A Phase II, Randomized, Open-Label Study of Neoadjuvant Degarelix versus LHRH Agonist in Prostate Cancer Patients Prior to Radical Prostatectomy

Neoadjuvant treatment with androgen deprivation therapy (ADT) is not typically indicated for patients with localized prostate cancer eligible for radical prostatectomy. However, several randomized trials have shown that treatment with luteinizing hormone–releasing hormone (LHRH) agonists prior to radical prostatectomy (RP) can significantly improve pathologic variables associated with poor prognosis. Neoadjuvant hormones before RP have been shown to decrease the positive surgical margin rate, reduce the size of tumor (30%–55% after 3 months of treatment), and lead to tumor downstaging (1). Although these trials did not ultimately improve disease-free survival, inclusion of large cohorts of men at low risk for progression may have diluted the benefits of this approach in intermediate and high-risk patients. Indeed, subset analyses from these trials have demonstrated trends toward improvement in patients with intermediate risk features, particularly in higher volume centers (2). These observations, as well as the emergence of novel androgen axis inhibition strategies, raise the specter of revisiting this concept in the clinical space.

LHRH therapy, despite ultimately leading to chemical castration, initially stimulates the release of luteinizing hormone (LH), causing a surge in serum testosterone levels that can precipitate a “flare” phenomenon, particularly in patients with bone metastatic disease. The initial response to treatment with LHRH can result in up to a 10-fold increase in LH levels and up to a 2-fold increase in serum testosterone levels, leading to an increase in PSA, a marker of tumor growth (3). This flare could be serious in nature, with exacerbation of pain, increase in uremia, development of neurologic sequelae, and possibly death (4, 5). For this reason, LHRH is usually used in combination with antiandrogens, such as flutamide and bicalutamide, which are androgen receptor (AR) blockers.

Degarelix is a gonadotropin-releasing hormone (GnRH) antagonist recently approved in the management of advanced, hormone-sensitive prostate cancer. Pivotal trials have demonstrated favorable testosterone suppression using this compound as well as better follicle-stimulating hormone (FSH) and PSA-free survival at one year (6). These promising results, coupled with the theoretical advantages of steeper FSH suppression, led us to investigate neoadjuvant use of this agent prior to RP and compare clinical, pathologic, and endocrine-related outcomes.

This study was a two-center, randomized, open-label, parallel group trial (trial registration ID: NCT01674270). All procedures were approved by the Research Ethics Boards of Princess Margaret Cancer Centre (Toronto, Ontario, Canada) and University of British Columbia (Vancouver, British Columbia, Canada), and all subjects signed written informed consent. Eligible men had surgically resectable, localized adenocarcinoma of the prostate, confirmed histologically by transrectal ultrasound–guided prostate biopsy performed within 6 months of study enrolment, and Gleason grade ≥7 and/or clinical stage ≥T2 disease.

Prior to radical retropubic prostatectomy, patients were block-randomized 1:1:1 to 3 months of neoadjuvant therapy with either degarelix alone, degarelix with bicalutamide, or LHRH and bicalutamide (ClinicalTrials.gov ID NCT01674270). Degarelix was administered as one 240 mg subcutaneous injection (starting dose), followed by two monthly maintenance doses of 80 mg each, bicalutamide as a once daily, 50 mg tablet, and LHRH as a 3-month injectable dose of leuprorelin 22.5 mg, leuprolide 22.5 mg, or goserelin acetate 10.8 mg.

The primary endpoint under investigation was the treatment effect on intratumoral dihydrotestosterone (DHT) levels. Secondary outcomes included treatment effect on prostate tumor morphology, pathologic outcomes (i.e., stage, extraprostatic extension, margin positivity, node positivity), change in baseline Gleason score, RP failure rate, PSA failure rate (defined as two consecutive PSA values above 0.2 ng/mL or use of adjuvant ADT or radiotherapy), intratumoral androgen precursor levels, serum PSA and hormone levels (testosterone, FSH, LH, GnRH, inhibin B, and PTH), and immunohistochemical staining outcomes (AR, PSA staining, PSAP staining, Ki-67, AMACR, chromogranin A, synaptophysin, and CD56).

