Unraveling the Mechanism of Action of Bispecific T-Cell Engagers in B-Cell Acute Lymphoblastic Leukemia Using Advanced Single-Cell Multiomics

Background

Bispecific T-cell engagers (BiTEs) stimulate T lymphocytes proximate to tumor cells, inducing T-cell stimulation and tumor killing. Blinatumomab, an anti CD3-CD19, was the first to be approved with overall response rates of up to 80% in refractory B-cell Acute Lymphoblastic Leukemia (B-ALL). To overcome resistance and to develop next-generation BiTEs, we need to build on blinatumomab's success and better understand BiTEs mode of action and interaction with the tumor microenvironment (TME).

Here, we apply advanced multiomics to examine blinatumomab's activation signatures and the sustained dynamics in the tumor and bone marrow (BM) immune TME.

Methods

Physically interacting Cells sequencing (PIC-seq) is an advanced multiomics technology deciphering the crosstalk between the cellular components of an immunological synapse (Cohen et al., 2022; Giladi et al., 2020). To assess Immediate activation signatures, we applied PIC-seq on a model of B-ALL cell lines (RS4;11, NALM-6) and lymphocytes obtained from healthy subjects, activated by blinatumomab, compared to no-treatment and activation with anti CD3-CD28 antibodies.

To examine the Sustained effects on the immune TME, we performed a longitudinal single-cell RNA sequencing of the bone BM of adult B-ALL patients, before and after treatment with blinatumomab in the Division of Hematology in the Tel Aviv Sourasky Medical Center in 2021-22.

Results

Immediate activation

Applying PIC-seq on n=16,543 CD3 and CD19+ cells, we identified the distinct gene expression of ‘blinatumomab-activated lymphocytes’ (Figure 1). Blinatumomab induced cellular synapses composed of specific T lymphocytes subsets along the activation continuum, namely: effector T cells (CCL5, GNLY, NKG7), early IFN-G activation (IRF1, GBP5), activated cytotoxic (Xcl1-2, IFN-G) and proliferating T cells (CD69, CCND2) .

‘Blinatumomab-activated lymphocytes’ were characterized by a unique activation signature: upregulation of cytotoxicity (Xcl1, IFN-G, GNLY), co-stimulation and B-T cells crosstalk (CD40, CD27, TNFRSF18), immune recruitment (CXCL10, CCL4) and antigen-presentation (TNFSF14, ITGB2) pathways. Late synapses also expressed immunoregulatory molecules (TNFAIP8L2). This signature differed from classical anti CD3-CD28 induced activation and is independent of the donor and the target-cell identity. The malignant cells' response varied: RS4;11 cells manifested cellular senescence, while NALM-6 upregulated INFG-induced apoptotic pathways, suggesting indirect killing in the mechanism of action of BiTEs.

Data on the introduction of primary B-ALL patients' cells will be presented at the ASH.

Sustained effects

We profiled >11,000 cells from three patients: Patient 1, 31-year-old (y/o) female, primary refractory B-ALL (50% BM blasts), attaining transient MRD negativity. Patient 2, 71-y/o male, primary refractory Ph+ B-ALL (74% BM blasts) achieving reduction in MRD following first cycle. Patient 3, 71 y/o, Ph+ B-ALL, treated as consolidation therapy, achieving MRD negativity.

Blinatumomab induced a dramatic perturbation in the lymphocytic and myeloid compartments of the BM and was highly patient-specific ( Figure 2). The BM lymphocytic compartment prior to initiation of blinatumomab in molecular remission (patient 3) was characterized by active T and Natural Killer cell states, compared to an abundancy of naïve lymphocytes in active disease (patients 1 and 2). There were no signs of T-cell exhaustion pre or post treatment in all patients. Post-treatment, CD8 T-Central Memory cells proliferated in all patients. Patients treated for active disease (patients 2 and 3), had an expansion of the Tregs population post-treatment.

In the myeloid compartment, patients with active disease had more non-classical monocytes, known to correlate with poor prognosis. These were replaced by classical monocytes following treatment, in addition to the emergence of a macrophage-precursor population.

Data on the BM and peripheral blood dynamics of an additional six patients will be presented at the ASH.

Conclusions

We present an advanced single cell dissection of blinatumomab-induced target-effector interaction, providing novel insights into the mechanism of action and TME dynamics to guide patient selection and next generation BiTEs.

A larger patients' cohort and introduction of patients' cells to the PIC-seq in vitro model will be presented at the ASH.