CELL INTELLIGENCE · 肿瘤细胞治疗研究

T cell-directed immunotherapies have transformed the treatment of hematological malignancies, but durable benefit remains limited by relapse, poor persistence, incomplete immune reconstitution and infection. These outcomes depend not only on target-antigen expression but also on the functional quality of the T cell compartment. In this Review, we define T cell fitness as a multidimensional capacity that includes cellular availability, memory reserve, proliferative competence, cytotoxic function, metabolic resilience, resistance to chronic stimulation and persistence. We distinguish exhaustion, senescence and terminal differentiation as overlapping but non-equivalent states and propose a measurable, modality-specific assessment framework rather than reliance on a single marker. We then compare how these states arise in multiple myeloma, lymphoma, acute lymphoblastic leukemia and acute myeloid leukemia, emphasizing the effects of age, tissue niche, disease burden and prior therapy. We critically appraise evidence from CAR-T cell therapy, bispecific antibodies and checkpoint blockade, including the limitations of predominantly retrospective, correlative and disease-specific datasets. Finally, we translate the fitness framework into clinical questions: when to collect cells, how to select bridging therapy, which biomarkers merit prospective testing, how treatment duration and sequencing may preserve immune competence, and where gene-edited products, CAR-NK cells and metabolic interventions may reduce dependence on compromised autologous T cells. A fitness-based approach may support more precise biomarker development and safer treatment selection, but prospective validation and modality-specific thresholds are still required.