A detailed standard operating procedure for acquisition and processing of prostatectomy specimens was developed and followed at both sites. At the time of surgery, RP specimens were sent from the operating room to the surgical pathology laboratory in fresh state. Transit time between the operating room and the pathology laboratory was tracked. Specimens were weighed, measured, and inked according to standard hospital protocol. Specimens were sectioned and inspected for grossly identifiable tumor. Tissue samples from the grossly identifiable tumor and normal prostate parenchyma were removed and snap frozen as per standard tissue banking procedure. For the purposes of the tissue androgen analysis, a single <1 gm (0.5 unit of measure; ref. 3) of fresh tissue from the peripheral zone was harvested and snap frozen in liquid nitrogen within 5 minutes of procurement and stored at −80°C prior to homogenization and solvent extraction. After procurement of the frozen tissue samples, the prostatectomy specimens were serially sectioned, divided into quadrants (left and right posterior, left and right anterior), and submitted in toto for histologic examination. Specifically, one 5-μm H&E section was cut from each paraffin block for each prostatectomy as per standard clinical guidelines (7). If residual prostatic adenocarcinoma was not identified in the initial H&E sections, additional H&E levels were cut from any paraffin blocks containing high-grade prostatic intraepithelial neoplasia, suspicious or atypical foci as well as those corresponding to areas of the prostate from which positive needle biopsy cores were obtained (8). Immunohistochemical staining with basal cell markers (p63 and 34βE12) and AMACR was performed to confirm the presence of small foci of residual adenocarcinoma at the discretion of the pathologist reviewing the case. The presence or absence of treatment effect in residual carcinoma was assessed according to standard criteria (9). If appreciable treatments effects were noted, Gleason scores were not assigned. Histologic examination and standard reporting were carried out by an experienced urologic pathologist at University Health Network (UHN; Toronto, Ontario, Canada) or Vancouver General Hospital (VGH; Vancouver, British Columbia, Canada).

The cryopreserved specimens were thawed on ice and homogenized with 3 to 5 volume equivalents water. Deuterated testosterone and deuterated DHT were added to 300 μL of homogenate, which was then extracted twice with 1 mL 70/30 hexane/ethyl acetate or 80/20 MTBE/MeOH. Pooled extracts were dried under vacuum, and the residue dissolved in 100 μL of 50 mmol/L hydroxylamine and incubated for 1 hour at 60°C to generate oxime derivatives of testosterone and DHT for substantially increased sensitivity.

The derivatized samples were analyzed with a Waters Acquity UPLC coupled to a Waters Quattro Premier XE triple quadrupole mass spectrometer (MS). Separation was carried out on a Waters BEH 1.7 μmol/L C18 column with a 20% to 70% acetonitrile gradient, followed by acetonitrile flush prior to reequilibration. MS parameters were optimized for ES⁺ ionization and fragmentation of the protonated T-oxime and acetonitrile adducted DHT-oxime with m/z 304 > 124 and 347 > 306, respectively, used for testosterone and DHT maximum reference matches. m/z 307 > 112 and 350 > 309 were used for the corresponding trideuterated testosterone and DHT internal standards. Integration and quantitation were carried out with Waters QuanLynx software using calibration samples prepared with testosterone and DHT ranging from 10 to 10,000 pg/mL and is equivalent to expected final sample levels. Processed data were further normalized as needed by exporting to Excel spreadsheets. Extraction and matrix variability were virtually eliminated with this isotopic normalization, and sensitivity was in the low pg/mL (low to mid femtomolar tissue levels). Tissue androgen and androgen precursor levels were measured at The University of British Columbia for all patients.

Tissue microarray (TMA) blocks were constructed independently by UHN and VGH pathologists according to standard protocols. Briefly, triplicate 1.0-mm cores were obtained from donor paraffin blocks with the largest tumor foci or from the largest foci in both the left and right prostatic lobes in cases with bilateral disease.

Sections from the TMA were studied by immunohistochemical staining with antibodies directed against the following antigens: PSA, prostatic specific acid phosphatase (PSAP), AR, Ki67, Alpha-Methylacyl-CoA Racemase (AMACR), chromogranin A (CGA), synaptophysin (SYP), and CD56. Immunohistochemical analysis was performed at The Princess Margaret Cancer Centre for all patients.

In this study, the primary outcome was intratumoral DHT level. Guided by results from a study by Mizokami and colleagues (10), we assumed the DHT levels among patients treated with LHRH/bicalutamide to be 0.62 ng/g. Eleven patients in each arm were associated with a power of approximately 77% for detecting a mean of 0.2 ng/g in the degarelix arm as significantly different from the LHRH/bicalutamide arm, under the assumption that intratumoral androgen levels are exponentially distributed and tested using rank sum method with statistical significance set to 0.05 (due to the exploratory nature of the study). With an expected drop-out rate of 10%, a total of 39 patients (n = 39) were included to allow for at least 35 evaluable cases.

