Alkaline Phosphatase And Prostate-Specific Antigen Dynamics And Survival Outcomes In Patients With Metastatic Prostate Cancer Receiving Radium-223

Abstract

Radium-223 (Ra-223) is a targeted alpha-emitter for metastatic castrate-resistant prostate cancer (mCRPC) patients that demonstrated overall survival (OS) benefit compared to placebo in the ALSYMPCA trial. A secondary analysis of the ALSYMPCA trial demonstrated that decreased alkaline phosphatase (ALP) values at 12 weeks after treatment initiation were associated with improved OS. We further investigate prostate-specific antigen (PSA) and ALP dynamics in a large real-world cohort of patients with mCRPC receiving Radium-223. The Flatiron Health database is a nationwide longitudinal de-identified database derived from electronic health record (EHR) data from 280+ cancer clinics (∼800 sites of care) representing 2.2 million patients in the United States. We included patients with mCRPC treated with Ra-223 between 2013 and 2017 with complete ALP and PSA data. We specified a priori the following factors in a Cox proportional hazards regression for OS: age, performance status, year of treatment, baseline PSA and ALP, occurrence of a PSA or ALP decrease at any point during treatment or within 12 weeks of last administration, timing of response (before cycle 3 versus after cycle 3). Date of first Radium-223 administration was considered the index date for time-dependent analyses. We analyzed data from 586 patients with mCRPC receiving Ra-223 at a median follow-up of 9.8 months, with 252 patients (29.7%) having received prior docetaxel or cabazitaxel. In total, 447 patients (76.3%) achieved any ALP decrease and 250 patients (42.3%) achieved an ALP decrease ≥30%. Of those who achieved an ALP decrease, the response occurred before cycle 3 in 301 patients (67.3%). The median percent change from baseline to ALP nadir was 27.9% (interquartile range [IQR] 10.3% to 46.8%). With respect to PSA, 192 patients (32.7%) achieved any PSA decrease and 124 patients (21.1%) achieved a PSA decrease ≥30%. The median percent change from baseline to PSA nadir was 43.9% (IQR 19.9% to 80.4%). In our survival analysis, we identified age, performance status, ALP value at baseline, PSA value at baseline, presence of an ALP decrease (hazard ratio [HR] 0.61; 95% CI 0.42-0.90; p = 0.012), and presence of a PSA decrease (HR 0.47; 95% CI 0.33-0.65; p = 0.001) as significant predictors of OS. Notably, timing of ALP decrease before or after cycle 3 (HR 0.76; 95% CI 0.57-1.02; p = .066) was not significantly associated with survival. We analyzed PSA and ALP dynamics in the largest reported series of patients receiving Ra-223 based on real-world data. We identified baseline ALP and PSA values, presence of an ALP decrease, and presence of a PSA decrease as predictors of OS. Approximately one-third of patients will have a delayed ALP response after cycle 3; however, timing of ALP response was not significantly associated with OS.