
therapy in prostate carcinoma: Long-term results of
phase III RTOG 85-31. Int J Radiat Oncol Biol Phys
61:1285-1290, 2005
3. Pilepich MV, Winter K, John MJ, et al: 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 Radiat Oncol Biol Phys 50:1243-1252,
2001
4. Hanks GE, Pajak TF, Porter A, et al: Phase III
trial of long-term adjuvant androgen deprivation after
neoadjuvant hormonal cytoreduction and radiother-
apy in locally advanced carcinoma of the prostate:
The Radiation Therapy Oncology Group protocol
92-02. J Clin Oncol 21:3972-3978, 2003
5. American Joint Committee on Cancer: Pros-
tate, in Beahrs OH, Henson DE, Hutter RVP, et al
(eds): Manual for the Staging of Cancer (ed 4).
Philadelphia, PA, JB Lippincott, 1992, pp 181-186
6. Zelen M: The randomization and stratification
of patients to clinical trials. J Chronic Dis 27:365-
375, 1974
We thank all of the patients, physicians, and support staff whose care and effort made the successful completion of this study possible.
2504 © 2008 by American Society of Clinical Oncology
Downloaded from ascopubs.org by ASTRAZENECA on December 16, 2017 from 202.126.172.042
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
Horwitz et al
7. American Society for Therapeutic Radiology
and Oncology Consensus Panel: Consensus state-
ment: Guidelines for PSA following radiation ther-
apy. Red Journal 37:1035-1041, 1997
8. Kaplan EL, Meier P: Nonparametric estima-
tion from incomplete observations. J Am Stat Assoc
53:457-481, 1958
9. Kalbfleisch JD, Prentice RL: The Statistical
Analysis of Failure Time Data. New York, NY, John
Wiley & Sons, 1980
10. Mantel N, Haenszel W: Statistical aspects of
the analysis of data from retrospective studies of
disease. J Natl Cancer Inst 22:719-748, 1959
11. Gray RJ: A class of K-sample tests for com-
paring the cumulative incidence of a competing risk.
Ann Stat 16:1141-1154, 1988
12. Cox JD, Stetz J, Pajak TF: Toxicity criteria of
the Radiation Therapy Oncology Group (RTOG) and
the European Organization for Research and Treat-
ment of Cancer (EORTC). Int J Radiat Oncol Biol
Phys 31:1341-1346, 1995
13. Bolla M, Collette L, Blank L, et al: Long-term
results with immediate androgen suppression and
external irradiation in patients with locally advanced
prostate cancer (an EORTC study): A phase III
randomised trial. Lancet 360:103-106, 2002
14. Bolla M, van Tienhoven G, De Reijke TM, et al:
Concomitant and adjuvant androgen deprivation
(ADT) with external beam irradiation (RT) for locally
advanced prostate cancer: 6 months versus 3 years
ADT—Results of the randomized EORTC phase III
■ ■ ■
Acknowledgment
trial 22961. J Clin Oncol 25:238s, 2007 (suppl; abstr
5014)
15. Rosenthal SA, Linstadt DE, Leibenhaut MH,
et al: Flutamide-associated liver toxicity during treat-
ment with total androgen suppression and radiation
therapy for prostate cancer. Radiology 199:451-455,
1996
16. D’Amico AV, Denham JW, Crook J, et al:
Influence of androgen suppression therapy for pros-
tate cancer on the frequency and timing of fatal
myocardial infarctions. J Clin Oncol 25:2420-2425,
2007
17. Feigenberg SJ, Hanlon AL, Horwitz EM, et al:
Long-term androgen deprivation increases grade 2
and higher late morbidity in prostate cancer patients
treated with three-dimensional conformal radiation
therapy. Int J Radiat Oncol Biol Phys 62:397-405,
2005
18. Pollack A, Zagars GK, Antolak JA, et al: Pros-
tate biopsy status and PSA nadir level as early
surrogates for treatment failure: Analysis of a pros-
tate cancer randomized radiation dose escalation
trial. Int J Radiat Oncol Biol Phys 54:677-685, 2002
19. Zietman AL, DeSilvio ML, Slater JD, et al:
Comparison of conventional-dose vs high-dose con-
formal radiation therapy in clinically localized adeno-
carcinoma of the prostate: A randomized controlled
trial. JAMA 294:1233-1239, 2005
20. Sandler H, DeSilvio M, Pienta K, et al: Prelim-
inary analysis of RTOG 92-02: Increased toxicity
observed with the use of adjuvant chemotherapy.
Int J Radiat Oncol Biol Phys 63:123, 2005
JOURNAL OF CLINICAL ONCOLOGY