Statistical analyses were performed by treatment arm using all patients with evaluable data. For safety analyses, all patients who received at least one dose of safety drug were evaluated.

Intratumoral DHT levels were compared using the Kruskal–Wallis test with pair-wise comparisons using the Mann–Whitney approach. All hypothesis testing was two-sided with P ≤ 0.05 deemed as significant. Secondary analyses testing the effects of baseline covariates on the relationship between treatment groups were performed via linear regression model using treatment, baseline Gleason score, and baseline testosterone as covariates. Immunohistochemical staining outcomes and serum markers were similarly compared using the Kruskal–Wallis test, followed by the Mann–Whitney U tests to detect differences between two specific treatment groups. The percentage of extraprostatic extension and positive surgical margins were compared using Fisher exact test. Intention-to-treat analysis principle was followed. All analyses were performed using SPSS v.22.0.

Thirty-nine patients with Gleason score ≥7 prostate adenocarcinoma were enrolled (13 in degarelix only arm, 14 in degarelix/bicalutamide arm, and 12 in LHRH/bicalutamide arm). Baseline characteristics were well balanced between the three treatment arms (Table 1). Four patients discontinued bicalutamide prior to prostatectomy due to elevated liver function test results (3 from degarelix/bicalutamide arm and 1 from LHRH/bicalutamide arm), and 3 patients (2 from degarelix only arm and 1 from degarelix/bicalutamide arm) received additional injections of degarelix because of delay of surgery (due to patient personal reasons). All 39 patients subsequently underwent radical prostatectomy.

Stage pT3 disease was more prevalent at the time of prostatectomy (61.5%, 57.1%, and 50% in degarelix only, degarelix/bicalutamide, and LHRH/bicalutamide arms, respectively). The proportion of patients who had extraprostatic extension was similar among the three groups (P = 0.84). Likewise, the rate of margin positivity and lymph node positivity was similar (P = 0.62 and P = 0.14, respectively). There was one case of radical prostatectomy failure in each treatment arm (i.e., PSA did not reach undetectable levels after surgery). No differences existed in rates of PSA failure or use of adjuvant ADT or radiotherapy (P = 0.17). Only 10 cases were assigned a pathologic Gleason score (due to effects of neoadjuvant hormonal therapy). Among those assigned a score, there was not enough evidence to declare a difference in rate of change of baseline Gleason score among the three treatment arms (P = 0.09). Cancer mass in prostatectomy specimens was similar overall (P = 0.41; Table 2). Two patients in the degarelix/bicalutamide arm and 1 in the LHRH/bicalutamide arm had complete responses. In terms of failure rates, defined as PSA ≥0.2 ng/mL or receipt of adjuvant ADT or radiotherapy, no significant differences were observed between all three arms (Table 2; Fig. 1).

Serum PSA and hormone levels were measured at regular intervals throughout the study duration. Serum PSA levels were similar at baseline; however, at week 12 of treatment administration, significant differences existed among the three groups (P < 0.01), with the degarelix alone arm having higher PSA values compared with the degarelix/bicalutamide (5.02 vs. 0.29, P < 0.01) and LHRH/bicalutamide arms (5.02 vs. 0.48, P < 0.01). However, postoperatively, mean PSA values were similar again among the three groups (P = 0.99; Table 3). Serum testosterone and DHT levels were consistently similar at all measured intervals. Likewise, serum LH levels were similarly impacted by the three treatments. Serum FSH levels were different between the three treatment groups after 12 weeks of receiving treatment (P < 0.01), with higher FSH levels in LHRH/bicalutamide arm only (Table 3). Regularly measured GnRH levels were similar between the three arms; however, mean inhibin B levels were higher in the LHRH/bicalutamide arms after 12 weeks of therapy (Table 3).

Tumoral tissue analysis for PSA, PSAP, and AR positivity revealed these markers to be positive in all patients in across the three treatment arms (Table 4). These markers all had similar staining intensity in all tumor blocks analyzed. There were no significant differences in mean Ki67 levels, a marker of cellular proliferation, among the treatment arms (P = 0.21). However, we must note that there was a considerable, albeit not statistically significant difference, between the degarelix only and LHRH/bicalutamide arms (2.38 vs. 0.39, P = 0.07; Table 4; Supplementary Fig. S1). AMACR was significantly higher in the degarelix only arm compared with the LHRH/bicalutamide arm (87.27 vs. 43.91, P = 0.01; Table 4; Supplementary Fig. S2). No significant differences existed for levels of CD56, CGA, or SYP, markers of neuroendocrine differentiation (Table 4).

