SUMMARY

• No prospective, randomized, head-to-head trials comparing the efficacy and safety of ERLEADA with darolutamide have been conducted.
• Apalutamide is an androgen receptor (AR) inhibitor that is structurally distinct from the AR inhibitor darolutamide.^1,2
• MAIC Studies: Results are described below for matching-adjusted indirect comparison (MAIC) studies that were conducted, utilizing data from the phase 3 SPARTAN (ERLEADA plus androgen deprivation therapy [ADT] vs placebo plus ADT) and ARAMIS (darolutamide plus ADT vs placebo plus ADT) studies³ as well as utilizing data from SPARTAN, ARAMIS, and the phase 3 PROSPER (enzalutamide plus ADT vs placebo plus ADT) study⁴ in patients with non-metastatic castration-resistant prostate cancer (nmCRPC).
• Retrospective Real-World Studies: Several studies have compared the use of ERLEADA vs darolutamide in patients with metastatic castration-sensitive prostate cancer (mCSPC) and nmCRPC.
  • In patients with androgen receptor pathway inhibitor (ARPI)-naïve mCSPC, comparisons of undetectable PSA0.2 response by 6 months and treatment patterns and clinical outcomes in patients initiated on ERLEADA vs darolutamide have been conducted. In the more recent analysis,⁵ PSA0.2 response was achieved in 71.0% vs 55.2% in the ERLEADA group vs darolutamide group, respectively (weighted HR, 1.44; 95% CI, 1.03-2.02; P=0.032). In a second analysis,⁶ PSA90 by 6 months among patients with baseline prostate-specific antigen (PSA) >0.2 ng/mL included 77.1% (95% CI, 72.2-81.0) in the ERLEADA group, 68.0% (95% CI, 51.2-79.1) in the darolutamide plus chemotherapy group, and 58.2% (95% CI, 45.8-67.7) in the darolutamide without chemotherapy group and by 12 months postindex, the proportion of patients progressing to castration resistance was 8.6% (95% CI, 6.3-10.8), 26.0% (95% CI, 12.4-37.5), and 14.2% (95% CI, 7.3-20.6), respectively.
  • In a descriptive analysis to describe the central nervous system (CNS) outcomes among United States-based patients with nmCRPC using ERLEADA vs darolutamide, a lower proportion of patients in the ERLEADA group experienced time-to-new onset of any CNS-related conditions at both 12-months (ERLEADA: 25.7%, darolutamide: 31.4%) and 24-months (ERLEADA: 46.6%, darolutamide: 54.6%) post-index. The patients in the ERLEADA group had a longer median time-to-new onset of any CNS-related conditions (29.2 months) compared with the darolutamide group (21.3 months) and the rates of commonly observed new onset CNS-related conditions were lower in the ERLEADA group compared with the darolutamide group.⁷ 
• Several additional real-world studies have evaluated efficacy and safety outcomes in patients with prostate cancer treated with ERLEADA or darolutamide, including:
  • A retrospective, observational chart review of the DEAR-EXT study and its race subgroup analysis that included a comparison of ERLEADA and darolutamide treated patients with nmCRPC.^8,9
  • A retrospective cohort study that compared clinical use and outcomes of ERLEADA and darolutamide in patients with nmCRPC.¹⁰ 
  • A prevalence study of potential drug-drug interactions that included a comparison of ERLEADA and darolutamide in patients with nmCRPC.¹¹
  • Studies reporting adverse events (AEs) in patients with prostate cancer treated with medications including ERLEADA and darolutamide.^12,13 
  • A retrospective, descriptive analysis that evaluated healthcare resource utilization and medical costs among patients with mCSPC initiated on ERLEADA vs darolutamide.¹⁴

CLINICAL DATA

No prospective, randomized, head-to-head clinical trials comparing the efficacy and safety of ERLEADA with darolutamide have been conducted.

ADDITIONAL DATA

MAIC Studies

Indirect treatment comparisons (ITCs) can be limited by cross-trial differences. By combining individual patient data (IPD) with published aggregate patient data, MAIC studies may reduce observed cross-trial differences and provide decision makers with timely comparative evidence.¹⁵

Chowdhury et al (2022)³ conducted an anchored MAIC study with a Bayesian framework that compared the efficacy and tolerability of ERLEADA plus ADT and darolutamide plus ADT in patients with nmCRPC.

