CAR-T cells are generated by genetically modifying T cells to reactivate them to recognize tumors by containing chimeric antigen receptors (CARs).CAR-T cell therapy has been developed for more than 30 years with encouraging therapeutic results in the clinic. Currently, five generations of CAR-T with different properties have been developed.The first generation of CAR-T had relatively low efficiency of intracellular activation. Second-generation CAR-Ts are produced by adding co-stimulatory molecules to improve activation efficiency. The third generation CAR-T was also developed based on the first generation by integrating multiple signaling domains as co-stimulatory molecules. However, third-generation CAR-T did not exhibit better activation efficiency than second-generation. Moreover, the fourth generation was generated based on the second generation CAR-T by integrating some specific cytokines (e.g., IL-12), which not only enhanced the anti-cancer effect. Thereafter, the anti-tumor efficacy of CAR-T cells was enhanced by inserting the interleukin 2 β receptor chain between the structural domains of CD3ζ and CD28, and adding the transcription factors STAT3 and STAT5 to the end of CD3ζ to generate the fifth generation CAR-T cells.

Interventions implemented during the expansion phase of CAR-T cells are critical for long-term survival after infusion. With sophisticated engineering tools, we have adapted the growth and differentiation pathways of CAR-T cells with the aim of enhancing their anti-tumor efficacy. Overall, the cells demonstrate superior proliferation and differentiation potential during the cultivation phase. There is a delicate balance between effector function and long-term survival, and the quality attributes and characteristics of the T-cells are decisive for the efficacy of the treatment. Researchers have observed that myostatin is able to inhibit extracellular acidification and shift the metabolic pattern from an acidic stress state to an oxidative high-energy state. Accordingly, the addition of myostatin was shown to enhance the efficiency of lentiviral gene transfer in activated T cells. In addition to these specific metabolic substances, growth factors are able to participate.

Viral vectors are the primary delivery platform for CAR genes. Viral vectors, such as gamma retroviruses and lentiviral vectors, can efficiently infect cells. However, packaging of viral particles is complex, expensive and time-consuming, and carries a high risk of infection contamination. In addition, viral vectors have limitations in helping to express multiple genes at the same time due to their low load capacity and a higher risk of random insertions. Researchers have developed various non-viral vector delivery methods, including transposon delivery. Transposon-mediated CAR gene delivery has the advantages of being safer, more efficient in expression, less costly, and less immunogenic than viral vectors. In recent years, nanomaterials such as lipid nanoparticles (LNPs) have been investigated for CAR gene delivery. However, they have drawbacks, including low persistence due to transient protein expression and complex preparation processes.CAR-T cells can also be obtained by gene editing techniques (e.g., TALEN and CRISPR/Cas9), which can facilitate the binding of the CAR gene to any part of the genome at a fixed point. Gene editing-mediated generation of CAR-T cells is safer and more efficient. Many of the delivery strategies that have been developed need to be further optimized to effectively deliver CAR genes and facilitate the clinical application of CAR-T cell therapies in human diseases.

The heterogeneity of tumor cells leads to difficulties in target selection, and there may be differences in antigen expression between different patients or even different tumor cells of the same patient. Certain targets may also be expressed in normal tissues, leading to off-target effects and potential side effects. False-positive or false-negative results may be encountered during the screening process, affecting the final antibody selection. We use high-throughput screening techniques, such as flow cytometry and fluorescence resonance energy transfer (FRET), to improve screening efficiency. Design and construction of mutant antibody libraries based on antibody sequence and structure information. Rigorous biological activity assessment of humanized antibodies to ensure that their function is not compromised. Use automated cell separation, activation and amplification equipment to improve preparation efficiency and consistency. Optimize gene modification methods, such as the use of non-viral vectors or CRISPR/Cas9 gene editing technology. Conduct long-term clinical trials to track and evaluate the safety and efficacy of CAR-T cell therapy. Establish a comprehensive monitoring and treatment mechanism for side effects such as cytokine release syndrome (CRS) to ensure patient safety. To gain an in-depth understanding of the behavior and mechanism of CAR-T cells in vivo through technical means such as single-cell sequencing.

In the temporal dimension, prolonged or excessive CAR-T cell activity leads to excessive release of cytokines and over-activation of the immune system.CRS is the main toxic response , and the incidence of grade 3 or above CRS has reached 30% in some clinical trials. In the spatial dimension, CAR-T cells may localize to normal cells expressing target antigens and activate them uncontrollably, leading to OTOT toxicity. All of these antigens show some degree of expression in normal tissues. Although the expression of TAA in normal cells is substantially lower than in tumor cells, CAR-T cells can respond to low expression of antigens due to their high sensitivity. When exposed to antigen, infused CAR-T cells undergo several orders of magnitude of activation and proliferation. Incorporation of regulatory modules into CAR-T cells can be used to achieve regulatory activation. Genetic circuits can be created by combining standardized regulatory elements using the logic-gating principle of circuitry. Integration of these genetic circuits into T cells or modification of chassis T cells can be used to effectively address safety concerns. Since the tumor microenvironment is characterized by hypoxia, modifying CAR-T cells to function only in a hypoxic microenvironment could significantly reduce the potential for OTOT toxicity and expand the range of target antigens available for therapeutic applications. In the future, CAR-T cells are expected to evolve into more complex engineered cells.