ICANS, Corticosteroid Use Identified As Risk Factors for NRM-Associated Infections After Axi-Cel Infusion in R/R NHL

Infections After Axi-Cel Infusion in R/R NHL

| Image Credit: © Mikkel – stock.adobe.com

Among patients with relapsed/refractory non-Hodgkin lymphoma (NHL), a notable percentage of non-relapse mortality (NRM) events were attributed to infections that delayed immune recovery following treatment with axicabtagene ciloleucel (axi-cel; Yescarta), which most commonly occurred within the first 30 days after infusion, according to findings from a retrospective analysis.1

The analysis included 54 patients, 36 (67%) of whom experienced at least 1 infectious episode during follow-up. Data presented at the 51st Annual EBMT Meeting showed that infections were primarily observed within the first year following axi-cel infusion, and the highest severity of infections was seen within the first 90 days. However, the rates of viral infections—predominantly respiratory in nature—increased after the first year. Fungal infections were infrequent and were most frequently observed during the early post-infusion period.

“Infections remain a major cause of NRM in CAR T-cell [therapy] recipients, with the first 30 days post-infusion being the most critical period,” lead study author Ana Benzaquén, MD, and coauthors wrote in a poster of the data. “Immune effector cell–associated neurotoxicity syndrome [ICANS] and corticosteroid use are significant risk factors, likely contributing to delayed immune reconstitution.”

Benzaquén is a member of the Hematology Department at the Hospital Clínico Universitario at the INCLIVA Research Institute in Valencia, Spain.

From days 0 to +30 following axi-cel infusion, 15 patients experienced at least 1 any-grade infection, translating to a cumulative incidence rate of 28% (95% CI, 16%-40%). Infection types include bacterial (fatal septic shock, n = 1; fatal pneumonia, n = 1; critical septic shock and pneumonia, n = 2; severe bacteremia, n = 2; moderate urinary tract infection [UTI], n = 3; moderate acute gastroenteritis, n = 2), viral (critical encephalitis, n = 1; moderate upper respiratory tract infection [URTI], n = 1; moderate influenza, n = 1), and fungal (fatal C. krusei, n = 1; fatal probable invasive fungal infection, n = 1; severe proven rhino-sinus aspergillosis, n = 1). Risk factors for any-grade infection during this period included activated B-cell lymphoma (41% vs 17%; P < .001) and an EASIX LS score of more than 30 (41% vs 15%; P = .03). During this time period, 5 patients experienced at least 1 grade 3 to 5 infection, translating to a cumulative incidence rate of 9% (95% CI, 2%-17%); no risk factors were identified for high-grade infections.

From days +31 to +90, 14 patients experienced at least 1 any-grade infection, translating to a cumulative incidence rate of 29% (95% CI, 16%-40%). Infection types included bacterial (fatal septic shock, n = 1; critical septic shock, n = 1; severe neutropenic fever, n = 1; severe bacteremia, n = 1; moderate acute gastroenteritis, n = 2; moderate otitis media, n = 1; moderate UTI, n = 1), viral (severe URTI, n = 2; moderate URTI, n = 1; mild URTI, n = 2; mild hemorrhagic cystitis, n = 1), and fungal (severe probable aspergillosis, n = 1). Risk factors for any-grade infection during this period included severe ICANS (54% vs 19%; P = .02). During this time period, 3 patients experienced at least 1 grade 3 to 5 infection, translating to a cumulative incidence rate of 6% (95% CI, 1%-13%); no risk factors were identified for high-grade infections.

From days +91 to +365, 12 patients experienced at least 1 any-grade infection, translating to a cumulative incidence rate of 33% (95% CI, 18%-49%). Infection types included bacterial (critical septic shock, n = 1; critical acute gastroenteritis, n = 1; severe pneumonia, n = 1; severe cellulitis, n = 1; severe bacteremia, n = 1; moderate UTI, n = 1; moderate acute gastroenteritis, n = 2; moderate prostatitis, n = 1; and dental infection, n = 2), viral (severe pneumonia, n = 2; moderate URTI, n = 2; moderate cutaneous varicella-zoster virus, n = 1; mild URTI, n = 3), and fungal (severe probable aspergillosis, n = 1). Risk factors for any-grade infections during this period included ICANS (50% vs 7%; P = .01) and corticosteroid use (48% vs 13%; P = .04). During this time period, 4 patients experienced at least 1 grade 3 to 5 infection, translating to a cumulative incidence rate of 11% (95% CI, 1%-22%); severe ICANS was identified as a risk factor for high-grade infections (32% vs 4%; P = .02).