The anchored MAIC was conducted using individual patient-level data from the phase 3 SPARTAN study (ERLEADA plus ADT vs placebo plus ADT) and aggregated patient data from the phase 3 ARAMIS study (darolutamide plus ADT vs placebo plus ADT) via the following steps:

• Application of inclusion/exclusion criteria from ARAMIS to SPARTAN prior to inclusion in MAIC
• Reweighting patients in SPARTAN to match aggregate baseline data of ARAMIS
• Recalculation of hazard ratios (HRs; for efficacy endpoints) and odds ratios (ORs; for tolerability endpoints) from the SPARTAN study based on the reweighted population
• Comparison of the reweighted SPARTAN data with ARAMIS data using a Bayesian framework with ADT as the common comparator across both studies

The primary efficacy endpoint was metastasis-free survival (MFS). Secondary efficacy endpoints included PSA progression, progression-free survival (PFS), and overall survival (OS). Tolerability endpoints included the overall incidence of any AEs and serious AEs. Efficacy endpoints were estimated using HRs and 95% credible intervals (CrIs), and tolerability endpoints were estimated using ORs. For the efficacy (except OS) and tolerability endpoints, the first interim analyses from both the SPARTAN and ARAMIS studies were used to compare ERLEADA and darolutamide. For OS, both the first and second interim analyses of the SPARTAN study were compared with the final analysis of the ARAMIS study.

In the first interim analysis, the median duration of follow-up for the SPARTAN and ARAMIS studies was 20.3 months and 17.9 months, respectively. Clinically relevant baseline characteristics were broadly comparable among patients in both studies (Table: Baseline Characteristics).

After matching, analysis of efficacy endpoints suggested that patients with nmCRPC treated with ERLEADA plus ADT had more favorable MFS, PFS, and PSA progression compared with those treated with darolutamide plus ADT. The OS benefit of ERLEADA plus ADT was similar to that of darolutamide plus ADT in the first and second interim analyses (Table: Efficacy Results in the ARAMIS and SPARTAN Studies and in the MAIC Study).

Analysis of tolerability endpoints suggested that both treatments have a similar tolerability profile based on incidence of any AE or serious AE (Table: Tolerability Results Before and After Matching and in MAIC).

Halabi et al (2021)⁴ conducted an anchored MAIC study that evaluated safety outcomes and MFS at primary analysis of patients with nmCRPC treated with darolutamide, ERLEADA, and enzalutamide in the ARAMIS, SPARTAN, and PROSPER studies, respectively. IPD from ARAMIS were selected and reweighted to match the inclusion criteria and baseline characteristics published in SPARTAN and PROSPER. The darolutamide vs ERLEADA MAIC matched on 7 covariates that included: age, PSA level and doubling time, Eastern Cooperative Oncology Group (ECOG) performance status, Gleason score, bone-sparing agent use, and prior surgery. For darolutamide vs ERLEADA, the effective sample size for the darolutamide and its placebo groups were 604 and 391, respectively, after matching. Of the AEs evaluated, there was no statistically significant difference between darolutamide and ERLEADA for dizziness, mental-impairment, hypertension, seizure, diarrhea, nausea, fatigue, and severe fatigue. Compared to ERLEADA, fall (risk difference [RD], -6.3%; OR, 0.6; Benjamini-Hochberg-adjusted p-value [p_BH]<0.05), fracture (RD, -6.2%; OR, 0.4; p_BH<0.05), and rash (RD, -16.0%; OR, 0.5; p_BH<0.0001) were significantly lower for darolutamide. Monitoring for AEs occurred more frequently in the SPARTAN study compared to the ARAMIS study and the impact of monitoring frequency on overall safety outcomes is unknown.^16,17 No MFS differences were observed before or after matching. The pre-match and post-match ratios of MFS HRs between darolutamide and ERLEADA were 1.29 (95% CI, 0.95-1.74) and 1.14 (95% CI, 0.83-1.58), respectively. Additionally, in the SPARTAN study the median treatment duration for the ERLEADA and placebo groups was 16.9 months and 11.2 months, respectively, and in the ARAMIS study was 14.8 months and 11.0 months in the darolutamide and placebo groups, respectively.^18,19

