Patterns of treatment (tx) with bone-modifying agents (BMA) in patients with bone metastases (mets).
e20690 Background: Bone mets can lead to serious and debilitating skeletal-related events (SREs: fracture, spinal cord compression, and surgery or radiation to bone). Three BMAs are approved in the US for prevention of SREs in cancer patients (pts) with bone mets: 2 intravenous bisphosphonates (IV BP) with recommended dosing every 3-4 weeks (pamidronate [PAM] and zoledronic acid [ZA]), and 1 RANK ligand inhibitor with subcutaneous injection every 4 weeks (denosumab). Monitoring real-world tx trends is important, particularly as low persistence has been linked to higher rates of SREs (Hatoum et al 2008). Methods: A retrospective cohort was defined from the Oncology Services Comprehensive Electronic Records (OSCER) database, which includes medical records from >500K pts treated at 76 US oncology practices. Adult solid tumor pts with bone mets who initiated tx with a BMA between Jan 2011 and Oct 2011 were included and followed for 12 mos after tx initiation. Pts were either naïve (no BMA in previous 6 mos) or transition (different BMA received in previous 6 mos). Results: The cohort was composed of 1445 denosumab pts (62% naïve) and 1807 IV BP pts (99% naïve). Pt characteristics were similar, although denosumab pts were somewhat more likely to have prostate cancer than IV BP pts (30% vs 25%). Denosumab pts were more likely to receive regular tx during follow-up and less likely to switch tx than IV BP pts (Table). Trends were consistent across tumor types and for naïve vs transition pts. Conclusions: This descriptive analysis reports on tx patterns for pts with bone mets since the availability of denosumab in Dec 2010. Denosumab represented almost half (44%) of pts initiating a BMA and one-third of pts naïve to tx. Uninterrupted tx was more common for denosumab pts than IV BP pts, regardless of tumor type. Trends, including reasons for and consequences of tx interruption, should be monitored as the pt experience develops. [Table: see text]