Chimeric Antigen Receptor (CAR) T-cell therapy has transformed the treatment of Relapsed/Refractory (R/R) B-cell malignancies. However, post-infusion cytopenias are common, often requiring significant transfusion support. Data on transfusion burden, predictors, and implications for transfusion services remain scarce.

To assess the frequency, timing, and predictors of blood product transfusions in CAR-T cell recipients and explore potential associations with clinical outcomes.

We conducted a sub-analysis of a prospective, single-center Phase 1 trial evaluating autologous CD19-directed CAR-T cell therapy for R/R B-cell malignancies. CAR-T cells were manufactured onsite under GMP-compliant conditions using the CliniMACS Prodigy system (Miltenyi Biotec). Patient-derived T-cells were transduced with a lentiviral vector encoding an anti-CD19 scFv, a CD8 hinge, the transmembrane domain of TNFRSF19, and intracellular signaling domains from 4-1BB and CD3ζ. Standard 3+3 dose-escalation design was employed, with target CAR+ cell doses of 0.5‒2×106 cells/kg, administered following lymphodepletion. Data on baseline characteristics, inflammatory biomarkers, toxicity (CRS, ICANS, HLH), mEASIX, CAR- and ALL- Hematotox scores, and transfusion requirements were collected. Outcomes included number of Red Blood Cell (RBC) and Platelet (PLT) units transfused, prolonged transfusion dependence (>30-days), N-ICAHT and T-ICAHT (neutrophil and Thrombocyte Immune effector Cell-Associated Hematotoxicity) grades, and correlations with predictive scores. Comparisons were performed using Mann-Whitney test, and correlations with Spearman’s rho.

Eleven patients were included (72.7% male). Median age was 42-years old (9‒69). Diseases included ALL (36.3%), CLL (18.2%) and NHL (45.5%). Cytopenias were universal: N-ICAHT was mostly grade 2 and T-ICAHT mostly grade 3. Two patients presented severe cytopenias after 30-days of infusion, requiring prolonged G-CSF and thrombopoietin receptor agonists use. The median number of transfused units was 3 for RBC and 6 for PLT. Overall, 90.9% of patients required RBC and PLT transfusions, and 54.5% required transfusions beyond day +30. High CAR-HEMATOTOX (≥ 2) was not associated with increased transfusion burden, while high ALL-Hematotox (≥ 4) trended toward higher needs (rho = 0.949, p = 0.051). T-ICAHT grades correlated strongly with RBC (rho = 0.880, p < 0.001) and PLT transfusions (rho = 0.919, p < 0.001), as expected. EASIX showed moderate correlation with RBC needs (rho = 0.54, p = 0.089). No significant associations were found with CRS or ICANS severity, likely due to limited sample size. One patient received granulocyte transfusions due to infection and severe neutropenia.

Transfusion support is common and clinically relevant after CAR T-cell therapy, particularly in the first 30-days, with over half of patients remaining dependent beyond this period. Early identification of high-risk patients can guide supportive care and optimize resource use. Occasional granulocyte transfusions add logistical challenges for blood banks. These findings underscore the need for proactive coordination with transfusion services to ensure timely access to compatible components, especially for patients with alloantibodies or rare blood types. Clear protocols for inventory management and rapid response are essential to avoid delays and ensure safety.