CELL INTELLIGENCE · 肿瘤细胞治疗研究

BACKGROUND: Chimeric antigen receptor (CAR) T-cell therapy in solid tumors is hindered by the immunosuppressive tumor microenvironment and by toxicities associated with viral-vector manufacturing. Non-viral gene delivery platforms have emerged as a potential alternative, though clinical evidence remains fragmented. METHODS: Following an a priori protocol registered on the Open Science Framework (OSF; https://doi.org/10.17605/OSF.IO/2TPQS) and adhering to JBI/PRISMA-ScR guidelines, a systematic search was conducted across four databases from inception through May 15, 2026. Patient-level data were extracted to describe cellular persistence and clinical outcomes across strictly non-viral delivery platforms. RESULTS: Four early-phase studies met the inclusion criteria, encompassing 28 heavily pretreated patients with metastatic solid tumors. Two non-viral platforms were identified: mRNA electroporation (n=19; intravenous in 13, intratumoral in 6) and the piggyBac transposon system (n=9). Across both mRNA routes, transient CAR-T persistence (<7 days) was observed, with no objective responses (ORR 0%), though disease stabilization yielded a disease control rate (DCR) of 53%; cross-route comparison is limited by differing distribution profiles. The piggyBac system showed longer persistence (~28 days) and a DCR of 78%, including the only documented objective response (ORR 11%). No Grade 3 cytokine release syndrome or neurotoxicity was reported in any of the 28 patients, and no tocilizumab or systemic corticosteroids were required. CONCLUSIONS: Within this limited early-phase evidence base, no severe toxicities attributable to non-viral platforms were reported, and the evidence identifies knowledge gaps warranting prospective investigation. mRNA platforms showed transient persistence and disease stabilization in 53% of patients. One partial response was documented with the piggyBac platform in a single patient; however, this outcome cannot be attributed to the delivery platform given simultaneous differences in target antigen, tumor histology, route of administration, and geographic setting. No firm conclusions regarding comparative platform performance can be drawn from this evidence base. SYSTEMATIC REVIEW REGISTRATION: https://doi.org/10.17605/OSF.IO/2TPQS, identifier OSF.IO/2TPQS.