ICT01, an Investigational γ9δ2 T Cell Activator, Added to Azacitidine-Venetoclax Achieves Frequent and Early Complete Remissions in Adults with AML Unfit for Intensive Induction Chemotherapy: Interim Results from the Ongoing Open-Label, Randomized Phase 1 Study Eviction

Background: ICT01 is a first-in-class humanized, Fc-disabled anti-butyrophilin 3A (BTN3A) monoclonal antibody that selectively activates γ9δ2 T cells, leading to both direct cytotoxicity against AML blasts and indirect immune modulation through activation of CD8 T and NK cells to mount an anti-leukemic response. It is known that venetoclax (V) overcomes granzyme B/perforin resistance of AML blasts, which enhances the anti-leukemic efficacy of cytolytic γ9δ2 T, NK and CD8 T cells and that azacidine (A) improves immune-effector-cell recognition of AML blasts. In preclinical studies, we have shown that ICT01-mediated γ9δ2 T-cell activation protected both γ9δ2 T and NK cells against V-induced cell death. In addition, we demonstrated that blast killing capacity of AV significantly increased in the presence of ICT01. In a xenograft mouse model with adoptive γ9δ2 T-cell transfer, ICT01‒AV significantly delayed tumor growth and improved median survival of animals compared to either treatment alone (Wieduwild et al. Blood 2023;142: abstr #1552). Previously, we also reported that ICT01 monotherapy dose escalation up to 75 mg Q3W in R/R AML is well-tolerated without any dose-limiting toxicities (Garciaz et al. Ann Oncol 2023;34,suppl 2: abstr #543), supporting further testing of ICT01‒AV combination in newly diagnosed patients with AML.

Methods: In two dose-optimizing, efficacy-estimating cohorts of this Phase 1 study, adults with newly diagnosed AML ≥ 75 years old or unfit to receive induction chemotherapy were randomized to receive ICT01 at a dose of 10 mg (ICT01^low) or 75 mg (ICT01^high) in combination with standard AV Q4W; sentinel dosing of ICT01^ow (N=3) was followed by sentinel dosing of ICT01^high (N=3), and subsequent patients were randomized 1:1 between both dose levels. Baseline assessments included cytogenetics, NGS, BTN3A expression on AML blasts in bone marrow (BM), and number of γ9δ2 T cells in blood and BM. We evaluated treatment-emergent adverse events (TEAE), anti-leukemic efficacy (per ELN 2022 criteria), target occupancy, and changes in γ9δ2 T cells.

Results: Of 20 patients randomized as of 19-Jul-2024, 16 had at least one post-baseline BM aspirate for efficacy evaluation. Among 20 patients enrolled, median age was 77 years (range 64‒84) with 75% of patients ≥ 75 years with 12% and 62% of patients having ELN intermediate and adverse risk AML, respectively. Median number of BM blasts was 43% (range 9‒95%) with only three patients having < 20% AML blasts at baseline. For 16 patients evaluable for efficacy at the data cut-off, median follow up was 4.0 months (range 0.7-7.4). 30-day mortality was 0%, no Grade 5 drug-related TEAE and no dose-limiting toxicity were reported. All patients had at least one TEAE. Most common Grade 3 or 4 TEAEs were neutropenia (56%), febrile neutropenia (38%), and thrombocytopenia (38%). Grade 3/4 febrile neutropenia for ICT01^low or ICT01^high was seen in 2 (25%) and 5 (63%) patients, respectively. Overall, the CR was 75% and CR/CRi was 94%. Rates of CR and CR/CRi were 75% and 100% for ICT01^low and 75% and 88% for ICT01^high. The proportion of patients with CR/CRi at the end of cycles 1, 2, 3, and 4 were 44%, 81%, 88% and 94%, respectively. Overall, 5 (31%) of 16 patients attained a CR at the end of Cycle 1. Notably, CR and CR/CRi were high in patients with adverse risk disease (mtTP53, CR and CR/CRi 100% each, N=3; mtASXL1, CR 50%, CR/CRi 100%, N=2) and low/intermediate risk disease (mtNPM1, CR and CR/CRi 100% each, N=4; mtIDH1/2, CR 67% and CR/CRi 100%, N=3). Overall, the duration of response was 2.5 months (95% CI, 1.8‒NR) and ongoing in all except patients as of the cut-off date. Median overall survival was immature with at the time of data cut-off. High baseline γ9δ2 T cells showed a good correlation between blood and BM. ICT01^low and ICT01^high resulted in rapid transient and sustained γ9δ2 T cell activation, respectively. Updated results with additional patients will be presented at the meeting.

Conclusions

In this ongoing Phase 1 study, both ICT01^low‒AV and ICT01^high‒AV were safe and very well tolerated and generated very high CR and CR/CRi rates in older/unfit patients newly diagnosed with AML.