Comparative Effectiveness of Antibiotic Prophylaxis Regimens in Pediatric Acute Myeloid Leukemia.

BACKGROUND: There are limited real-world comparative effectiveness data on different antibiotic prophylaxis regimens of children receiving treatment for acute myeloid leukemia (AML). PROCEDURE: We compared the effectiveness of third/fourth-generation cephalosporin (cephalosporin) or quinolone prophylaxis with no antibiotic prophylaxis to prevent bacteremia during frontline AML therapy using an established cohort from 17 institutions. The prophylaxis regimen was at the discretion of the treating clinician. Secondary outcomes were Clostridioides difficile infection (CDI), time to next chemotherapy course, and mortality. RESULTS: A total of 550 pediatric patients with AML contributed data from 1758 chemotherapy courses. Incident bacteremia was lower following cephalosporin and quinolone prophylaxis compared to no prophylaxis in post-induction I chemotherapy courses (6.9% cephalosporin vs. 33.9% no prophylaxis, risk ratio [RR] 0.20, p < 0.01; 20.8% quinolone vs. 33.9% no prophylaxis, RR 0.61, p < 0.01). Reduction in bacteremia risk was greater with cephalosporin than with quinolone prophylaxis (RR 0.33, p < 0.01). CDI was rare and was not increased in those receiving prophylaxis (2.5% quinolone vs. 3.6% cephalosporin vs. 5.7% no prophylaxis; p = 0.05). Time to next course of chemotherapy was similar between groups except during intensification I, where patients receiving cephalosporin prophylaxis had a longer time to next chemotherapy (50 days for cephalosporin vs. 41 days for no prophylaxis, p < 0.01). Deaths during treatment were rare. CONCLUSIONS: Cephalosporin and quinolone prophylaxis reduced bacteremia without an increased risk of CDI. Additional work is needed to assess the differential impact of prophylaxis regimen on subsequent use of empiric therapy and development of resistance.