When non-Whiteness becomes a condition

There is a long history in medicine of using healthy White men as the standard for normal, ascribing anything different as abnormal, other, or requiring adjustments. Many clinical algorithms include race-based corrections with the implicit assumption that those with non-White skin are inherently “other.”¹  Before sequencing of the human genome, it was argued that racial “differences” were likely based in genetic variations concentrated within certain communities. On the contrary, it is now categorically clear that there is no genetic basis by which we can classify humans into discrete races, and race is simply a socially construed grouping of humans based on shared physical or social qualities.²  In fact, there is often more genetic variation within people of the same self-identified race than between those of different races. Thus, race as a proxy for genetics is woefully inadequate. Although our understanding of and access to genetics has improved significantly in the past decades, our reliance on race as a proxy for genetic markers has not changed, suggesting that medicine as a profession, a system, and an institution has yet to erase racist ideologies acquired from the past.

One example of non-whiteness being ascribed the label of a condition in hematology is the designation of benign ethnic neutropenia (BEN). BEN is an inherited cause of an absolute neutrophil count (ANC) <1500 cells/μL with no clinical sequelae or increased risk of infection that is most often seen in people of sub-Saharan African descent.³  However, BEN has also been described in people of other ethnicities, including people of Arab, West Indian, Egyptian, and Jordanian descent.^4,5  Ethnicity and race are not synonymous, but there is often a significant overlap. Ethnicity is the grouping of people based on national or cultural similarities, whereas race is the social grouping of people based on skin color that has no genetic basis. Nevertheless, medicine routinely categorizes people based on physical features such as skin tone, better defined as race, but the implications for ethnicity are similar. Although these terms have different definitions, the biases in medicine and research toward people of non-European ancestry or non-White skin remain the same.

BEN is most commonly observed in people of African ancestry, and it is well known that people of European background often have higher peripheral ANC levels than people of African ancestry.^6-8  Data from the National Health and Nutrition Examination Survey show that African-American individuals have a white blood cell count that is on average 700 cells/μL lower than that in White American subjects.⁹  Moreover, ANC reference ranges for African adults are variably reported to be 1050 to 4080 cells/μL,¹⁰  910 to 4720 cells/μL,¹¹  and 500 to 5400 cells/μL,¹²  all below the conventional adult ANC reference range used in most countries outside of Africa.^13,14  Importantly, <1% of healthy White American subjects are neutropenic (ANC <1500 cells/μL),¹⁵  but this clinically insignificant condition of an ANC <1500 cells/μL resulting in the label of BEN is identified in up to 25% to 50% of people of African ancestry in the United States.¹⁶  This elicits concern that the current ANC reference ranges in many countries are not representative of the entire population. An accurate ANC reference range is crucial to adequate care because ANC levels below the reference range frequently prompt expensive evaluation for neutropenia and can also prevent appropriate treatment with chemotherapy in the setting of malignancy.¹⁷ 

Although BEN was historically linked to ethnicity, we now know that BEN is correlated with homozygosity of a single nucleotide polymorphism (SNP) in the promoter region of the Duffy antigen receptor for chemokines(DARC) gene (also known as atypical chemokine receptor[1ACKR1; rs2814778]) gene.¹⁸  Homozygosity at this SNP results in a red blood cell membrane antigen phenotype termed Duffy null or Fy(a-b-). Aligning with the distribution of BEN, the Fy(a-b-) phenotype is found in <1% of those with European or Asian ancestry but is very common in individuals from sub-Saharan Africa (80%-100%) and the Arabic peninsula (50%-70%).^{8,11,16,19,20}  In addition to being a membrane glycoprotein that acts as a chemokine receptor for proinflammatory cytokines,²¹  Duffy antigens are used by malaria merozoites to penetrate and invade red blood cells. In the absence of Duffy antigens [the Fy(a-b-) phenotype], Plasmodium vivax merozoites attach to red blood cells but are unable to invade the cells.²²  Thus, the Fy(a-b-) phenotype is protective against malaria and enriched in individuals of sub-Saharan African and Arabic ancestry as this was an advantageous trait.

Several studies have confirmed that the Fy(a-b-) phenotype is the most robust method to identify BEN, and we know that the Fy(a-b-) phenotype has stronger diagnostic utility for BEN than self-identified race or ethnicity [Fy(a-b-) phenotype, 97.36% sensitivity, 95.65% specificity; self-identified race, 65.7% sensitivity, 48.8% specificity].^23-25  In addition, other more recently identified polymorphisms, such as the lead SNP rs9131 on the CXCL2 gene seen in some people of African ancestry, are implicated as drivers of lower ANCs.²⁶  Although it is well known that genetic polymorphisms are much more strongly associated with lower ANCs than ethnicity, the practice of ascribing the condition of BEN to people solely based on a physician’s assessment of skin tone and a comparison with current ANC reference ranges has not changed.

BEN, although carrying no clinically important ramifications, is a label that illustrates how common phenotypes seen in non-White individuals continue to be considered nonstandard in the medical community. Although ANC reference ranges from Africa prove that an ANC <1500 cells/μL is common, healthy Black people in the United States with an ANC <1500 cells/μL are described as abnormal and thus worthy of the diagnosis “benign ethnic neutropenia” simply because they do not fall within the ANC reference range of 2000 to 8000 cells/μL of the average White person. Medicine’s classification of common, harmless, non-White characteristics as aberrant is not unique to BEN. It is seen in diagnoses such as Voigt-Futcher lines, the normal demarcations between areas of differing pigmentation such as between the anterior and posterior arm,²⁷  or steatopygia, which describes a higher accumulation of adipose tissue in the buttocks than what is typically seen in White people.²⁸ 

It is no surprise that systemic racism, structures of discriminatory policies and practices that assign privilege on the basis of skin color, is insidiously embedded in systems that we blithely invoke in our daily practice. We need to examine why we ascribe labels to clinically insignificant variants, labeling as conditions differences that predominantly affect non-White communities. Systemic racism was purposely construed; therefore, deliberate actions are crucial to its abolishment. Thus, we advocate for “benign ethnic neutropenia” to be referred to as “typical neutrophil count with Fy(a-b-) status.” This is a more accurate description of the genetic driver of the lower ANCs and avoids the suggestion that ethnicity alone is causal or that this common variant is abnormal. In addition, we anticipate that further research will allow for the development of new comprehensive ANC references ranges that account for the Fy(a-b-) phenotype, which should eliminate the need for a qualifying label altogether. We must actively confront and correct the racist labeling of common and harmless characteristics in non-White people as abnormal conditions simply because they differ from White phenotypes. Rejecting the term “benign ethnic neutropenia” is a simple way to start.

Contribution: L.E.M. developed the idea and wrote the manuscript; and M.A. wrote and edited the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Lauren E. Merz, 75 Francis St, Boston, MA 02115; e-mail: lmerz@bwh.harvard.edu.