Efficacy and Safety of TAK-007, Cord Blood-Derived CD19 CAR-NK Cells, in Adult Patients with Relapsed/Refractory (R/R) B-Cell Non-Hodgkin Lymphoma (NHL)

Autologous chimeric antigen receptor (CAR)-T cell therapies have provided significant benefit to patients (pts) with R/R B-cell NHL, but options remain limited for pts who are not eligible for autologous CAR-Ts or high-dose chemotherapy followed by autologous stem cell transplantation. Logistical limitations and safety concerns of autologous CAR-Ts may be overcome by allogeneic products with a more favorable safety profile. TAK-007 is an off-the-shelf, allogeneic, cryopreserved, umbilical cord blood-derived, CD19-targeting CAR natural killer (NK) cell therapy product with a dual mode of action including CD19 CAR-mediated and innate NK receptor-mediated (CAR-independent) killing of malignant B-cells. An open-label, multi-center phase 2 trial was conducted to evaluate the safety and preliminary efficacy of TAK-007 in adult pts with R/R large B cell lymphoma (LBCL) or indolent NHL (iNHL) (NCT05020015).

The study included a dose escalation phase evaluating 2 dose levels (200 x 10⁶ [200M] and 800 x 10⁶ [800M] CD19 CAR+ viable NK cells per pt), and an expansion phase at the dose level selected from dose escalation. Treatment included 3 consecutive days of IV lymphodepleting chemotherapy (LDC) with 300 mg/m² cyclophosphamide and 30 mg/m² fludarabine, followed by a single dose of TAK-007 >2 days later. Treatment was administered in a human leukocyte antigen (HLA)-agnostic manner. The primary endpoint was incidence of treatment-emergent adverse events (TEAEs). Response was assessed per Lugano 2014 criteria.

At data cut-off (Feb 23, 2024), 27 pts (LBCL, n=18; iNHL, n=9) had been enrolled into dose escalation (n=9) and expansion (n=18) at 9 sites across the US. Of these, 26 pts received TAK-007, with 1 LBCL pt in expansion withdrawn prior to TAK-007 administration due to adverse events. Pts had received a median of 5 (range, 2–11) prior lines of systemic therapy; 10 (59%) LBCL pts and 1 (11%) iNHL pt had received prior anti-CD19 CAR-Ts. At screening, 17 (65%) pts had grade ≥2 cytopenia. Median time from enrolment to LDC was 8 days. Half of the pts received TAK-007 in the outpatient setting. All 26 pts who received TAK-007 experienced ≥1 TEAE. Among non-hematological (heme) TEAE events, 77% were grade 1–2; 18 (69%) pts had grade ≥3 non-heme TEAEs. The most common grade ≥3 TEAEs (heme + non-heme) included neutropenia (73%), leukopenia (50%), anemia (38%), thrombocytopenia (38%), sepsis (27%), and febrile neutropenia (12%). No dose-limiting toxicities were observed. Serious TEAEs were reported in 20 (77%) pts but were not attributable to LDC or TAK-007 in 14 (54%). One pt had a grade 1 infusion-related reaction after TAK-007 administration. Cytokine release syndrome (CRS) was observed within 10 days of TAK-007 administration in 3 pts (grade 1, n=2; grade 2, n=1). No ICANS was reported within 60 days from TAK-007 administration. No significant increases in the levels of pro-inflammatory cytokines were observed. While no responses were seen in the 3 LBCL pts who received 200M TAK-007, 7 of 14 (50%) LBCL pts who received 800M TAK-007 had a response, including 3 (21%) complete responses (CRs). Among 8 pts who previously received CD19 CAR-Ts and received TAK-007, 2 (25%) had a response including 1 (13%) CR. Among 9 iNHL pts, 7 (78%) had a response including 5 (56%) CRs. Median duration of response was 3.4 months among LBCL pts and 4.7 months among iNHL pts. Clinical responses were seen across a range of CD19 antigen densities on pre-treatment biopsies. Cellular kinetics (average time course) exhibited multiphasic disposition, representing an initial cellular contraction followed by cellular expansion. Peak levels were observed on Days 7–10 post-TAK-007 administration, which were followed by a terminal elimination phase with decreasing levels on Days 10–28. The exposure-response analysis indicated a positive correlation between higher exposures (Cmax) and response (CR and overall response rate). Treatment-emergent humoral immunogenicity, specific to donor HLA or the extracellular domain of CAR molecules on TAK-007, was not reported.

TAK-007, an allogeneic off-the-shelf CD19 CAR-NK cell therapy, demonstrated early efficacy at 800M and a favorable safety profile in a heavily pretreated pt population, and may be a safer and simpler treatment option with a shorter turnaround time compared with autologous CD19-targeting CAR-T cells. The use of >1 dose of TAK-007 at 800M may further increase efficacy and durability of response.