Comorbidities Associated with Early Mortality after CD19 CAR-T Cell Therapy

Objective: CD19 chimeric antigen receptor (CAR) T-cell therapy has changed the standard for treatment of relapsed and/or refractory (r/r) large b-cell lymphoma (LBCL). Currently, pre-treatment evaluation of patients and disease factors do not accurately predict outcomes. This study aims to identify risk factors associated with early mortality prior to day 100 after CAR-T infusion.

Methods: We performed a single-centered retrospective adult patients (pts) with r/r LBCL who received CAR-T cell therapy between 2018-2021. CAR-T cell toxicities graded according to either CAR T-Cell Therapy–Associated Toxicity (CARTOX) grading system prior to May, 2019 or American Society of Transplantation and Cellular Therapy (ASTCT) consensus grading thereafter for cytokine release syndrome (CRS) and immune cell associated neurotoxicity syndrome (ICANS). Comorbidities defined as per Hematopoietic Cell Transplantation-specific Comorbidity Index (HCT-CI). (Sorror et al., 2009). Hepatic dysfunction defined as infection with hepatitis B or C or history of liver cirrhosis or elevated total bilirubin, AST, or ALT on at least two values on two different days from day -24 till day -10 prior to CAR-T. Cardiovascular (CV) comorbidity determined by presence of coronary heart disease, congestive heart failure or reduced ejection fraction detected by echocardiogram. Psychiatric disturbance defined as presence of depression and/or anxiety requiring treatment or consultation. Patient demographics, disease characteristics and clinical outcomes summarized through descriptive statistics. Wilcoxon rank sum test used to evaluate difference in a continuous variable between patient groups. Kaplan-Meier method used to estimate time-to-event endpoints including progression free survival (PFS), and overall survival (OS).

Results: Two hundred and fifty patients included in analysis. Median age of 61years (range: 18-89) and 30.8% of patients were female (n=77). Histology was categorized as LBCL (65.9%) and included diffuse LBCL or high grade LBCL, transformed follicular lymphoma (26.9%), primary mediastinal b-cell lymphoma (6.4%) or follicular lymphoma (0.8%). Majority of pts received axicabtagene ciloleucel 95.6% (n=239) while 4.4% (n=11) received tisagenlecleucel and 52% (n=130) of pts received bridging therapy. CRS of any grade was 90.4% ; ICANS any grade was 57.2% while grade 3 CRS/ICANS occurred 6.4% and 38.4% respectively. Median follow-up 24.7 months (mo) (95% CI: 20.7 ~ 26.4 mo). One-hundred twenty-eight pts died during follow-up, 48 pts died within 100 days, and most common cause of death was progression of lymphoma (29.2%; n=73). Early 30-day mortality occurred in 4.8% pts and causes of death due to CAR-T related toxicity in 6 pts, disease progression in 3 pts, and infection in 1 pts. 100-day mortality rate was 24.6% (n=32) for pts >60 years old vs 13.3% (n=16) for <60 (p-value=0.02). 100-day mortality rate was 36.5% for ECOG 2-4 vs 14.9% for ECOG 0-1 (p-value=0.0005). In terms of non-relapsed mortality (NRM), risk stratification according to cumulative HCT-CI score was not associated with higher 100-day NRM (p=0.24). Pts with cardiac score of 1 had a higher NRM (26.9% vs. 8.1%; p-value=0.005). The 100-day NRM rates for hepatic score of 0, 1, and 3 were 8.9%, 14.3%, and 50% (p-value=0.04). Pts with psychiatric disturbance score of 1 had higher 100-day NRM (19.6% vs. 7.8%; p-value=0.03).

Conclusion: This single center retrospective study observed an increase in 100-day mortality in correlation with pre-treatment performance status, and older age. Pts with cardiac comorbidity, hepatic dysfunction or psychiatric disturbance had a higher 100-day NRM. Similar to other studies, the combined HCT-CI score was not predictive of NRM, but individual variables were associated with higher non-CART related toxicity. These findings require validation with further studies.