[摘 要] 嵌合抗原受体T细胞（CAR-T细胞）疗法在血液系统恶性肿瘤中显示出显著疗效，但传统体外制备存在流程复杂、周期长和成本高等局限。体内CAR-T细胞通过递送载体在患者体内直接重编程T细胞，有望简化制造流程并提高治疗可及性。慢病毒载体具有载荷容量较大、可稳定整合和生产体系相对成熟等优势，但其天然或假型化包膜的广谱嗜性可能导致非靶细胞转导和安全性风险。本文综述水疱性口炎病毒G蛋白（VSV-G）、Cocal病毒包膜蛋白（COCV-G）、辛德比斯病毒及麻疹病毒/尼帕病毒包膜等假型化与去受体化工程策略，讨论CD3、TCR、CD4、CD8和CD7等靶向受体的选择及CD2/CD28共刺激信号整合，并梳理相关临床前研究和早期临床转化进展。现有证据表明，靶向慢病毒可在体内生成具有抗肿瘤活性的CAR-T细胞，但仍需解决插入突变、脱靶转导、免疫原性、剂量控制和规模化生产等问题。本文旨在为下一代体内CAR-T细胞递送系统的设计与转化提供参考。

[Abstract] Chimeric antigen receptor T (CAR-T) cell therapy has shown remarkable efficacy in hematologic malignancies, but conventional ex vivo manufacturing is limited by procedural complexity, long production cycles, and high cost. In vivo CAR-T cell generation directly reprograms T cells in patients through gene-delivery vectors and may simplify manufacturing and improve accessibility. Lentiviral vectors offer relatively large payload capacity, stable genomic integration, and mature manufacturing systems; however, broad tropism of native or pseudotyped envelopes may cause off-target transduction and safety risks. This review summarizes envelope pseudotyping and receptor- detargeting strategies based on the envelopes of vesicular stomatitis virus glycoprotein (VSV-G), Cocal virus glycoprotein (COCV-G), Sindbis virus, measles virus, and Nipah virus; discusses the selection of CD3, TCR, CD4, CD8, and CD7 as targeting receptors and the integration of CD2/CD28 costimulatory signals; and reviews preclinical studies and early clinical translation. Available evidence indicates that targeted lentiviral vectors can generate functional antitumor CAR-T cells in vivo, although insertional mutagenesis, off-target transduction, immunogenicity, dose control, and scalable manufacturing remain major challenges. This review provides a technical reference for the design and translation of next-generation in vivo CAR-T cell delivery systems.

[基金项目] 河南省科技攻关计划项目（242102311031）