SUMMARY

• MFS has been shown to be a strong surrogate endpoint for overall survival (OS) in localized prostate cancer as well as castration-resistant prostate cancer (CRPC).^1,2
• In SPARTAN, the phase 3 study that evaluated the efficacy and safety of ERLEADA plus androgen deprivation therapy (ADT) compared to placebo plus ADT in patients with high-risk non-metastatic castration-resistant prostate cancer (nmCRPC; N=1207), a significant improvement in the primary endpoint MFS was observed.³
  • At the final analysis (following 378 events) for MFS, significant improvement in median MFS was observed: 40.5 months in the ERLEADA group vs 16.2 months in the placebo group (HR, 0.28; 95% CI, 0.23-0.35; P<0.001). At the first interim analysis for overall survival (OS; secondary endpoint), ERLEADA was associated with longer median OS compared to placebo (HR, 0.70; 95% CI, 0.471.04; P=0.07), however this data was considered immature (104 events; 24% of required number of events).^3,4
  • On the basis of the efficacy and safety data from the primary analysis, the independent data and safety monitoring committee unanimously recommended unblinding the study and offering patients assigned to the placebo group the option to receive ERLEADA.³
  • At the final analysis (following 428 events) for OS, after a median follow-up of 52 months, a statistically significant improvement in median OS was observed: 73.9 months in the ERLEADA group vs 59.9 months in the placebo group (HR, 0.78; 95% CI, 0.64-0.96, P=0.016).⁵
  • Adverse events (AEs) that occurred in ≥15% of patients in the ERLEADA group included: fatigue, hypertension, diarrhea, fall, arthralgia, nausea, weight decreased, back pain, and hot flush.⁵
• Additional analyses, including the second interim analysis for OS, a subgroup analysis, post-hoc analyses, and multivariate analyses have been conducted.^6-17
• PROTEUS (NCT03767244) is an ongoing phase 3, randomized, double-blind, placebo controlled, multicenter study evaluating the efficacy and safety of ERLEADA plus ADT compared to placebo plus ADT before and after radical prostatectomy (RP) in patients with localized or locally advanced high-risk prostate cancer. The dual primary endpoints are pathologic complete response (pCR) rate and MFS. The study is expected to enroll approximately 2000 patients from >203 sites in 18 countries. Safety and efficacy results from the PROTEUS study have not been published.^18,19

BACKGROUND

Surrogate Endpoints in Prostate Cancer Clinical Trials

Due to prolonged survival, surrogate endpoints for OS are needed in CRPC to assess clinical efficacy within a feasible time frame.^2,20 A number of endpoints have been proposed for clinical trials and are being validated in clinical studies.^21,22 Proposed surrogate endpoints for OS include MFS, biochemical failure, prostate-specific antigen (PSA) doubling time (PSADT), PSA nadir, end of treatment PSA, disease-free survival, and radiographic progression-free survival (PFS).^1,20,22

MFS assesses the time interval from PSA recurrence (or start of trial/randomization) to first radiographic metastasis or death.^1,2,23 MFS has been shown to correlate with OS in men with localized and biochemically recurrent prostate cancer in several studies, including a recent study by the International Clinical Endpoints for Cancer of the Prostate (ICECaP) Working Group.^1,2,22 The aim of the ICECaP Working Group study (to accelerate adjuvant prostate cancer study readouts) resulted in rigorous methodology that evaluates potential intermediate clinical endpoints for OS surrogacy, using individual patient data from international randomized controlled trials.²²

In this systematic review of 19 randomized controlled trials evaluating 21 study units in 12,712 patients with localized prostate cancer, the correlation between OS and MFS was 0.91 (95% CI, 0.91-0.91) at the patient level, as measured by Kendall’s τ. At the trial level, R² was 0.83 (95% CI, 0.71-0.88) from weighted linear regression of 8-year OS rates vs 5year MFS rates. The treatment effect (measured by log HR) for MFS and OS was well correlated (R², 0.92 [95% CI, 0.81-0.95]).¹ Similarly, in an analysis of patients with nmCRPC at a single cancer center, MFS was found to be strongly correlated with OS in a real world population.²⁴

