Infectious complications of car T-cell therapy: A longitudinal risk model.

BACKGROUND: CAR T-cell therapy, where a patient's own T cells are re-engineered to express a receptor to a target of interest, is becoming an increasingly utilized cancer-directed therapy. There are significant toxicities that contribute to a novel state of immunocompromise, leading to new patterns of infectious complications that require further detailed study. METHODS: We created a single-center cohort of adult recipients of CD19-directed CAR T-cell therapy and assessed infectious outcomes, supportive care received, toxicities, and markers of immune function up to 2 years following CAR T-cell therapy. Descriptive statistics were used as appropriate for analysis. We additionally conducted time-to-event analysis assessing time-to-first infection with either log-rank testing or Cox regression with univariate analysis, before including significant predictors into a multivariate Cox model of time to infection. RESULTS: We identified 73 patients who received CD19-directed CAR T-cell therapy who predominantly had diffuse large B-cell lymphoma. Within 30 days of cell infusion, bacterial and Candida infections were the most common, with 64% of infections due to these organisms. Between 30 days and 2 years postinfusion, respiratory viruses and pneumonia were the most frequent infections, with 68% of infections due to these etiologies. Receipt of tocilizumab, development of immune effector cell-associated neurotoxicity syndrome (ICANS), or lower neutrophil count were associated with quicker onset of infection in a multivariate Cox model. CONCLUSIONS: Respiratory viruses remain an important infectious complication of CAR T-cell therapy following the first year. The model may be a useful tool to identify patients at the highest risk of infection.

The effect of broader, directed antimicrobial prophylaxis including fungal coverage on perioperative infectious complications after radical cystectomy.

OBJECTIVES: Radical cystectomy (RC) with urinary diversion has a significant risk of infection. In an effort to decrease the rate of infectious complications, we instituted a broader, culture-based preoperative antimicrobial regimen, including fungal coverage, and studied its effect on infectious complications after RC. MATERIALS AND METHODS: In May 2013, antimicrobial prophylaxis for RC was changed at our institution after review of previous positive cultures. Ampicillin-sulbactam 3g, gentamicin 4mg/kg, and fluconazole 400mg replaced cefoxitin. Patients undergoing RC from May 2011 to May 2014 were included. Before and after implementation of the new regimen, 30-day infectious complications (positive blood culture, urinary tract infection, wound infection, abscess, and pneumonia) and adverse events (Clostridium difficile, readmission, and mortality) were compared. Multivariate logistic regression was used to identify independent risk factors for infection while controlling for covariates. RESULTS: In total, 386 patients were studied (258 before the change and 128 after). The overall infection rate decreased with the new regimen (41% vs. 30%, P = 0.043) with improvements in wound (14% vs. 6%, P = 0.025) and fungal (10% vs. 3%, P = 0.021) infections. Median length of stay decreased from 8 (interquartile range [IQR]: 7-12) to 7 (IQR: 7-10) days (P = 0.008). On multivariate analysis, the new regimen decreased the risk of infections (odds ratio [OR] = 0.58, 95% CI [0.35-0.99], P = 0.044) whereas body mass index, operating room time, smoking, and total parenteral nutrition increased the risk (all P< 0.05). CONCLUSIONS: Risk factors for infection after RC include body mass index, operating room time, smoking, and total parenteral nutrition use. Changing from cefoxitin to broader, culture-directed antimicrobial prophylaxis, based on institutional data to include antifungal coverage, decreased postoperative infections.