Articles

High-dose radiotherapy with short-term or long-term

androgen deprivation in localised prostate cancer
(DART01/05 GICOR): a randomised, controlled, phase 3 trial
Almudena Zapatero, Araceli Guerrero, Xavier Maldonado, Ana Alvarez, Carmen Gonzalez San Segundo, Maria Angeles Cabeza Rodríguez,
Victor Macias, Agustí Pedro Olive, Francesc Casas, Ana Boladeras, Carmen Martín de Vidales, Maria Luisa Vazquez de la Torre, Salvador Villà,
Aitor Perez de la Haza, Felipe A Calvo

Summary
Lancet Oncol 2015; 16: 320–27 Background The optimum duration of androgen deprivation combined with high-dose radiotherapy in prostate cancer
Published Online remains undefined. We aimed to determine whether long-term androgen deprivation was superior to short-term
February 19, 2015 androgen deprivation when combined with high-dose radiotherapy.
http://dx.doi.org/10.1016/
S1470-2045(15)70045-8
Methods In this open-label, multicentre, phase 3 randomised controlled trial, patients were recruited from ten
This online publication has
been corrected. The corrected
university hospitals throughout Spain. Eligible patients had clinical stage T1c–T3b N0M0 prostate adenocarcinoma
version first appeared at with intermediate-risk and high-risk factors according to 2005 National Comprehensive Cancer Network criteria.
thelancet.com/oncology on Patients were randomly assigned (1:1) using a computer-generated randomisation schedule to receive either 4 months
May 28, 2015
of androgen deprivation combined with three-dimensional conformal radiotherapy at a minimum dose of 76 Gy
See Comment page 244 (range 76–82 Gy; short-term androgen deprivation group) or the same treatment followed by 24 months of adjuvant
Hospital Universitario de la androgen deprivation (long-term androgen deprivation group), stratified by prostate cancer risk group (intermediate
Princesa, Madrid, Spain
(A Zapatero PhD,
risk vs high risk) and participating centre. Patients assigned to the short-term androgen deprivation group received
C Martín de Vidales PhD); 4 months of neoadjuvant and concomitant androgen deprivation with subcutaneous goserelin (2 months before and
Hospital Son Espases, Palma de 2 months combined with high-dose radiotherapy). Anti-androgen therapy (flutamide 750 mg per day or bicalutamide
Mallorca, Spain 50 mg per day) was added during the first 2 months of treatment. Patients assigned to long-term suppression
(A Guerrero MD); Hospital
Universitari Vall d’Hebron,
continued with the same luteinising hormone-releasing hormone analogue every 3 months for another 24 months.
Barcelona, Spain The primary endpoint was biochemical disease-free survival. Analysis was by intention to treat. This study is registered
(X Maldonado MD); Hospital with ClinicalTrials.gov, number NCT02175212.
General Universitario Gregorio
Marañón, Madrid, Spain
(A Alvarez MD,
Findings Between Nov 7, 2005, and Dec 20, 2010, 178 patients were randomly assigned to receive short-term androgen
C Gonzalez San Segundo PhD, deprivation and 177 to receive long-term androgen deprivation. After a median follow-up of 63 months (IQR 50–82),
Prof F A Calvo PhD); Hospital 5-year biochemical disease-free survival was significantly better among patients receiving long-term androgen
Universitario 12 de Octubre, deprivation than among those receiving short-term treatment (90% [95% CI 87–92] vs 81% [78–85]; hazard ratio [HR]
Madrid, Spain
(M A Cabeza Rodríguez MD);
1·88 [95% CI 1·12–3·15]; p=0·01). 5-year overall survival (95% [95% CI 93–97] vs 86% [83–89]; HR 2·48 [95% CI
Hospital General de Catalunya, 1·31–4·68]; p=0·009) and 5-year metastasis-free survival (94% [95% CI 92–96] vs 83% [80–86]; HR 2·31 [95% CI
Sant Cugat del Vallès and 1·23–3·85]; p=0·01) were also significantly better in the long-term androgen deprivation group than in the short-term
Hospital Universitario de androgen deprivation group. The effect of long-term androgen deprivation on biochemical disease-free survival,
Salamanca, Salamanca, Spain
(V Macias PhD); Hospital Plató,
metastasis-free survival, and overall survival was more evident in patients with high-risk disease than in those with
Barcelona, Spain low-risk disease. Grade 3 late rectal toxicity was noted in three (2%) of 177 patients in the long-term androgen
(A Pedro Olive MD); Hospital deprivation group and two (1%) of 178 in the short-term androgen deprivation group; grade 3–4 late urinary toxicity
Clinic, Barcelona, Spain
was noted in five (3%) patients in each group. No deaths related to treatment were reported.
(F Casas PhD); Institut Català
d’Oncologia, Barcelona, Spain
(A Boladeras MD, S Villà PhD); Interpretation Compared with short-term androgen deprivation, 2 years of adjuvant androgen deprivation combined
Hospital Do Meixoeiro, Vigo, with high-dose radiotherapy improved biochemical control and overall survival in patients with prostate cancer,
Spain
particularly those with high-risk disease, with no increase in late radiation toxicity. Longer follow-up is needed to
(M L Vazquez de la Torre MD);
and Apices Data Management determine whether men with intermediate-risk disease benefit from more than 4 months of androgen deprivation.
and Biostatistics Centre,
Madrid, Spain Funding Spanish National Health Investigation Fund, AstraZeneca.
(A Perez de la Haza MSc)
Correspondence to:
Dr Almudena Zapatero,
Introduction radiotherapy have enabled dose escalation with substantial
Radiation Oncology Department, Several randomised trials done during the past two decades improvements in biochemical outcome.10–15 Because
Hospital Universitario de la have shown a significant improvement in biochemical randomised trials showing a significant clinical benefit
Princesa, Health Research control and overall survival with the combination of with androgen deprivation and radiotherapy use exclusively
Institute IIS, 28006 Madrid,
androgen deprivation and conventional-dose radiotherapy conventional dose levels of 65–70 Gy, the optimum
Spain
almudena.zapatero@salud. (≤70 Gy) in patients with high-risk1–7 and intermediate-risk duration of androgen deprivation to use in combination
madrid.org prostate cancer.8,9 Similarly, advances in external beam with high-dose radiotherapy remains unresolved.16 Thus,

