Evidence for continued dose escalation of plerixafor for hematopoietic progenitor cell collections in sickle cell disease

Even with the advent of plerixafor mobilization as a common strategy for autologous hematopoietic progenitor cell (HPC) mobilization for ex vivo gene therapy of sickle cell disease (SCD), consistent adequate CD34+ mobilization to achieve the optimal target cell dose for HPC collection of at least 10 x 10⁶ CD34+ cells/kg (1, 2) remains problematic, especially in patients with a low baseline peripheral blood (PB) CD34+ count (1, 3). Changing the route of plerixafor administration from subcutaneous to intravenous does not appear to increase CD34+ mobilization (4). One possibility to address suboptimal CD34+ mobilization is to increase the dose of plerixafor. In normal volunteer donors in a crossover trial of plerixafor mobilization alone, escalating the dose from the standard 240 ug/dose to 480 ug/kg increased the pre-apheresis peripheral blood (PB) CD34+ count, as well as the CD34+ area under the curve over 24 hours (5). There is equipoise regarding the nature of the benefit of dose escalation above 160 ug/kg plerixafor, as our study found a significant association only between dose and the fold increase in PB CD34+ count (3), whereas the one other study with dose escalation found a significant association only between dose and the absolute increase in PB CD34+ count (6). This discrepancy is likely related to the strong correlation between baseline and pre-apheresis PB CD34+ or CD34+ yield (3, 7), which is stronger than any correlation (1) other than the well-known correlation between pre-apheresis CD34+ count and CD34+ yield (8).

Our purpose in this letter is to describe the efficacy and safety results from our IRB-approved study of plerixafor mobilization in SCD patients (NCT02193191) in two patients who underwent plerixafor mobilization at 320 μg/kg. Because of the strong correlation between baseline and pre-apheresis PB CD34+ count or CD34+ yield, we controlled for inter-individual variation by comparing dose response in patients with similar baseline PB CD34+ counts. In our study, there were 2 subsets of unique patients with the same PB CD34+ count who were given different dose levels of plerixafor: 4 patients with a baseline of 1/uL and 3 patients with a baseline of 2/uL (Figure 1A and B). In the patients with a baseline of 1/uL, the one mobilized at 320 μg/kg had the highest CD34+ response, both in regard to the absolute level and fold increase of the PB CD34+ count. In the patients with a baseline of 2/uL, the one mobilized at 320 μg/kg had neither the highest absolute level or fold increase. This patient, however, also happened to be one of a subset of 4 patients who underwent two mobilizations at two different dose levels of plerixafor (Figure 1C and 1D). It is advantageous to compare patients to themselves due to the aforementioned inter-individual variation, at a given dose of plerixafor, in increases of absolute PB CD34+ count with plerixafor (5). Although two patients, including the one mobilized at 320 μg/kg, had a lower absolute PB CD34+ count at their higher dose level, all patients had a higher fold increase in PB CD34+ count with their higher dose level. The lower absolute PB CD34+ count at their higher dose level in two patients was likely related to their lower baseline PB CD34+ counts at the time of their higher compared to lower plerixafor dose.

Figure 1.

Post-plerixafor increases from baseline in PB CD34 concentration. Patients with hatching were on hydroxyurea. A and B. Seven unique patients with the same starting PB CD34+ concentration (1/uL: 4 pts, 2/uL: 3 pts) who received different dosing of plerixafor. A shows the absolute values, B shows the fold increases. C and D. Four unique patients who underwent repeat mobilization at higher plerixafor doses. C shows the absolute values, D shows the fold increases. The second patient shown underwent HPC collection (2.6 total blood volumes) prior to the 11 hr post-plerixafor (240 ug/kg) blood draw. The 3^rd patient shown, when dosed at 240 ug/kg, had a baseline CD34 count of 0, so a baseline of 0.25/uL was used to obtain a fold-increase value.

The two patients mobilized at 320 ug/kg did not have any clinical adverse events. Lab-based safety parameters of the 320 compared to the 240 μg/kg dose of plerixafor are shown in Figure 2 and Supplementary Table 1, as per analysis of white blood cell (WBC) numbers and cell activation markers, as previously described (3). The two patients mobilized at 320 μg/kg had relatively high fold-increases in WBCs compared to patients dosed at 240 μg/kg, although not significantly different given the small sample size. They had no significantly increased cell activation markers compared to patients dosed at 240 μg/kg. As we previously reported, correlation between the fold increases in absolute numbers of neutrophils and in neutrophils expressing activated β2⁺ integrins appeared high.

Figure 2.

Fold changes from baseline to 12 hr post-plerixafor in white blood cells (WBC) and cell activation markers in the two patients mobilized at 320 ug/kg (■ and ●) compared to our previously published 240 ug/kg cohort (column bar means with standard deviations). A. Fold increases in WBC, absolute neutrophil count (ANC), absolute lymphocyte count (ALC), and absolute monocyte count (AMC). B. Fold increases in absolute numbers and percentages of cell activation markers. TF⁺ mono: tissue factor-positive monocytes, E-Sel⁺ PMN: E-selectin positive neutrophils, aβ2⁺ PMN: activated β2 integrin-positive neutrophils, aMac-1⁺ PMN: activated Mac-1-positive neutrophils, L-Sel⁻ PMN: L-selectin negative neutrophils, PMA:plateletmonocyte aggregates, PNA: platelet-neutrophil aggregates.

The enumeration of more primitive CD34+CD38− HPCs and their CXCR expression is shown in Supplementary Figure 1. CXCR4 expression in plerixafor-mobilized HPCs is associated with increased in vivo homing capacity (9) and thus potentially higher HPC engraftment. This is still an important question because although SCD patients undergoing gene therapy may not have delayed engraftment, the percent of infused HSCs that home to the bone marrow and contribute to long-term engraftment is still unknown. Importantly, there was no significant difference between dosing levels in the % of CD34+ cells that were CD38 negative. There was a significant decrease in the % of CD34+CD38−CXCR4+ cells with increasing dose of plerixafor, but the 12G5 antibody was used for staining which competes with plerixafor for binding (10). This antibody clone choice may explain why other publications also observed a decrease in CXCR4 expression with plerixafor (11-13). CXCR4 expression may actually increase with plerixafor, as was found when a CXCR4 clone not affected by plerixafor was used (14). Also, plerixafor is a reversible inhibitor of the CXCR4 receptor, so it is possible that with further manufacturing steps, CXCR4 expression on mobilized HPCs would normalize prior to infusion.

In summary, our data are compatible with the hypothesis that, at a given baseline PB CD34 count in a given patient, higher dosing of plerixafor may achieve higher fold-increases in PB CD34+ mobilization, and is safe. Although funding issues preclude our further patient enrollment on plerixafor dosing above 240 ug/kg, our results support further studies of continued dose escalation. CXCR4 expression on HPC with plerixafor mobilization should also be elucidated, given its crucial role in maintaining HSC quiescence (15, 16)