CELL INTELLIGENCE · 肿瘤细胞治疗研究

Glioblastoma remains the most lethal primary malignancy of the central nervous system, and the modest gains achieved with maximal surgery, radiotherapy and temozolomide have not been matched by the immune checkpoint inhibitors and antigen-specific vaccines that reshaped the treatment of many extracranial cancers. The recurrent disappointment of these approaches has been attributed less to a single molecular lesion than to a confluence of obstacles: profound intratumoural heterogeneity, a densely immunosuppressive and myeloid-rich microenvironment, sequestration and exhaustion of conventional T cells, and the practical difficulty of delivering effectors across the blood-brain barrier. Against this background, γδ T cells have attracted interest as an unconventional effector population that recognises transformed cells through stress-associated and metabolic cues rather than peptide-major histocompatibility complex (MHC) complexes, that kills in an MHC-unrestricted manner, and that can be expanded from healthy donors for allogeneic, off-the-shelf use with little expectation of graft-versus-host disease. This narrative review examines, with a deliberately critical lens, the biological rationale and the experimental evidence for γδ T cell-based immunotherapy of glioblastoma. We summarise the developmental biology and functional subsets of human γδ T cells, the natural killer group 2 member D (NKG2D)-, DNAX accessory molecule 1 (DNAM-1)- and T-cell-receptor-dependent mechanisms through which they engage glioblastoma cells and glioma stem-like cells, and the in vitro and animal-model studies that underpin the field, taking care not to overstate efficacy that has so far been demonstrated only in preclinical or early-phase settings. We then weigh the principal opportunities-locoregional and repeated dosing, combination with chemoradiotherapy, checkpoint blockade and antibody-based redirection-against barriers that include limited persistence, uncertain intratumoural trafficking, donor and manufacturing variability, and the unsettled requirements of potency testing and trial design. We give particular weight to what becomes of γδ T cells inside a hostile tumour-the exhaustion-like dysfunction that follows chronic stimulation, the oxygen and glucose dependence of their effector programme, the interleukin-17-polarising signals generated by activated microglia and by genotoxic therapy, and the confounding effect of corticosteroids-together with the engineering and pharmacological strategies proposed to counter them. The first peer-reviewed phase 1 report of intracranially delivered, drug-resistant γδ T cells has now appeared and documents tolerability in a small, single-arm cohort without establishing survival benefit. Throughout, γδ T cells are presented as a biologically plausible but still investigational strategy whose clinical value will be determined by adequately powered trials rather than by mechanistic appeal alone.