Complement inhibition in pediatrics: Comprehensive infection analysis from a large pediatric single center

Abstract

Abstract Background Complement inhibition (Ci) has revolutionized outcomes in disorders, such as paroxysmal nocturnal hemoglobinuria (PNH) and atypical hemolytic uremic syndrome (aHUS), by improving prognosis and quality of life. However, Ci increases susceptibility to infection, despite use of immunization and prophylactic antibiotics. While infection risk with encapsulated bacteria is well documented, the true extent of this risk is not well characterized in children. Methods In this IRB-approved, retrospective study, we identified patients treated with Ci at Children's Healthcare of Atlanta from 2010 to 2025. Patients with solid organ transplant and hematopoietic stem cell transplant were excluded. Infections were identified via positive culture, and data were reviewed from 1st dose of Ci to 6 months following last dose. Infection rate was calculated as # infections/days at risk. Results Infection rates were assessed in 96 patients over 227 patient years. Median age at first Ci dose was 15 years (range 4 months to 20 years) and 52% were male. Ci indications included a cohort of children who were otherwise immune competent: aHUS (n=23), C3 glomerulopathy (n=6), Shiga-toxin producing Escherichia coli (STEC)-HUS (n=4), classic PNH (n= 9), and SARS-CoV-2 (n=1). The second cohort included patients who received additional immune suppression or had functional/surgical asplenia, including sickle cell disease (SCD) hyperhemolysis (n=29), autoimmune hemolytic anemia (AIHA) (n=4), PNH with bone marrow failure (BMF) (n= 12), refractory myasthenia gravis (MG) (n=3), thrombotic microangiopathy secondary to lupus (n=3), and other (n=2). All patients with infections were on antibiotic prophylaxis and up to date with meningococcal and pneumococcal vaccines. Among the immune competent cohort (n= 43), median duration of Ci treatment was 27.7 months (IQR 6.7, 80.0). Only 1 patient with STEC-HUS developed an infection, pseudomonas bacteremia, 8 days after first dose. Among the immunocompromised cohort (n= 53), median duration of Ci was 6.37 months (IQR 6, 17). There was no association of duration of Ci dosing (< 5 doses compared to ≥5 doses) and infection rate (p = 0.99). Nine immunocompromised patients developed significant infection(s), including bacteremia (n=5), candidemia (n= 1), invasive mold (n= 1), and pneumonia (n=2). Two children with SCD had bacteremia/fungemia; infections were identified within 1 day of first Ci dose and both patients died. In contrast, median time from first Ci dose in the remaining immunocompromised cohort to first infection was 122.5 days (IQR 61, 1068). Cumulative incidence of infection was significantly higher in the immunocompromised patients (1-yr estimate 12% vs 2%, p=0.008.) and they had significantly higher infection rate (p=0.016). There were ten deaths: 9 immunocompromised patients and 1 immune competent patient with a complex medical history. Two deaths were attributed to encapsulated bacterial infection. A patient with AIHA on several immunomodulators and splenectomy died from necrotizing pancreatitis and Klebsiella pneumoniae bacteremia 40 days after last Ci dose. A patient with SCD, on immunosuppression for a rare immune disorder, received one dose of eculizumab for hyperhemolysis 4 days prior to dying from Streptococcal pneumonia sepsis and meningitis; these infections were present prior to eculizumab administration. Conclusion This comprehensive institutional review demonstrates the favorable safety profile of Ci in pediatrics. Notably, no cases of meningitis were linked to Ci, duration of Ci was not associated with increased infection, and patients with aHUS and classic PNH had no significant infections. There were significantly more infections in the immunocompromised cohort, but there may be additional important confounding factors. Among those with SCD, given that all infections were identified within 1 day of first Ci, it is likely that these infections were drivers of hyperhemolysis (the Ci indication) and present prior to the drug. Thus, it's unlikely that Ci itself contributed to bacteremia. The underlying diagnosis in the remaining immunocompromised cohort, degree of immune suppression, and neutropenia are important infection risk factors; additional analyses accounting for degree of immune suppression and risk are pending. These data emphasize the importance of ongoing vigilance despite immunizations and antibiotics and education regarding infection risk and early symptom recognition.