320 www.thelancet.com/oncology Vol 16 March 2015

 Articles

we aimed to determine whether long-term androgen programme (version 9.1) and an interactive web response
deprivation was superior to short-term androgen system. The research coordinator then faxed the
deprivation for patients receiving high-dose radiotherapy. investigator of the participating centres, reported the
number, and informed the investigator about the
Methods assigned treatment. No blocks were used. Neither the
Study design and participants participants nor the investigators were masked to
In this open-label, multicentre, phase 3 randomised treatment allocation, because blinding was not feasible.
controlled trial, patients were recruited from ten university
hospitals throughout Spain (appendix). Patients aged Procedures See Online for appendix
18 years or older with histologically confirmed clinical stage Radiotherapy was administered with three-dimensional
T1c–T3b adenocarcinoma of the prostate, N0, M0 with conformal radiotherapy techniques done with a six-field
intermediate-risk and high-risk factors according to isocentric beam setup based on a CT scan. The target
National Comprehensive Cancer Network criteria, serum volume included the prostate and the seminal vesicles.
PSA concentration less than 100 ng/mL, a Karnofsky In view of the controversy regarding the role of
performance score of 70 or greater, and a life expectancy of prophylactic pelvic radiotherapy and the absence of
more than 5 years were eligible for inclusion. Patients with definitive data, elective pelvic radiotherapy was left to the
T4 tumours, regional lymph-node involvement, distant criteria of each participating centre. The radiation dose
metastatic disease, previous pelvic radiotherapy or surgery, was specified at the intersection of the beam axes
neoadjuvant hormonal treatment for more than 3 months, (isocentre) according to the guidelines of the International
or concomitant use of chemotherapy were excluded from Commission on Radiation Units.17 Treatment was
the trial. Severe psychiatric or medical conditions that could provided in daily 2 Gy fractions at a minimum dose of
hamper both treatment and follow-up and major 76 Gy (range 76–82 Gy). The median isocentre radiation
malignancies were also considered exclusion criteria. dose to the prostate was 78 Gy for both groups, and the
Patients with a previous history of cancer that had been corresponding dose to the seminal vesicles was 56 Gy.
controlled for 5 years or more and patients with cutaneous Beams were shaped with multileaf collimators or
basal cell or squamous-cell carcinoma were not excluded. customised shaped blocks, and treatment was delivered
Pretreatment evaluation included a digital rectal with 6–18 MV photons. Dose constraints for normal
examination, transrectal ultrasound, abdominal-pelvic CT, tissues have been described elsewhere.18 Treatment was
and bone scan. Review of pathology specimens was not verified with electronic portal image devices according to
centralised. the quality assurance protocols at each centre.
The study was approved by the independent review The hormone therapy regimen was based on that used
board at each participating centre and conducted in the RTOG 9202 trial4 and on the usual clinical practice
according to the provisions of the Declaration of Helsinki in Spain. Patients assigned to the short-term androgen
and the Good Clinical Practice Guidelines of the deprivation group received 4 months of neoadjuvant and
International Conference on Harmonization. All patients concomitant androgen deprivation with subcutaneous
provided written informed consent before participating goserelin (in both groups, goserelin was given
in the trial. The full study protocol can be viewed online. subcutaneously at 3·6 mg; after 1 month, it was given For the protocol see
subcutaneously at 10·8 mg subcutaneously every http://www.gicor.es/invest003/
resumen.pdf
Randomisation and masking 3 months). Treatment started 2 months before high-dose
Before randomisation, patients were screened to verify radiotherapy, and was then given for 2 months combined
the study selection criteria and stratified based on with radiotherapy. Anti-androgen therapy (flutamide
prostate cancer risk subgroups (intermediate risk: T1–T2 750 mg per day or bicalutamide 50 mg per day) was
with a Gleason score of 7, or PSA concentration of added during the first 2 months of treatment. Patients
10–20 ng/mL, or both; high risk: T3 with Gleason score assigned to long-term suppression continued with the
of 8–10, or PSA concentration of >20 ng/mL, or both) same luteinising hormone-releasing hormone analogue
and the participating centre. every 3 months for another 24 months.
Patients were randomly assigned (1:1) to receive Follow-up visits were at intervals of 3 months after
4 months of neoadjuvant and concomitant androgen radiotherapy during the first year, every 6 months for
deprivation combined with three-dimensional conformal 5 years, and yearly thereafter. PSA concentrations, serum
radiotherapy (short-term androgen deprivation group) or testosterone concentration, and a complete blood count
the same treatment followed by 24 months of adjuvant were obtained at every visit. Specifically, 12 PSA
androgen deprivation (long-term androgen deprivation measurements were obtained during the first 5 years of
group). Randomisation was centralised at the Health follow-up to enable systematic assessment of the lowest
Research Institute of Hospital Universitario de la PSA value achieved (PSA nadir) after completion of
Princesa (Madrid, Spain). After eligibility screening, the treatment. Imaging (abdominal-pelvic CT and bone scan)
research coordinator assigned eligible patients using a was repeated in cases in which clinical or biochemical
randomisation schedule generated by means of the SAS progression was suspected. Decisions on salvage therapy

