Adjuvant Leuprolide With or Without Docetaxel in Patients With High-Risk Prostate Cancer After Radical Prostatectomy (TAX-3501)

Abstract

BACKGROUND

The current trial evaluated 2 common therapies for patients with advanced prostate cancer, docetaxel and hormonal therapy (HT), in the surgical adjuvant setting.

METHODS

TAX-3501 was a randomized, phase 3, adjuvant study post-radical prostatectomy (RP) in high-risk patients with prostate cancer (n = 228) comparing 18 months of HT with (CHT) without docetaxel chemotherapy either immediately (I) or deferred (D). High-risk disease was defined as a 5-year freedom-from-disease-progression rate of ≤60% as predicted by a post-RP nomogram. Progression-free survival (PFS), including prostate-specific antigen disease recurrence, was the primary endpoint. The authors also assessed the accuracy of the nomogram and analyzed testosterone recovery in 108 patients treated with HT who had at least 1 posttreatment testosterone value.

RESULTS

Between December 2005 and September 2007, 228 patients were randomized between the treatment cohorts. TAX-3501 was terminated prematurely because of enrollment challenges, leaving it underpowered to detect differences in PFS. After a median follow-up of 3.4 years (interquartile range, 2.3–3.8 years), 39 of 228 patients (17%) demonstrated PSA disease progression, and metastatic disease progression occurred in 1 patient. The median time to baseline testosterone recovery after the completion of treatment was prolonged at 487 days (95% confidence interval, 457–546 days). The nomogram’s predicted versus observed freedom from disease progression was significantly different for the combination D(HT) and D(CHT) group (P < .00001).

CONCLUSIONS

TAX-3501 illustrated several difficulties involved in conducting postoperative adjuvant systemic trials in men with high-risk prostate cancer: the lack of consensus regarding patient selection and treatment, the need for long follow-up time, nonvalidated intermediate endpoints, evolving standard approaches, and the need for long-term research support. Except for selected patients at very high-risk of disease recurrence and death, surgical adjuvant trials in patients with prostate cancer may not be feasible.

INTRODUCTION

Prostate cancer is the second most common cancer in men worldwide, accounting for >900,000 new cases and >250,000 deaths per year.¹ Although 10-year overall survival (OS) rates after radical prostatectomy (RP) or radiotherapy have been reported to be >80%, the clinical course of patients with prostate cancer is variable.^2,3 A smaller yet significant percentage of patients will subsequently develop disease recurrence after local treatment, develop distantmetastasis, and eventually die of their disease.

Several models have been developed to identify risk factors for the development of recurrent prostate cancer after prostatectomy.^4–8 Although men at high-risk who are treated with definitive radiotherapy appear to benefit from neoadjuvant, concurrent, and adjuvant hormonal therapy (HT) (ie, luteinizing hormone-releasing hormone agonists/antagonists), there has yet to be an established role for systemic adjuvant therapy in men after RP.^9–15 Trials addressing this group of men have been limited in size and number, and although there is some suggestion that higher-risk individuals may benefit from HT, this has yet to be established as a standard.^16–19 Furthermore, at the time of this trial’s inception, adjuvant docetaxel, the first agent shown to prolong survival in patients with metastatic castration-resistant prostate cancer, had to the best of our knowledge never been tested in a group of patients other than those with castration-resistant prostate cancer.²⁰

Herein we report the experience of the TAX-3501 trial, a multinational, randomized clinical trial. TAX-3501 was designed to determine differences in progression-free survival (PFS) for high-risk men with an undetectable prostate-specific antigen (PSA) level after RP who were treated with immediate versus deferred HT with or without docetaxel. This study represents a global effort to determine the adjuvant roles of docetaxel and HT after RP, the 2 most accepted treatments for advanced prostate cancer at the time of its design.

