Abstract Background Use of anti-anaerobic antibiotics has been associated with disruptions of gut microbiota composition and adverse outcomes, including acute graft-vs-host disease in allogeneic hematopoietic cell transplant (allo-HCT) patients (pts). To test the hypothesis that sparing gut anaerobes would confer clinical benefit, we randomized pts undergoing allo-HCT to different regimens for treatment of fever and neutropenia (FN) (NCT03078010). Here we analyzed antibiotic utilization patterns in the trial to evaluate the feasibility of complete avoidance of anti-anaerobic agents. Methods Pts were randomized to the institutional standard piperacillin-tazobactam (PTZ) or cefepime (CPM) for initial FN. Subsequent antibiotic changes for FN were left to provider discretion. Pts on the CPM arm could be de-escalated to aztreonam (or to a quinolone later in the trial) if clinically stable and afebrile x 72 hours, whereas no formal de-escalation was built into the PTZ arm. The primary outcome was a microbiome endpoint, analysis of which is ongoing. Here, we analyzed anti-anaerobic days of therapy (DOT), which was the sum of duration of days exposed to anti-anaerobic drugs. In contrast, length of therapy (LOT) was analyzed as the duration of days exposed to antibiotics, irrespective of the number of drugs. Results There were 54 and 46 evaluable pts in the CPM and PTZ arms, respectively (Table 1). The majority (29/46, 63%) in the PTZ arm remained on PTZ until engraftment. In contrast, more antibiotic modifications were seen in the CPM arm, including 10 (18.5%) switched to PTZ or meropenem for persistent FN, 1 (1.9%) narrowed to ceftriaxone for targeted infection treatment, and 31 (57.4%) with de-escalation. The remaining 12 (22.2%) stayed on CPM until neutrophil recovery. The median (interquartile range, IQR) anti-anaerobic DOT was 12 days (range 7, 18) and 4 days (range 3, 8) for PTZ and CPM arms, respectively. Conclusion Antibiotic modifications for FN therapy were common in the CPM arm. Complete avoidance to anti-anaerobic drugs was not possible in the CPM arm, but anaerobic duration of treatment was shorter in the CPM than the PTZ arm. The ability to de-escalate likely mitigated the duration of exposure to anti-anaerobic agents. Disclosures Marcel van den Brink, MD, PhD, Ceramedix: Honoraria|Da Volterra: Honoraria|DKMS: Board Member|Garuda: Honoraria|GSK: Honoraria|Juno: IP Licensing|Lygenesis: Honoraria|Nektar Therapeutics: Honoraria|Notch Therapuetics: Honoraria|Pluto Therapeutics: Honoraria|Rheos: Honoraria|Seres: Grant/Research Support|Seres: Honoraria|Seres: IP Licensing|Seres: Stocks/Bonds (Public Company)|Smart Immune: Board Member|Thymofox: Honoraria|Thymofox: Stocks/Bonds (Private Company)|Vor Biopharma: Honoraria|Wolters Kluwer: Royalties Jonathan Peled, MD, PhD, Canaccord Genuity, Inc: Advisor/Consultant|Crestone Inc: Advisor/Consultant|CSL Behring: Advisor/Consultant|DaVolterra: Advisor/Consultant|MaaT Pharma: Advisor/Consultant|Memorial Sloan Kettering Cancer Center: MSKCC Cancer Center Core Grant NCI P30 CA008748|NIH: NHLBI NIH Award K08HL143189|Probiotics Plus Research: Advisor/Consultant|Probiotics Plus Research: Stocks/Bonds (Private Company)|Prodigy Biosciences: Advisor/Consultant|Prodigy Biosciences: Stocks/Bonds (Private Company)|RA Capital: Advisor/Consultant|Seres Therapeutics: Grant/Research Support
Arya et al. (Thu,) studied this question.