www.thelancet.com/oncology Vol 16 March 2015 321

 Articles

STAD (n=178) LTAD (n=177) 498 assessed for eligibility

Age (years) 72 (54–85) 71 (56–82)
Risk subgroup 121 refused to participate
Intermediate 81 (45%) 85 (48%) 15 did not have pretreatment
data
High 97 (55%) 92 (52%)
T stage
T1 42 (24%) 38 (21%) 362 registered
T2 103 (58%) 100 (57%)
T3 33 (18%) 39 (22%) 7 did not meet inclusion criteria
3 had inadequate disease
PSA (ng/mL) 11·0 (3·4–66) 11·1 (3·1–72)
stage
<10 74 (42%) 80 (45%) 1 histopathological finding
10–20 64 (36%) 61 (35%) 1 synchronous malignancy
2 neoadjuvant hormonal
>20 40 (22%) 36 (20%) treatment for more than
Gleason score 3 months

≤6 30 (17%) 21 (12%)
7 103 (58%) 110 (62%) 355 randomised
8–10 45 (25%) 46 (26%)
Positive biopsy samples 4 (1–16) 4 (1–13)
Duration of androgen 2·2 (0·3– 7·7) 2·2 (0·5 –7·8)
178 allocated to STAD 177 allocated to LTAD
deprivation before 169 received allocated
170 received allocated
randomisation (months) intervention
intervention
Prostate radiotherapy dose (Gy) 78·0 (64·0–82·2) 78·0 (30·6–88·4) 8 did not receive allocated 8 did not receive allocated
intervention intervention
<78 51 (29%)* 43 (25%)†
2 investigator decision 4 patient decision
≥78 122 (71%) 128 (75%) 2 patient decision 2 major deviation
4 major deviation 2 intolerance
Pelvic radiotherapy
Yes 28 (16%) 20 (12%)
No 145 (84%)* 148 (86%)† 5 lost to follow-up 3 lost to follow-up
Missing 0 3 (2%)
Pelvic radiotherapy by risk
178 analysed 177 analysed
group
Intermediate risk 2 (7%) 2 (10%)
Figure 1: Trial profile
High risk 26 (93%) 18 (90%)
STAD=short-term androgen deprivation. LTAD=long-term androgen deprivation.
Pelvic radiotherapy dose (Gy) 46·0 (34·7–56·0) 46·0 (30·6–46·0)
Salvage treatment
Androgen deprivation 14 (8%) 8 (5%) cause, or censoring at the date of the last contact. The
Other treatments 1 (1%) 0 RTOG-ASTRO Phoenix Consensus Conference
Time from randomisation to 25·9 (7·1–71·4) 54·7 (41·6–66·0) definition20 (an increase in the PSA concentration of
salvage treatment (months)
≥2 ng/mL above the nadir) was used to define biochemical
Data are n (%) or median (range). Intermediate-risk disease was defined as a clinical failure. Secondary endpoints included overall survival,
stage of T1 to T2 with a Gleason score of 7, or a PSA concentration of 10–20 ng/mL, distant metastasis-free survival, and cause-specific
or both. High-risk disease was defined as a clinical stage of T3, Gleason score of
survival. Overall survival was defined as the time from
8–10, PSA concentration of more than 20 ng/mL, or both. STAD=short-term
androgen deprivation. LTAD=long-term androgen deprivation. PSA=prostate- randomisation to death from any cause or censoring at the
specific antigen. *173 evaluable patients. †171 evaluable patients. date of the last contact. Metastasis-free survival was
defined as the time from randomisation to the occurrence
Table 1: Patient baseline clinical and treatment characteristics
of metastatic disease (documented by imaging studies) or
death from any cause, and patients who received salvage
were based on the criteria of each participating centre. therapy were censored. Cause-specific survival included
Baseline and annual or every 2 year bone mineral all deaths from prostate cancer or treatment complications,
densitometry was recommended but not mandatory. and deaths from unknown causes in patients with either
Radiation-related complications were assessed with active cancer or a previously documented relapse. Cause
EORTC/RTOG radiation morbidity scoring criteria.19 of death was recorded by the treating physician and was
subject to central independent review.
Outcomes
The primary endpoint was biochemical disease-free Statistical analysis
survival, defined as the time from randomisation to On the basis of previous studies, we estimated that
progression of biochemical disease, or death from any biochemical disease-free survival at 5 years in the

322 www.thelancet.com/oncology Vol 16 March 2015

 Articles

short-term androgen deprivation group would be 60% in

the high-risk group and 70% in the intermediate-risk A
100 LTAD
group. The addition of 2 years of androgen deprivation
STAD
was expected to increase this to 75% in the high-risk
group and 85% in the intermediate-risk group. Assuming

Biochemical disease-free survival (%)

80
that the risks between the two groups were proportional
and accepting a two-tailed α risk of 0·05 with a power
(1–β) of 0·80, we estimated that we would need to enrol 60

