A collaborative team led by Professor Huji Xu from Naval Medical University, China, together with Professors Xuanming Yang and Yujia Cai at Shanghai Jiao Tong University, China, published a comprehensive review in Volume 2, article number 33 of the journal Immunity & Inflammation on August 31, 2026. The article systematically summarizes the development history, key engineering strategies, and clinical translation progress of in vivo chimeric antigen receptor T cells (CAR-T) technology, providing a theoretical framework for next-generation in vivo immune cell therapies and offering critical translational insights.
The advent of CAR-T therapy has profoundly transformed the landscape of cancer immunotherapy. However, conventional CAR-T therapy relies on ex vivo modification of a patient's autologous T cells—a complex process involving leukapheresis, T cell activation, genetic modification, expansion, quality testing, and reinfusion. This lengthy, costly workflow is constrained by manufacturing capacity and healthcare resources, severely limiting clinical accessibility. Even as "universal" off-the-shelf CAR-T products continue to improve, the requirement for complex ex vivo manipulation and lymphodepletion conditioning remains an unavoidable barrier.
The core breakthrough of in vivo CAR-T lies in transferring the entire T cell engineering process into the patient's body. This strategy drastically reduces both time and cost—from weeks of manufacturing to practically immediate availability. Additionally, in vivo transduction can occur while T cells remain in a more native, undifferentiated state, such as the memory stem cell phenotype, potentially endowing CAR-T cells with superior self-renewal capacity and anti-tumor persistence.
Achieving efficient and specific in vivo reprogramming depends critically on the precision of delivery systems. Viral vector platforms include targeted lentiviral vectors with engineered envelope proteins and engineered adeno-associated viruses (AAV). Lentiviral vectors offer large cargo capacity and genomic integration for persistent expression, while AAV vectors, existing as episomes, reduce the risk of insertional mutagenesis. Non-viral platforms are represented primarily by lipid nanoparticles (LNPs), polymeric nanoparticles, and virus-like particles. LNPs achieve T cell targeting through surface modification or optimized lipid composition. The transient expression of mRNA cargos avoids genotoxicity concerns and has demonstrated remarkable efficacy in non-tumor disease models, including cardiac fibrosis and systemic lupus erythematosus.
"In vivo CAR-T clinical translation is now accelerating," the authors point out. Both LNP-mRNA and targeted lentiviral systems have achieved efficient T cell transduction and deep target cell clearance in non-human primate models. Early-phase clinical trials for relapsed/refractory multiple myeloma and autoimmune diseases have shown promising safety and preliminary efficacy.
The review objectively addresses significant challenges and explores frontier solutions. Regarding persistence limitations, the transient expression of mRNA may be insufficient for complete tumor eradication, making circular RNA and self-amplifying RNA promising strategies for extended expression. For off-target and on-target toxicity, the development of more sophisticated "logic-gated" dual-recognition strategies and T cell-specific promoters is critical. Immunogenicity and repeat administration issues require stealth vector design and transient immunomodulation. The functional state of patient T cells—often exhausted in advanced cancer—may necessitate cytokine pre-treatment or co-delivery of anti-exhaustion genes. Looking forward, the authors highlight: "With advances in artificial intelligence (AI)-assisted molecular design, more precise T cell subset targeting, multi-target combination therapy, and vector engineering for the tumor microenvironment, in vivo CAR-T is poised to mature into a reliable off-the-shelf genetic medicine, dramatically expanding the boundaries of cellular immunotherapy."
In summary, in vivo CAR-T represents a significant evolutionary direction in cell therapy. As delivery material engineering, vector design, gene editing technologies, and AI-assisted drug development continue to advance, this strategy promises to further enhance treatment precision and safety, overcome the limitations of conventional CAR-T therapy, and provide new therapeutic options for cancer, autoimmune diseases, and beyond—transitioning cell therapy from individualized manufacturing to a scalable, off-the-shelf paradigm.
