Abstract
Acute myeloid leukemia (AML) is sustained by a subpopulation of leukemic stem cells (LSCs) that are resistant to conventional therapies and contribute to disease relapse. Eradicating LSCs remains a major therapeutic challenge. Our group previously demonstrated that CD84 is upregulated in over 70% of AML blasts and is enriched on the surface of LSCs compared to normal hematopoietic stem cells (HSCs) and early progenitors. CD84 supports leukemic cell survival, proliferation, and engraftment by maintaining redox homeostasis. Building on these findings, we developed a CD84-targeting T-cell engager.
To identify a lead CD84-binding monoclonal antibody (mAb) for T-cell engager development, we screened eight candidates for antibody-dependent cellular cytotoxicity (ADCC). ADCC was assessed by co-culturing THP1-GFP⁺ AML cells with healthy donor-derived natural killer (NK) cells (effector:target [E:T] ratio = 10:1) in the presence of each mAb (5 µg/mL) for 4 hours. Target cell lysis was quantified by flow cytometry using 7-AAD staining. Five mAbs were selected for bispecific antibody development.
Among these, BT1, an Fc-null T-cell engager, emerged as the most efficacious candidate. BT1 induced potent T-cell activation and CD84-dependent killing of AML cell lines (THP1, MOLM13, U937) and primary AML blasts (n=3) across various E:T ratios. The IC₅₀ values for BT1 were 6.14 pM (95% CI: 1.65–19 pM) for THP1, 7.18 pM (95% CI: 4.94–10.53 pM) for MOLM13, and 33.69 pM (95% CI: 11.18–90.63 pM) for U937 at a 5:1 E:T ratio using resting T-cells over 48 hours.
To assess antigen specificity, we used K562 cells lacking endogenous CD84 and generated a CD84-overexpressing (CD84^OE) line via lentiviral transduction. In 48-hour co-culture assays with primary human T cells, BT1 induced robust cytotoxicity and T-cell activation—measured by CD69 and CD25 expression—only in K562 CD84^OE cells, not in parental CD84-negative K562 cells (E:T ratio 20:1, p=0.046, n=2), confirming CD84-dependent specificity.
BT1 also abrogated the clonogenic capacity of primary AML blasts, outperforming control IgG and teclistamab, an Fc-null T-cell engager that is not specific for AML but is instead FDA-approved for the treatment of multiple myeloma. In vivo efficacy was evaluated using Luc+ MOLM13 and Luc+ THP1 xenograft models in NSG mice. Mice were intravenously engrafted with Luc+ AML cells (1×10⁶ cells/mouse) and monitored via bioluminescent imaging. BT1 treatment (MOLM13: 5 µg/mouse/week; THP1: 20 µg/mouse/week) co-injected with human T cells (3×10⁶ cells/mouse/week) significantly reduced leukemic burden and extended survival compared to levels in control treatments including teclistamab or vehicle (MOLM13: ctrl 34 days vs. BT1 77 days, p=0.0026; THP1: ctrl 30 days vs. BT1 68 days, p=0.012).
BT1 is a potent and selective CD84-targeting T-cell engager that effectively eliminates AML cells and impairs LSC function. Its specificity, robust activity in primary patient samples, and significant survival benefit in xenograft models support its further preclinical development and potential clinical translation as a novel immunotherapeutic for AML.