MATERIALS AND METHODS

Participants

Men were eligible for this study if they had an Eastern Cooperative Oncology Group performance status ≤1, a life expectancy of ≥5 years, histologically confirmed prostatic adenocarcinoma, undergone a RP <120 days before randomization, and were deemed to be at high risk of disease recurrence. Pathology of whole mounted specimens was reviewed centrally (at Johns Hopkins Hospital), at which point subjects were classified as being at high risk if they had a 5-year freedom-from-disease-progression (FFP) rate of ≤60% according to the postoperative predictive probability model developed by Kattan et al.^5,21,22 This cutoff was based on consensus agreement among the trial’s executive steering committee. It should be noted that a minority of patients (n = 23) did not have all of the pathologic variables necessary to calculate their FFP rate. The consensus agreement among the steering committee was to assume said variables were negative. Subjects were also required to have a postoperative PSA level ≤0.2 ng/mL at least 30 days after RP and within 7 days of randomization; normal hematologic, renal, and hepatic function; and normal serum testosterone (ie, ≥150 ng/dL) within 6 months of randomization. In addition, patients were required to have a bone scan, chest x-ray, and abdominal computed tomography (CT) scan without evidence for metastasis within 6 months of randomization. Prior systemic therapies for prostate cancer were not allowed.

This study was conducted in 108 countries. Thirty-one of the participating centers were in the United States, 45 were in Europe, 7 were in Australia, 6 were in Canada, 1 was in Mexico, 2 were in South America, 2 were in Asia, 7 were in the Middle East, 4 were in Russia, and 3 were in South Africa. Patients were registered after signing an Institutional Review Board-approved informed consent specific to this protocol.

Trial Design

This trial was a prospective, multinational, 2 × 2 factorial-designed, open-label, randomized phase 3 study performed in subjects with prostate cancer who were at high risk of disease recurrence after RP (ClinicalTrials.gov identifier: NCT00283062). Eligible patients were randomized 1:1:1:1 to receive hormonal therapy with chemotherapy (CHT) or without chemotherapy (HT) either immediately (I) after RP or deferred (D) until the time of disease progression (Fig. 1). Stratification was according to age (aged ≥65 years vs aged <65 years), predicted 5-year FFP rate per the nomogram (60%-40% vs 40%-20% vs 20%-0%), and country of the study site (Fig. 2). The primary endpoint of this trial was PFS after systemic treatment. Secondary endpoints were OS and safety.

Figure 1.

Study scheme is shown. I(HT) indicates immediate hormonal therapy without chemotherapy; I(CHT), immediate hormonal therapy with chemotherapy (docetaxel); D(HT), deferred hormonal therapy without chemotherapy; D(CHT), deferred hormonal therapy with chemotherapy (docetaxel).

Figure 2.

CONSORT (Consolidated Standards Of Reporting Trials) trial flow diagram is shown. ITT indicates intent-to-treat; PSA, prostate-specific antigen; I(CHT), immediate hormonal therapy with chemotherapy (docetaxel); HT, hormonal therapy; I(HT), immediate hormonal therapy without chemotherapy; D(CHT), deferred hormonal therapy with chemotherapy (docetaxel); D(HT), deferred hormonal therapy without chemotherapy; chemo, chemotherapy.

Interventions

Docetaxel was administered at a dose of 75 mg/m² intravenously over 1 hour repeated every 3 weeks for a total of 6 cycles. Leuprolide at a dose of 22.5 mg was given subcutaneously every 3 months for a total duration of 18 months of treatment. Standard dose modification criteria for docetaxel toxicity were used. Leuprolide dosing was constant for the duration of treatment. The length of treatment was based largely on consensus within the trial’s executive steering committee. The duration of docetaxel therapy was influenced by pilot data demonstrating the feasibility of 6 cycles of adjuvant therapy.²³ Eighteen months of androgen deprivation therapy was agreed on given that this roughly corresponded to the interval required to administer 6 cycles of docetaxel.

