2544 Background: CD19 CAR T cell therapy (CAR-19) has been a breakthrough for relapsed/refractory (R/R) B-cell non-Hodgkin lymphoma (B-cell NHL). However, both relapse and toxicity remain major challenges. We utilized novel programming conditions aimed at enhancing the function and persistence of our CAR T products by improving metabolic fitness, via promotion of a hybrid Th1/Th17 phenotype. Additionally, CD34 selection was employed, via CD34 tag, to yield a more purified product and reduce potential toxicity. Methods: This single institution IRB approved phase 1B trial (NCT 05702853) evaluated a novel CD28 co-stimulated CAR-19 product, with metabolic programming and CD34 selection (CD19-CAR-CD34t hybrid T cells). Eligible patients had R/R B-cell NHL and were enrolled at dose level (DL) 1, 2, or 3 (1x10 6 , 1.5x10 6 , 2x10 6 CAR T cells/kg with max dose of 2.0x10 8 ). Dose escalation (Desc) used a model-assisted keyboard design. Dose-limiting toxicity (DLT) evaluation occurred in the first 28 days following infusion per CTCAE v5.0, with CRS and ICANS assessed per ASTCT criteria. The primary objectives were to evaluate safety and determine recommended dose for expansion. Results: Desc completed enrollment with 15 patients across dose levels DL1-3, DL2-3, DL3-9. NHL subtypes included DLBCL (8), mantle cell (4), follicular (2), and marginal zone lymphoma (1). Patients received a median of two prior systemic therapies (range 1-5), and 11 (73%) received bridging therapy. All patients underwent successful leukapheresis, manufacturing, CAR T infusion, and completion of the 28-day DLT evaluation. Common AEs included neutropenia (100%), fatigue (53%), and nausea (47%). One DL3 patient experienced a DLT (grade 4 lung infection). CRS was noted in 9 patients (60%; (grade 1: 7 pts; grade 2: 2 pts), with no grade 3+ events. No ICANS was observed. Six patients (40%) received tocilizumab (toci) and two (13%) received corticosteroids for CRS management with cumulative administration of 8 doses of toci (mean 0.5 doses/patient) and 20 mg of dexamethasone (mean 1.3 mg/patient) for entire cohort. With 11.1-month median follow-up, ORR/CR rates were 73%/67% (1-year PFS 0.67; 95% CI 0.47-0.95). All 10 patients with CR remained relapse-free as of 1/1/26; 2 deaths occurred due to lymphoma progression. The median percentage of CD34 CAR T cells at infusion was 67%. Median total T-cell doses were 165.1 (DL1), 262.8 (DL2), and 236.2 (DL3) x 10 6 . Data on persistence/expansion of Desc cohort are forthcoming. Conclusions: CD19-CAR-CD34t hybrid T cells showed promising efficacy and durability, with no relapses among patients with CR. There were no occurrences of either ICANS or grade 3+ CRS and limited CRS intervention needed. This is a promising toxicity profile for a CD28 co-stimulated CAR-19, likely owing to a more purified product via CD34 selection. DL3 was chosen for dose expansion and currently ongoing. Clinical trial information: NCT05702853 .
Hess et al. (2026) studied this question.