Impact of enzalutamide on patient-reported fatigue in patients with prostate cancer: data from the pivotal clinical trials

Fatigue is one of the most common and burdensome symptoms experienced by men with prostate cancer, driven by the cancer itself, its treatments (including ADT, chemotherapy, and novel hormonal agents), and patient factors such as age. Symptoms often worsen as disease advances toward metastatic castration-resistant prostate cancer. Discrepancies between patient-reported and clinician-reported symptoms are well documented, particularly for subjective symptoms like fatigue. Given enzalutamide’s widespread use across mHSPC and CRPC, the study aimed to evaluate baseline fatigue across the disease continuum and quantify the impact of enzalutamide plus ADT versus placebo plus ADT on patient-reported fatigue using FACT-P item GP1 across four phase 3 trials.

Prior work shows fatigue is prevalent throughout the prostate cancer trajectory and may worsen with systemic therapies including abiraterone, enzalutamide, docetaxel, PARP inhibitors, radium-223, and cabazitaxel. Studies have observed inconsistencies between patient- and clinician-reported adverse events. A recent report highlighted wide variability in patient-reported fatigue rates across similar trials, including placebo arms, suggesting the influence of data collection methods, patient selection, and regional differences.

Design: Post hoc exploratory analysis of patient-reported outcomes from four double-blind, randomized, placebo-controlled phase 3 trials comparing enzalutamide 160 mg/day plus ADT to placebo plus ADT: ARCHES (mHSPC), PROSPER (high-risk nmCRPC), PREVAIL (chemotherapy-naïve mCRPC), and AFFIRM (post-docetaxel mCRPC). Ethics: Conducted per GCP and Declaration of Helsinki with informed consent. PRO instruments: FACT-P, BPI-SF, and EQ-5D were used in all studies; BFI-SF in PREVAIL and AFFIRM; EORTC QLQ-PR25 in ARCHES and PROSPER. Fatigue measure: FACT-P item GP1 ("I have a lack of energy"), scored 0–4, administered at baseline, week 13 (or week 17 in PROSPER), and every 12 weeks until progression. Analysis population and data handling: Intention-to-treat with analyses based on observed values (no imputation). Completion rates were calculated for expected assessments. Descriptive statistics summarized responses and change from baseline; distributions categorized as improvement by 1, 2, or ≥3 units; unchanged; or worsening by 1, 2, or ≥3 units. Minimal clinically important difference is not established for GP1 and was not estimated. Longitudinal analysis: Restricted maximum likelihood-based mixed-model repeated measures (MMRM), adjusting for baseline characteristics, evaluated change from baseline and from week 13/17. Assessments were included up to the last visit with <10% missing data per arm (for MMRM; a threshold of 40% missing was also referenced for sensitivity).

- Baseline fatigue prevalence (GP1 ≥1): 58% (PROSPER, nmCRPC) to 85–86% (AFFIRM, post-docetaxel mCRPC); higher in later-stage and previously treated populations. - Baseline severity: Median GP1 score was 1.0 (mild) in ARCHES, PROSPER, PREVAIL; AFFIRM mean was ~2.0. Severe fatigue (GP1 ≥3) at baseline: 29–31% in AFFIRM vs 8–13% in other trials. - Early trajectory: Both enzalutamide and placebo arms showed a small increase in fatigue by week 13 (week 17 in PROSPER), with stabilization thereafter; initial increases were <1 GP1 unit. - Distribution of change: >50% of patients had stable or improved fatigue (≥1-unit improvement) across all trials and arms over time. - Worsening at end of study vs baseline: Higher with enzalutamide than placebo in ARCHES (+9% for ≥1-unit worsening; +12% for ≥2 units), PROSPER (+1%, +5%), and PREVAIL (+1%, +3%); lower with enzalutamide in AFFIRM (−6%, −8%). Differences for ≥3-unit worsening were small across trials. - MMRM adjusted change at week 13/17 (enzalutamide minus placebo): ARCHES +0.26 (95% CI −0.14 to 0.38), PROSPER +0.30 (95% CI 0.18 to 0.41), PREVAIL −0.07 (95% CI −0.17 to 0.03), AFFIRM −0.62 (95% CI −0.41 to −0.11).

The analysis shows that patient-reported fatigue is common throughout the prostate cancer continuum and is most prevalent and severe in post-chemotherapy mCRPC. Initiation of therapy (enzalutamide or placebo) is associated with a modest early increase in fatigue that stabilizes thereafter, suggesting an initial adjustment period. Relative differences between treatment arms varied by disease state: enzalutamide was associated with slightly more worsening than placebo in earlier or less symptomatic settings (ARCHES, PROSPER, PREVAIL), whereas in more symptomatic, post-docetaxel mCRPC (AFFIRM), enzalutamide was associated with less worsening, consistent with effective disease control potentially mitigating cancer-related fatigue. The correspondence between GP1 scores and fatigue adverse events underscores the value of systematic patient-reported measures, which may capture symptom burden more sensitively than clinician-reported AEs. Given the early transient increase, proactive fatigue management early in treatment (e.g., counseling, exercise, psychoeducation, selective steroid use in metastatic settings) may improve patient experience and adherence.

Across four pivotal trials, patient-reported fatigue increased modestly in the first 13–17 weeks after starting therapy and then stabilized, with patterns varying by disease stage. Enzalutamide was associated with slightly greater fatigue than placebo in less symptomatic settings, but less worsening in heavily pretreated, symptomatic mCRPC. Most patients reported stable or improved fatigue over time. Early, proactive fatigue management is warranted to help patients cope during treatment initiation. Future research should employ multidimensional fatigue instruments, establish clinically meaningful change thresholds for single-item measures like GP1, better align patient- and clinician-reported outcomes, and assess fatigue trajectories within contemporary treatment sequences.

- Trials do not follow the same patients across the full disease continuum; comparisons are across studies with differing populations. - Post hoc, exploratory nature focused on enzalutamide vs placebo; faster progression in placebo arms may bias comparisons. - Earlier trial treatment histories (AFFIRM, PREVAIL) may not reflect current intensified treatment paradigms. - Fatigue assessed with a single item (FACT-P GP1), a unidimensional proxy for a multidimensional construct; no established MCID for GP1. - Potential ceiling/floor effects for GP1; more patients had no fatigue at baseline, which may constrain observed improvements. - Observed-data analyses without imputation; patients with missing PROs (potentially due to fatigue or other AEs) were not accounted for, though completion rates were high.