OBJECTIVE: To compare the efficacy and safety of panipenem/betamipron with cefepime as empirical monotherapy for adult cancer patients with febrile neutropenia, a randomized, open-label, comparative trial was performed. METHODS: All enrolled patients were randomly assigned to receive either panipenem or cefepime. All febrile episodes were classified as microbiologically defined infection (MDI), clinically defined infection (CDI) or unexplained fever (UF). Clinical responses to antibiotic therapy were defined as success, initial response but regimen modified or failure. RESULTS: A total of 116 patients were enrolled: 55 patients in the panipenem group and 61 patients in the cefepime group. Demographic and clinical characteristics were similar in the two groups (P > 0.05). In the final evaluation, the success rate for the panipenem group (89.1%) was similar to that of the cefepime group (91.8%) (non-inferiority, P = 0.002, 95% confidence interval: -13.48%, 10.35%). Of the 18 bacterial isolates, nine (50%) were gram-positive and nine (50%) were gram-negative. The prevalence of adverse events in the panipenem group (23.6%) were similar to those in the cefepime group (23.0%) (P = 0.93). All of the adverse events were well tolerated and transient. CONCLUSIONS: Although larger studies are necessary, panipenem appeared to be as effective and safe as cefepime for empirical monotherapy in the treatment of adult cancer patients with febrile neutropenia.