This study is a prospective, multicenter, open-label, randomized phase 2 study investigating the effect of neoadjuvant apalutamide or apalutamide with abiraterone acetate plus prednisone and a GnRH agonist (Lupron [AbbVie] or Zoladex [AstraZeneca]) on post-RP potency and continence in men with high-risk localized PCa. We propose that the use of apalutamide alone or apalutamide with abiraterone acetate plus prednisone and a GnRH agonist for 3 months preoperatively will downstage the local tumor such that bilateral NVB sparing becomes more feasible, resulting in the improvement of postsurgical potency without compromising oncologic outcomes. The protocol-specified study duration was 5 years, including 2 years of accrual, 2 years of follow-up, and 1 year of analysis.

This protocol was developed at and sponsored by the Rutgers Cancer Institute of New Jersey (CINJ) and received funding and medication support from Janssen Scientific Affairs. The protocol version date was April 13, 2021 (CINJ #081603).

This study was approved by the institutional review boards of Rutgers (Pro20160000563) and Yale (Human Investigation Committee number: 2000031291), and all patients provided written informed consent prior to enrollment. The protocol specified that informed consent must be obtained before any study-specific procedures. Additional safeguards were included for non–English-speaking participants through interpreter-assisted consent procedures and institutional review board–approved short-form documentation, when applicable.

Eligible participants were men with histologically confirmed prostate adenocarcinoma and high-risk localized disease. High-risk disease was defined as a Gleason score of 8 or higher or a prostate-specific antigen (PSA) level greater than or equal to 20 ng/mL with 2 or more positive cores. Patients were required to have clinical stage T3 or less; no evidence of metastasis; an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; and adequate baseline laboratory values, including hemoglobin, platelet count, albumin, renal function, potassium, bilirubin, and hepatic transaminases within protocol-defined limits. Clinical stage T3 or less remained eligible for the protocol-directed nerve-sparing approach; palpable disease and magnetic resonance imaging (MRI)–detectable lesion location were not specified as absolute contraindications to nerve sparing.

Key exclusion criteria included clinical stage T4 disease, metastatic PCa, a seizure history or conditions that predisposed patients to seizures, a recent major cardiovascular or thromboembolic event, clinically significant ventricular arrhythmias, moderate or severe hepatic impairment, and warfarin use. Given the known association between androgen receptor antagonists and seizure threshold reduction, patients taking medications known to lower the seizure threshold were required to discontinue or substitute those agents prior to study entry.

Participants were recruited from urologic and medical oncology clinics at CINJ (New Brunswick, New Jersey) and Yale Cancer Center (New Haven, Connecticut). Centralized enrollment procedures used the OnCore clinical trials management system. Following signed consent, patients were registered, screened using a study-specific eligibility checklist, and randomized only after confirmation of protocol eligibility.

Patients were randomized 1:1:1 to 1 of 3 treatment arms, with the full schema presented in Figure 1. Surgery consisted of RP with extended pelvic lymph node dissection (ePLND). Prior to surgery, study participants randomized to arm 1 received apalutamide (240 mg; 4×60 mg tablets) orally once daily for 3 months. Patients randomized to arm 2 received apalutamide (240 mg; 4×60 mg tablets), abiraterone acetate (1000 mg; 4×250 mg tablets), and prednisone (10 mg; 2×5 mg tablets to be taken in the morning) orally once daily without food for 3 months. Patients in arm 2 also received a subcutaneous injection of a GnRH agonist (Lupron, 22.5 mg intramuscularly, or Zoladex, 10.8 mg subcutaneously). Patients in arm 2 took apalutamide 1 week (±2 days) before receiving the GnRH agonist and abiraterone acetate with prednisone (AAP). Both apalutamide and abiraterone acetate were securely stored in the pharmacy in charge of dispensing the investigational products. Surgery was performed as early as possible but not earlier than 2 weeks after completion of apalutamide treatment in the treatment arms (arms 1 and 2).

The selected drug doses reflected established clinical dosing and available safety and pharmacokinetic data at the time of protocol development. Apalutamide 240 mg once daily corresponded to the established clinical dose subsequently used in pivotal phase 3 studies, including the Targeted Investigational Treatment Analysis of Novel Anti-Androgen study [6]. The combination of apalutamide with abiraterone acetate 1000 mg and prednisone 10 mg daily was selected in consultation with the study sponsor and was supported by pharmacokinetic data evaluating concurrent apalutamide and abiraterone acetate plus prednisone [8]. The GnRH agonist regimens, leuprolide 22.5 mg intramuscularly or goserelin 10.8 mg subcutaneously, represented standard 3-month depot formulations. The 3-month neoadjuvant duration was selected to allow sufficient androgen pathway suppression and potential local tumor downstaging while minimizing delays to definitive RP and, based on the principal investigator’s surgical experience, limiting the potential for periprostatic fibrosis associated with more prolonged preoperative systemic therapy.

