Background: Cytokine release syndrome (CRS) is a frequent and potentially life-threatening complication of CAR-T cell therapy. While clinical trials report rates and severities, limited real-world data exists. We performed this study to identify potential risk factors and describe the hospitalization burden and mortality in a nationally representative population.

Methods: We conducted a retrospective analysis using the 2021–2022 National Inpatient Sample (NIS) database to identify a cohort of hospitalizations involving CAR-T cell therapy for hematologic malignancies, based on relevant ICD-10-PCS procedure codes. Eligible cases included patients diagnosed with diffuse large B-cell lymphoma (DLBCL), acute lymphoblastic leukemia (ALL), multiple myeloma (MM), mantle cell lymphoma (MCL), or follicular lymphoma (FL).

The primary outcome was CRS, identified using ICD-10-CM code D89.83. CRS severity was categorized as grade 1–4 based on secondary coding, with grades ≥3 defined as severe. Secondary outcomes included neurotoxicity (ICANS), in-hospital complications (e.g., sepsis, AKI, stroke), length of stay (LOS), total charges, and in-hospital mortality. We compared characteristics and outcomes between CRS and non-CRS groups using chi-square and t-tests.

Multivariate logistic regression was performed to assess the association of CRS with malignancy type, age, sex, race/ethnicity, primary payer, income quartile by ZIP code, Charlson Comorbidity Index (CCI), and hospital teaching status. A separate model among CRS patients evaluated predictors of severe CRS (grade ≥3). Significance was set at p < 0.05.

Results: We identified 1,791 unweighted (8,960 weighted) hospitalizations with hematological malignancies who received CAR-T cell therapy between 2021 and 2022. The median age was 64 years (IQR: 54–71), with 41% aged ≥ 65 years. Of these, 896 (64.1%) were male. The racial distribution included 963 (70.7%) White, 190 (13.9%) Hispanic, 100 (7.3%) Black, and 59 (4.3%) Asian/Pacific Islander (API) patients. The most common malignancies were DLBCL (635, 45.4%), followed by MM (346, 24.8%), ALL (155, 11.1%), FL (140, 10.0%), and MCL (122, 8.7%).

CRS occurred in 62.2% of patients, with 55.0% experiencing grade 1 and 34.2% grade 2 events. Only 7.5% developed severe CRS (grade ≥3). In unadjusted analysis, CRS was more frequent in MCL (68.9%, p=0.028), Hispanic (67.5%), and API patients (76.3%) compared to White patients (61.9%, p=0.044). Adjusted analysis confirmed higher odds of CRS among Hispanic (aOR 1.51, 95% CI 1.05–2.19) and API patients (aOR 2.01, 95% CI 1.07–3.79). Higher comorbidity burden (CCI ≥3) was associated with lower odds of CRS (aOR 0.63, 95% CI 0.44–0.88). No associations were found for sex, age group, insurance, or hospital teaching status.

Although severe CRS was less common among patients with higher CCI scores in unadjusted analysis (p=0.042), multivariable analysis showed that CCI 1–2 (vs. 0) was associated with higher odds of severe CRS (aOR 2.28, 95% CI 1.25–4.17).

Patients with CRS had significantly higher rates of ICANS (79.7% vs. 20.3%, p<0.001) and neutropenic fever (76.6% vs. 23.4%, p<0.001). In-hospital mortality was 58.54% among CRS patients and 41.46% among non-CRS patients, but this difference was not statistically significant (p = 0.627). No significant differences were observed in other complications, including respiratory failure, use of mechanical ventilation, sepsis, AKI, DIC, pneumonia, thrombotic events, or cardiac complications. CRS was significantly associated with prolonged hospitalization (p < 0.001), with higher proportions of patients staying beyond 10 days. Total hospital charges did not differ significantly between groups (p = 0.10).

Conclusion: Among patients receiving CAR-T cell therapy for hematologic malignancies, 62% developed CRS. Hispanic and API patients were more likely to develop CRS, while underlying malignancy type was not independently associated. CRS did not impact in-hospital mortality but was significantly associated with higher rates of ICANS, neutropenic fever, and prolonged hospitalization. Notably, a higher comorbidity burden (CCI ≥3) was associated with reduced odds of CRS. These real-world rates of CRS are broadly consistent with those reported in clinical trials, although population differences and coding limitations may account for some variation. These data add to our knowledge of this condition and may help guide future management. More research is needed.

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