In a retrospective study using the Optum electronic health record database in patients with nmCRPC, MFS independently predicted mortality risk. Patients developing metastasis within 1 year had a 4.4-fold greater risk for mortality (95% CI, 2.2-8.8) than those who remained metastasis-free at year 3.²⁵

Delaying metastases, or extending MFS, may also delay symptomatic progression, morbidity, mortality, costs, and healthcare resource utilization.^26,27

MFS - Regulatory History and Selection as Primary Endpoint in the SPARTAN Study

MFS was the primary endpoint of the pivotal study, SPARTAN, used to support regulatory agency approval of ERLEADA.^3,28

The appropriate primary efficacy endpoint to use in trials for study patients with nmCRPC was addressed at two meetings of the Food and Drug Administration’s (FDA) Oncology Drugs Advisory Committee. The first meeting was non-product specific and discussed general issues related to the development of drugs for the treatment of nmCRPC, while the second meeting was focused on the review of a specific drug for potential approval in the nmCRPC setting. There was recognition that the transition to a state where the patient developed measurable metastatic disease represented a significant event for patients with nmCRPC. A consensus emerged that an MFS endpoint would likely be clinically meaningful if it was of sufficient magnitude and was accompanied by data from supportive secondary endpoints such as OS. Approval of ERLEADA based on MFS, the primary efficacy endpoint of SPARTAN, established a new regulatory precedent.²⁸

Additionally, in November 2018, the FDA issued a draft guidance which provides recommendations to sponsors regarding the use of MFS as an endpoint in clinical trials for nmCRPC development programs for drug or biological products regulated by the Center for Drug Evaluation and Research (CDER) and the Center for Biologics Evaluation and Research (CBER). The FDA has recognized that a prolonged delay in development of metastatic disease is an objective and clinically relevant measure. Furthermore, FDA has stated that future agents may be approved on the basis of MFS only if substantial effects on MFS are demonstrated and the safety profile is acceptable for a medication taken long-term.^29,30 This endpoint also has broad support within the oncologic community as an intermediate endpoint that is well correlated with OS.³¹

Results from the SPARTAN study have been summarized in the CLINICAL DATA section.

CLINICAL DATA

SPARTAN Study

The SPARTAN study evaluated the efficacy and safety of ERLEADA compared to placebo in patients with high-risk (defined as PSADT ≤10 months) nmCRPC (N=1207).³

Study Design/Methods

SPARTAN (NCT01946204) was a phase 3, randomized, double-blind, placebo-controlled, multicenter study (Figure: SPARTAN Study Design). A total of 332 sites in 26 countries in North America, Europe, and Asia-Pacific were included.

• Disease assessments using technetium-99m bone scans and computed tomography (CT) scans were performed every 16 weeks for the pelvis, abdomen, and chest and additionally if distant metastases were suspected.³
• Radiographic evidence of distant metastasis was determined based on Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 and all imaging was assessed by blinded independent central review (BICR). When new bone lesions were detected by bone scan, a second imaging modality (CT or magnetic resonance imaging) was required to confirm metastasis.
• Select exclusion criteria: presence of detectable distant metastases; prior treatment with second-generation antiandrogens, cytochrome P450 (CYP) 17 inhibitors, radiopharmaceutical agents, immunotherapy, or any other investigational agent(s) for nmCRPC;³² prior chemotherapy for prostate cancer, except if administered in the adjuvant/neoadjuvant setting;³² history of seizure or condition that may predispose to seizure³²
• Primary endpoint: MFS, defined as the time from randomization to the time of first evidence of BICR-confirmed distant metastasis, defined as new bone or soft tissue lesions or enlarged lymph nodes above the iliac bifurcation, or death due to any cause, whichever occurred first³²
• Select secondary endpoints:
  • Time to metastasis, defined as time from randomization to first detection of distant metastasis involving bone or soft tissue on imaging, as assessed by BICR
  • PFS, defined as time from randomization to first detection of local or distant metastatic disease on imaging, assessed by BICR, or death from any cause, whichever occurred first
  • Time to symptomatic progression, defined as time from randomization to skeletalrelated event, pain progression, or worsening of disease-related symptoms leading to initiation of a new systemic anticancer therapy, or the time to development of clinically significant symptoms due to local or regional tumor progression requiring surgery or radiation therapy
  • OS
  • Safety and tolerability³²
• Other exploratory endpoints were additionally assessed.
• After the first detection of distant metastasis, patients were eligible to receive sponsorprovided abiraterone acetate plus prednisone as a treatment option. All poststudy treatment for mCRPC was at the treating physician’s discretion.

