Fracture Rate and Overall Survival in mCRPC Patients Treated with Radium-223 and Concomitant Abiraterone

Abstract

Both radium-223 (Ra) and abiraterone/prednisone (Abi/Pred) prolong overall survival (OS) in metastatic castrate resistant prostate cancer (mCRPC). Recent prospective randomized data (ERA-223 trial) combining Abi/Pred and Ra (starting simultaneously) in mCRPC patients (pts) indicate that this combination increases fractures and deaths compared to Abi/Pred alone. In ERA-223, risks were partially mitigated by bone health agents (BHA) (i.e denosumab). Herein, we analyzed our institutional pts to assess fractures/OS in pts treated with Ra +/- Abi/steroid (Abi/S). After institutional review board approval, all pts in our institutional database were assessed for fractures/OS provided that they received Ra therapy between 2010 and 2017 and had one or more CT, MRI, or bone scan performed post-Ra. Categorical data were analyzed by chi-square. OS was evaluated using KM and log-rank calculations. Of 82 identified pts, 12 (13%) had radiographically documented fractures. Fracture data is summarized in Table 1. Forty (49%) of the Ra pts were treated with concomitant Abi/S, and 7 (17.5%) had fractures as compared to 5 (12%) pts of the 42 (51%) Ra pts treated with no Abi/S. Of the 82 pts, 34 (41%) also received BHA and 5 (15%) had fractures. There were 48 (59%) pts not receiving BHA and 7 (15%) had fractures. In the Ra group without Abi/S, 10 (12%) pts received BHA and 1 (10%) of these patients had a fracture. Within the Ra + Abi/S group, 10 (12%) pts were started on Ra and Abi/S simultaneously and 3 (30%) of these pts had fractures. Non-simultaneous Ra and Abi/S fracture rates were 4/30 (13%). None of the fracture data were statistically distinct between groups. Overall, pts on Ra and Abi/S had better OS (median 727 days) compared to pts on Ra without Abi/S (353 days, p = 0.0346). Fractures rates are higher in pts with imaging as compared to ALSYMPCA. The highest fracture rates we observed were with pts simultaneously starting Ra and Abi/S. Survival analyses are likely biased when evaluating Abi/S as a univariate factor. Larger data sets are required to provide sample sizes adequate for appropriately powered statistical analyses.Abstract SU_33_2328; Table 1Fracture rate in patients on Ra +/- Abi/S +/-BHACategoryNNumber of patients with fracturesPatients on BHA? (X=All Inclusive)Total patients8212 (15%)XPatients on Ra and Abi/S regardless of BHA status407 (18%)XPatients on Ra, no Abi/S regardless of BHA status425 (12%)XPatients on Ra and Abi/S and BHA244 (17%)YesPatients on Ra and Abi/S and no BHA163 (19%)NoPatients on Ra and Abi/S started simultaneously103 (30%)XPatients on Ra and Abi/S started previously304 (13%)XPatients on Ra, no Abi/S and BHA101 (11%)YesPatients on Ra, no Abi/S and no BHA324 (13%)No Open table in a new tab