Follow-Up and Endpoints Definition

Patients receiving chemotherapy were seen before each 3-week treatment until completion and then every 3months whereas those receiving leuprolide alone were seen every 3 months. Follow-up procedures included blood counts (before chemotherapy and then every 3 months), blood chemistries, PSA, and serum testosterone every 3 months. Bone scans and CT scans were repeated once a year until disease progression and then every 6 months and/or as clinically indicated.

Disease progression was defined as PSA, radiographic, or histological disease progression after systemic treatment or death from any cause, whichever came first, as measured from the time of surgery. PSA progression was defined as the date of the first PSA increase to ≥0.4 ng/mL. A confirmatory PSA was obtained within 2 weeks from the initially elevated value of ≥0.4 ng/mL. For subjects treated in the deferred treatment arms whose PSA nadir did not reach <0.4 ng/mL, the date of PSA progression was taken to be the date of the nadir. Local disease progression was considered evidence of disease progression.

Survival was measured from the date of surgery to the date of death from any cause. PFS was measured from the date of surgery to the date of disease progression as defined above. Adverse events were defined per the National Cancer Institute (NCI) guidelines in the Common Terminology Criteria for Adverse Events (CTCAE) Version 3.0.

Statistical Analysis

This trial was designed to detect a 30% difference in the median PFS for both comparisons (I(HT) vs D(HT) and I(CHT) vs D(CHT)). A total of 622 events would yield a 90% power at a 2-sided type I error level of α = .05. Assuming a 10% dropout rate, a total of 1696 patients were required with a planned 2-year accrual period and a 4-year maximum follow-up (median follow-up of 3 years).

RESULTS

Patients

Of a total of 571 subjects screened from December 2005 through September 2007, 399 were deemed potentially eligible and 228 were eventually randomized after a central pathology review (Fig. 2). A total of 110 subjects were randomized to receive immediate treatment (55 with HT and 55 with CHT), whereas 118 were randomized to deferred treatment (56 with CHT and 62 with HT). Characteristics of the randomized cohort are shown in Table 1. Reasons for screen failure (n = 172) were not captured. The most common reason for nonrandomization (n = 171) was that subjects were determined to be at low risk after the central pathology review (n = 62). Additional reasons for nonrandomization are listed in the CONSORT (Consolidated Standards Of Reporting Trials) diagram shown in Figure 2. Ninety randomized patients (39.5%) did not receive treatment. Reasons for nontreatment are listed in the CONSORT diagram (Fig. 2). Although the current study was not designed to identify subjects who did not receive treatment because of the absence of disease progression, it should be noted that 73 patients (32%) in the deferred group did not progress by the end of the study and therefore were not treated.

TABLE 1.

Demographics and Baseline Disease Characteristics

	Randomized Treatment Group
Parameter	I(CHT) N = 55	I(HT) N = 55	D(CHT) N = 56	D(HT) N = 62	Total N = 228
Age, y
Mean (SD)	61.2 (7.4)	61.6 (7.0)	62.1 (7.0)	62.9 (7.5)	61.9 (7.2)
Race, no. (%)
White	48 (87.3)	49 (89.1)	43 (76.8)	59 (95.2)	199 (87.3)
Black	6 (10.9)	3 (5.5)	7 (12.5)	2 (3.2)	18 (7.9)
Asian	0	1 (1.8)	4 (7.1)	0	5 (2.2)
Other	1 (1.8)	2 (3.6)	2	1 (1.6)	6 (2.6)
Preoperative PSA, ng/mL
Mean (SD)	10.69 (7.88)	11.95 (10.85)	11.12 (6.26)	13.05 (13.08)	11.74 (9.96)
Mean pathology Gleason sum (range)	7.8 (7–10)	8.0 (7–9)	7.9 (7–10)	7.8 (7–10)	7.9 (7–10)
Pathology Gleason sum, no. (%)
5–7	28 (50.9)	23 (41.8)	26 (46.4)	33 (53.2)	110 (48.2)
8–10	27 (49.1)	32 (58.2)	30 (53.6)	29 (46.8)	118 (51.8)
Prostate capsule invasion, no. (%)
None	0	0	0	0	0
Invasive capsule	2 (3.6)	0	0	0	2 (0.9)
Focal	5 (9.1)	2 (3.6)	3 (5.4)	3 (4.8)	13 (5.7)
Established	48 (87.3)	53 (96.4)	53 (94.6)	59 (95.2)	213 (93.4)
Surgical margins
Positive	37 (67.3)	32 (58.2)	38 (67.9)	50 (80.6)	157 (68.9)
Positive seminal vesicle invasion, no. (%)	25 (45.5)	30 (54.5)	31 (55.4)	29 (46.8)	115 (50.4)
Positive lymph node involvement, no. (%)	14 (25.5)	11 (20.0)	15 (26.8)	5 (8.1)	45 (19.7)
Predicted probability of 5-y freedom from recurrence, mean (SD)	23.3 (17.7)	21.9 (18.3)	18.3 (18.4)	20.5 (16.9)	20.9 (17.8)