Results

Patient Characteristics

• Patient demographics and disease characteristics were well balanced between the 2 groups, and there were no significant differences.
• Six patients (3 per group) were randomized, but never received study treatment.
• At the clinical cutoff date for the primary analysis, driven by metastasis events or death, the median follow-up was 20.3 months, with 60.9% of patients still on treatment in the ERLEADA group vs 29.9% in the placebo group.
• On the basis of the efficacy and safety data from the primary analysis, the independent data and safety monitoring committee unanimously recommended unblinding the study and offering patients assigned to the placebo group the option to receive ERLEADA.
  • After unblinding of the study, 76 patients (19%) in the placebo group received therapy with ERLEADA plus ADT (crossover group).⁵
• At the clinical cutoff date for the final OS analysis, 30% (237/803) and 61% (46/76) of the patients were still on treatment in the ERLEADA and crossover groups, respectively.⁵

Efficacy

• At the final analysis for MFS performed after 378 events of distant metastasis or death (184 [22.8%] patients in the ERLEADA group and 194 [48.4%] patients in the placebo group), significant improvement in median MFS was observed: 40.5 months vs 16.2 months in the ERLEADA group vs the placebo group, respectively (HR, 0.28; 95% CI, 0.23-0.35; P<0.001).
• The treatment effect of ERLEADA on MFS was consistently favorable across prespecified subgroups, including patients with PSADT ≤6 months vs >6 months, use of bonesparing agents, and local-regional disease.
• At the primary analysis, ERLEADA was associated with improvements in all secondary endpoints, with significant improvement observed in time to metastasis, PFS, time to symptomatic progression (all P<0.001).³ At the final analysis, ERLEADA was associated with significant improvement in the secondary endpoint of OS (P=0.016).⁵
  • At the final analysis (following 428 events) for OS, after a median follow-up of 52 months, a statistically significant improvement in median OS was observed: 73.9 months in the ERLEADA group vs 59.9 months in the placebo group (HR, 0.78; 95% CI, 0.64-0.96, P=0.016).⁵
• A total of 48% of ERLEADA-treated patients and 71% of placebo-treated patients received first subsequent systemic therapy for prostate cancer, with abiraterone acetate plus prednisone as the most common first subsequent therapy (73% in the ERLEADA group and 72% in the placebo group). Of the 401 patients randomized to the placebo group, 338 (84%) patients received either life-prolonging active therapy as the first subsequent therapy upon disease progression or ERLEADA as a crossover treatment option without progression after study unblinding.⁵

