Clinical implications and prevalence of benign ethnic neutropenia (BEN) in breast cancer patients of Middle Eastern ethnicity.

Abstract

1601 Background: Benign ethnic neutropenia (BEN) commonly affects patients of African and Middle-Eastern descent and is not a true neutropenic state. Patients with BEN have the Duffy-null phenotype on red cells and Duffy phenotyping has been used as a surrogate marker for diagnosis. There is evidence that cancer patients with BEN are not at increased risk of infection with chemotherapy. The primary aim of this study was to assess the prevalence of BEN among breast cancer patients in Bahrain using Duffy antigen phenotyping on red cells. The secondary aims were to study treatment delays and infectious complications in BEN patients. Methods: We conducted this retrospective study after obtaining IRB approval. We reviewed records of 493 consecutive breast cancer patients treated in our setting from January 2018 to January 2024. We included patients with neutropenia at presentation (defined as having an absolute neutrophil count [ANC] of < 1.5 ×10 3 /µL). Patients with Duffy-null phenotype and no identifiable secondary causes of neutropenia were presumed to have BEN. Clinical details studied included drug, and family history, treatment interruptions for neutropenia, filgrastim responsiveness, and episodes of febrile neutropenia. Overall survival (OS) and progression-free survival (PFS) estimation using the Kaplan-Meier method and Cox regression analysis of prognostic factors were performed using R software version 4.2.0. Results: Of 493 patients, 72 (14.6%) had a presumed diagnosis of BEN. The median age at presentation was 45 yrs, and the median follow-up duration was 3.6 yrs. 13% patients had metastatic disease at presentation and 11% had triple-negative breast cancer (TNBC). The median ANC at diagnosis was 1.2 × 10 3 /µL (range 0.4─2.1 × 10 3 /µL). The 4-yr OS was 95% (95% CI, 89–100%) and the 4-yr PFS was 75% (95% CI, 63–89%). Treatment was interrupted due to low ANC in 65% of patients, and the median ANC at which treatment was delayed was 0.8×10 3 /µL. 89% patients had received filgrastim and all were filgrastim responsive. Only one patient had uncomplicated neutropenic fever. On multivariable analysis, inferior PFS was seen in patients with metastatic disease (HR, 6.2; 95% CI, 2.17–17.9; p < 0.001), and TNBC (HR, 7.73; 95% CI 1.79–33.3; p = 0.006). We did not find any effect of treatment delay on the PFS. Conclusions: Ethnic neutropenia is prevalent among breast cancer patients in Bahrain. Duffy phenotyping can be used in place of more invasive tests to identify these patients. Treatment delays due to apparent neutropenia are common, however, response to filgrastim is universal, and febrile neutropenia episodes are rarely seen. Since these patients are not at increased risk of infection, larger studies to identify unique neutrophil thresholds for holding chemotherapy in BEN can help avoid compromising therapy. This can have far-reaching implications in populations with a high prevalence of BEN.