From 1 year to 2 years following infusion, 6 patients experienced at least 1 any-grade infection, translating to a cumulative incidence rate of 43% (95% CI, 15%-72%). Infection types included bacterial (moderate pneumonia, n = 1; severe pneumonia, n = 1; moderate URTI, n = 1), viral (moderate pneumonia, n = 1; critical pneumonia, n = 1; mild URTI, n = 1; severe URTI, n = 1; mild hemorrhagic cystitis, n = 1), and fungal (critical pneumonia, n = 1). Corticosteroid use (67% vs 14%; P = .03) was identified as a risk factor for any-grade infections during this period. Three patients experienced at least 1 grade 3 to 5 infection, translating to a cumulative incidence rate of 19% (95% CI, 1%-39%); no risk factors were identified for high-grade infections.

Summarizing the Study Rationale and Design

Infections are the primary cause of NRM among CAR T-cell therapy recipients. Risk factors for NRM include immune-mediated toxicities like cytokine release syndrome (CRS) and ICANS, the effects of which are exacerbated by immunosuppression associated with prior treatments and underlying disease.

The present study evaluated the type of severity of infectious episodes among adult patients with relapsed/refractory NHL who received axi-cel at Hospital Clinico Universitario de Valencia between December 2019 and April 2024. Study follow-up was discontinued at relapse or death. Asymptomatic cytomegalovirus reactivations from systematic monitoring were excluded from the infections noted in this analysis. Investigators graded infection severity according to established criteria. Prophylaxis included fluconazole, posaconazole, and quinolones based on the presence of immunosuppression or neutropenia. Additionally, treatment with cotrimoxazole and acyclovir was maintained for at least 6 months following axi-cel infusion and was discontinued if patients had CD4-positive T-cell levels above 200/mm3.

Zooming in on Baseline Characteristics and Causes of Death

The median age of patients enrolled onto the study was 58 years (range, 21-76). Most patients were female (52%), and had diffuse large B-cell lymphoma (DLBCL; 85%). Patients had received a median of 2 prior lines of therapy (range, 2-4). Among 50 evaluable patients, 52% had germinal center B-cell DLBCL. Among 46 evaluable patients, 18.5% displayed MYC translocations. In total, 73% of 48 evaluable patients had Ann Arbor stage 3-4 disease prior to lymphodepletion. Among 53 evaluable patients, 47% had an International Prognostic Index score of at least 3.

Among all evaluable patients, 54% had primary refractory disease and 26% had an ECOG performance score of at least 2 prior to lymphodepletion. A total of 93% of patients had received bridging therapy, and 30% of patients achieved a complete response or partial response with bridging therapy. Prior to lymphodepletion, 52% of patients had an HCT-CI score of at least 3, the median EASIX score was 3.03 (range, 0.5-128.2), and 52% of patients had a high HEMATOTOX score. At baseline, 96% and 64.8% of patients had CRS and ICANS, respectively. Additionally, 50% and 57% of patients, respectively, had received tocilizumab or corticosteroids; the median cumulative dose of corticosteroids was 193 mg (range, 8-3720).

At a median follow-up of 479 days (range, 28-1528), 50% of patients were still alive. Causes for the 27 deaths observed on the study included disease progression or relapse (n = 19), infections (9.2%), ICANS (n = 2), and CRS (n = 1).

“These findings emphasize the importance of implementing strategies for reducing the incidence of ICANS and corticosteroid use, and conducting long-term infection monitoring,” the authors concluded. “Further research is needed to optimize infection prevention and management in this vulnerable population.”

Reference

Benzaquén A, Hernani R, Pérez-Bravo M, et al. Infections after CAR T-cell therapy with axicabtagene ciloleucel in patients with refractory or relapsed non-Hodgkin lymphoma. Presented at: EBMT 51st Annual Meeting;