Patterns and characteristics of infection episodes in the first year following CD19 CAR T-cell treatment for diffuse large B-cell lymphoma

CAR T-cell therapy represents a novel treatment for the management of several hematological malignancies, including diffuse large B-cell lymphoma. As well as significant toxicities, such as cytokine release syndrome and neurological events (e.g. immune effector cell-associated neurotoxicity syndrome), immune compromise through B-cell aplasia and hypogammaglobulinemia is a recognised sequalae to treatment. While infectious complications have been reported following CART T-cell therapy, most patients have been evaluated within the controlled context of clinical trials. Identifying treatment-associated infections and factors that increase the risk of significant infection in a real-world setting will inform both prophylactic and treatment choices to reduce associated morbidity and mortality in these patients.

Kitsada Wudhikarn and colleagues present real-world data from their cohort study exploring the nature and severity of infection in patients receiving CAR T-cell therapy for diffuse large B-cell lymphoma.¹ This was published in Blood Cancer Journal in October 2020, and here we summarize their findings.

Study design

• A single-centre, retrospective cohort study.
• All patients received axicabtagene ciloleucel or tisagenlecleucel.
• A total of 60 consecutive patients were treated between January 2018 and June 2019.
• All infections were documented from the day of CAR T-cell infusion to whichever timepoint came first out of
  • 1-year post-CAR T-cell treatment;
  • last follow-up; or
  • relapse or progression.

Data collection

Core demographic data were collected, including age, sex, histopathological diagnosis, stage of disease, performance status, prior treatment including previous hematopoetic stem cell therapy, history of infection within 30 days prior to CAR T-cell infusion, and previous lymphodepletion chemotherapy.

Severe infection was classed as infection requiring intravenous antibiotics. Life-threatening infection was defined as the presence of end-organ or cardiovascular compromise.

Results

• Median follow-up was 6 months (range, 0.8–12).
• Median age was 63 years (range, 19.5–85.9)
• The proportion of male patients was 42/60 (70%).
• The number of patients who had received prior hematopoetic stem cell therapy was 16/60 (26.7%).
• A total of 43/60 patients (71.7%) received axicabtagene ciloleucel, and 17/60 patients (28.3%) received tisagenlecleucel.
• In total, 13/60 patients (21.7%) received CAR T-cell treatment after introduction of a new institutional antimicrobial prophylaxis protocol.

After CAR T-cell infusion, the following observations were made:

• Cytokine release syndrome was identified in 48/60 patients (80%), with grade ≥ 3 in 7/60 patients (11.7%).
• Immune effector cell-associated neurotoxicity syndrome was identified in 24/60 patients (40%), with grade ≥ 3 in 13 patients (21.6%).
• A total of 52/60 patients (86.7%) developed neutropenic fever within 30 days.
• There were 101 episodes of infection in 40 patients (60 bacterial, 38 viral, 2 fungal, 1 protozoal):
  • 22/101 (22.8%) classified as severe
  • 1/101 was life-threatening (Escherichia coli biliary sepsis)
  • 1/101 was fatal (Influenza A pneumonia).

Key findings

Within the first 30 days, 37 infection episodes occurred, 25 of which were bacterial (in 20 patients). After day 30, 35 bacterial infections were identified in 16/60 patients, six of whom had infection before day 30.

The main risk factors identified in multivariate analysis for all infections were as follows:

• Impaired performance status (≥ 2 vs 0–1; HR, 1.87; 95% CI, 0.91–3.84; p = 0.09).
• Systemic corticosteroid during CAR T-cell therapy (yes vs no; HR, 2.22; 95% CI, 1.05–4.67; p = 0.03).

The main risk factors identified in multivariate analysis for severe bacterial infections were the following:

• Impaired performance status (≥ 2 vs 0–1; HR, 2.84; 95% CI, 1.0–8.06; p = 0.05).
• Infection prior to CAR T-cell therapy (yes vs no; HR, 3.98; 95% CI, 1.30–12.20; p = 0.01).

Conclusion

CAR T-cell therapy continues to hold promise for patients with large B-cell lymphoma, both treatment naïve and those with refractory disease.

During the period after infusion, bacterial and viral infections are common, with bacterial infections the most predominant, especially during the first 20 days. Antiviral immunity is preserved in most patients, and infectious events of all etiologies are mild in most patients and are treatable with conventional therapies.

This study adds to our understanding of patterns of infection during treatment in this patient group, but does have several limitations: (1) limited power due to the small study size, (2) the retrospective nature of the cohort and associated bias, (3) the fact that certain parameters such as immune function were not measured in a consistent and systematic manner, and (4) that the microbial prophylaxis being routinely used for these patients changed at a midpoint in the study.

Prospective clinical trials of antimicrobial prophylaxis are needed to reduce the incidence of treatment-related infection and associated morbidity to minimize the impact of these events on the quality of life of patients.