307 patients, roughly equally distributed between the two

subgroups. Assuming a loss to follow-up of 15%, the 40
estimated required sample size to be 350 patients. All
analyses were done on an intention-to-treat basis, with
patients analysed according to the treatment group. No 20
formal stopping rules were specified in the protocol.
The χ² test was used to evaluate differences in toxicities
0
and the overall worst degree of toxicity. Survival analyses 0 20 40 60 80 100 120
were done with Kaplan-Meier curves21 and the log-rank Number at risk
LTAD 173 166 98 46 0 0
(Mantel-Cox) test was used to compare survival between STAD 170 152 86 42 2 0
groups.22
Univariate analysis was done to assess the relation B
between potential prognostic factors with biochemical 100
progression-free survival. Variables included in the
analysis were patient age, T stage (T3 vs T1–2),
80
pretreatment PSA (>20 ng/mL vs ≤20 ng/mL), Gleason
score (>7 vs ≤7), number of positive prostate biopsies,
Overall survival (%)

treatment group (short-term vs long-term androgen 60

deprivation), radiation dose, pelvic radiotherapy (yes or
no), and PSA nadir. Patient age, radiation dose, and PSA
40
nadir were analysed as continuous variables. PSA nadir
was treated as a time-dependent covariable. Variables with
a statistical significance less than 0·25 were taken into 20
account in a multivariate Cox regression analysis.23 The
Wald forward method was used to select variables in the
Cox proportional hazard model. All hazard ratios (HRs) 0
0 20 40 60 80 100 120
were calculated with Cox proportional hazard models and Number at risk
expressed relative to the control group. LTAD 173 167 102 50 0 0
STAD 174 162 94 46 2 0
We did a planned subgroup analysis of the efficacy
endpoints within the prostate risk categories used in C
stratification—ie, intermediate-risk and high-risk 100
prostate cancer. A forest plot was generated to explore the
treatment effects across risk groups.
Analyses were done with SPSS for Windows version 80
Metastasis-free survival (%)

19. This trial is registered with ClinicalTrials.gov, number

NCT02175212, and the EU Clinical Trials Register, 60
number 2005-000417-36.

Role of the funding source 40

AZ was the sponsor of the trial, because in 2004 GICOR
still had no legal entity. Further funding was provided by
20
the Spanish National Health Investigation Fund and
AstraZeneca. Neither of these funding bodies had a role
in trial design, data collection, statistical analysis, or 0
0 20 40 60 80 100 120
Time (months)
Figure 2: Kaplan-Meier estimates of biochemical disease-free survival (A),
Number at risk
overall survival (B), and metastasis-free survival (C) LTAD 172 167 88 45 0 0
STAD=short-term androgen deprivation. LTAD=long-term androgen STAD 170 154 86 41 2 0
deprivation.