Safety

• The safety population included all patients who received at least 1 dose of study drug.
• At the final analysis for OS, the median treatment duration was 32.9 months in the ERLEADA group, 11.5 months in the placebo group, and 26.1 months in the crossover group.⁵
• AEs that occurred at ≥15% in the ERLEADA group (all grades) included: fatigue (30.4%), hypertension (24.8%), rash (23.8%), diarrhea (20.3%), nausea (18.1%), weight loss (16.1%), arthralgia (15.9%), and falls (15.6%). Other AEs of interest included fracture (11.7%), dizziness (9.3%), hypothyroidism (8.1%), mental impairment disorder (5.1%), and seizure (0.2%).³
  • In the final OS analysis, AEs that occurred in ≥15% of patients in the ERLEADA group (all grades) included: fatigue (33%), hypertension (28%), diarrhea (23%), fall (22%), arthralgia (20%), nausea (20%), weight decreased (20%), back pain (18%), and hot flush (15%).⁵
• Treatment discontinuation due to progressive disease was reported in 155 (19.3%) patients in the ERLEADA group and in 210 (52.8%) patients in the placebo group.³
  • In the final OS analysis, treatment discontinuation due to progressive disease was reported in 343 (43%) patients in the ERLEADA group, 238 (74%) patients in the placebo group, and 11 (14%) patients in the crossover group.³³
• Treatment discontinuation due to AEs was reported in 85 (10.6%) and 28 (7.0%) patients who received ERLEADA or placebo, respectively.³ Rash was the most common AE leading to treatment discontinuation (2.4% vs 0%), dose reduction (2.7% vs 0.3%), and dose interruption (6.8% vs 1.3%) in the ERLEADA group vs the placebo group, respectively.³⁴
  • In the final OS analysis, treatment discontinuation due to AEs was reported in 122 (15%), 27 (8.4%), and 8 (11%) patients in the ERLEADA, placebo, and crossover groups, respectively.³³
• AEs were associated with death in 10 patients in the ERLEADA group: prostate cancer (n=2), sepsis (n=2), acute myocardial infarction (n=1), cardiorespiratory arrest (n=1), cerebral hemorrhage (n=1), myocardial infarction (n=1), multiple organ dysfunction (n=1), and pneumonia (n=1).³

Landmark and Correlation Analysis of MFS and OS

• A landmark and correlation analysis was conducted to determine the relationship between MFS and OS in patients with nmCRPC included in the SPARTAN study.³⁵
• MFS was analyzed at 6, 9, and 12 months after treatment initiation.
  • Patients were analyzed according to their metastatic status at each time point.
  • The Kaplan-Meier method was used to estimate OS by metastatic status at each landmark time point.
• A multivariate proportional hazards model evaluated the significance of metastasis to predict OS at 6, 9, and 12 months.
• A parametric statistical model developed by Fleischer and colleagues was implemented to account for dependence of MFS on OS.
  • The model explains the correlation between endpoints in terms of the proportion of variability of OS attributable to the variability of MFS, and accounts for censoring.
• Results of the landmark analysis showed that patients who developed metastases at 6, 9, and 12 months had a significantly shorter median OS compared with those patients without metastasis. In the ERLEADA group, for patients who developed metastases by 6, 9, and 12 months compared with patients who did, the adjusted for prognostic factors HR were 6.55 (95%, 2.94-14.49), 7.63 (95% CI, 4.12-14.08), and 9.17 (95% CI, 5.0816.67), respectively.
  • Overall, 5, 14, and 26 patients died before 6, 9, and 12 months, respectively, and were excluded from the analyses.
• Results from the parametric Fleischer’s statistical model confirmed the positive association between MFS and OS, with an estimated correlation coefficient of 0.69 (bootstrap standard error, 0.002; 95% CI, 0.69-0.70).

Additional Information

Additional information regarding the SPARTAN study, including the clinical study report, protocol, and statistical analysis plan, can be found at: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2018/Erleada_210951_toc.cfm (scroll to the “Sponsor Clinical Study Reports ARN-509-003 SPARTAN NCT # 01946204” section at the bottom of the web page).

Literature Search

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, and Derwent Drug File (and/or other resources, including internal/external databases) was conducted on 31 July 2024. Summarized in this response are relevant data from the phase 3 SPARTAN study and a landmark and correlation analysis.