Increased numbers of circulating hematopoietic stem/progenitor cells are chronically maintained in patients treated with the CD49d blocking antibody natalizumab

Halvard Bonig, Annette Wundes, Kai-Hsin Chang, Sylvia Lucas and Thalia Papayannopoulou


Blockade of CD49d-mediated lymphocyte trafficking has been used therapeutically for certain autoimmune diseases, such as multiple sclerosis (MS). In addition to negative effects on the trafficking of mature lymphocytes to sites of inflammation, CD49d blockade in mice and monkeys rapidly mobilizes hematopoietic stem/progenitor cells (HSPCs) capable of short- and long-term engraftment. Here we aimed to ascertain the effects of treatment with antifunctional anti-CD49d antibody in humans (MS patients receiving infusions of the CD49d-blocking antibody natalizumab) on levels of circulating HSPCs after a single dose of antibody or after long-term treatment. On average, 6-fold elevated levels of circulating CD34+ cells and colony-forming unit-culture (CFU-C) were achieved within 1 day of the first dose of natalizumab, and similar levels were continuously maintained under monthly natalizumab infusions. The blood of natalizumab-treated subjects also contained SCID-repopulating cells. The fate of these circulating HSPCs and their clinical relevance for MS patients remains to be determined.


In nonhuman primates, CD49d-blockade with humanized antifunctional antibodies results in rapid (peak < 1 day) and prolonged (> 10 days) increase in circulating hematopoietic stem/progenitor cells (HSPCs).1,2 Elevated circulating HSPCs were likewise observed in anti-CD49d antibody-treated mice.2 Furthermore, mice conditionally ablated for CD49d sustain more than 8-fold elevated levels of circulating HSPCs without evidence of progressive accumulation or of bone marrow (BM) depletion for their life span.3 Even though the homing efficiency of anti-CD49d-mobilized HSPC was reduced, if sufficient numbers of cells were used, these cells provided short- and long-term engraftment. Studies in humans treated with anti-CD49d antibody, particularly after protracted blockade of CD49d, have not been reported, nor have any other studies of prolonged administration of mobilizing agents been performed. A clinical-grade humanized mouse-antihuman function-blocking CD49d antibody (natalizumab, Tysabri;Biogen/Idec, Cambridge, MA) is available under a special restricted distribution program for treatment of relapsing-remitting multiple sclerosis (MS) patients who failed to respond to or did not tolerate first-line therapeutics.4 As trials of natalizumab in healthy volunteers are not justifiable, because of prolonged immune-modulating effects of the antibody, including the possibility of rare, but potentially fatal progressive multifocal leukencephalopathy,5 we made observations in a cohort of MS patients receiving/scheduled to receive disease-modifying monotherapy with natalizumab.


Human subjects and protocol

Adults with MS receiving/scheduled to receive disease-modifying therapy with natalizumab (300 mg intravenously once every month) at the University of Washington Departments of Neurology/Rehabilitation Medicine were eligible for participation. Exclusion criteria were other disease-modifying therapy, steroids, or lithium. After written informed consent was obtained in accordance with the Declaration of Helsinki, immediately preceding the next scheduled natalizumab infusion blood was drawn from “untreated” patients (before the first infusion) and “chronic” patients (≥ 5 prior doses). In some patients, a second blood draw was done after infusion, generally on the subsequent day. A cohort of healthy controls was also recruited. Blood draws were anonymous; except for classification as “untreated/first-dose” or “chronic” recipient, no subject information was collected. The study was approved by the University of Washington Internal Review Board.

HSPC assays

Colony-forming unit-culture (CFU-C) assays were performed as described.6 Side-scatter low/CD34bright (CD34+) cells were quantified by flow cytometry, as described.7 The presence of competitive repopulating units (CRU) was tested in xenotransplants 9 to 10 weeks posttransplantation, as described. Transwell migration of CFU-C toward SDF-1 (100 ng/mL) was enumerated as described.6 Cell-cycle status on flow-sorted CD34+ cells was analyzed by Acridine Orange staining.8

Results and discussion

Before the first natalizumab dose, circulating CD34+ cells and CFU-C in “untreated” MS patients (Figure 1A,D) were within the range reported for healthy subjects9 and our healthy controls studied concurrently (ie, MS per se is not associated with elevated circulating HSPCs). Subjects on “chronic” natalizumab treatment had 5- to 7-fold elevated circulating CD34+ cells and CFU-C (Figure 1A,D) 1 month after infusion. In a subgroup of “chronic” subjects, circulating HSPCs were analyzed before and 1 day after the monthly dose of natalizumab. Renewed infusion did not result in significant further augmentation of circulating CD34+ cells or CFU-C (analyzed 1 day after the infusion, Figure 1C,F), indicating continuous functional satiation of CD49d on BM-HSPC with standard natalizumab dosing.

