Development of autologous humanized leukemia models for immunotherapy testing

About This Grant

PROJECT SUMMARY Immunotherapies such as bispecific T-cell engagers (BiTEs) and chimeric antigen receptor (CAR) T cells have shown significant promise in treating hematologic malignancies. However, therapeutic responses vary between patients and across the different subtypes of leukemia and lymphoma. A key challenge in optimizing these therapies is the lack of in vivo preclinical models that accurately reflect both the patient's immune and cancer cell biology. Current patient-derived xenograft (PDX) models lack functional immune systems, while humanized mouse models typically involve healthy donor immune cells paired with cancer cells from a different donor. These immunologically mismatched, or allogeneic, models fail to replicate autologous immune-cancer cell dynamics and the effect the cancer microenvironment and therapy have on immune cell function. To overcome these limitations, we aim to develop innovative autologous humanized PDX models using leukemia and immune cells derived from the same patient. We will collect paired bone marrow (BM) samples from pre-B acute lymphoblastic leukemia (B-ALL) patients at diagnosis and remission. Hematopoietic stem and progenitor cells (HSPCs) from the remission BM will be expanded and transplanted into immunodeficient mice to generate humanized mice with intact immune systems. These mice will then be engrafted with diagnostic leukemia cells, creating PDX models with autologous immune and leukemia cells. In parallel, a second cohort of models will be developed using autologous peripheral blood mononuclear cells (PBMCs) and leukemia cells from the same patient. We hypothesize that these fully patient-derived autologous models will provide deeper insights into immune-leukemia interactions and enhance preclinical testing of immunotherapies. To test this hypothesis, we will pursue three specific aims. First, we will establish autologous PBMC- and HSPC- humanized models using samples from B-ALL patients with diverse genetic and risk subtypes. We will evaluate disease progression and immune responses longitudinally in these models. Second, we will assess the efficacy of a CD3/CD19 BiTE and CD19 CAR-T cells in these autologous models. Third, we will analyze the immunophenotypic and transcriptional profiles of cells from autologous models, allogeneic models, and patient samples before and after immunotherapy treatment to validate the translational relevance of the models. These studies aim to create robust, patient-specific models that can be used to test and optimize immunotherapies, ultimately improving their clinical impact in the treatment of hematologic malignancies.