www.thelancet.com/oncology Vol 16 March 2015 323

 Articles

170 (96%) patients in the short-term androgen deprivation

Number of events 5-year rate (%, 95% CI) Hazard ratio p
(95% CI) value group and in 169 (95%) in the long-term androgen
N STAD LTAD STAD LTAD deprivation group (figure 1).
Biochemical disease-free survival Median follow-up was 63 months (IQR 50–82)—
High risk 189 23 13 76 (71–80) 88 (84–92) 1·91 (0·97–3·77) 0·054 61 months (IQR 50–81) for the short-term androgen
Intermediate risk 166 14 8 88 (84–91) 92 (89–95) 1·82 (0·76–4·33) 0·174 deprivation group, and 64 months (IQR 49–83) for the
Overall survival long-term androgen deprivation group. 5-year
High risk 189 17 5 82 (77–86) 96 (94–98) 3·43 (1·26–9·32) 0·015 biochemical disease-free survival was 90% (95% CI
Intermediate risk 166 10 6 91 (88–95) 94 (91–96) 1·67 (0·61–4·60) 0·318 87–92) for patients receiving long-term androgen
Metastasis-free survival
deprivation compared with 81% (78–85) for those
High risk 189 20 9 79 (74–83) 94 (91–96) 2·27 (1·04–5·01) 0·041
receiving short-term androgen deprivation (HR 1·88,
Intermediate risk 166 13 6 89 (85–93) 94 (91–96) 2·14 (0·81–5·66) 0·124
95% CI 1·12–3·15; p=0·01; figure 2A). In the subgroup
0·1 1 10 analysis by prostate cancer risk, the benefit in 5-year
biochemical disease-free survival was more evident in
Favours STAD Favours LTAD the high-risk population than in the intermediate-risk
Figure 3: Effects of duration of androgen deprivation stratified by risk group population (figure 3).
STAD=short-term androgen deprivation. LTAD=long-term androgen deprivation. 5-year overall survival with long-term androgen
deprivation was 95% (95% CI 93–97) compared with
86% (83–89) with short-term androgen deprivation
Univariate analysis Multivariate analysis
(HR 2·48 [95% CI 1·31–4·68]; p=0·009; figure 2B). In
HR (95% CI) p value HR (95% CI) p value the subgroup analysis by prostate cancer risk, the
Patient age 0·946 (0·907–0·988) 0·012 0·941 (0·900–0·985) 0·008 benefit in overall survival with long-term deprivation
Treatment group (STAD vs 1·881 (1·101–3·215) 0·021 2·171 (1·178–4·002) 0·013 was more evident for patients with high-risk disease,
LTAD) but not for those with intermediate-risk disease
Radiation dose 0·950 (0·902–1·000) 0·051 0·943 (0·899–0·988) 0·014 (figure 3). 5-year metastasis-free survival was 94%
PSA nadir 6·211 (2·296–16·799) <0·001 5·123 (1·399–18·757) 0·014 (95% CI 92–96) in the long-term androgen deprivation
Pelvic radiotherapy 0·734 (0·380–1·419) 0·35 0·953 (0·414–2·197) 0·91 group compared with 83% (80–86) in the short-term
T stage (T3 vs T1–2) 1·243 (0·681–2·270) 0·47 1·552 (0·699–3·445) 0·28 androgen deprivation group (HR 2·31 [95% CI
Pre-treatment PSA 1·794 (1·037–3·105) 0·037 1·841 (0·809–4·187) 0·14 1·23–3·85]; p=0·01; figure 2C). This benefit in
(>20 ng/mL vs ≤20 ng/mL)
metastasis-free survival was greater in patients with
Gleason score (>7 vs ≤7) 1·287 (0·731–2·267) 0·38 1·398 (0·626–3·120) 0·41
high-risk disease than in those with intermediate-risk
Number of positive biopsy 1·064 (0·967–1·172) 0·21 1·067 (0·966–1·179) 0·19 disease (figure 3).
samples
The results of the multivariate analysis showed that the
HR=hazard ratio. STAD=short-term androgen deprivation. LTAD=long-term androgen deprivation. PSA=prostate- independent prognostic factors affecting biochemical
specific antigen. failure were patient age, radiation dose, PSA nadir, and
Table 2: Univariate and multivariate analysis of biochemical progression-free survival treatment group (table 2).
At the date of analysis, 38 (11%) of 355 patients had
died; 27 in the short-term androgen deprivation group
interpretation of the results. The drugs were not supplied and 11 in the long-term androgen deprivation group. The
by the manufacturer. AZ had full access to the data and cause of death was prostate cancer in only five patients,
responsibility for the decision to submit for publication. all of whom were in the short-term androgen deprivation
group. 17 patients died of cancer, but not of the prostate
Results (14 [8%] in the short-term group vs three [2%] in the long-
Between Nov 7, 2005, and Dec 20, 2010, 498 men were term group), eight of cardiac failure (three [2%] vs five
screened (121 refused to participate and 15 did not have [3%]), and eight of other causes (five [3%] vs three [2%]).
pretreatment data), and 362 patients were registered. Of 61 cardiovascular events occurred: 36 (20%) in
these, seven did not meet the inclusion criteria because of 177 patients in the long-term androgen deprivation group
inadequate disease stage (three patients), histopathology and 25 (14%) in 178 in the short-term group, but only
findings (one patient), synchronous malignancies (one eight (2%) were fatal (five [3%] in the long-term group
patient), and patient refusal of allocated treatment (two and three [2%] in the short-term group). Late rectal
patients). The final trial population thus consisted of toxicity of grade 2 or worse occurred in 21 (12%) of
355 men, of whom 178 were randomly assigned to the 177 patients in the long-term androgen deprivation group
short-term androgen deprivation group and 177 to the and 15 (8%) of 178 in the short-term group; late urinary
long-term androgen deprivation group. The treatment toxicity of grade 2 or worse occurred in 18 (10%) patients
groups were well balanced in terms of demographic, in the long-term group and in 17 (10%) in the short-term
tumour-related, and treatment characteristics (table 1). group (table 3). Three (2%) patients in the long-term
Adherence to the treatment protocol was confirmed in androgen deprivation group and two (1%) patients in the