Abbreviations: D(CHT), deferred hormonal therapy with chemotherapy (docetaxel); D(HT), deferred hormonal therapy without chemotherapy; I(CHT), immediate hormonal therapy with chemotherapy (docetaxel); I(HT), immediate hormonal therapy without chemotherapy; PSA, prostate-specific antigen; SD, standard deviation.

Disease Progression

After a median follow-up of 3.4 years (interquartile range [IQR], 2.3–3.8 years), a total of 41 of 228 patients (17.9%) developed disease progression. Twenty-four of 110 patients (22%) in the immediate-treatment group developed disease progression (10 in the I(CHT) group and 14 in the I(HT) group), whereas 17 of 118 patients in the deferred group (14%) progressed (second progression) (9 in the D(CHT) group and 8 in the D(HT) group) (Table 2) (Fig. 3). One patient randomized to the I(HT) arm developed bone metastasis; the remaining patients demonstrated PSA progression only. The study was underpowered to detect significant differences between the study arms for all endpoints.

TABLE 2.

Disease Progression by Treatment Subgroups

	Randomized Treatment Group
Outcome	I(CHT) N = 55	I(HT) N = 55	D(CHT) N = 56	D(HT) N = 62
Progression, no. (%)	10 (18.2)	14 (25.5)	9 (16.1)	8 (12.9)
PSA progression	10 (18.2)	12 (21.8)	9 (16.1)	8 (12.9)
Bone metastasis	0	1 (1.8)	0	0
Other	0	1 (1.8)	0	0
Death, no. (%)	0	1 (1.8)	0	1 (1.6)

Abbreviations: D(CHT), deferred hormonal therapy with chemotherapy (docetaxel); D(HT), deferred hormonal therapy without chemotherapy; I(CHT), immediate hormonal therapy with chemotherapy (docetaxel); I(HT), immediate hormonal therapy without chemotherapy; PSA, prostate-specific antigen.

Figure 3.

Kaplan-Meier progression-free survival curves are shown for all groups. The P value was not found to be significant between groups. I(CHT) indicates immediate hormonal therapy with chemotherapy (docetaxel); I(HT), immediate hormonal therapy without chemotherapy; D(CHT), deferred hormonal therapy with chemotherapy (docetaxel); D(HT), deferred hormonal therapy without chemotherapy. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Safety

Of the 138 patients who received treatment, 118 (86%) experienced at least 1 adverse event (AE) considered to be possibly drug related. The majority of AEs were grade 1 or grade 2, and were similar to those observed when either drug was given alone. Grade 3 and 4 AEs attributed to the study drugs were uncommon (reported to occur in 21 patients). These occurred more frequently in the group of patients who received chemotherapy, with the most common AE being febrile neutropenia, occurring in 5 of 21 patients (23.8%).

