Targeting CD99 in acute myeloid leukemia with the novel antibody DIA 3.2-3: dissecting receptor dynamics and functional implications

Acute myeloid leukemia (AML) is characterized by profound genetic heterogeneity and poor prognosis, necessitating the identification of novel therapeutic targets. The transmembrane protein CD99 is highly expressed on leukemic blasts and leukemic stem cells (LSCs), representing a promising candidate for targeted strategies. This study provides a comprehensive preclinical characterization of DIA 3.2-3, a novel fully-human IgG4 anti-CD99 monoclonal antibody, elucidating its selectivity and mechanism of action. Flow cytometry analysis on healthy donors highlighted the distinct binding properties of DIA 3.2-3 compared to the reference clone 3B2/TA8. While 3B2/TA8 exhibits broad reactivity towards both lymphocytes and monocytes, DIA 3.2-3 shows strict specificity for the monocytic lineage within the hematopoietic compartment. Crucially, ex-vivo profiling of circulating blasts from AML patients at first diagnosis demonstrated that the antibody significantly targets the leukemic bulk, with a preferential recognition of both the CD34+CD38+ progenitor fraction and the CD34-CD38+ population. To elucidate the internalization route, we utilized DIA 3.2-3 conjugated with AlexaFluor488 and cathepsin-dependent probes LysoLight. Confocal microscopy and flow cytometry revealed rapid uptake and lysosomal sorting via a non-canonical route, strictly dependent on cholesterol-rich lipid rafts and amiloride-sensitive pathways, pointing to a mechanism associated with macropinocytosis. To dissect the molecular impact of this perturbation, we performed untargeted surface and global proteomics on AML models with distinct genetic backgrounds: HL-60 (FLT3 wild-type) and MV4-11 (FLT3-ITD mutated). Surfaceome analysis highlighted a massive, antibody-driven downregulation of membrane proteins in both models. A notable exception was the Protein kinase A regulatory subunit (PKA-RIIα), which was uniquely enriched at the membrane of MV4-11. This accumulation, contrasting with the kinetics observed in total lysate, suggests a specific subcellular imbalance of this regulatory subunit. Global proteomics revealed a divergent response. While HL-60 engaged metabolic adaptation, MV4-11 exhibited a signature of ribosomal stress and signaling impairment. Specifically, DIA 3.2-3 induced a blockade of the NF-κB pathway in FLT3-ITD mutated cells, driven by the failure to degrade the inhibitor IκBα. Finally, combinatorial treatments with Bafilomycin A1 and Rapamycin identified a critical vulnerability in a FLT3-ITD mutated cell line. Despite correct ER stress sensing, DIA 3.2-3 induces the failure to respond with a proportional upregulation of the effector chaperone BiP/GRP78. This functional uncoupling strongly suggests a defect in the unfolded protein response (UPR). In conclusion, DIA 3.2-3 induces a specific translational impairment that prevents the resolution of proteotoxic stress, unveiling a selective vulnerability that could be exploited to target high-risk FLT3-ITD leukemic cells.