CD25 Enables Oncogenic BCR-Signaling and Represents a Therapeutic Target in B Cell Malignancies

Background and hypothesis: B cells critically depend on continuous survival and proliferation signals from a functional B cell receptor (BCR). In >50% of B cell malignancies, the tumor clone is driven by an oncogenic BCR-mimic. Oncogenic mimics of BCR-signaling include BCR-ABL1 ( Ph⁺ ALL) and other ABL1-fusion genes ( Ph -like ALL), viral oncoproteins (e.g. EBV; KSHV), BRAF- (hairy cell leukemia) and NF-κB-pathway lesions (Hodgkin's lymphoma, primary mediastinal B cell lymphoma). B cell tumors driven by oncogenic BCR-signaling represent a highly heterogeneous group of various cellular origins, genetic lesions and divergent clinical outcomes. Despite this heterogeneity, our correlative analyses showed that these B cell malignancies are uniformly characterized by high expression levels of CD25.

Features of CD25 function in B cell malignancies: We found out that adults with Ph⁺ ALL (ECOG; MDACC) and children with Ph -like ALL (P9906) with high CD25 expression at the time of diagnosis have a particularly poor outcome (n=416; P =0.005). While CD25 mediates IL2 signaling on T cells and is highly expressed on regulatory T cells (Tregs), CD25 on B-lineage cells did not pair with IL2Rβ and was not responsive to IL2. Our genetic studies showed that BCR signaling rapidly induces CD25 cell surface translocation by PKCδ-dependent phosphorylation of the CD25 cytoplasmic tail (S268), which is further supported by transcriptional upregulation of CD25 by activation of NF-κB. Antibody mediated stimulation of surface exposed CD25 in response to BCR signaling shuttles inhibitory phosphatases (e.g. SHIP1) from the cytoplasm to the cell membrane. CD25-mediated membrane recruitment of inhibitory phosphatases limits kinase signaling (e.g. SYK, SRC and ERK) downstream of the BCR or its oncogenic mimics. Consistent with a role as negative feedback regulator, our genetic studies showed that CD25 deficient pre-B cells ( Il2ra^-/- ) have autonomous pre-BCR signals, resulting in uncoordinated Ca²⁺ oscillations of variable duration. Genetic ablation of CD25 subverted the ability of B cell leukemia cells to balance oncogenic signal strength, resulting in p53 checkpoint activation, cell cycle arrest, therefore failed initiation of B cell leukemia in transplant recipients. Importantly, reconstitution of myristoylated CD25 tail but not a mutant construct lacking the serine/threonine motif (S268A/T271A) rescued proliferation and survival defects of Il2ra^-/- ALL cells.

CD25 as novel therapeutic target in BCR-mimic B cell malignancies: Despite heterogeneity of B cell malignancies, we found that CD25 is an integrated biomarker of tumor clones driven by oncogenic BCR-mimics. BCR-dependent B cell tumors further exposed CD25 on the surface in response to the PKCδ activator, PEP005 (also known as Ingenol Mebutate) with rapid phosphorylation of PKCδ^T505, followed by CD25^S268. As a strategy to eradicate CD25-expressing B cell malignancies, combined treatment of PEP005 with ADCT-301, an antibody drug conjugate (ADC) targeting CD25 with a pyrrolobenzodiazepine (PBD) dimer as the cytotoxic warhead, synergistically delayed leukemia development in recipient mice xenografted with patient-derived Ph⁺ ALL cells. Notably, synergistic effects of ADCT-301 were further enhanced by treatment with AZD5363, an AKT-inhibitor, which further upregulates CD25 on patient-derived Ph⁺ ALL cells by imitating signaling pressure leading to differentiation of large circulating pre-B to small pre-B cell fraction.

Conclusion: Based on these finding, we propose CD25 as an integrated biomarker of BCR-dependent B cell tumors that are highly sensitive to established BCR signaling antagonists (e.g. Ibrutinib, Entospletinib) and as a therapeutic target in refractory B cell tumors.