Abstract
Background: Cyclophosphamide plus G-CSF (C+G-CSF) is the most widely used stem cell (SC) mobilization regimen in multiple myeloma (MM) patients. Plerixafor plus G-CSF (P+G-CSF) has demonstrated superior SC mobilization efficacy when compared to G-CSF alone in phase II and III studies and has been shown to rescue patients who fail mobilization with G-CSF with or without cyclophosphamide. Despite the proven efficacy of P+G-CSF in upfront SC mobilization, its use for this indication has been limited, mostly due to concerns of high cost of the drug. Investigators have proposed "on demand" use of plerixafor in patients identified to have inadequate SC mobilization with G-CSF with or without cyclophosphamide, with the assumption that such an approach promotes cost containment by limiting plerixafor use. However, a comprehensive comparison of the cost effectiveness of SC mobilization using C+G-CSF versus P+G-CSF has not been performed. The goal of this retrospective study was to conduct a cost analysis between these two approaches.
Methods: Using the pharmacy database, we identified all MM patients treated at Memorial Sloan Kettering Cancer Center between 11/2008 and 6/2012 who received C+G-CSF or P+G-CSF for upfront SC mobilization. Patients collecting <5 x 106 CD34+ cells/kg were considered mobilization failures and had a second attempt at SC mobilization using an alternative approach. For salvage mobilization, patients received P+G-CSF after failing C+G-CSF-based mobilization or were re-mobilized with C+G-CSF along with plerixafor after failing upfront P+G-CSF mobilization. Mobilization costs included in the analysis were those associated with upfront mobilization, those associated with salvage mobilization in patients failing an initial mobilization, and those associated with complications directly related to the mobilization procedures. Cost calculations included the following: cost of cyclophosphamide 3000 mg/m2, plerixafor 0.24 mg/kg, and G-CSF 10 mcg/kg and their administration prior to and during pheresis sessions; pheresis sessions; laboratory tests on pheresis days; re-hospitalization occurring within 15 days of either mobilization approach and considered directly related to the mobilization procedure. All costs were calculated using the institution’s ratio of cost to charges, and were normalized and adjusted based on institutional charges and costs for 2012.
Results: A total of 223 patients undergoing upfront mobilization were identified, with 111 patients receiving C+G-CSF, and 112 patients receiving P+G-CSF. Thirteen patients (12%) were re-hospitalized due to C+G-CSF-related complications, with an average hospital stay of 6.5 days. No patients in the P+G-CSF arm were hospitalized. Nineteen patients (17%) in the C+G-CSF group failed first mobilization and received P+G-CSF as salvage regimen, with four (3.6%) failing salvage collection and ultimately deemed collection failures. Seven patients (6.2%) in the P+G-CSF group failed upfront mobilization and received C+G-CSF along with plerixafor as salvage regimen, with two (1.8%) subsequently failing salvage mobilization. The average number of pheresis sessions performed was 3.29 and 2.42 in the C+G-CSF and P+G-CSF upfront groups, respectively (p=0.373). In total, the average cost of stem cell collection per patient was 1.3 times greater in the C+G-CSF group than in the P+GCSF upfront group (p=0.017). When the costs associated with salvage pheresis are discounted for the 19 patients in the C+G-CSF upfront group who failed first SC mobilization, assuming that these patients could have been salvaged by plerixafor-on-demand, the cost per patient in the C+G-CSF group remains 1.26 times greater (p=0.019) than that of the P+G-CSF group.
Conclusion: The use of P+G-CSF upfront for SC mobilization is more cost effective than the more widely used approach employing C+G-CSF. This difference is likely due to several factors including: 1) higher rate of hospitalization in the C+G-CSF group due to expected complications such as febrile neutropenia and catheter-related infections; 2) higher rate of mobilization failure leading to increased need for salvage mobilization in the C+G-CSF group; 3) reduced G-CSF use in the upfront P+G-CSF group. Overall, this single institution study provides additional rationale for the standard use of P+G-CSF as upfront mobilization regimen in MM patients.
Disclosures
No relevant conflicts of interest to declare.
Discrete-variable quantum key distribution with homodyne detection