Retrospective Real-World Studies

Lowenritt et al (2026)⁵ conducted a retrospective, real-world, longitudinal analysis comparison of undetectable PSA0.2 response (defined as a decline in PSA value ≤0.2 ng/mL) by 6 months in patients with ARPI-naïve mCSPC treated with ERLEADA without docetaxel (n=714) vs darolutamide without docetaxel (n=145). Data was linked from a clinical urology database (Precision Point Specialty Analytics [PPS]) and administrative claims database (Komodo Research Database [KRD]) and the study period was from August 2021-April 2025. Patient selection criteria included having a detectable PSA (>0.2 ng/mL) in the most recent measurement within the 13-week period prior to and including the index date and no evidence of docetaxel use within 90 days prior to and 30 days after the index date. Inverse-probability of treatment weighting (IPTW) was used to balance potentially confounding baseline demographics. The median time between metastasis and index was 1.9 months vs 2.0 months and concurrent ADT use was 98.1% vs 97.4% in the ERLEADA group vs darolutamide group, respectively. Achievement of a PSA0.2 response was 71.0% vs 55.2% in the ERLEADA group vs darolutamide group, respectively (weighted HR, 1.44; 95% CI, 1.03-2.02; P=0.032). There were 285 patients vs 63 patients at 3 months and 114 patients vs 33 patients at 6 months that achieved PSA0.2 response in the ERLEADA group vs darolutamide group, respectively.

Burbage et al (2025)⁶ conducted a retrospective, real-world, descriptive analysis of treatment patterns and clinical outcomes in patients with ARPI-naïve mCSPC initiated on ERLEADA (n=940) vs darolutamide (n=239; stratified by receipt [n=65] or absence [n=174] of concomitant chemotherapy) was conducted. Data was linked from PPS and KRD and the study period was from October 2015-October 2024. Concurrent ADT use was 88.6% vs 98.5% vs 92.5% and mean PSA level was 21.9 ng/mL vs 94.0 ng/mL vs 60.5 ng/mL in the ERLEADA group, darolutamide plus chemotherapy group, and darolutamide without chemotherapy group, respectively. Achievement of a 90% PSA reduction by 6 months among patients with baseline PSA >0.2 ng/mL included 77.1% (95% CI, 72.2-81.0) in the ERLEADA group, 68.0% (95% CI, 51.2-79.1) in the darolutamide plus chemotherapy group, and 58.2% (95% CI, 45.8-67.7) in the darolutamide without chemotherapy group. By 12 months postindex, the proportion of patients progressing to castration resistance was 8.6% (95% CI, 6.3-10.8), 26.0% (95% CI, 12.4-37.5), and 14.2% (95% CI, 7.3-20.6), respectively.

Lledo et al (2025)⁷ conducted a retrospective, real-world, descriptive analysis of central nervous system (CNS) outcomes among United States-based patients with nmCRPC, which included ERLEADA (n=253) and darolutamide (n=544) cohorts. Data was linked from PPS and KRD and the study period was from January 2016-August 2024. Concomitant use of ADT was not required for inclusion. The mean duration of the on-treatment observation period was 12.4 months for ERLEADA group and 14.0 months for the darolutamide group. A lower proportion of patients in the ERLEADA group experienced time-to-new onset of any CNS-related conditions at both 12-months (ERLEADA: 25.7%, darolutamide: 31.4%) and 24-months (ERLEADA: 46.6%, darolutamide: 54.6%) post-index. The patients in the ERLEADA group had a longer median time-to-new onset of any CNS-related conditions (29.2 months) compared with the darolutamide group (21.3 months) and the rates of commonly observed new onset CNS-related conditions (fatigue, falls, dizziness, pain, and weakness) were lower in the ERLEADA group compared with the darolutamide group.

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 08 January 2026. Due to methodological differences and other potential limitations including scope, timing, study designs and statistical approaches, network meta-analyses and meta-analyses have been excluded in this response. Additional data beyond these parameters may be available in the literature.