Testosterone Recovery

Testosterone recovery kinetics were analyzed for those patients completing an entire course of therapy (ie, 18 months of treatment) and with at least 1 posttreatment testosterone value (n = 108). Time to a testosterone level >150 ng/dL and baseline testosterone recovery were analyzed given that this trial was designed under the assumption that PSA disease recurrences would occur within the context of full gonadal recovery. The median posttreatment follow-up for that group was 676 days (IQR, 478–847 days). The median time to testosterone recovery to >150 ng/dL and baseline was 306 days (95% confidence interval [95% CI], 294–345 days) and 487 days (95% CI, 457–546 days), respectively. Ninety patients (83%) and 64 patients (59%) had their testosterone level recover to >150 ng/dL and to baseline, respectively. There were no statistically significant differences in testosterone recovery between the treatment groups. The time to baseline testosterone recovery in the combined chemotherapy (ie, I(CHT) and D(CHT)) and HT arms (ie, I(HT) and D(HT)) was not statistically significantly different at 458 days (95% CI, 336–529 days) and 535 days (95% CI, 457–749 days), respectively (hazards ratio, 1.4; P = .18) (Fig. 4).

Figure 4.

Testosterone recovery to (A) >150 ng/dL and (B) baseline is shown. Testosterone recovery to (C) >150 ng/dL and (D) baseline is shown stratified by treatment. 95% CI indicates 95% confidence interval; HT, hormonal therapy; chemo, chemotherapy.

Multivariable Cox hazard models were built to evaluate variables associated with recovery of testosterone to baseline and to >150 ng/dL. Variables were selected based on clinical relevance. Log transformation was applied to baseline testosterone to reduce skewness. Two variables were found to independently predict for increased time to baseline testosterone recovery: age ≥65 years and log-transformed baseline testosterone. Three variables were found to independently predict for time to testosterone recovery >150 ng/dL: age ≥65 years, white race, and log-transformed baseline testosterone (Table 3). It should be noted that individuals of nonwhite race only constituted approximately 12.7%of the total randomized population.

TABLE 3.

Multivariable Cox Proportional Hazards Models for Time to Testosterone Recovery to Baseline and to >150 ng/dL

Variable	HR for Time to Baseline Testosterone Recovery (95% CI)	P
Age ≥65 y	0.55 (0.31–0.95)	.032
White race	0.60 (0.28–1.26)	.18
Receipt of docetaxel	1.21 (0.73–1.99)	.47
Log-transformed baseline testosterone	0.36 (0.18–0.75)	.006
Variable	HR for Time to Testosterone Recovery >150 ng/dL (95% CI)	P
Age ≥65 y	0.62 (0.39–0.98)	.04
White race	0.5 (0.25–0.98)	.043
Receipt of docetaxel	0.84 (0.55–1.29)	.42
Log-transformed baseline testosterone	2.22 (1.22–4.03)	.0091

Abbreviations: 95% CI, 95% confidence interval; HR, hazards ratio.

Observed Versus Predicted FFP

Of the 228 patients randomized to treatment, only 205 had all the variable values to compute their predicted FFP using the postoperative predictive probability model developed by Kattan et al.^5,21,22 One individual did not have the seminal vesicle site marked adequately for determination of seminal vesicle invasion. The other 22 individuals did not have their lymph nodes sampled by the surgeon. These variables were all assumed to be negative for the purposes of stratification. Men in the deferred groups (ie, D(HT) and D(CHT)) were the only subjects not treated until the time of disease progression after RP, and were the focus of our assessment of the predictive ability of the nomogram. The combined deferred group’s predicted compared with observed FFP was significantly different (P < .00001). The concordance index between the observed-to-predicted FFP for the combined deferred group was 0.658. A calibration plot for the combined deferred group is presented in Figure 5.

Figure 5.

Predictive nomogram calibration curve is shown for the combined deferred treatment groups (ie, deferred hormonal therapy without chemotherapy [D(HT)] and deferred hormonal therapy with chemotherapy [docetaxel] [D(CHT)]).