324 www.thelancet.com/oncology Vol 16 March 2015

 Articles

short-term androgen deprivation group had grade 3 late

Grade 1 Grade 2 Grade 3 Grade 4 Grade 5
rectal complications, whereas five (3%) in the long-term
group and five (3%) in the short-term group had LTAD (n=177)
grade 3–4 late urinary complications (table 3). We did not Rectal 39 (22%) 18 (10%) 3 (2%) 0 0
note significant differences in rectal complications Urinary 32 (18%) 13 (7%) 3 (2%) 2* (1%) 0
(p=0·54) or urinary complications (p=1·00) between the STAD (n=178)
treatment groups. Rectal 35 (20%) 13 (7%) 2 (1%) 0 0
Urinary 31 (17%) 12 (7%) 4 (2%) 1* (<1%) 0
Discussion Data are n (%). All complications were resolved except two cases of urethral stenosis
The results of this trial show that long-term androgen (one in the LTAD group and one in the STAD group) and one case of urinary
deprivation, in combination with high-dose radiotherapy, incontinence in the STAD group. LTAD=long-term androgen deprivation.
STAD=short-term androgen deprivation. *Urethral stenosis.
significantly improved biochemical disease-free survival
for men with localised prostate cancer compared with Table 3: Chronic late radiation toxicity
short-term androgen deprivation and high-dose
radiotherapy. Subgroup analyses suggest that the greatest
clinical benefit with long-term androgen deprivation was Panel: Research in context
in the high-risk subgroup of patients. This benefit was
accompanied by a significant improvement in the Systematic review
secondary endpoints of overall survival and metastasis- We searched PubMed and Medline between Jan 1, 1988, and
free survival, with no significant increase in late radiation Dec 30, 2003, using the search terms “randomized trial”,
toxicity (panel). Although the results of our trial are “androgen deprivation therapy”, “androgen suppression
encouraging, whether high-dose radiotherapy is necessary therapy”, “hormone therapy”, “dose escalation radiotherapy”,
to prolong survival in men with high-risk prostate cancer and “conformal radiotherapy”. Only English language articles
given long-term androgen deprivation remains to be were included.
determined. Several randomised trials done during the past two decades
Although it is common practice for patients with high- have shown that androgen deprivation combined with
risk prostate cancer to receive 2–3 years of adjuvant conventional-dose radiotherapy improves overall survival,
androgen deprivation and for patients with intermediate- mainly in patients with intermediate-risk and high-risk
risk prostate cancer to receive 4–6 months of adjuvant prostate cancer.1,7 Similarly, clinical outcomes have also
androgen deprivation, the optimum duration of androgen improved substantially with high-dose radiotherapy.10
deprivation, especially in the setting of high-dose However, in view of the absence of specific randomised trials,
radiotherapy at 76 Gy or higher, has not been established. the optimum duration of androgen deprivation remained
Furthermore, results from non-randomised studies unresolved in the era of dose-escalated radiotherapy.16,24
assessing the role and optimum duration of androgen
deprivation alongside high-dose radiotherapy have been Interpretation
mixed (table 4). To our knowledge, this is the first randomised trial to report
Our findings indicate that the effect of long-term that long-term androgen deprivation is superior to short-
androgen deprivation on biochemical disease-free term androgen deprivation in patients given high-dose
survival and overall survival was more evident in patients radiotherapy in terms of biochemical control and overall
with high-risk disease than in those with low-risk disease. survival, particularly in men with high-risk prostate cancer.
The role of androgen deprivation in the management of The optimum duration of androgen deprivation in
intermediate-risk disease is more controversial in the intermediate-risk disease remains to be defined. Further
context of dose-escalated radiotherapy,29,31 at least in part follow-up is needed to determine the effect of long-term
because of the heterogeneous nature of the patient androgen deprivation in this subgroup.
population in this group. Further stratification of patients
with intermediate-risk disease may go some way to An unexpected finding of our trial was that almost five
clarifying the situation. We believe that longer follow-up times as many patients died of cancers other than of the
and more events will enable us to provide more prostate in the short-term androgen deprivation group
information about the effect of long-term versus short- than in the long-term androgen deprivation group. We
term androgen deprivation in patients with intermediate- cannot provide a satisfactory explanation for this finding,
risk prostate cancer. Nevertheless, larger sample sizes will although we do recognise its potential effect on the
be needed to show significant benefits in unfavourable interpretation of the results. An association between the
intermediate-risk prostate cancer. Several ongoing hormonal environment, immune tolerance, and the
randomised trials (GETUG 14, EORTC 22991, and RTOG immune response to cancer cannot be excluded.
0815) are investigating the role of short-term androgen At the time that this study was designed, we felt that
deprivation combined with high-dose radiotherapy in the decision to use pelvic radiotherapy should be based
patients with intermediate-risk disease.24,32,33 on the criteria of the participating institution. Evidence

www.thelancet.com/oncology Vol 16 March 2015 325

 Articles

Number of Patients Androgen deprivation therapy Radiotherapy dose Outcome

patients
Phase 3 trials of ADT and conventional-dose radiotherapy
RTOG 92–024 1554 T2c–4 N0–1 M0 4 months NAD with or without 65–70 Gy ADT for 2 years improved overall survival
2 years LHRHa only in patients with a Gleason scoreof
8–10 (31·9% vs 45·1%; p=0·006)
EORTC 229615 970 T1c–2ab N1 M0; 6 months NAD with or without 70 Gy ADT for 6 months provides worse overall
T2c–4 N0–1 M0 2·5 years LHRHa survival compared with 3 years
Nabid et al6 630 T3–4, N0 36 months vs 18 months 70 Gy No significant differences in overall
(NCT00223171) PSA >20 ng/mL, adjuvant LHRHa survival
or Gleason >7
Retrospective studies of ADT and HDRT
Zelefsky et al25 1980 T1–T3 N0 3–6 months LHRHa 64·8–86·4 Gy HDRT and ADT for 6 months improved
biochemical disease-free and metastasis-
free survival, but not overall survival
Nguyen et al26 741 NCCN criteria 295 ADT for 2 or more years Median 70 Gy ADT and HDRT had a positive effect on
(range 2–18) (range 60–79) overall survival (p=0·003)
Zapatero et al27 306 NCCN criteria 231 patients, 28 months LHRHa; Median 78 Gy ADT for 28 months and >78 Gy improved
59 patients, 6 months; (range 66–84) overall survival; 96–89% at 5–10 years
16 patients, no hormones
Feng et al28 234 NCCN criteria No ADT, 48; Median 77 Gy Long-term ADT improved overall survival
STAD (<1 year), 84; (range 75–79) (p=0·001)
LTAD (≥1 year), 102
Krauss et al29 262 NCCN criteria 40% ADT duration not specified, EBRT 75·6 Gy, or No benefit with the addition of ADT
60% no ADT % no ADT brachytherapy, or EBRT
plus b brachytherapy
Tendulkar et al30 585 NCCN criteria 95% 6 months LHRHa, Median 78 Gy (range No benefit with the addition of ADT
5% no ADT 74-80)

NAD=neoadjuvant androgen deprivation. ADT=androgen deprivation therapy. LHRHa=luteinising hormone-releasing hormone analogue. PSA=prostate-specific antigen.
NCCN=National Comprehensive Cancer Network. STAD=short-term androgen deprivation. LTAD=long-term androgen deprivation. HDRT=high-dose radiation therapy.
EBRT=external beam radiation therapy.

