Introduction

The rate of carbapenem-resistant Enterobacteriaceae (CRE) detection in hematology-oncology units has shown a significant upward trend, but the relevant data for immunocompromised CAR T cell recipients is still scarce. Furthermore,emerging evidence is illuminating correlations between the gut microbiome and CAR T cell Therapy. Here we report the CRE intestinal colonization, subsequent infection, and related mortality in CAR T cell recipients, with its impact on efficacy and toxicity assessed.

Methods

This study enrolled lymphoma patients receiving CAR T cell infusion with CRE surveillance. Exclusions: (i) CRE-negative patients with inadequate surveillance (< 3 screenings); (ii) absence of CRE screening during hospitalization for CAR T cell infusion; (iii) pre-infusion CRE clearance with sustained negativity; (iv) new CRE colonization post-discharge.Stratification by CRE colonization defined the exposure (CRE+) and control cohorts (CRE-). The severity of inflammation and response were compared using the Chi-square, Fisher's exact, and Mann-Whitney U tests. Long-term outcomeswere estimated by the Kaplan-Meier method with log-rank comparison, while Cox proportional hazards models were employed to adjust for potential confounders.

Results

Following exclusions, 200 patients were enrolled, with a total of 1,097 anal swab screenings performed. In the exposure cohort (n=57), carbapenem-resistant Klebsiella pneumoniae (CRKP) colonization occurred in 38.60% (22/57) of patients, while 31.58% (18/57) had carbapenem-resistant Escherichia coli (CRECO) colonization.36.84% (21/57) of patients had CRE detected within 48 hours of admission.Compared with these community-acquired cases,hospital-onset colonization was more frequent across all isolates. CRECO demonstrated higher nosocomial colonization rates than CRKP (72.22% vs 59.09%). During follow-up, 64.91% (37/57) of patients achieved CRE clearance,but 56.76% (21/37) had recurrence. The decolonization rate (≥ 3 consecutive negative screenings) was observed in 15.79%. CRE infection occurred in 12.28% (7/57) of the exposure cohort versus 0.70% (1/143) in the controls. CRE-attributable mortality was 3.51% (2/57) exclusively in the colonized cohort, with none in controls.

No significant differences were observed between exposure and control groups in baseline/peak levels and time-to-peak of inflammatory markers (IL-6, ferritin, CRP), nor in cytokine release syndrome (CRS)/immune effector cell-associated neurotoxicity syndrome (ICANS) grade distribution or onset timing (P > 0.05).Similarly, peak levels and time-to-peak for lentiviral copy numbers, and efficacy at 30 days after CAR T cell infusion did not differ statistically significantly (P > 0.05). Regarding long-term outcomes, the colonized cohort demonstrated significantly higher 1-year cumulative mortality (33.33% vs 20.98%, P = 0.016) and shorter median PFS (125 days vs 428 days, P < 0.001). On univariate analysis, CRE colonization was associated with reduced OS (HR 1.994, P = 0.019) and PFS (HR 2.543, P < 0.001). After adjusting for confounders, including age, performance status (ECOG score), LDH, tumor size, prior lines of therapy, and sequential hematopoietic stem cell transplantation (HSCT),multivariable Cox regression showed persistent association between CRE colonization and PFS (HR 1.988, P=0.013), but not OS (HR 1.379, P=0.325).

Conclusions

In lymphoma patients receiving CAR T cell therapy, CRE intestinal colonization increased infection risk and attributable mortality, and independently predicted progression risk.

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