DISCUSSION

Surgical adjuvant prostate cancer trials face numerous obstacles, including a lack of validated proximal endpoints, the need for long follow-up to capture meaningful endpoints, a lack of consensus on the definition of high-risk disease, the most appropriate local treatment for high-risk patients, the potential establishment of new treatment standards during the course of the trial, and long-term commitment for funding (Table 4). TAX-3501 was designed to address the role of adjuvant/salvage systemic HT and chemotherapy for patients considered to have high-risk prostate cancer after RP. Unfortunately, the study was terminated by the sponsor approximately 21 months after its activation due in part to early enrollment challenges. It is interesting to note that although the rate of subject randomization for TAX-3501 was below expectations, it was comparable to that during the early stages of other multiinstitutional adjuvant trials.^18,24–26 In the end, TAX-3501 serves to outline these difficulties, and raises serious doubts that future surgical adjuvant prostate cancer trials can produce meaningful results.

TABLE 4.

Barriers to Completing Postoperative Adjuvant Prostate Cancer Trials

Long follow-up time needed for completion Lack of validated early endpoints Confounding relationship between testosterone and PSA in trials using PSA-based endpoints Lack of consensus regarding what constitutes high-risk disease New standards that may compete for enrollment Required long-term monetary support

Abbreviation: PSA, prostate-specific antigen.

PFS from the time of surgery to first (immediate treatment arms) or second (deferred treatment arms) disease progression was the main endpoint. As anticipated, PSA recurrences occurred in the vast majority of cases, with 40 of 41 patients exhibiting PSA progression only. Short follow-up time and the low incidence of biochemical and clinical/radiological progression events did not allow us to examine the relationship between biochemical disease recurrence and more clinically meaningful objective endpoints. A major assumption in designing this trial was that disease recurrence would occur within the context of full gonadal function recovery (ie, a testosterone level ≥150 ng/dL), approximately 6 months after the last administration of HT.^27,28 It is interesting to note that in the 108 patients who completed the 18-month treatment period, only 64 (59%) recovered gonadal function to baseline after a median posttreatment follow-up of 676 days (IQR, 478–847 days). The median time to baseline serum testosterone recovery after treatment was longer than anticipated at 487 days (95% CI, 457–546 days). A more recent retrospective report in patients receiving combined HT (16 months of treatment) and primary radiotherapy also suggests longer times to recovery than initially assumed on this study.²⁹ Because PSA expression is androgen-responsive, concomitant testosterone monitoring after treatment discontinuation should be routine in all trials using PSA-based endpoints in patients with hormone-sensitive disease. Studies evaluating extended courses of HT will in turn necessitate relatively long follow-up times.

Prostate cancer has a variable clinical course, with the OS in patients with biochemically disease recurrence ranging from approximately 4 years to >20 years.^30–32 Several clinical and pathologic factors predict the outcome of patients undergoing RP, and complex stratification processes and large sample sizes are necessary to ensure the balanced distribution of these factors between study arms.³³ TAX-3501 used a widely referenced postoperative nomogram to define individual patients’ eligibility and to stratify them based on their calculated risk of disease recurrence.^5,21,22 We prospectively assessed the accuracy of the nomogram in the group of patients randomized to initial observation. The nomogram predicted versus observed disease recurrence rates for patients in the deferred treatment arm suggest that the nomogram may have overpredicted disease recurrences. Aside from possible issues with the nomogram itself, the most likely explanation for the difference in the predicted-to-observed FFP was that the nomogram considered adjuvant therapy as failure when it was constructed. The nomogram also did not assume an undetectable PSA level 30 days after RP. Because patients on the deferred treatment arms of this trial were not allowed to receive adjuvant therapy, it may have resulted in fewer anticipated events as defined by the nomogram. Because of limited patient numbers and relatively short follow-up, these data cannot be used conclusively to assess the validity of the nomogram. However, if disease progression rates were underestimated, it remains possible that even had TAX-3501 met its target enrollment and follow-up time, it may still have been underpowered to detect differences in its primary endpoint.