Table 4: Studies addressing the duration of androgen deprivation combined with conventional-dose and high-dose radiotherapy in high-risk prostate cancer

from randomised trials showing a significant benefit in affect the power of the subgroup analyses and the
overall survival with pelvic radiotherapy was absent, analysis of late grade 3 radiation toxicity.
and the use of pelvic radiotherapy was a controversial In conclusion, our results show that long-term
issue. The results of our univariate and multivariate androgen deprivation plus high-dose radiotherapy is
analyses did not show a significant difference in superior to short-term androgen deprivation plus
biochemical disease-free survival between patients high-dose radiotherapy in terms of biochemical disease-
who were or were not given pelvic radiotherapy. free survival and overall survival, particularly in patients
Nevertheless, the clinical effect of pelvic radiotherapy with high-risk prostate cancer. Longer follow-up is
in the context of combined treatment with androgen required to confirm these results and to determine the
suppression remains highly controversial, and we effect of long-term androgen deprivation in patients with
await the results of the RTOG 0924 trial to confirm intermediate-risk prostate cancer.
whether there is any benefit of pelvic radiotherapy in a Contributors
high-risk population. AZ was the sponsor and chief investigator. AZ developed the protocol and
The incidence of late grade 2 or higher rectal or urinary wrote the report. FAC coordinated the study. APdH did the statistical
analysis. All the investigators enrolled the patients, did the literature search,
complications in our study was acceptably low in both and were involved in the interpretation of data and the review of the report.
treatment groups. Despite the use of higher radiation All investigators approved the final version of the report for publication.
doses, our results compare favourably with those of RTOG Declaration of interests
9202. However, we should be cautious when comparing All authors declare no competing interests.
these two trials because of relevant differences in Acknowledgments
technique (conventional vs three-dimensional conformal Funding was provided by a government grant (No. 04/2506) from the FIS
radiotherapy) and treatment volume. (Spanish National Health Investigation Fund) and AstraZeneca. We thank
Juan Luis Sanz (APICES, Madrid, Spain) for his participation in data
We recognise the limitations inherent in our study—
management, Marta Bonet for her review of the clinical data, and
namely, the fairly short follow-up, the low number of Feliciano Garcia-Vicente for his review of the radiotherapy schedule.
events, and the small sample size. We intend to report Editorial assistance was provided by Thomas O’Boyle (Health Research
10-year outcomes in due course. These limitations also Institute IIS from Hospital Universitario de la Princesa).

