Impact of Broad-Spectrum Antibiotic Exposure on Outcomes Following Autologous Stem Cell Transplantation for Patients with Relapsed Diffuse Large B-cell Lymphoma or Hodgkin Lymphoma

Autologous hematopoietic stem cell transplantation (ASCT) remains a curative option for relapsed diffuse large B-cell lymphoma (DLBCL) and Hodgkin Lymphoma (HL), though relapse occurs in up to 50%. The gut microbiome influences immunity and treatment efficacy, and antibiotic-induced microbial loss has been linked to inferior cancer outcomes. Broad-spectrum antibiotics (BSA) and anaerobic antibiotics (AA) are frequently used during ASCT but may impair outcomes through microbiome disruption, an effect not well studied in lymphoma. 1. Evaluate the association between BSA exposure and ASCT outcomes in lymphoma. 2. Interpret how confounding factors, such as age, may influence overall survival after ASCT. 3. Discuss the potential implications of antibiotic stewardship on optimizing long-term outcomes and infection management in ASCT recipients. We retrospectively analyzed adults with DLBCL or HL undergoing first ASCT at our institution (2010-2022). Antibiotic exposure was defined >1 day of IV BSA (cefepime, piperacillin-tazobactam, cephalosporins, or monobactams) or vancomycin; BSA were stratified by anaerobic coverage. Primary outcomes were overall survival (OS) and cumulative incidence of relapse. OS was assessed with Kaplan-Meier and Cox models. Progression free survival (PFS) used competing risk analysis. Among 296 patients: 183 (62%) had DLBCL and 113 (38%) HL; 237 (80%) received BSA and 74 (25%) AA, with median BSA exposure of 6 days. In multivariate analysis of relapse, DLBCL had higher risk than HL (HR 1.7, p = 0.046), while BSA exposure was not significant. In univariate analysis, BSA and AA exposure were associated with worse OS (5 year 68% vs 75%, p = 0.04; 61% vs 72%, p = 0.0004; figure 1). OS also varied by diagnosis (DLBCL 60% vs HL 85%, p < 0.001), and age quartile by years (17-36, 94%; 37-53, 67%; 54-61, 67%, 62-79, 49%, p < .001). On multivariate regression accounting for AA, age, relapse, and disease, only age and relapse remained significant (table 1), suggesting age was confounding the initial observation of AA exposure on OS and older patients may be more likely to require antibiotics. Early antibiotic exposure correlated with decreased OS in univariate analysis, but the effect of AA exposure on OS appeared to be confounded by age, where age at time of transplant was the major driver of OS. Although the microbiome is influenced by antibiotics, other factors can drive microbiota changes which may be more important than antibiotic exposures alone. Prospective studies should clarify whether antibiotic stewardship can improve ASCT outcomes without compromising infection control.