Figure 1

Elevated numbers of circulating HSPCs in the blood of natalizumab-treated MS patients. (A,D) Circulating HSPCs in healthy controls, not natalizumab-treated MS patients and long-term natalizumab-treated MS patients: Circulating CD34+ cells (1.8 ± 0.4/μL, P = .36 vs control) and CFU-C (638 ± 128/mL, P = .4 vs control) were normal in MS patients before the first natalizumab infusion (“untreated”) and significantly elevated in patients who had received at least 5 prior doses of natalizumab, measured immediately before application of the next dose (“chronic”; 9.0 ± 1.2/μL CD34+ cells, 3243 ± 332/mL CFU-C, P < .005 vs control). (Normal controls [“nl. ctrl.”]: 1.3 ± 0.1/μL CD34+ cells, 608 ± 129/mL CFU-C, on the left.) (B,E) First-dose natalizumab patients: After the first natalizumab infusion (“after”), peripheral blood CD34+ cells and CFU-C were significantly increased over pretreatment values (“before”; 1.6 ± 0.2/μL vs 8.0 ± 2.1/μL CD34+ cells, 414 ± 161/mL vs 2560 ± 726/mL CFU-C, P < .005). (C,F) Chronic natalizumab patients: Renewed natalizumab infusion (“after”) in “chronic” natalizumab recipients did not result in additional mobilization, compared with CD34+ cell and CFU-C values just before that infusion (7.9 ± 1.7/μL vs 7.9 ± 0.9/μL CD34+ cells, P = .99; 3133 ± 335/mL vs 3525 ± 305/mL CFU-C, P = .27). CD34+ cells/μL (A-C) or CFU-C/mL (D-F) are plotted on the y-axis. Each diamond represents values from one patient (for CFU-C: mean values from replicates from one patient); bars and whiskers indicate mean values plus or minus SEM. (G) CFU-C migration. In contrast to normal BM HSPCs, peripheral blood CFU-C from natalizumab recipients did not migrate toward SDF-1 in in vitro transwell assays (P < .001). (H) Cell-cycle status of natalizumab-mobilized HSPCs. natalizumab-mobilized CD34+ cells were almost exclusively quiescent and overwhelmingly in G0 phase of cell cycle (flow cytometric histogram; RNA displayed on the x-axis, DNA on the y-axis).

Comparison of natalizumab-recipients before and 1 day after the first infusion (“1st dose”) revealed 5- to 6-fold increased CD34+ cells and CFU-C after the first infusion (Figure 1B,E). Mean postinfusion values after the first dose were no different from those in “chronic” natalizumab-recipients before or after repeated natalizumab infusions, documenting achievement of maximal levels of circulating HSPCs within 24 hours of a single natalizumab infusion. Because circulating CD34+ cells and CFU-C in untreated MS patients were normal, the increased frequency of circulating HSPCs in natalizumab-treated MS patients is apparently the result of drug effects.

CD34+ counts in natalizumab-treated MS patients were thus approximately one-sixth of those in MS patients mobilized with granulocyte colony-stimulating factor (G-CSF).10 However, the relative frequency of clonogenic cells appeared to be higher among natalizumab-mobilized than among G-CSF-mobilized CD34+ cells (1 in 3 CD34+ cells for natalizumab vs 1 in 10 for G-CSF mobilized CD34+ cells),11 likely because the G-CSF induced proliferation/differentiation dilutes clonogenic cells with more mature CD34+ subsets. Thus, the difference in HSPC mobilization potency may be less pronounced than CD34+ numbers suggest. CD34+ counts in natalizumab patients were one-third of those achieved with the CXCR4-antagonist AMD3100 in healthy volunteers.9

