7004 Background: A substantial proportion of patients (pts) with relapsed/refractory (R/R) non-Hodgkin lymphomas (NHL) will not derive a long-term benefit from the existing anti-CD19 chimeric antigen receptor (CAR) T cells. To enhance therapeutic efficacy, we have engineered huCART19-IL18, a 4 th generation 4-1BB anti-CD19 construct, armored with the ability to secrete the pro-inflammatory cytokine, IL-18. Methods: This is a first-in-human trial using huCART19-IL18 for CD19+ B-cell malignancies (NCT04684563). Expedited 3-day manufacturing is utilized to limit T-cell exhaustion. To be eligible for the NHL cohort, pts must be R/R to prior anti-CD19 CAR T cells if indicated by FDA label. Dose levels (DL) between 3x10 6 and 3x10 8 of huCART19-IL18+ cells are administered as a single IV infusion following lymphodepleting chemotherapy. Bridging therapy is optional. Responses are first assessed at 3 months (mo) using Lugano criteria. Results: As of January 20, 2024, 21 pts with CD19+ NHL were infused with huCART19-IL18. Characteristics include median age 64 yrs (47-74), 76% male, 9 (43%) DLBCL, 6 (29%) FL, 3 (14%) MCL, 2 (10%) tFL, 1 (5%) HGBCL. Median number of prior Rx was 7 (4-14) with 20 (95%) pts R/R to prior anti-CD19 CAR T cells. Manufacturing of DL5 (3x10 8 ) was not feasible due to inability to achieve the target dose in 4/6 (67%) pts assigned to DL5. 18 (86%) pts received bridging. 3 pts received DL1 (3x10 6 ), 4 pts DL2 (7x10 6 ), 1 pt non-defined dose (2.8x10 7 ), 6 pts DL3 (3x10 7 ), 5 pts DL4 (7x10 7 ), 2 pts DL5 (3x10 8 ). No study-related deaths occurred in 21 safety-evaluable pts. CRS occurred in 15 (71%) pts: G1 in 8 (38%), G2 in 4 (19%), G3 in 3 (14%). ICANS occurred in 3 (14%) pts: G1 in 2 (10%), G2 in 1 (5%). The most common G3 adverse events at least possibly related to huCART19-IL18 included fatigue (38%), hypotension (29%), and low fibrinogen (23%). 20 pts are efficacy evaluable with median (m) follow-up of 15 mo (3-31). The 3 mo ORR was 80% (90% CI: 60-93%), with CR 50% (90% CI: 30-70%) and PR 30% (90% CI: 14-51%). mDOR was 10 mo (5.5-NR). mPFS was 8.7 mo (90% CI 5-NR), and mOS was NR (90% CI 25 mo-NR). We detected continued persistence of huCART19-IL18 in pts with 24 mo follow-up. No correlation between cell dose and outcome was identified, but response rates and mean expansion (copies/µg gDNA) were higher in pts previously exposed to CD28 CAR than those who had prior 4-1BB CAR (Table). Conclusions: Treatment with huCART19-IL18 has an acceptable safety profile and produced durable remissions in heavily pre-treated pts with R/R NHL despite prior CAR T-cell therapy. The subtype of the preceding CAR product may influence the expansion and effectiveness of huCART19-IL18. Clinical trial information: NCT04684563 . [Table: see text]