A new analysis of the phase 3 PEACE-3 trial evaluated alkaline phosphatase (ALP) and prostate-specific antigen (PSA) responses in patients with metastatic castration-resistant prostate cancer (mCRPC) and bone metastases treated with enzalutamide alone or in combination with radium-223.
The analysis, published in European Urology Oncology on August 28, 2026, was titled “Alkaline Phosphatase and Prostate-specific Antigen Response in Metastatic Castration-resistant Prostate Cancer Treated with Enzalutamide Alone or in Combination with Radium-223: Ad Hoc Analysis of the PEACE-3 Consortium Trial.”
Authors: Andrey Soares, Silke Gillessen, Ananya Choudhury, Fred Saad, Enrique Gallardo, Yohann Loriot, Raymond S. McDermott, Alejo Rodriguez-Vida, Pedro Isaacsson Velho, Franco Nolè, Felipe Melo Cruz, Thierry André Roumeguère, Gedske Daugaard, Rosely Yamamura, Coralie Poncet, Corneel Coens, Béatrice Fournier, and Bertrand F. Tombal.
The PEACE-3 Trial
EORTC 1333/PEACE-3 was an international, randomized, open-label phase 3 trial that enrolled 446 patients with progressive mCRPC, at least two bone metastases, and no visceral metastases. Patients were asymptomatic or mildly symptomatic and were randomly assigned 1:1 to enzalutamide 160 mg daily alone or enzalutamide combined with six intravenous injections of radium-223 given monthly at 55 kBq/kg.
Following safety findings from the ERA-223 trial, the protocol was amended in April 2018 to require a bone-protecting agent, either zoledronic acid or denosumab, in both treatment groups.
Previous PEACE-3 analyses showed that adding radium-223 to enzalutamide prolonged radiographic progression-free survival from 16.4 to 19.4 months (HR 0.69; 95% CI, 0.54–0.87; p<0.001). In the final overall survival analysis, median OS was 38.2 months with the combination and 32.6 months with enzalutamide alone (HR 0.76; 95% CI, 0.60–0.96; p=0.01). The current exploratory post hoc analysis focused specifically on ALP and PSA responses.
Evaluating ALP and PSA Responses
Of the 446 randomized patients, 441 were evaluable for ALP, including 217 receiving enzalutamide plus radium-223 and 224 receiving enzalutamide alone. A total of 436 patients were evaluable for PSA, including 215 and 221 patients, respectively. ALP response was defined as a reduction of at least 30% from baseline, referred to as ALP-30. Among patients with elevated baseline ALP, normalization was defined as an ALP level below 115 U/L.
PSA responses were defined as reductions of at least 50% (PSA-50) or at least 90% (PSA-90) from baseline. Confirmed responses required consecutive measurements at least 21 days apart. The 6- and 12-month response rates were calculated only among patients with a corresponding biomarker measurement available at each landmark.
ALP Responses
At 6 months, confirmed ALP-30 responses were observed in 56.5% of patients receiving enzalutamide plus radium-223 and 50.8% receiving enzalutamide alone. At 12 months, the corresponding rates were 50.0% and 48.1%. Among patients with elevated ALP at baseline, confirmed ALP normalization occurred in 76.5% of patients receiving the combination versus 48.9% receiving enzalutamide alone at 6 months. At 12 months, the rates were 77.8% and 61.3%, respectively.
Median time to confirmed ALP-30 response was 2.4 months with enzalutamide plus radium-223 versus 3.7 months with enzalutamide alone (HR 1.41; 95% CI, 1.12–1.78; p=0.003). Median time to confirmed ALP normalization was also shorter with the combination, at 2.0 versus 4.5 months (HR 2.05; 95% CI, 1.46–2.88; p<0.001).
Deep PSA Responses
Differences were also observed in deep PSA responses. At 6 months, a confirmed PSA-90 response was achieved in 50.5% of patients receiving enzalutamide plus radium-223 compared with 34.1% receiving enzalutamide alone. At 12 months, confirmed PSA-90 response rates were 54.9% and 37.6%, respectively. The differences were statistically significant at both time points.
Median time to confirmed PSA-90 response was 5.6 months with the combination compared with 22.1 months with enzalutamide alone (HR 1.48; 95% CI, 1.13–1.93; p=0.004). Confirmed PSA-50 responses at 6 months occurred in 77.1% versus 69.8% of patients, and at 12 months in 76.8% versus 66.2%, respectively. However, median time to confirmed PSA-50 response was 2.8 months in both treatment groups, with no difference between treatments (HR 1.00; 95% CI, 0.80–1.24; p=0.967).
Bone-Protecting Agents and ALP
The investigators also explored ALP responses according to bone-protecting agent use at study entry. Median time to confirmed ALP-30 response was 2.3 months among patients who had started a bone-protecting agent less than 4 weeks before protocol treatment, compared with 6.5 months among those who had not previously received one. Median time to confirmed ALP normalization was 2.6 versus 7.6 months, respectively.
The authors noted that bisphosphonates or denosumab can suppress bone turnover and reduce total ALP. ALP suppression is greatest during the first months of treatment, which may contribute to differences in ALP dynamics according to bone-protecting agent use.
Limitations
The investigators emphasized that this was an exploratory post hoc analysis. The ALP and PSA analyses were not supported by prior power calculations, and p values were not adjusted for multiplicity.
Importantly, the analysis did not assess whether the observed biomarker responses were associated with clinical outcomes such as progression-free survival or overall survival. The authors therefore stated that these biomarker changes should be interpreted descriptively and should not be overinterpreted in terms of clinical benefit.
Takeaway
In this exploratory analysis of PEACE-3, adding radium-223 to enzalutamide was associated with faster ALP-30 responses and ALP normalization, higher PSA-50 and PSA-90 response rates, and a shorter time to confirmed PSA-90 response compared with enzalutamide alone.
The authors concluded that these findings support additional antitumor activity of radium-223 when combined with enzalutamide. However, the analysis did not evaluate the relationship between these biomarker changes and clinical outcomes, and further analyses are needed to determine how these biomarker changes relate to patient outcomes.