326 www.thelancet.com/oncology Vol 16 March 2015

 Articles

References 18 Zapatero A, García-Vicente F, Modolell I, et al. Impact of mean

1 Bria E, Cuppone F, Giannarelli D, et al. Does hormone treatment rectal dose on late rectal bleeding after conformal radiotherapy for
added to radiotherapy improve outcome in locally advanced prostate prostate cancer: Dose-volume effect. Int J Radiat Oncol Biol Phys
cancer? Meta-analysis of randomized trials. Cancer 2009; 115: 3446–56. 2004; 59: 1343–51.
2 Roach M 3rd, Bae K, Speight J, et al. Short-term neoadjuvant 19 Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy
androgen deprivation therapy and external-beam radiotherapy for Oncology Group (RTOG) and the European Organization for
locally advanced prostate cancer: long-term results of RTOG 86-10. Research and Treatment of Cancer (EORTC).
J Clin Oncol 2008; 26: 585–91. Int J Radiat Oncol Biol Phys 1995; 31: 1341–46.
3 Bolla M, Van Tienhoven G, Warde P, et al. External irradiation with 20 Roach M, Hanks G, Thames H, et al. Defining biochemical failure
or without long-term androgen suppression for prostate cancer with following radiotherapy with or without hormonal therapy in men
high metastatic risk: 10-year results of an EORTC randomised with clinically localized prostate cancer: recommendations of the
study. Lancet Oncol 2010; 11: 1066–73. RTOG-ASTRO Phoenix Consensus Conference.
4 Horwitz EM, Bae K, Hanks GE, et al. Ten-year follow-up of Int J Radiat Oncol Biol Phys 2006; 65: 965–74.
radiation therapy oncology group protocol 92-02: a phase III trial of 21 Kaplan EL, Meier P. Nonparametric estimation from incomplete
the duration of elective androgen deprivation in locally advanced observations. J Am Stat Assoc 1958; 53: 447–57.
prostate cancer. J Clin Oncol 2008; 26: 2497–504. 22 Mantel N. Evaluation of survival data and two new rank order
5 Bolla M, de Reijke TM, Van Tienhoven G, et al. Duration of statistics arising in its consideration. Cancer Chemother Rep 1966;
androgen suppression in the treatment of prostate cancer. 50: 163–70.
N Engl J Med 2009; 360: 2516–27. 23 Cox DR. Regression models and life tables. J R Stat Soc 1972;
6 Nabid A, Carrier N, Martin A, et al. High-risk prostate cancer 34: 187–220.
treated with pelvic radiotherapy and 36 versus 18 months of 24 Bolla M. Current status of combined radiation therapy and
androgen blockade: Results of a phase III randomized study. androgen suppression in locally advanced prostate cancer: what is
Proc Am Soc Clin Oncol 2013; 31 (suppl): Abs 3. the way forward? Eur Urol 2010; 9: 788–93.
7 Schmidt-Hansen M, Hoskin P, Kirkbride P, Hasler E, Bromham N. 25 Zelefsky MJ, Pei X, Chou JF, et al. Dose escalation for prostate
Hormone and radiotherapy versus hormone or radiotherapy alone cancer radiotherapy: predictors of long-term biochemical tumor
for non-metastatic prostate cancer: a systematic review with meta- control and distant metastases–free survival outcomes. Eur Urol
analyses. Clin Oncol (R Coll Radiol) 2014; 26: e21–46. 2011; 60: 1133–9.
8 D’Amico AV, Chen M, Renshaw AA, Loffredo M, Kantoff PW. 26 Nguyen QN, Levy LB, Lee AK, et al. Long-term outcomes for men
Androgen suppression and radiation vs. radiation alone for prostate with high-risk prostate cancer treated definitively with external
cancer: a randomized trial. JAMA 2008; 299: 289–95. beam radiotherapy with or without androgen deprivation. Cancer
9 Jones CU, Hunt D, McGowan DG, et al. Radiotherapy and short- 2013; 119: 3265–71.
term androgen deprivation for localized prostate cancer. 27 Zapatero A, García-Vicente F, Martin de Vidales C, et al. Long-term
N Engl J Med 2011; 365: 107–18. results after high-dose radiotherapy and adjuvant hormones in
10 Viani GA, Stefano EJ, Afonso SL. Higher-than-conventional prostate cancer: how curable is high-risk disease?
radiation doses in localized prostate cancer treatment: A meta- Int J Radiat Oncol Biol Phys 2011; 81: 1279–81.
analysis of randomized, controlled trials. 28 Feng FY, Blas K, Olson K, Stenmark M, Sandler H, Hamstra DA.
Int J Radiat Oncol Biol Phys 2009; 74: 1405–18. Retrospective evaluation reveals that long-term androgen
11 Kuban DA, Levy LB, Cheung MR, et al. Long-term failure patterns deprivation therapy improves cause-specific and overall survival in
and survival in a randomized dose-escalation trial for prostate the setting of dose-escalated radiation for high-risk prostate cancer.
cancer. Who dies of disease? Int J Radiat Oncol Biol Phys 2011; Int J Radiat Oncol Biol Phys 2013; 86: 64–71.
79: 1310–17. 29 Krauss DK, Kestin L, Ye H, Brabbins D, Ghilezan M, Gustafson G.
12 Zietman AL, Bae K, Slater JD, et al. Randomized trial comparing Lack of benefit for the addition of androgen deprivation therapy to
conventional-dose with high-dose conformal radiation therapy in dose-escalated radiotherapy in the treatment of intermediate- and
early-stage adenocarcinoma of the prostate: long-term results from high-risk prostate cancer. Int J Radiat Oncol Biol Phys 2011;
Proton Radiation Oncology Group/American College of Radiology 80: 1064–71.
95-09. J Clin Oncol 2010; 28: 1106–11. 30 Tendulkar RD, Reddy CA, Stephans KL, et al. Redefining high-risk
13 Dearnaley DP, Jovic G, Syndikus I, et al. Escalated-dose versus prostate cancer based on distant metastases and mortality after
control-dose conformal radiotherapy for prostate cancer: long-term high-dose radiotherapy with androgen deprivation therapy.
results from the MRC RT01 randomised controlled trial. Int J Radiat Oncol Biol Phys 2012: 82: 1397–404.
Lancet Oncol 2014; 15: 464–73. 31 Castle KO, Hoffman KE, Levy LB, et al. Is androgen deprivation
14 Peeters STH, Heemsbergen WD, Koper PCM, et al. Dose response therapy necessary in all intermediate-risk prostate cancer patients
in radiotherapy for localized prostate cancer: results of the Dutch treated in the dose escalation era? Int J Radiat Oncol Biol Phys 2013;
multicenter randomized phase III trial comparing 68 Gy of 85: e693–99.
radiotherapy with 78 Gy. J Clin Oncol 2006; 24: 1990–96. 32 Dubray BM, Beckendorf V, Guerif S, et al. Does short-term
15 Beckendorf V, Guerif S, Le Prise YE, et al. 70 Gy versus 80 Gy in androgen depletion add to high-dose radiotherapy (80 Gy) in
localized prostate cancer: 5-year results of GETUG 06 randomized localized intermediate-risk prostate cancer? Intermediate analysis of
trial. Int J Radiat Oncol Biol Phys 2011; 80: 1056–63. GETUG 14 randomized trial (EU-20503/NCT00104741). J Clin Oncol
16 Smith MJ, Akhtar NH, Tagawa ST. The current role of androgen 2011; 29: abs 4521.
deprivation in patients undergoing dose-escalated external beam 33 Matzinger O, Duclos F, van den Bergh A, et al. Acute toxicity of
radiation therapy for clinically localized prostate cancer. curative radiotherapy for intermediate- and high-risk localized
Prostate Cancer 2012; 2012: 280278. prostate cancer in the EORTC trial 22991. Eur J Cancer 2009;
17 International Commission on Radiation Units and Measurements. 45: 2825–34.
Report 29. Dose specification for reporting external beam therapy
with photons and electrons. Washington, DC: International
Commission on Radiation Units and Measurements, 1978.

www.thelancet.com/oncology Vol 16 March 2015 327