Early discontinuation of empirical antibiotics versus extended treatment in children with cancer and high-risk febrile neutropenia in Denmark: an open-label, randomised controlled trial

Background: Early discontinuation of empirical antibiotics in adults with high-risk febrile neutropenia based on clinical stability and defervescence, irrespective of neutrophil count, has been shown to reduce antibiotic exposure without compromising safety. We aimed to evaluate this approach in children with cancer, a population for which randomised trials in high-income settings are lacking. Methods: We conducted an investigator-initiated, multicentre, open-label, randomised, superiority trial in three paediatric haematology–oncology centres in Denmark. Patients aged 0–17 years receiving intensive chemotherapy and presenting with fever during high-risk neutropenia (absolute neutrophil count <0·5 × 10⁹ cells per L anticipated to last ≥7 additional days) were eligible. Patients with microbiologically documented infection were excluded. Fever episodes were randomly assigned (1:1) to either early discontinuation of intravenous antibiotics after 48 h of defervescence and clinical stability (short treatment group), or continuation until neutrophil recovery with a maximum of 10 days per episode (long treatment group). Randomisation was done using a computer-generated sequence embedded in a dedicated electronic data capture system, stratified by site, via block randomisation with random block size (of four, six, or eight). The randomisation sequence was concealed until allocation, and investigators were unaware of group assignment until completion of randomisation. The primary endpoint was the number of days on antibiotics within 28 days after the first treatment initiation. Main analyses were done in the intention-to-treat population. Safety was assessed by infection-related serious adverse events, including bacteraemia, focal infection, sepsis, and death. The trial was registered with ClinicalTrials.gov, NCT04637464, and is completed. Findings: Between Nov 26, 2020, and Jan 28, 2025, febrile episodes were systematically screened for eligibility; 88 febrile episodes in 70 children were randomly assigned (45 to the short treatment group and 43 to the long treatment group) and included in the intention-to-treat analysis. 37 (42%) episodes were in female participants and 51 (58%) were in male participants. The time on antibiotic treatment within the 28-day follow-up was a mean of 8·1 days (SD 5·9) in the short treatment group versus 13·5 days (6·2) in the long treatment group, leading to an estimated difference of –4·0 days (95% CI –6·0 to –1·9; p=0·0002). Serious adverse events occurred in ten (22%) of 45 episodes in the short treatment group and nine (21%) of 43 in the long treatment group (risk ratio 1·06 [95% CI 0·46–2·47]). No deaths occurred. Interpretation: Early discontinuation of intravenous antibiotics regardless of neutrophil recovery reduced unnecessary antibiotic exposure without indications of compromised safety. Although the study was not powered to detect differences in rare safety outcomes, it provides evidence for this approach in the management of children with cancer and high-risk febrile neutropenia. Funding: Danish Childhood Cancer Foundation, Danish Cancer Society, Innovation Fund Denmark.