Analysis of the Risk of Infection Stratified By Absolute Neutrophil Counts during Deferiprone-Associated Severe Neutropenia or Agranulocytosis

Introduction

Patients with transfusional iron overload receiving iron chelation therapy with deferiprone may experience a decrease in absolute neutrophil counts (ANC) that range from mild (ANC <1.5 x 10 ⁹/L) to severe neutropenia/agranulocytosis (typically defined as ANC <0.5 x 10 ⁹/L), and which put them at risk for infections. In patients taking deferiprone, the risk of infection and/or mortality associated with ANC of <0.2 or <0.1 x 10 ⁹/L is unknown. Our objective was to analyze data from patients taking deferiprone in clinical trials and the postmarketing setting to determine the incidence and severity of infection stratified by ANC levels. s

Methods

The clinical trial setting involved patients with systemic iron overload who had thalassemia, sickle cell disease, or other anemias and received deferiprone across 15 clinical trials. Data associated with episodes of severe neutropenia/agranulocytosis in patients who had been treated with at least 1 dose of deferiprone were reviewed.

For the postmarketing setting, all individual case safety reports since the first market approval of deferiprone on August 25, 1999 to August 31, 2021, were reviewed; cases coded to agranulocytosis were extracted; cases without any reported ANC were removed; and cases with at least 1 ANC value were analyzed. Obtaining confirmed ANC value was not always possible in the postmarketing setting. All infections occuring within ±2 weeks of agranulocytosis onset were identified, graded for severity, and classified as serious, including fatal or life-threatening. Serious events were defined as any event that resulted in death, was life-threatening, required hospitalization or prolonged existing hospitalization, caused disability or permanent damage, caused congenital anomaly, or was classified as an important medical event. Infection events associated with ANC values were categorized into 3 groups based on the nadir ANC (all x 10 ⁹/L): Group A, ANC ≥0.2 to <0.5; Group B, ANC ≥0.1 to <0.2; or Group C: ANC <0.1.

Results

A total of 977 patients with systemic iron overload received deferiprone in sponsored clinical trials. Among those, 9 patients were in Group A, 3 in Group B, and 10 in Group C. Of the 12 patients in Groups A and B, none had serious, severe, or fatal infections ( Table 1). Three of the 10 (30.0%) patients in Group C experienced serious or severe infections that included (1) parvovirus infection accompanied by diarrhea, a maculopaular rash, and pyrexia resulting in hospitalization; (2) a severe but nonserious respiratory tract infection with esophageal candidiasis in a patient with a history of neutropenia prior to initiating deferiprone; and (3) a fatal case of Pseudomonal sepsis in a patient who had a history of multiple cardiac and pulmonary comorbidities. Deferiprone was permanently discontinued in all 3 patients.

From postmarketing data, of a total of 238 adverse reactions that were coded as agranulocytosis, 176 had at least 1 ANC value. Among those, 65 reactions were in Group A, 20 in Group B, and 91 in Group C. Among the reactions in Group A, 5/65 (7.7%) had serious infections and 3 of them (4.6%) had fatal outcomes due to sepsis ( Table 2). Among the reactions in Group B, 2/20 (10.0%) had serious infections, with a fatal outcome in one of them (5.0%); the death was from sepsis and the life-threatening infection involved pneumonia with multiple pulmonary complications. Among the reactions in Group C, 12/91 (13.2%) had serious infections and 8 of them (8.8%) had fatal outcomes; the 8 deaths were associated with sepsis. The 4 serious non-fatal infections involved sepsis; pancytopenia with ANC nadir of 0 accompanied by oropharyngeal pain, fever, and diarrhea that required hospitilization; angioinvasive mucormycosis; and disseminated herpes simplex accompanied by sepsis, resulting in hospitalization.

Conclusion

Patients receiving deferiprone who become neutropenic are generally at a higher risk of infections or mortality with lower neutrophil levels (ie, ANC <0.2 or <0.1 x 10 ⁹/L compared with ANC ≥0.2 to <0.5 x 10 ⁹/L). Our findings are consistent with the recently published guidelines from the European Hematology Association and the European Network for Innovative Diagnosis and Treatment of Chronic Neutropenias that used agranulocytosis to describe events with ANC of ≤0.2 x 10 ⁹/L in association with a high risk of severe and life-threatening infections.