The primary outcome under assessment was whether mean intratumoral DHT levels were different between the three treatment arms. Analyses revealed significant differences between degarelix alone and degarelix/bicalutamide arms (0.87 vs. 0.26, P < 0.01) as well as degarelix alone versus LHRH/bicalutamide (0.87 vs. 0.23, P < 0.01; Table 5; Supplementary Figs. S3–S6). No differences in intratumoral testosterone levels or those of any other androgen precursors were noted (Table 5; Supplementary Fig. S7).

A multiple regression analysis was run to predict intratumoral DHT levels from treatment arm (degarelix vs. degarelix/bicalutamide vs. LHRH/bicalutamide), baseline serum testosterone, and Gleason score (based on biopsy findings). These variables statistically significantly predicted DHT levels, F(3,34) = 7.887, P < 0.01, R² = 0.410. The treatment arm (with degarelix alone arm displaying higher levels) and baseline serum testosterone variables added statistically significantly to the prediction, P < 0.05.

Safety was evaluated in the neoadjuvant and perioperative period, up to 6 weeks after surgery. Overall, all three treatment modalities were well tolerated (Table 6). There were 9 incidences of grade ≥3 treatment-emergent adverse events (TEAE), distributed evenly across the three treatment groups. The most frequent (TEAEs) were similar, with hot flashes and fatigue being the two most common. There were four cases of TEAEs (elevated LFTs) that necessitated early discontinuation of bicalutamide (3 in degarelix/bicalutamide arm and 1 in LHRH/bicalutamide arm; Supplementary Table S1).

Degarelix is a GnRH antagonist recently approved in the management of advanced, hormone-sensitive prostate cancer. Unlike LHRH, this compound does not result in a “flare” phenomenon and rapidly suppresses testosterone levels (11), which could translate to improved clinical outcomes. A large randomized trial comparing degarelix and leuprolide, an LHRH, demonstrated that degarelix produces faster PSA and testosterone reductions within the first month of therapy (6). Treatment with degarelix produced immediate and sustained reductions in LH and FSH and a greater percentage decrease from baseline of FSH at one year compared with leuprolide (89% vs. 55%; ref. 6). Importantly, degarelix led to improved 1-year PSA progression-free survival (P = 0.04) and reduced the probability of PSA failure at one year in patients with baseline PSA >20, as compared with leuprolide (29% vs. 40%, P = 0.01; ref. 12). Degarelix was shown to be as effective as leuprolide in maintaining low levels of testosterone ≤0.5 ng/mL during the first year of treatment, without the need for flare protection (6).

In designing this trial, we decided to assess three arms: degarelix alone, degarelix plus bicalutamide, and LHRH plus bicalutamide. We included the degarelix/bicalutamide arm, as we felt that although this combination is not used in clinical practice, testosterone and DHT suppression may be influenced by adrenal sources (13).

The study was designed to determine whether neoadjuvant treatment with degarelix was superior to a combination of an LHRH and bicalutamide in terms of intratumoral DHT levels. The results of this study were in contradiction to our hypothesis, whereby those patients receiving degarelix alone had significantly higher mean DHT levels (0.87 ng/mL vs. 0.23 ng/mL, P < 0.01) compared with those receiving a combination of LHRH/bicalutamide or degarelix/bicalutamide (Table 5). Interestingly, the values of the other intratumoral androgen precursors (i.e., testosterone, DHEA, androstenedione, etc.) and serum DHT levels were similar in all treatment arms. Increased levels of the potent androgen DHT would suggest a risk of increased cellular proliferation, which is supported by the immunohistochemical analyses demonstrating a trend toward tumor aggressiveness in terms of increased Ki67 (P = 0.07) and AMACR (P = 0.01) values in those receiving degarelix only (Table 4). Ki67 is a known marker of disease aggressiveness in prostate cancer (14). Increased AMACR staining has also been associated with most virulent oncogenic properties (15). However, in spite of higher mean values of these negative predictor markers, time-to-event analysis, albeit with short-term follow-up, showed no significant differences in the rates of PSA failure or use of adjuvant ADT or radiotherapy (Fig. 1). Indeed, we strongly feel that the survival data are underpowered and should not guide clinical decision making.

