Background
International, open-label, randomized phase 3 trial (ANZUP 1304) in 1125 men with metastatic hormone-sensitive prostate cancer (mHSPC). All received testosterone suppression and were randomized to enzalutamide or a first-generation non-steroidal anti-androgen (NSAA: bicalutamide, nilutamide, or flutamide); concurrent early docetaxel was permitted at clinician discretion. Overall survival benefit was previously reported at median follow-ups of 34 and 68 months. This analysis reports 8-year outcomes with a focus on long-term adverse events, treatment duration, and causes of death.
Results
Interventions and follow up: Arm A: Enzalutamide 160 mg orally daily plus testosterone suppression (n=563)
Arm B: First-generation NSAA (bicalutamide, nilutamide, or flutamide) plus testosterone suppression (n=562)
Primary endpoint: OS
mFollow up: 8.1 years (97 months)
Results: OS: median 7.9 vs 5.8 years; 8-year OS 50% vs 40%; HR 0.73 (95% CI 0.63–0.86), P=.0001
Deaths: 285/563 (51%) vs 337/562 (60%)
Clinical PFS: 8-year 43% vs 20%; HR 0.49 (95% CI 0.42–0.57), P<.001
Cause-specific mortality: of 622 deaths, 468 attributed to prostate cancer — 207 vs 261; non–prostate-cancer deaths 78 vs 76
Treatment duration: median 4.8 vs 1.9 years; 185/563 (33%) still on enzalutamide at cutoff, 163/185 (88%) at the full 160 mg dose
Arm B: First-generation NSAA (bicalutamide, nilutamide, or flutamide) plus testosterone suppression (n=562)
Primary endpoint: OS
mFollow up: 8.1 years (97 months)
Results: OS: median 7.9 vs 5.8 years; 8-year OS 50% vs 40%; HR 0.73 (95% CI 0.63–0.86), P=.0001
Deaths: 285/563 (51%) vs 337/562 (60%)
Clinical PFS: 8-year 43% vs 20%; HR 0.49 (95% CI 0.42–0.57), P<.001
Cause-specific mortality: of 622 deaths, 468 attributed to prostate cancer — 207 vs 261; non–prostate-cancer deaths 78 vs 76
Treatment duration: median 4.8 vs 1.9 years; 185/563 (33%) still on enzalutamide at cutoff, 163/185 (88%) at the full 160 mg dose
Adverse events
Grade 3–5 cardiac (per 100 person-years): 2.2 vs 2.2
Grade 3–5 nervous system (per 100 person-years): 2.3 vs 2.0
Falls (per 100 person-years): 0.70 vs 0.24
Non–prostate-cancer mortality: no excess with enzalutamide (78 vs 76 deaths)
Dose maintenance: 88% of continuing patients remained at full dose, arguing against cumulative dose-limiting toxicity
Grade 3–5 nervous system (per 100 person-years): 2.3 vs 2.0
Falls (per 100 person-years): 0.70 vs 0.24
Non–prostate-cancer mortality: no excess with enzalutamide (78 vs 76 deaths)
Dose maintenance: 88% of continuing patients remained at full dose, arguing against cumulative dose-limiting toxicity
Conclusions
At 8.1 years of follow-up, adding enzalutamide to testosterone suppression continues to deliver a substantial and durable overall survival advantage over a first-generation non-steroidal anti-androgen in mHSPC (median 7.9 vs 5.8 years; 8-year OS 50% vs 40%). The survival gain is driven by fewer prostate-cancer deaths with no increase in deaths from other causes, and exposure-adjusted cardiac and neurologic toxicity is comparable between arms. Falls are the one consistently increased long-term harm.
Key Limitations
Open-label design. The comparator — a first-generation NSAA — was a defensible standard when ENZAMET opened but is not a contemporary control; the trial therefore does not compare enzalutamide against ADT alone, against another androgen receptor pathway inhibitor, or against triplet therapy with docetaxel. Concurrent early docetaxel was given at clinician discretion rather than by randomization, so docetaxel-stratified comparisons are non-randomized and confounded by selection. Long-term safety comparison is complicated by markedly unequal exposure (median 4.8 vs 1.9 years); person-year adjustment mitigates but does not eliminate this. Cause-of-death attribution in advanced prostate cancer is subject to investigator judgment, and adverse event ascertainment attenuates with time on study.
Clinical Context
Enzalutamide is FDA and EMA approved for metastatic castration-sensitive prostate cancer (US approval 2019, supported by ARCHES and ENZAMET). NCCN, ASCO, and ESMO all recommend intensifying androgen deprivation with an androgen receptor pathway inhibitor for mHSPC; testosterone suppression alone, or with a first-generation anti-androgen, is no longer an acceptable standard. These 8-year data address the durability question that mature follow-up is uniquely able to answer — the survival curves do not converge, most long-term responders stay on full-dose therapy, and there is no signal of delayed excess non-cancer mortality. The falls signal is actionable and supports routine fall-risk assessment in older men on long-term enzalutamide, consistent with the geriatric assessment recommendations in current guidelines.