Benign Ethnic Neutropenia in Individuals of African Descent: Incidence, Granulocyte Mobilization, and Gene Expression Profiling.

Introduction: Benign ethnic neutropenia (BEN) is a condition observed in individuals of African descent, and is characterized by a reduced absolute neutrophil count (ANC) less than 1500/uL in the absence of secondary causes. In contrast to other causes of neutropenia, BEN does not increase risk of oral or systemic infections. BEN has been described in up to 40% of certain ethnic groups, but its incidence in African Americans is unknown. Additionally, the pathogenesis is currently unknown. Through this study, we wish to determine the incidence of BEN, compare the neutrophil increment after dexamethasone (Dex) and granulocyte-colony stimulating factor (G-CSF), and compare peripheral blood (PB) granulocyte gene expression profiles by microarray and PB protein array to controls in order to better understand BEN.

Methods: Complete blood count (CBC) data were obtained from the National Health and Nutrition Examination Survey (NHANES) III. White blood cell and absolute neutrophil counts were compared in non-Hispanic whites, non-Hispanic blacks, Mexican-Americans, and others. Normal volunteers of African descent were recruited through an IRB approved study at NIH. Each volunteer underwent three PB leukaphereses: baseline, one day after 8mg of Dex, and one day after 5ug/kg G-CSF. CBCs were measured before and after each drug treatment. Granulocytes were enriched by gravity sedimentation with 6% hetastarch, hypotonic saline lysis of erythrocytes, and density gradient centrifugation using Ficoll-Pague. Granulocyte mRNA were applied to Affymetrix U133 plus 2.0 chips following extraction using a phenol-chloroform based method (RNA Stat 60, Tel-Test) and DNA removal. The gene expression pattern of BEN was compared to that of normals (ANC >4000/uL).

Results: The NHANES III sample size included 25,925 individuals. Mean ANC for the each ethnic groups were 4.55 (white), 3.67 (black), 4.80 (Mex-Am), and 4.52 (others) k/uL. The incidence of neutropenia (<1500/uL) was 0.25% (white), 4.05% (black), 0.35% (Mex-Am), and 0.98% (others). Our local incidence of BEN was 5.7% (4 of 70 volunteers screened). The mean WBC in our BEN were 4.13 +/− 0.72 (baseline), 8.57 +/− 0.59 (Dex), and 16.9 +/− 2.3 (G-CSF) k/uL, compared to those of normals, 7.56 +/− 0.95 (baseline), 12.17 +/− 2.13 (Dex), and 29.22 +/− 4.8 (G-CSF) k/uL. The absolute neutrophil increment of BEN volunteers was 4.95 k/uL (94.7% of neutrophil increment in normal volunteers) after Dex, and 12.7 k/uL (60.2% of normal) after G-CSF. Gene expression profile comparisons showed that at baseline, several genes were upregulated more than 10 fold in BEN, including apoptotic factors, metalloproteases 8 and 13, TREML 3&4, TGF-beta, and endothelial factors (endothelin 1, endothelin receptor B, and fibroblast growth factors). These differences were less pronounced after Dex or G-CSF. Additionally, gene expression pattern is markedly different after Dex vs. G-CSF, regardless of the neutrophil count.

Conclusions: Our results establish the incidence of BEN in the US at approximately 4%, significantly lower than that suggested by prior small reports. Additionally, the neutrophil increment is much lower in BEN following stimulation with G-CSF, suggesting the mechanism of granulocyte mobilization differs between Dex and G-CSF. Neutrophil mRNA in BEN showed increased expression of apoptotic, endothelial, and some leukocyte specific transcripts. Real-time PCR of candidate genes, accrual of more BEN subjects, and analyses of protein array are ongoing.