These unexpected results, especially in the context of similar androgen precursor levels, might suggest activation of an alternate pathway for de novo androgen synthesis. This analogy is drawn from castration-resistant prostate cancer cells that continue to achieve high intra-tissue levels of androgens despite castrate levels of serum androgens. These cells upregulate the levels of enzymes necessary for androgen synthesis and are capable of de novo conversion of [¹⁴C]acetic acid to DHT, with production of six other steroids upstream of DHT (16). We hypothesize that degarelix, through its immediate suppression of gonadotropins, perhaps leads to a similar phenomenon with activation of this alternate androgen production pathway, thus possibly explaining the elevated DHT and normal androgen precursor levels in the degarelix only–treated arm. An alternate hypothesis that may explain these findings is that bicalutamide could have theoretical off-target properties, such as DHT-synthesis inhibition or perhaps may influence DHT glucuronidation, resulting in higher DHT levels (17). We have however found no support for such an effect in the published literature. Given that intraprostatic DHT levels were elevated in the degarelix only arm, while testosterone levels were similar, this may suggest hormonal effects on tissue 5-a reductase activity. Future studies evaluating potential changes in this enzymatic activity are necessary. Our results may also simply suggest that degarelix is inferior to hormonal combinations in suppressing intraprostatic DHT. Future studies are necessary to evaluate the validity of such a hypothesis and the consistency of our results.

Despite the increased levels of intratumoral DHT and markers of cellular proliferation, patients receiving degarelix did have lower serum levels of FSH and inhibin B after 12 weeks of therapy (Table 3). There has been mounting preclinical data suggesting that FSH may play a role in the pathogenesis of prostate cancer. FSH receptors have been found in both prostate cancer cell lines and cancer specimens (18). It has further been shown that FSH stimulates proliferation and suppresses apoptosis in androgen-independent PC-3 cells in vitro (19). Similarly, inhibin, an endocrine hormone, is a member of the TGFβ superfamily, which is recognized to have regulatory roles in both normal and malignant prostate tissue (20). It is thus possible that the decreased levels of both FSH and inhibin B could balance out the impact of increased DHT on the prostate cancer, explaining the similarity of results in the time-to-failure analysis.

The major limitation of our study is the small patient sample size (n = 39), which entails a higher risk of imbalanced covariates at baseline, and thus, any clinical results obtained need to be validated in larger phase III trials. Another limitation was the open-label nature of our study and its associated inherent biases. The relatively short duration of follow-up precludes adequate comparison of long-term treatment outcomes among patients across the three study arms. In addition, our study was not adequately powered to assess for statistically significant differences in our reported secondary clinical endpoints.

Our data suggest that further studies may still be needed to fully evaluate the potential role of degarelix in the neoadjuvant setting. These studies need to evaluate the possible enzymatic and endocrinologic mechanisms behind increased intratumoral DHT levels in prostate cancer patients receiving degarelix.

K. Hersey is a consultant/advisory board member for Ferring. G. Kulkarni is a consultant/advisory board member for Astellas, Bayer, and Janssen. A.R. Zlotta is a consultant/advisory board member for 3D Biopsy, Ferring, Sanofi-Aventis, and Sanofi-Pasteur. R. Hamilton reports receiving speakers bureau honoraria from AbbVie and is a consultant/advisory board member for Astellas and Janssen. N.E. Fleshner reports receiving commercial research grants from Amgen, Astellas, Bavarian Nordic, Bayer, Canadian Cancer Society Research Institute, Diagnocure, Ferring, Janssen, Medivation, Nucleix, Progenics, Prostate Cancer Canada, Sanofi, and Spectracure AB and is a consultant/advisory board member for AbbVie, Amgen, Astellas, Bayer, Ferring, Hybridyne Imaging Technologies, Janssen, and Sanofi. No potential conflicts of interest were disclosed by the other authors.

As the sole sponsor, the investigator assumed responsibility for the conduct of this trial.

Conception and design: A. Evans, G. Kulkarni, A. Finelli, A.R. Zlotta, M. Gleave, N.E. Fleshner

Development of methodology: A. Evans, R. Maloni, G. Kulkarni, A. Finelli, N.E. Fleshner

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R.K. Sayyid, A. Evans, K. Hersey, R. Maloni, A. Hurtado-Coll, A.R. Zlotta, R. Hamilton, M. Gleave, N.E. Fleshner

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R.K. Sayyid, A. Evans, G. Kulkarni, R. Hamilton, M. Gleave, N.E. Fleshner

Writing, review, and/or revision of the manuscript: R.K. Sayyid, A. Evans, K. Hersey, G. Kulkarni, A. Finelli, A.R. Zlotta, R. Hamilton, M. Gleave, N.E. Fleshner

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): R.K. Sayyid, K. Hersey, R. Maloni, A. Hurtado-Coll

Study supervision: A. Evans, K. Hersey, R. Maloni, N.E. Fleshner

Financial support for this investigator-initiated trial was provided by Ferring.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.