Large language model-guided CAR-T in silico platform for cytokine optimization in liver cancer with low antigen density

Clinical Snapshot

Key Findings

• P Value: Not reported in abstract

• Effect Size: Not reported in abstract; specific quantitative effect sizes for cytotoxicity, proliferation, or tumour volume reduction are not provided

• Primary Outcome: IL-15-engineered GPC3-targeted CAR-T cells demonstrated superior proliferation, persistence, and serial cytotoxicity against GPC3-low liver cancer cells compared with conventional CAR-T cells and those armoured with IL-2, IL-7, IL-12, or IL-18

• Nnt Or Sensitivity: Not applicable at preclinical stage; no NNT, hazard ratio, or diagnostic accuracy metrics reported. In vivo xenograft tumour control described as 'improved' without quantification in the abstract

• Confidence Interval: Not reported

Clinical Application

Clinical translation is distant. CAR-T manufacturing for solid tumours remains technically complex, costly, and logistically demanding. IL-15 armoured CAR-T products would require IND-enabling studies, Phase I dose-escalation trials with rigorous cytokine release and lymphoproliferation monitoring, and GMP-compliant manufacturing scale-up before any clinical use Hepatocellular carcinoma is a significant burden in Australia, with approximately 2,000 new cases annually, disproportionately affecting populations with chronic viral hepatitis and Indigenous Australians. No GPC3-targeted CAR-T therapy is currently TGA-approved or PBS-listed in Australia. The only TGA-approved CAR-T products (axicabtagene ciloleucel, tisagenlecleucel, lisocabtagene maraleucel, idecabtagene vicleucel, ciltacabtagene autoleucel) are restricted to haematological malignancies. RACGP and COSA guidelines do not currently address CAR-T therapy for solid tumours. This research is at a stage where it would be relevant to Australian academic oncology centres and clinical trial networks (e.g., AGITG, ANZCHOG) for future early-phase trial design, but has no immediate clinical practice implications. Patients with hepatocellular carcinoma (HCC) expressing low levels of GPC3 — a subgroup with particularly poor prognosis and limited therapeutic options. Potentially relevant to other GPC3-expressing solid tumours (hepatoblastoma, squamous cell lung carcinoma) in future studies

Abstract

CAR-T cell therapy has shown remarkable success in hematologic malignancies but remains limited in solid tumors such as liver cancer due to antigen heterogeneity, low target antigen density, and an immunosuppressive tumor microenvironment (TME). Cytokine engineering can enhance CAR-T persistence and effector function; however, the optimal cytokine payload may vary depending on tumor type, target antigen expression level, and microenvironmental context, making systematic experimental comparison time-consuming and labor-intensive. Here, we applied a large language model (LLM)-based CAR-T in silico platform to systematically evaluate cytokine engineering strategies, including IL-2, IL-7, IL-12, IL-15, and IL-18, in glypican-3 (GPC3)-targeted CAR-T cells for liver cancer. We used cytokine selection as a biologically grounded benchmark to test whether the platform could recover known CAR-T cell-relevant cytokine biology and support future novel predictions. Computational predictions identified IL-15 as the most effective enhancer, particularly against tumor cells with low GPC3 expression. Guided by these results, we generated cytokine-armored GPC3 CAR-T cells and performed in vitro and in vivo validation. IL-15-engineered CAR-T cells exhibited superior proliferation, persistence, and serial cytotoxicity against GPC3-low liver cancer cells. In human liver cancer xenograft models, IL-15-enhanced CAR-T cells achieved improved tumor control compared with conventional and other cytokine-engineered CAR-T cells. The recovery of IL-15 served as a positive benchmark supporting the validity of the LLM-guided CAR-T in silico workflow. Collectively, this study establishes an LLM-guided framework, schema-constrained for rational cytokine selection in CAR-T engineering and identifies IL-15 as a potent enhancer for targeting antigen-low liver cancers.

References

1. 1.Nan, H., Shen, X., Yang, Y., Wang, S., & Li, Y.-R. (2026). Large language model-guided CAR-T in silico platform for cytokine optimization in liver cancer with low antigen density. Journal of Hematology & Oncology. https://doi.org/10.3389/fmed.2023.1135468