Patients randomized to arm 3 immediately underwent surgery without receiving neoadjuvant treatment.

Patients deemed surgically eligible underwent robot-assisted RP with ePLND. Bilateral NVB sparing was the planned surgical approach and was performed in all patients undergoing RP; unilateral or wide NVB resection was not performed. Pathological assessment after surgery included margin status, pathological stage, and nodal status, with central genitourinary pathology review within each institution’s department.

Perioperative outcomes included major and minor complications based on the Clavien-Dindo classification system, estimated blood loss, operative time, and length of hospital stay.

Each patient underwent postoperative follow-up through 24 months. The follow-up schedule included visits at week 1; months 1, 2, and 3; and months 6, 9, 12, 15, 18, 21, and 24. Follow-up assessments included physical examination; adverse event documentation; PSA; selected laboratory studies, including serum testosterone and androgen levels and urinalysis; ECOG performance status; and QoL assessments using the AUA Symptom Score, SHIM, and EPIC models (Table 1).

The primary end point was potency rate by 12 months after RP. Secondary end points included complete pathological response rate (pathologic stage T0), positive surgical margin rate, biochemical recurrence rate, postoperative continence rate, change in tumor volume on pelvic MRI after neoadjuvant therapy, and postsurgical QoL.

Continence was defined using EPIC question 5 as the use of no pads, and potency was defined based on EPIC question 19 and SHIM questions 2 and 4 as the ability to achieve penetration and complete sexual intercourse in more than half of attempts.

Biochemical recurrence was defined using PCa Clinical Trials Working Group 2 criteria as a rising PSA level greater than 0.2 ng/mL, confirmed by a second value at least 3 weeks later.

Biospecimens were prospectively collected and banked to permit future exploratory translational studies. Potential correlative analyses include whole-exome sequencing, RNA sequencing, array comparative genomic hybridization, and methylation profiling of biopsy and postprostatectomy specimens. These analyses have not yet been initiated and are not included in the present clinical analysis. If undertaken, the exploratory objective will be to characterize genomic, transcriptional, and epigenetic features potentially associated with biological response to neoadjuvant androgen pathway inhibition and pathological outcomes. Specific laboratory and bioinformatic methods will be defined before initiation of these analyses and reported separately. Surgical and biopsy specimens, as well as serum, urine, plasma, and whole blood, were stored through the CINJ biospecimen bank under deidentification procedures.

Toxicity monitoring and adverse event reporting were protocol-defined for both apalutamide and abiraterone acetate. Dose modifications were specified for general toxicity, rash, liver function test abnormalities, hypokalemia, hypertension, and edema or fluid retention. Serious adverse events were reported within 24 hours according to institutional and sponsor procedures, and the Human Research Oversight Committee at CINJ served as the data and safety monitoring body for the trial.

Patients in the neoadjuvant arms were monitored prospectively for adverse events using protocol predefined dose holds, reductions, and discontinuation rules, with institutional oversight for safety and compliance.

The trial was designed around the primary end point of potency recovery by 12 months after RP. On the basis of the principal investigator’s historical surgical outcomes, the surgery-only potency recovery rate was estimated at 13%. Assuming 30 participants per arm, the planned sample size of 90 participants provided 80% power to detect an increase in potency recovery from 13% in the surgery-only arm to 45% in either experimental arm using 1-sided hypothesis testing. Because 2 prespecified experimental-vs-control comparisons were planned (arm 1 vs arm 3 and arm 2 vs arm 3), a Bonferroni-adjusted α level of .025 was applied to each comparison. A sample size of 30 participants per experimental arm was considered sufficient to detect a complete pathological response rate of 20%.

Baseline and study characteristics were summarized using descriptive statistics. Continuous variables were compared using the Wilcoxon rank-sum test, and categorical variables were compared using the Fisher exact test or chi-square test, as appropriate. For the primary end point, arm 1 and arm 2 were each compared separately with the immediate-surgery control arm using 1-sided Fisher exact tests with a prespecified Bonferroni-adjusted significance threshold of α=.025. Unless otherwise specified, additional statistical tests were 2-sided, with P<.05 considered statistically significant.

As the study intervention was designed to affect postoperative functional recovery through the surgical procedure, the primary analysis included participants who received the assigned treatment and proceeded through the surgical treatment pathway. To evaluate the robustness of the primary functional analyses to postrandomization withdrawal, a conservative intention-to-treat sensitivity analysis was additionally performed including all randomized participants. Participants who withdrew following randomization were classified as failing the functional end points, not achieving potency recovery at 12 months, and not achieving continence recovery at the assessed postoperative time points.

Biochemical recurrence-free survival was estimated using Kaplan-Meier methods and compared between treatment arms using the log-rank test. Statistical analyses were performed using Stata statistical software (version 18; StataCorp).