Clozapine and Benign Ethnic Neutropenia (Duffy-Null): Thresholds, Testing, and Evidence

Creator: Ryan S. Sultan, MD
Published: 2026-11-12

By Ryan S. Sultan, MD
Psychiatrist · Columbia University Irving Medical Center
Published November 12, 2026 · 6-minute read

Benign ethnic neutropenia (BEN), also called Duffy-null associated neutrophil count, is a lower baseline neutrophil count. The Clozaril label recommends against starting clozapine below an absolute neutrophil count (ANC) of 1,500/µL for most patients and sets that line at 1,000/µL for patients with BEN.¹ The prescriber applies those thresholds.

In a six-month prospective trial of clozapine in 274 people of African descent, 227 (82.8%) completed treatment and one participant (0.36%) developed severe neutropenia, defined as a count below 500.²

What benign ethnic neutropenia is, and how common

A 2026 UK guideline describes the cause as a homozygous variant in the ACKR1 gene, commonly called the Duffy-null genotype. It puts the genotype's prevalence at approximately 80% in people of African ancestry, 25% in people of Middle Eastern ancestry, and around 1% for both European and East Asian ancestry.³ In a study of 120 self-identified Black adults, 66.7% had the Duffy-null phenotype, and their median ANC was 2,820/µL, compared with 5,005/µL in the others.⁴

In a US national survey of 25,222 participants from 1999 to 2004, a neutrophil count under 1.5 × 10⁹/L (1,500/µL) was found in 4.5% of Black participants, 0.79% of white participants, and 0.38% of Mexican-American participants. Each person was measured once, so the survey couldn't capture fluctuation over time.⁵

A 2026 study of 1,412 healthy Duffy-null participants from 23 sites set the adult reference interval at 1.03 to 7.30 × 10⁹/L, and 25% of the adults had a count under 2.00 × 10⁹/L.⁶

What the label says about BEN

The label states that patients with BEN generally have lower baseline neutrophil counts, are not at higher risk of infections, and are not at increased risk of clozapine-induced neutropenia.¹ It gives them a separate table. These are the thresholds prescribers work from.

ANC thresholds for patients with documented BEN
ANC result	What the label recommends
1,000/µL or higher	Continue treatment on the usual testing schedule
500 to 999/µL	Continue treatment. Hematology consultation is recommended, and testing moves to three times weekly
Below 500/µL	Discontinue treatment, with hematology consultation and daily testing

Source: Clozaril prescribing information, Table 2, revised June 2025.¹ For patients without BEN, the label interrupts treatment at 500 to 999/µL.

Within each group's normal range, the routine testing schedule is the same with or without BEN, and it's laid out on the page about clozapine blood tests. For documented BEN the label asks for at least two baseline counts. For every patient it says to consider a hematology consultation before or during treatment.¹

The label refers to "documented" BEN and names no test for documenting it. Its text contains no instruction on genotyping or phenotyping.¹

How Duffy-null status is tested

Two kinds of test appear in the published studies. Phenotyping is a serologic test for the Fya and Fyb antigens; Duffy-null is the Fy(a-b-) result.⁴ Genotyping reads the rs2814778 variant in ACKR1, and the prospective clozapine trial used a TaqMan assay.²

The UK guideline recommends genotype testing for all people starting clozapine and after a below-threshold blood result, with a recommended laboratory turnaround of 1 week.³ At one clozapine clinic that tested 108 patients, 26 (24%) who had never been diagnosed with BEN by standard blood monitoring had the Duffy-null genotype.⁷

What happened when patients of African descent took clozapine

The prospective trial enrolled people with schizophrenia spectrum disorders at sites in the United States and Nigeria. Of 249 participants with a known genotype, 199 (79.9%) were Duffy-null. A count under 1,500 occurred in 33% of the Duffy-null group (65 of 199) and 6% of the others (3 of 50). The one case of severe neutropenia came over 1,467.6 person-months of clozapine exposure, and the authors described rates of infection and fever as low.²

