Harnessing Negative B Cell Selection to Overcome Drug-Resistance in Acute Lymphoblastic Leukemia

Background: B cells are selected for an intermediate level of (pre-) B cell receptor (BCR) signaling strength: attenuation below minimum (e.g. non-functional BCR) or hyperactivation above maximum (e.g. autoreactive BCR) thresholds of signaling strength causes negative selection and cell death. About 25% of B cell acute lymphoblastic leukemia (ALL) is driven by oncogenic tyrosine kinases (e.g. BCR-ABL1 in Ph⁺ ALL), which mimics constitutively active pre-BCR signaling and defines the ALL subgroup with the worst clinical outcome. Currently more potent tyrosine kinase inhibitors (TKI) are developed for Ph⁺ ALL to suppress oncogenic signaling below a minimum threshold for survival. However Ph⁺ ALL cells invariably develop resistance against TKI. Here, we tested the hypothesis that targeted hyperactivation of oncogenic signaling above a maximum threshold will trigger B cell-inherent mechanisms of negative selection and selectively kill Ph⁺ALL cells.

Results: The Ph⁺ ALL cells don not express a functional pre-BCR and BCR-ABL1 oncogene mimics a constitutively active pre-BCR by phosphorylating SYK, LYN, BTK and PLCg2. An incremental increase of pre-BCR downstream signaling (ITAM or SYK overexpression) was indeed sufficient to induce cell death in Ph⁺ ALL. TKI-treatment, while designed to kill leukemia cells, seemingly paradoxically rescued Ph⁺ALL cells in this experimental setting.

Ph⁺ ALL cells differ from normal pre-B cells by expression of high levels of ITIM containing inhibitory receptors including PECAM1 (CD31), CD300A and LAIR1. Importantly, high expression levels of ITIM-receptors are predictive of poor outcome in two clinical trials. In the COG trial (P9906; n=207) for children high-risk ALL, mRNA levels of PECAM1, CD300A and LAIR1at diagnosis positively correlated with early minimal residual disease (MRD) findings on day 29 (p<0.0005), and negatively correlated with overall survival (OS) rate (p<0.02) or relapse free survival (RFS) rate (p<0.05). In the ECOG trial (E2993; n=215) for adults ALL, PECAM1 mRNA level negatively correlated with OS rate (p=0.0285).

Genetic studies revealed that Pecam1, Cd300a and Lair1 receptors are critical to calibrate pre-BCR signaling strength through recruitment of the inhibitory phosphatases Ptpn6 (SHP1) and Inpp5d (SHIP1). Deletion of Pecam1, Cd300a or Lair1 in Ph⁺ ALL cells caused increased ROS levels, G₀/G₁cell cycle arrest, decreased colony formation capacity and cellular senescence. Phosphorylation of pre-BCR downstream molecules (SYK, LYN, BTK and PLCg2) was increased after Lair1 deletion and this hyper-signaling could not be tolerated by Ph⁺ ALL cells. Lair1 deletion resulted in rapid leukemia regression and prolonged survival of recipient mice in a transplant experiment. Leukemia cell death caused by Lair1-deletion could be rescued by overexpression of the active inhibitory phosphatase Ptpn6 (CD8-SHP1) or Inpp5d (CD8-SHIP1). Genetic deletion of Ptpn6 and Inpp5d caused increased pre-BCR signaling and cell death in BCR-ABL1 ALL cells but not myeloid cells (normal and BCR-ABL1-transformed), which -unlike B cells- are not subject to negative selection of auto-reactive clones. Decreasing pre-BCR signaling by SYK inhibition rescued cell death after Ptpn6- or Inpp5d- deletion.

Blocking inhibitory receptors by using chimeric PECAM1, CD300A and LAIR1 receptor decoys inhibited proliferation and caused cell death in Ph⁺ ALL xenograft cells. More potently, a novel small molecule inhibitor of INPP5D (SHIP1) selectively killed Ph⁺ ALL xenograft cells through inducing hyper pre-BCR signaling, regardless of TKI resistance. We demonstrate that inhibitory phosphatase signaling represents a potential novel class of therapeutic targets for Ph⁺ALL.

Conclusions: These results indicated that inhibitory receptors and downstream phosphatases are critical regulators of pre-BCR signaling strength in Ph⁺ ALL, and identified ITIM-receptors and phosphatases as members of a potential novel class of therapeutic targets. The concept of pharmacological perturbance of oncogenic signaling equilibrium in leukemia cells by inhibition (e.g. TKI-treatment) or exaggeration of signaling strength (e.g. blockade of ITIM-receptors) may lead to the discovery of multiple additional therapeutic targets and broaden our repertoire of currently available therapeutic intervention.