Introduction:

Autologous hematopoietic stem cell transplantation (ASCT) has curative potential for relapsed diffuse large B-cell lymphoma (DLBCL) and Hodgkin Lymphoma (HL). Unfortunately, relapse rates post ASCT remain up to 50%. While chimeric antigen receptor T-cell therapy is emerging as a treatment option in DLBCL, ASCT remains a key option, particularly in patients responsive to salvage chemotherapy. The gut microbiome is emerging as an important modulator of host immunity with implications on both oncogenesis and treatment efficacy.Reduced microbial diversity, often resulting from antibiotic exposure, has been associated with inferior outcomes across cancer types. In ASCT, broad-spectrum antibiotics (BSA) and anaerobic antibiotics (AA) are frequently used but may negatively impact outcomes through microbiome disruption. While some studies suggest a link between antibiotics and poor outcomes in lymphoma, their effect specifically in the ASCT setting has not been well studied.

Methods:

We retrospectively analyzed adult patients with DLBCL or HL who underwent first ASCT at the University of Nebraska Medical Center between 2010-2022. Antibiotic exposure in the peri-transplant period was recorded, defined as receipt of greater than one day of: broad-spectrum intravenous antibiotics (cefepime, piperacillin-tazobactam, cephalosporins, or monobactams) or vancomycin. The BSA group was separately analyzed according to specific anaerobic coverage (metronidazole, carbapenems, piperacillin-tazobactam, or clindamycin) or not. The primary outcomes were overall survival (OS), defined as time from transplant to death from any cause, and cumulative incidence of relapse defined as time from transplant to relapse, with death from other causes analyzed as a competing event. Baseline characteristics were compared using chi-squared tests for categorical variables and non-parametric tests for continuous variables. OS was evaluated using Kaplan-Meier curves and then analyzed with Cox proportional hazards models. Variables with univariate p-values <0.20 were considered for multivariable modeling using backward selection or manual stepwise removal. PFS was analyzed using cumulative incidence models, treating relapse as the event of interest and death without relapse as a competing risk.

Results:

We included 296 patients in the study: 183 (62%) DLBCL and 113 (38%) HL. Two-hundred thirty-seven (80%) received BSA and 74 (25%) received AA. Median exposure to BSA was six days (range 2-18 days). Multivariate analysis of the cumulative incidence of relapse demonstrated an increased risk for DLBCL vs HL (HR 1.7, p = 0.046) but not for BSA exposure (HR 1.4, p = 0.35), race (white vs not-white, HR 1.7, p = 0.47), or gender (male vs female, HR 1.5, p = 0.10). In univariate analysis, BSA and AA exposure were negatively associated with OS with a 5 year OS rate of 68% vs 75% for those not receiving BSA (p = 0.04) and AA exposure (61% vs 72%, p = 0.004), respectively. Cancer type (DLBCL vs HL 60% vs 85%, p < 0.001), and age by quartile (17-36, 37-53, 54-61, 62-79 years) with 5 year survival rates of 94%, 67%, 67%, and 49%, p < 0.001 were also significantly associated with OS. Multivariate analysis incorporating these variables identified only age as significantly associated with OS (37-53 vs 17-36 years, HR 7.57 p < 0.001, 62-79 vs 17-36 years, HR 11.2 p < 0.001). Among patients with DLBCL, the average age was 69 for those who received BSA and 65 those who did not. Among patients with HL, the average age was 51 for those who received BSA and 42 for those who did not.

Conclusion:

In this retrospective cohort study of ASCT recipients, early exposure to BSA and AA were associated with decreased OS in univariate analysis, however in multivariate analysis only age remained significant. These findings suggest that while antibiotic exposure may correlate with outcomes, its effect is likely mediated by other clinical factors such as age, or may reflect underlying illness severity, rather than a direct causal relationship. Prospective studies are needed to clarify whether modification of antibiotic exposure can meaningfully improve long-term transplant outcomes without compromising infectious risk management.

This content is only available as a PDF.
2025
Sign in via your Institution