The sustained HSPC levels in natalizumab recipients contrast with mobilization kinetics after G-CSF, in which after the peak on day 5, circulating HSPCs numbers are regressive despite continued administration.12,13 It indicates that, without CD49d-mediated BM retention, a new equilibrium is established between HSPCs in BM and those in circulation. In natalizumab recipients, a total of approximately 50 × 10E6 CD34+ cells (8 CD34+ cells/μL × 6 L of blood) are in circulation at any given time. Assuming that the transit time of normal circulating HSPCs is short14 and not very different from normal in natalizumab-treated patients (supported by unpublished data for CD49d−/− mice), several million CD34+ cells are trafficking through the blood of the average natalizumab recipient every day. A similar pattern of sustained elevation of circulating HSPCs was observed in CD49d-deficient mice.3

Most CD34+ cells in blood from natalizumab patients are quiescent (Figure 1H), a phenotype associated with favorable homing and/or engraftment.15,16 Similar data were reported for unmobilized17 or G-CSF-mobilized CD34+ cells18 (ie, may represent a general feature of circulating HSPCs). Peripheral blood CFU-C from natalizumab-treated patients did not migrate in in vitro transwell assays, whether spontaneously or SDF-1–directed, in contrast to normal BM HSPC (Figure 1G). We attribute the lack of migration to continuous saturation by natalizumab, as several previous observations show absence of migration with anti-CD49d antibody-treated cells.

To establish whether peripheral blood of natalizumab-treated patients contained any CRU, xenotransplants of mononuclear cells from 7 to 8 mL of blood (containing 50 000-60 000 CD34+ cells) were performed. Human engraftment of more than 1% was observed in 3 of 3 recipients (not shown), although no quantitative conclusions about CRU can be drawn from these studies. The engraftment data are of interest in view of the cells' inability to migrate in vitro. Similar to our data, cells mobilized with the chemokine Groβ did not migrate in vitro but engrafted well,19 reinforcing the view that in vitro behavior of HSPCs cannot predict their in vivo performance.

The role, or clinical consequence, of elevated circulating HSPCs for natalizumab-treated MS patients is unclear. Potential effects of HSPC on remodeling/repair of nonhematopoietic organs have been reported.20,21 It is thus tempting to hypothesize that circulating HSPCs may contribute to the anecdotal cases of neurologic improvement under therapy, which may not be explained by the main mechanism of natalizumab in MS (ie, attenuated lymphocyte recruitment).22,23 Although in natalizumab-treated MS patients significant numbers of HSPCs continuously circulate outside the protective BM environment, their likelihood to acquire deleterious mutations is probably small, given their G0 cycling status, but this possibility needs to be considered. Thus far, an increased propensity of natalizumab patients for hematologic malignancies has not been reported; however, further long-term observations are needed.

The usefulness of natalizumab as a priming agent for HSPC peripheralization is likely restricted only to patients receiving natalizumab for their underlying illness, given the risk of immunization against natalizumab24 and prolonged immune-modulating effects. However, CD49d blockade with short-acting small-molecule CD49d inhibitors, if efficacious, could be theoretically considered as treatment strategy for patient groups intolerant or unresponsive to G-CSF,10,25 especially in combination with other mobilizing agents, such as the CXCR4 antagonist AMD3100.26


Contribution: H.B. conceived of the studies, wrote the IRB protocol, obtained informed consent, performed experiments, collected and analyzed data, and wrote the paper; A.W. wrote the IRB protocol and recruited subjects; K.-H.C. performed experiments; S.L. recruited subjects; T.P. analyzed data and wrote the paper.

Conflict-of-interest disclosure: S.L. serves on the Speaker's Bureau for Biogen/Idec, manufacturers of natalizumab. The other authors declare no competing financial interests.

Correspondence: Halvard Bonig, Johann-Wolfgang-Goethe University, Institute for Transfusion Medicine and Immune Hematology, Sandhofstrasse 1, 60528 Frankfurt; e-mail: hbonig{at} or hbonig{at}


The authors thank the nursing staff of the 8SE infusion station at University of Washington Medical Center for their support.

This work was supported by a Core Center of Excellence in Hematology grant from the FHCRC (National Institutes of Health [NIH] DK56465 to H.B.) and by NIH grant HL58734 (T.P.).

DK56465HL58734National Institutes of Health


  • The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

  • Submitted September 10, 2007.
  • Accepted January 10, 2008.


View Abstract