Abstract
Hematological malignancies such as acute myeloid leukemia (AML), chronic myeloid leukemia (CML), lymphomas, and multiple myeloma remain difficult to model ex vivo because conventional two-dimensional (2D) cultures and murine systems fail to reproduce the spatial, metabolic, vascular, and immune complexity of human bone marrow and lymphoid niches. Recent advances in three-dimensional (3D) platforms-including spheroids, engineered organoid-like marrow models, and microfluidic niche-on-a-chip systems-now allow for a more physiological replication of stromal, endothelial, and immune interactions that drive resistance and relapse. In this review, we introduce explicit definitions distinguishing spheroids, organoid-like constructs, true hematopoietic organoids, and microfluidic devices to establish a unified framework for hematologic 3D modeling. We synthesize applications across AML, CML, lymphoma, and myeloma, highlighting mechanistic insights, strengths, and limitations unique to each disease. Finally, we outline a translational roadmap that integrates bioprinting, perfusable vasculature, immune reconstitution, and AI-driven analytics toward next-generation patient-specific platforms. These innovations position 3D marrow-mimetic systems as essential tools for precision oncology in blood cancers.