Predicting Poor Peripheral Blood Stem Cell Mobilization in Multiple Myeloma (MM) Using Initial Peripheral CD34 Counts: Developing Target-Based Cut-Points for Early Intervention.

Peripheral blood stem cells (PBSC) are the preferred graft type for autologous stem cell transplantation (ASCT) and mobilization is typically done using growth factor (GF) alone or a combination of chemotherapy (CTX) and GF. The success of mobilization is influenced by several factors such as prior therapy and mobilization strategy. Patients (Pts) failing to mobilize successfully often have GF dose increased and if unsuccessful, mobilization reattempted with chemotherapy + GF or G-CSF (Granulocyte Colony-Stimulating Factor) + GM-CSF (Granulocyte-Macrophage Colony-Stimulating Factor). More recently, plerixafor, a CXCR4 inhibitor, has been found to enhance G-CSF based collection allowing potential early intervention in poor mobilizers. However, this approach requires valid cut-points for peripheral CD34 counts (pCD34) for cost efficient use.

We studied 910 first time, G-CSF based PBSC mobilizers with MM (565; 62%), or light chain amyloidosis (345; 38%). Salvage attempts were excluded, as were first attempts using CTX+GCSF, plerixafor or GF combinations. Sensitivity-specificity analysis (ROC) was used to define ideal cut-points for predefined outcomes. We used a CD34 yield of 2 million/kg as the minimum required collection and 4 million/kg as the ideal collection for this analysis. Our current practice is to initiate collection at a pCD34 of 10/uL.

We first determined the best day 1 pCD34 cutoff for being able to collect at least 2 million/kg PBSC to be 9.2/uL. Among those with ≥9.2/uL, only 1.6% failed to collect at least 2 million compared to 27% among the rest (P < 0.001). These numbers were 1.7% and 25% respectively using the traditional cutoff of 10/uL. We then examined how the pCD34 on day 2 predicted collection, among those failing to reach the current threshold of 10/uL. Among this group, a pCD34 on day 2 of 8/uL was the best cutoff to predict a collection of 2 million/kg. Among those with >8/uL, only 9% failed to collect at least 2 million compared to 48% among the rest (P < 0.001). Among those with day 1 pCD34 < 10, 65% achieved a count of 8/uL on day 2. We then examined what pCD34 on day 1 can best predict a pCD34 ≥ 8 on day 2 and found this to be 5/uL. Among those with day 1 pCD34 ≥ 5/uL, 92% reached day 2 pCD34 ≥ 8 compared to 36% among the rest (P < 0.001).We then determined cutoffs based on ideal targets using only those patients achieving pCD34 count > 10/uL by day 2, since these patients would have started collection without any change in the mobilization technique. We used a maximum of 5 collections to achieve the goal as being ideal, taking into account the most commonly used practice of collecting during weekdays. We examined cutoffs both for a target of 4, 8 and 12 million/kg for one, two or three transplants respectively. Among those with a day 1 or day 2 pCD34 > 10/uL 98% collected 4 million/kg in 5 collections. The best day 1 or day 2 pCD34 predicting a 8 million/kg yield in 5 collections was 21.4/uL and a 12 million/kg yield was 27.9/uL. Among those with day 1or 2 pCD34 ≥ 21.4/uL, 90% collected 8 million compared to 54% among those with pCD34 < 21.4/uL. Similarly, among those with day 1 or 2 pCD34 ≥ 27.9/uL, 79% reached 12 million in ≤ 5 days compared to 17% among those with pCD34 < 27.9/uL (P < 0.001).Finally we examined if poor collection on first day, despite a pCD34 > 10 predicts for poor overall yield. A ratio of day 1 collection (million/kg) to pCD34 day 1 or 2 < 5×10⁴ was highly predictive of failure to get eventual yield of 2 million/kg, with 38% below the cutoff failing to reach the goal compared to 2% for those above the cutoff.

This study provides an estimate of optimal pCD34 count thresholds for early identification of patients at high risk of collection failure. It is reasonable to use the first or second day pCD34 count, since a substantial proportion of patients will achieve the threshold value by day 2 and have similar outcome as those reaching the threshold by day 1. Based on the results here, interventions such as plerixafor can be considered if pCD34 count is < 5/uL on day 1 or < 9/uL by day 2, to achieve the minimum collection of 2 million/kg. However, in order to achieve the ideal targets, a higher threshold should be considered and the recommendation would be 17/uL, and 28/uL for target of 8 or 12 million respectively. Finally, identification of poor mobilizers after start of collection potentially provides another time point for early intervention, and this needs to be prospectively studied.

Micallef:Genzyme: Research Funding. Gertz:Genzyme: Research Funding. Kumar:Celgene,Genzyme,Millenium,Novartis,Bayer,: Honoraria, Research Funding.