In the past decade, chimeric antigen receptor T-cell (CAR-T) immunotherapy has demonstrated significant efficacy in patients with chemotherapy-refractory hematological malignancies, including lymphoma, leukemia, and multiple myeloma [1]. However, the success of CAR-T cell therapy in treating solid and brain tumors has been limited [2]. Recent clinical trials have further emphasized these limitations, including a restricted number of targetable antigens with uneven expression, limited adaptability of T cells, ineffective trafficking of T cells to tumor sites with insufficient penetration of tumor barriers, and the immunosuppressive nature of the tumor microenvironment [2,3,4,5].

Target selection is a crucial determinant of CAR-T therapy's efficacy. To ensure both safety and effectiveness, the ideal target antigen must exhibit specificity, stability, and broad malignant cell coverage. Unfortunately, most CAR-T targets for solid and brain tumors fail to meet these stringent criteria. Currently, target selection mainly focuses on differentially expressed antigens [6] that also exist at low levels in normal tissues, making the search for novel targets a promising solution. However, many newly identified antigens are derived from intracellular proteins, which are difficult to target using traditional CAR-T cells. Consequently, researchers are now turning to genomic and proteomic approaches to identify new CAR-T cell targets [7, 8].

Recently, Shaw and his team proposed an innovative pan-tumor targeting strategy using cancer-specific exons (CSEs) as potential targets for immunotherapy against pediatric solid and brain tumors [9]. Through RNA sequencing analysis of 16 types of pediatric tumors, they identified proteins encoded by 157 genes as highly tumor-specific, either at the gene level or as alternatively spliced (AS) isoforms. The majority of these CSE targets result from gene-level aberrations. Among the identified targets, 11 (CD83, CD276, FAP, FN1, GPC2, GPC3, IL1RAP, KDR, KIT, MET, and CD133) have already been investigated in preclinical or clinical studies, while the remaining 93% represent novel targets. Approximately 30% of these targets are highly expressed across both solid and brain tumors, underscoring the potential for pan-tumor target development.

Shaw et al. further categorized the targets into primary and secondary groups based on their expression in normal tissues and vital organs. Primary targets, which exhibit the least expression in normal tissues and bone marrow, are expected to minimize treatment-related side effects. In patient-derived xenograft models, three primary targets—FN1, VCAN1, and COL11A1—were widely expressed in pediatric solid and brain tumors. In vitro studies demonstrated that both COL11A1-CAR and FN1's alternative splicing extra domain CAR (EDB-CAR) had anti-tumor activity against pediatric sarcomas. Orthogonal experiments confirmed the antigen specificity of COL11A1-CAR. Additionally, Shaw's team used a mouse osteosarcoma model to validate the in vivo anti-tumor efficacy of COL11A1-CAR, showing a significant extension of median survival in CAR-treated mice, although tumors eventually recurred.

In conclusion, Shaw et al. demonstrated that CSEs can serve as candidate targets for CAR-T therapy through RNA-seq data mining, yielding a wealth of potential targets for future research. The team validated at least one target, COL11A1, which broadens the scope of immunotherapy options for pediatric solid and brain tumors.

The success of this research can be attributed to several key factors. First, the study employed exon-level analysis, enabling researchers to identify exons specifically expressed in tumor cells due to alternative splicing, thereby discovering more potential targets. Second, the team expanded the range of CAR-T cell targets by including proteins from the extracellular matrix, rather than focusing solely on membrane-bound proteins. Their results showed that extracellular matrix proteins, such as COL11A1, can also serve as effective CAR-T targets. Third, the team conducted an integrated analysis of large-scale RNA sequencing data from multiple databases, including contributions from St. Jude Children's Hospital, the University of Washington, and the National Cancer Institute. This expansive data set ensured the reliability and depth of their findings. Additionally, strict screening criteria were applied to ensure that CSE targets were highly cancer-specific, requiring significantly higher expression in tumor samples compared to normal tissues, and excluding targets highly expressed in critical organs such as the brain, liver, lungs, and bone marrow.

However, the study has some limitations. First, the team used normal adult tissue samples as controls to validate CSE targets in pediatric tumors, which may introduce bias due to the differences between pediatric and adult tissue expression. Additionally, pediatric tumors often arise from developmental origins and exhibit unique biological characteristics that allow them to evade treatment in unpredictable ways [10]. The recurrence of tumors in mice treated with COL11A1-CAR, accompanied by reduced COL11A1 expression, raises questions about the role of CAR-T cell exhaustion and tumor immune evasion, which require further investigation. Lastly, the occurrence of secondary T-cell tumors in some patients undergoing CAR-T therapy calls for careful evaluation of the long-term safety of Shaw's CAR-T strategy.

CAR-T therapy holds great promise for treating solid tumors. Continued exploration of suitable targets and optimization of design strategies are essential to overcoming the challenges of CAR-T therapy. Shaw et al.'s approach offers a new direction for pan-tumor target discovery, with potential applications beyond CAR-T, including immune cytokines and antibody-conjugated therapies. This research opens up vast possibilities for the future of pediatric solid tumor immunotherapy.