Addition of doxycycline to fluoroquinolones for bacterial prophylaxis in autologous stem cell transplantation for multiple myeloma.

Abstract

e18732 Background: Autologous stem cell transplant (ASCT) is standard therapy for patients (pts) with multiple myeloma (MM), yet the risks of such treatment include neutropenic fever (NF) and bacteremia. Fluoroquinolones (FQs) are a mainstay of bacterial prophylaxis (ppx) of NF in ASCT at many transplant centers and reduce the incidence of bacteremia. Nevertheless, ppx regimens for ASCT are not standardized and variability exists even within single institutions. While antibacterial ppx decreases infections, antibiotic use can also cause toxicity, resistance, and microbiome disturbance. The addition of doxycycline (doxy) to FQs was previously linked to reduced NF and bacteremia when utilized for ppx in MM pts at our institution, leading some providers to incorporate doxy into their ppx regimens. However, our prior study introduced multiple confounders, including differing pre-transplant conditioning regimens between ppx groups. We sought to further compare the incidence of NF and bacteremia in MM pts receiving ppx with FQ alone and FQ-doxy during a time when ppx strategies varied but transplant conditioning was consistent between groups. Methods: This is a single institution retrospective review. We analyzed all medical charts in the ASCT database of MM pts treated at our institution between Jan 2016 and Dec 2021. The primary objective was to determine the effect of bacterial ppx on the rate of NF and bacteremia within 30 days of ASCT. Data was also collected on several pt characteristics. Results: Out of 341 pts with MM who underwent ASCT following melphalan (mel) conditioning, 121 received FQ (ciprofloxacin or levofloxacin) ppx and 220 received FQ-doxy ppx. Pt characteristics and outcomes of interest are shown in Table 1. NF developed in 71 (58.7%) and 103 (46.8%) pts in the FQ and FQ-doxy groups, respectively (p = 0.042). The odds of NF in FQ-doxy compared to FQ is 0.62 (95% CI 0.4, 0.97). Bacteremia developed in 7 (5.8%) and 6 (2.7%) pts of the FQ and FQ-doxy groups (p = 0.235). The odds of bacteremia in FQ-doxy compared to FQ is 0.46 (95% CI 0.15, 1.39). Documented bacteremia was infrequent, yet resistance profiles demonstrated FQ-resistant E. coli strains in the 7 total Gram-negative bacteremia cases and 5 extended-spectrum beta-lactamases. Conclusions: Overall, there were fewer cases of NF in the FQ-doxy ppx group than FQ, however there was no significant difference in the cases of bacteremia. This finding does not support the addition of doxy to FQ ppx at our institution and a future randomized controlled trial investigating appropriate ppx for ASCT is warranted. [Table: see text]