How Radium‑223 Plus Enzalutamide Accelerates ALP and PSA Responses in Bone‑Metastatic mCRPC

An exploratory post hoc analysis of the international PEACE‑3 trialadds a compelling biomarker layer to the already positive story of combining radium‑223 with enzalutamide in men with bone‑metastatic, castration‑resistant prostate cancer. The parent study, a phase 3 randomised trial in asymptomatic or mildly symptomatic metastatic castration‑resistant prostate cancer (mCRPC) with bone metastases, had already shown that adding six monthly cycles of radium‑223 (55 kBq/kg) to enzalutamide 160 mg daily improved both radiographic progression‑free survival and overall survival compared with enzalutamide alone. The new ad hoc analysis focuses on two widely used blood markers, alkaline phosphatase (ALP), reflecting bone turnover, and prostate‑specific antigen (PSA), reflecting tumour burden, and asks whether the combination not only prolongs life but also drives faster and deeper biochemical responses.

Across the biomarker cohort, 441 patients were evaluable for ALP and 436 for PSA. ALP response was defined as at least a 30% decline from baseline (ALP‑30), while PSA responses were defined as at least 50% (PSA‑50) or 90% (PSA‑90) declines. To count as a confirmed response, patients needed two consecutive qualifying measurements at least 21 days apart. Time‑to‑event endpoints were analysed with standard Kaplan–Meier methods and Cox proportional‑hazards models.
At six months, confirmed ALP‑30 responses occurred in 56.5% of patients on the combination versus 50.8% on enzalutamide alone. More strikingly, the median time to reach that ALP‑30 response was 2.4 months with radium‑223 plus enzalutamide compared with 3.7 months with enzalutamide monotherapy. For ALP normalisation, the advantage was even clearer: median time to normal ALP was 2.0 months with the combination versus 4.5 months with enzalutamide alone.

Parallel patterns emerged for PSA. Confirmed PSA‑90 responses at six months were achieved by 50.5% of patients receiving the combination versus 34.1% on enzalutamide alone. The median time to confirmed PSA‑90 response was 5.6 months with radium‑223 plus enzalutamide compared with 22.1 months with enzalutamide alone (HR 1.48, 95% CI 1.13–1.93; p = 0.004). In other words, not only did more patients on the combination reach deep PSA declines, they did so substantially earlier. These findings align with the biological rationale for the regimen: radium‑223 is an alpha‑emitting radiopharmaceutical that mimics calcium and concentrates in areas of active bone remodelling, delivering highly localised radiation to bone metastases and their microenvironment, while enzalutamide blocks androgen‑receptor signalling throughout the body. Together, they attack the disease on two complementary fronts, which is consistent with the observation of accelerated and more profound drops in both bone‑specific and tumour‑derived markers.

The clinical context for these biomarker results matters. In metastatic prostate cancer, early and deep PSA declines have repeatedly been associated with better outcomes. A 2026 meta‑analysis in metastatic hormone‑sensitive disease showed that achieving a PSA nadir at or below 0.2 ng/mL was linked to markedly longer progression‑free and overall survival, and real‑world data in metastatic castration‑sensitive disease found that reaching PSA <0.2 ng/mL within nine months of starting androgen deprivation therapy cut the risk of death by about half. Although the PEACE‑3 ad hoc analysis did not directly tie ALP or PSA responses to survival, the magnitude and speed of the biomarker shifts observed here fit a pattern in which rapid, deep responses often presage improved disease control. ALP, in particular, carries specific relevance in bone‑predominant mCRPC: elevated ALP reflects high bone turnover driven by metastatic activity, and prior work has shown that ALP levels and their evolution under therapy carry prognostic information for overall survival.

Interpreting these results requires acknowledging their limitations. This was an exploratory post hoc analysis, not a pre‑specified primary or secondary endpoint of PEACE‑3. No multiplicity adjustment was applied, and the investigators did not formally model the relationship between biomarker response and clinical outcomes such as radiographic progression‑free survival or overall survival within this dataset.

Source.