A phase III randomized crossover trial of plerixafor versus G-CSF for treatment of WHIM syndrome
David H. McDermott, … , Michael P. Fay, Philip M. Murphy
David H. McDermott, … , Michael P. Fay, Philip M. Murphy
Published August 10, 2023
Citation Information: J Clin Invest. 2023;133(19):e164918. https://doi.org/10.1172/JCI164918.
View: Text | PDF
Clinical Research and Public Health Immunology

A phase III randomized crossover trial of plerixafor versus G-CSF for treatment of WHIM syndrome

Abstract

BACKGROUND Warts, hypogammaglobulinemia, infections, and myelokathexis (WHIM) syndrome is a primary immunodeficiency disorder caused by heterozygous gain-of-function CXCR4 mutations. Myelokathexis is a kind of neutropenia caused by neutrophil retention in bone marrow and in WHIM syndrome is associated with lymphopenia and monocytopenia. The CXCR4 antagonist plerixafor mobilizes leukocytes to the blood; however, its safety and efficacy in WHIM syndrome are undefined.METHODS In this investigator-initiated, single-center, quadruple-masked phase III crossover trial, we compared the total infection severity score (TISS) as the primary endpoint in an intent-to-treat manner in 19 patients with WHIM who each received 12 months treatment with plerixafor and 12 months treatment with granulocyte CSF (G-CSF, the standard of care for severe congenital neutropenia). The treatment order was randomized for each patient.RESULTS Plerixafor was nonsuperior to G-CSF for TISS (P = 0.54). In exploratory endpoints, plerixafor was noninferior to G-CSF for maintaining neutrophil counts of more than 500 cells/μL (P = 0.023) and was superior to G-CSF for maintaining lymphocyte counts above 1,000 cells/μL (P < 0.0001). Complete regression of a subset of large wart areas occurred on plerixafor in 5 of 7 patients with major wart burdens at baseline. Transient rash occurred on plerixafor, and bone pain was more common on G-CSF. There were no significant differences in drug preference or quality of life or the incidence of drug failure or serious adverse events.CONCLUSION Plerixafor was not superior to G-CSF in patients with WHIM for TISS, the primary endpoint. Together with wart regression and hematologic improvement, the infection severity results support continued study of plerixafor as a potential treatment for WHIM syndrome.TRIAL REGISTRATION Clinicaltrials.gov NCT02231879.FUNDING This study was funded by the Division of Intramural Research, National Institute of Allergy and Infectious Diseases.

Authors

David H. McDermott, Daniel Velez, Elena Cho, Edward W. Cowen, John J. DiGiovanna, Diana V. Pastrana, Christopher B. Buck, Katherine R. Calvo, Pamela J. Gardner, Sergio D. Rosenzweig, Pamela Stratton, Melissa A. Merideth, H. Jeffrey Kim, Carmen Brewer, James D. Katz, Douglas B. Kuhns, Harry L. Malech, Dean Follmann, Michael P. Fay, Philip M. Murphy

×

Figure 5

Effects of plerixafor versus G-CSF on circulating blood cell counts in patients with WHIM.

Options: View larger image (or click on image) Download as PowerPoint
Effects of plerixafor versus G-CSF on circulating blood cell counts in p...
(A) Plerixafor, but not G-CSF, reversed panleukopenia without affecting circulating platelet concentration. Baseline refers to the day 0 value of the equilibration phase after 2-day washout of G-CSF or plerixafor from the preceding phase. G-CSF and plerixafor values are the final values obtained for the approximately 3-hour postmorning dose at the end of each treatment phase. Dashed green horizontal lines indicate the prespecified thresholds for judging success for improving neutropenia and lymphopenia. Each symbol represents a different patient. Dashed red horizontal lines demarcate the normal range for adults for each parameter established by the NIH-CC Department of Laboratory Medicine. P values were determined by a 2-sided Wilcoxon’s matched pairs rank test. (B and C) Plerixafor is noninferior to G-CSF for durably reversing neutropenia and is superior to G-CSF for durably reversing lymphopenia for 1 year. Filled black circles indicate individual values at the indicated times; open black squares indicate missing values due to scheduling conflicts; and red triangles indicate missing data due to drug failure. The y axes are on a log scale from 50 to 5,000 for ANC and from 100 to 10,000 for ALC. The thresholds for judging success are indicated by dashed horizontal lines at 500 and 1,000 cells/μL for ANC and ALC, respectively. P values (at the top of each panel) were determined by a Wilcoxon’s matched pairs rank test, as specified in the Supplemental Statistical Analysis Plan and the Supplemental Statistical Analysis Plan Supplement.