A broader issue that arose during the course of this trial related to the emergence of new treatment approaches, all of which directly competed with TAX-3501. These included adjuvant radiotherapy in high-risk patients, primary combined HT plus radiotherapy in patients with intermediate-risk or high-risk disease, and neoadjuvant chemo-hormonal therapy.^{9,10,12,13,16,18,19,34–38} Table 5 illustrates the competitive landscape of clinical trials conducted in patients with early-stage prostate cancer that affected TAX-3501 enrollment.^{11–16,18,24–26,37–39} Some of these trials led to new standards of care that were not provided for in TAX-3501. Adjuvant radiotherapy, an exclusion criteria in TAX-3501, was perhaps the strongest argument cited by various sites as a factor impeding enrollment.^24,26,40,41

TABLE 5.

Early Prostate Cancer Trials That Affected Enrollment on TAX-3501

Trial	Study Details	Start of Enrollment	End of Enrollment	Median Follow-Up, Years
TAX-3501	HT ± docetaxel either immediately or deferred until biochemical disease recurrence after RP	12/2005	9/2007	3.4
SWOG 879424	± Adjuvant XRT after RP	8/1988	1/1997	10.9
EORTC 2291125	± Adjuvant XRT after RP	11/1992	12/2001	5
ARO/AUO26	± Adjuvant XRT after RP	4/1997	9/2004	4.5
Messing 199918	± Adjuvant HT in men with lymph node-positive disease after RP	1988	1993	7.1
SWOG S992116	Adjuvant HT vs adjuvant HT and mitoxantrone after RPa	2/2000	1/2007	4.4
RTOG 86-1037	± Adjuvant HT and antiandrogen therapy after XRT	4/1987	6/1991	6.7
RTOG 85-3139	± Adjuvant HT after XRT	2/1987	4/1992	4.5
EORTC 2286311	± Adjuvant HT after XRT	5/1987	9/1995	5
EORTC 2296112	± Adjuvant HT after XRT	4/1997	11/2001	6.4
RTOG 92-0213	± Adjuvant HT after XRT with neoadjuvant and concurrent HT and antiandrogen therapy	6/1992	4/1995	5.8
RTOG 941314	Compared whole-pelvic XRT vs prostate-only XRT with either neoadjuvant and concurrent HT or adjuvant HT	4/1995	6/1999	5
D’Amico 200415	± Adjuvant HT after XRT	12/1995	4/2001	4.5
CALGB 9020338	± Neoadjuvant HT and docetaxel followed by RP	12/2006	Open	NA

Abbreviations: ARO, Arbeitsgemeinschaft Radiologische Onkologie of the German Cancer Society; AUO, Urologische Onkologie of the German Cancer Society; CALGB, Cancer and Leukemia Group B; EORTC, European Organization for Research and Treatment of Cancer; HT, hormonal therapy; RP, radical prostatectomy; RTOG, Radiation Therapy Oncology Group; SWOG, Southwest Oncology Group; XRT, radiotherapy.

^aOnly the adjuvant HT arm was reported.

TAX-3501 was a carefully planned, surgical adjuvant chemo-hormonal therapy trial. Enrollment difficulties resulted in its premature termination by the sponsor; however, it does serve to highlight some of the key issues facing adjuvant prostate cancer trials. The most frequently used proximal endpoint in early prostate cancer trials, PSA progression, not only lacks clear clinical significance but is also confounded by slow androgen recovery after prolonged HT. Available nomograms may not be sufficient to prospectively identify high-risk patients who are the most likely to benefit from adjuvant treatment and, most importantly, given the prolonged amount of time needed to complete adjuvant trials, an evolving treatment landscape will inevitably compete for trial enrollment. TAX-3501 demonstrates the difficulty in conducting surgical adjuvant prostate cancer trials, and exemplifies why these trials may not be feasible.