A 2019 genetic study of 552 people with treatment-resistant schizophrenia and African ancestry found that Duffy-null patients were significantly more likely to develop neutropenia and have to stop clozapine, with an odds ratio of 20.4. Its abstract reports neutropenia and stopping clozapine, and it offers no estimate of agranulocytosis risk. The authors described a genetic test as dependent on additional safety studies.⁸

In an earlier study covering 1989 to 1999, 5.3% of African American patients (31 of 588) and 2.4% of Caucasian patients (31 of 1,287) had clozapine discontinued for leucopenia. No African American patient developed agranulocytosis, and 8 Caucasian patients (0.62%) did.⁹

The consensus threshold

A global Delphi panel published in Lancet Psychiatry recommended lowering the ANC threshold for considering clozapine cessation to 1.0 × 10⁹ cells per L, and to 0.5 × 10⁹ cells per L for Duffy-null individuals.¹⁰ The Duffy-null figure equals 500/µL, the count below which the label discontinues clozapine in BEN.¹ A 2016 systematic review concluded that the evidence suggests a count above 1,000/mm³ is safe for initiating or resuming clozapine in BEN.¹¹

Does a low baseline count mean more infections?

A 2020 analysis of the UK Biobank and a Danish case-cohort study found that 21% of people with the Duffy-null genotype had a neutrophil count below a commonly used treatment cutoff, compared with 1% of others. The genotype was not associated with an increased risk of viral or bacterial infection. These were population-based samples, and the authors left proof in Duffy-null people who have a low count to a prospective study.¹²

The UK guideline states, as background, that these patients have no increased risk of infection or of clozapine-induced agranulocytosis.³ The label's sentence refers to neutropenia.¹

Limitations of the evidence

This evidence has limitations. First, a PubMed search on November 12, 2026 found no randomized trial comparing BEN thresholds with standard thresholds. Second, the prospective trial recorded one case of severe neutropenia among 274 participants, too few to compare agranulocytosis rates by genotype, and the infection data come from population-based samples, which weren't studies of clozapine patients. Nevertheless, the label permits clozapine at lower counts in documented BEN, and a low baseline count in a patient who may be Duffy-null warrants a conversation with the prescriber about testing and hematology consultation before clozapine is ruled out.

Frequently asked questions

Can you take clozapine if you have benign ethnic neutropenia?

The label allows it, with the prescriber applying its separate thresholds. The Clozaril label states that patients with benign ethnic neutropenia are not at increased risk of clozapine-induced neutropenia, and it sets the baseline line for starting at 1,000/µL instead of 1,500/µL.¹

Can I take clozapine with a low white blood cell count?

It depends on the neutrophil count and its cause, and the prescriber makes that call. The label recommends against starting below an ANC of 1,500/µL for most patients and sets that line at 1,000/µL for documented benign ethnic neutropenia, with at least two baseline counts.¹

How do you get tested for Duffy-null?

Two tests appear in the published studies: a serologic phenotype test for the Fya and Fyb antigens, and a genotype test for the rs2814778 variant in the ACKR1 gene. The Clozaril label refers to documented benign ethnic neutropenia and names no test.^{4, 2, 1}

Does being Duffy-null make clozapine less safe?

In a six-month prospective trial of 274 people of African descent, one participant (0.36%) developed severe neutropenia, and 199 of 249 genotyped participants were Duffy-null. A count under 1,500 occurred in 33% of the Duffy-null group and 6% of the others. The trial was too small to compare agranulocytosis rates by genotype.²

Does benign ethnic neutropenia raise infection risk?

The label states that patients with benign ethnic neutropenia are not at higher risk of infections. A 2020 analysis of the UK Biobank and a Danish case-cohort study found the Duffy-null genotype was not associated with an increased risk of viral or bacterial infection. Those were population-based samples, and they weren't studies of clozapine patients.^{1, 12}

In a crisis: call or text 988, or go to the nearest emergency department. A first episode of psychosis belongs in a hospital-based or coordinated specialty care program.

I start and manage clozapine in an outpatient practice in Chelsea, Manhattan. Details are on the schizophrenia care in NYC page. Clozapine should not be stopped, skipped, or restarted without the prescriber. This page is general information and does not replace an evaluation.