Alternatively spliced cell surface proteins for CAR T cell therapy of Acute Myeloid Leukemia

Acute Myeloid Leukemia (AML) is a fatal malignancy derived from hematopoietic stem/progenitor cells. Patients with relapsed or refractory (R/R) AML have poor prognosis with a 5-year overall survival at 27%. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) represents the only curative approach mainly driven by alloreative donor immune cells recognizing leukemia cells. However, most patients are ineligible to allo-HSCT, related complications can be severe and relapse represents the major cause of mortality. Hence, there is an urgent need for novel therapeutic strategies in AML.
Relative to other cancers, AML presents with a low number of mutations. Immune recognition of canonical neoantigens arising from mutations in coding sequences is less probable and less likely to respond to immune checkpoint blockade. An alternative immunotherapeutic approach is represented by Chimeric Antigen Receptor (CAR) T-cell therapy that recognizes molecules expressed on the cell surface independently of antigen processing and HLA presentation, two mechanisms commonly downregulated in cancers like AML and contributing to immune escape. CAR therapy has shown remarkable clinical success in B-cell acute lymphoblastic leukemia (B-ALL) patients. In contrast with B-ALL, the clinical experience with CAR therapy in AML patients has shown moderate efficacy and significant toxicity. Most CAR T-cell trials in AML target CD33 or CD123 that are upregulated on AML cells but, are also expressed on normal cells, raising serious safety concerns. Thus, there is no FDA approval for CAR T therapy for AML patients. The foremost challenge in developing AML CAR therapy is the lack of specific targets and complex clonal heterogeneity.
To identify leukemia-specific targets for CAR therapy of AML, we developed a pipeline identifying alternatively spliced cell surface proteins in AML. AML cells presents with few but frequently recurring mutations in splicing factors and epigenetic modifiers and thus, we sought to investigate AML mutational quality rather than quantity. We identified unconventional targets on the cell surface of AML containing unique sequences derived from alternative mRNA splicing, thus generating unique domains that are not existent in otherwise normal cells. We generated isoform-specific antibodies validating the aberrant protein expression of these molecules in AML patient samples. We constructed novel CAR T cells recognizing one novel splice variant and found that they can eradicate leukemia in immunocompromised NSG mice. Given these exciting data, we propose to define the therapeutic role of alternatively spliced cell surface proteins and prepare for a phase-1 clinical trial with CAR T cells in R/R AML patients through the following Specific Aims:
SA1. Define the expression of alternatively spliced cell surface proteins in AML.
SA2. Assess the efficacy of multi-specific CAR T cell platforms targeting AML heterogeneity.