Abstract

The failure of peripheral blood stem cell (PBSC) mobilization and harvest is a critical issue for multiple myeloma (MM) patients undergoing high-dose chemotherapy. Plerixafor (PLX) is an effective mobilizing agent; however, its use for every MM patient undergoing high-dose chemotherapy has led to a notable increase in costs. We designed a highly specific and sensitive algorithm for identifying patients likely to fail PBSC mobilization after chemotherapy and G-CSF (Blood Transfusion 2013.11:94). The use of this algorithm thereby allows selective administration of PLX to patients predicted to fail mobilization after chemotherapy and G-CSF (on-demand PLX) and may reduce failure rate of PBSC mobilization while limiting cost.

We performed a multicenter phase II prospective study of on-demand PLX used according to our algorithm for patients with lymphoma or MM, who were mobilized by cyclophosphamide and G-CSF. The study was powered to demonstrate a reduction in mobilization failure from 14% to 7% for the entire population of treated patients. Here, we report the final results for MM patients. The inclusion criteria for MM patients were as follows: diagnosis of symptomatic MM, age 18-70 yr, achievement of any response after first-line treatment administered for 4-8 months, first mobilization attempt, cardiac and pulmonary function adequate for high-dose chemotherapy. Mobilization schedule was cyclophosphamide (CTX, 4 g/m2) and G-CSF (5-10 mcg/Kg), PLX (240 mcg/Kg) was administered only to patients selected by the algorithm. Estimation of costs was performed according to a previously reported study (BJH 2014, 164, 113).

There were 111 patients with MM who underwent treatment. Successful CD34+ cell mobilization (>20×109 cells/mL in PB) was achieved for 97.2% (108/111) of patients, and failure of mobilization occurred in the remaining 3 (2.8%); minimal apheretic harvest success (>2.0×106 CD34+ cells/Kg) was achieved for 97.2% (108/111); and optimal harvest success (=/>4.0×106 CD34+ cells/Kg) was achieved for 84.6% (94/111). On-demand PLX was needed for 8.2% of patients (9/111). After autologous hematopoietic transplantation, neutrophil (N) engraftment (N>0.5x109 cells/L) was reached at day +11.8 (range day +8 to +24). We compared these prospective results with the mobilization results obtained retrospectively in a control group of 183 MM patients who received the same mobilization schedule without PLX.

After the two groups were adjusted for unbalanced factors, multivariable logistic regression analysis revealed that on-demand PLX treatment according to the algorithm led to significant increases in the probabilities of achieving a successful minimal apheretic harvest (p=0.006; hazard ratio [HR] 5.624, 95% confidence interval [CI] 1.168-19.548) and optimal harvest (p=0.02; HR 2.121, 95% CI 1.118-4.025). The mean cost increase for the first mobilization in the PLX-on-demand prospective study, in respect to control group, was 615 /patient. The incremental cost-effectiveness ratio (ICER) was calculated as: (cost1-cost2)/(result1-result2). ICER was 47 /1% increase in probability of a minimal apheretic harvest while it was 68 /1% increase in probability of an optimal apheretic harvest.

In conclusion, the final analysis of our study found that on-demand PLX for MM patients, which was added to the mobilization schedule of CTX (4 g/m2) + G-CSF (5-10 mcg/Kg), allowed a successful harvest from the first mobilization treatment in > 97% of patients, with 85% of patients achieving a harvest sufficient for two rounds of high-dose chemotherapy. These results indicate that on-demand PLX added to mobilization chemotherapy is a significant improvement over the same type of mobilization chemotherapy without PLX. The limited use of PLX in this study allowed for a favorable incremental cost-effectiveness ratio of this expensive agent. On-demand PLX used according to a validated algorithm in addition to CTX plus G-CSF may be considered a new standard for PBSC mobilization and harvest in patients with MM.

Table 1.
 Failure of CD34+ Mobilization in PB  Failure of Minimal Harvest  Failure of Optimal Harvest  Cost per Patient ICER
(Minimal Harvest) 
ICER
(Optimal Harvest) 
PLX on Demand
(n 111) 
2.8% 2.8% 15.4% 3,969  47 / 1% increase in probability of a Minimal Harvest 68 / 1% increase in probability of an Optimal Harvest 
Control Cohort
(n 183) 
7.6% 15.8% 24.4% 3,354  
P
(adjusted for comparisons) 
NS 0.006 0.02  
 Failure of CD34+ Mobilization in PB  Failure of Minimal Harvest  Failure of Optimal Harvest  Cost per Patient ICER
(Minimal Harvest) 
ICER
(Optimal Harvest) 
PLX on Demand
(n 111) 
2.8% 2.8% 15.4% 3,969  47 / 1% increase in probability of a Minimal Harvest 68 / 1% increase in probability of an Optimal Harvest 
Control Cohort
(n 183) 
7.6% 15.8% 24.4% 3,354  
P
(adjusted for comparisons) 
NS 0.006 0.02  

Disclosures

Milone:Sanofi: Consultancy. Martino:Sanofi: Consultancy. Olivieri:Sanofi: Consultancy.

Author notes

*

Asterisk with author names denotes non-ASH members.