Cost Analysis of Mobilization and Autologous Transplantation in Patients Who Received AMD 3100 after Failing Standard Mobilization.

Abstract

AMD3100 administered with G-CSF has been demonstrated to mobilize hematopoietic stem cells in patients (pts) who do not collect sufficient cells for autologous transplant following other mobilization regimens. This retrospective study evaluated the difference in clinical outcomes and costs of mobilization and transplant between pts who achieved sufficient CD34+ cells to proceed to transplant using a standard mobilization regimen (control group) versus pts who had failed previous mobilization regimens and were remobilized with AMD3100 and G-CSF in the compassionate use protocol (CUP). Between 2005 and 2008, a total of 88 pts in the control group and 36 pts in CUP were evaluated. In the latter group, mobilization data were available for 35 pts, post-transplant care information was obtainable for 28 pts, and 23 pts had cost data. Prior mobilization attempts in this group included cyclophosphamide plus G-CSF (67%) or G-CSF alone (33%). Following AMD3100 administration, 33 (94%) pts achieved successful mobilization of > 1.9 million/kg CD34+ cells. A median of 3.36 x106 CD34+ cells/kg (range, 0.51–13.05) were collected in a median of 3 (range, 1–4) apheresis days. In comparison, control pts collected a median of 10.8 x106 CD34+ cells/kg (range, 1.95–69.32) in a median of 2 (range, 1–6) apheresis days. Twenty-eight CUP pts went on to transplant and received a median of 3.71 x106 CD34+ cells/kg (range, 1.91–13.05) compared to a median of 5.5 x106 CD34+ cells/kg (range, 1.95–18.76) infused into the control group. The CUP pts had a median day to ANC > 500/μl of 11 days (range, 9–17) and platelets > 20,000 /μl of 27 days (range, 13–37). At day 100, 92% of the CUP pts reached engraftment and there were no transplant related mortalities. The control population had a median day to ANC > 500/ ul of 11 days (range, 8–14) and platelets > 20,000/μl of 18 days (range, 9–83). The control group experienced 100% engraftment and one transplant related fatality. Relapse occurred in 14% of the CUP pts and 32 % of the control pts. With a median follow up of 385 days (range, 55–992) in the CUP pts and 560 days (range, 156–867) in the control pts, 86% of CUP pts and 91% of control pts are alive (p=NS). The median cost of remobilization with AMD3100 was $2,959 less in the CUP pts than the median cost of initial mobilization for the control group, but this was not significantly different between the groups. As expected, the median total cost of mobilization in the CUP pts, including initial mobilization costs, was $16,927 more than in the control pts (p<0.0001). Total cost of mobilization in the CUP pts does not include the cost of AMD3100 as drug was administered on study. In both groups of pts, greater than 95% of total transplant costs occurred in the first 30 days of the transplant course. The difference in the total median transplant costs at day 30 in the CUP pts was $7,373 more than in the control pts. (p=0.0024). The primary difference in cost between the groups was reflected in total inpatient costs; CUP pts incurred 91% of total costs and control pts 57% of total costs in the first 30 days. In conclusion, this single center review shows remobilization with AMD3100 + G-CSF allowed a high percentage of patients who failed standard mobilization to subsequently mobilize adequate CD34+ cells and proceed to autologous transplant with acceptable clinical outcomes. Pts who required remobilization also had an increased cost of transplantation because of increased inpatient costs. Additional pharmacoeconomic studies that compare the costs of utilizing AMD3100 in the first-line setting and comparing it to costs of cytokines with or without chemotherapy are warranted.