A Siriraj team compared second-, third- and fourth-generation anti-GPC3 CAR T cells. Among 195 HCC biopsies, 82.6% were GPC3-positive, and the fourth-generation construct showed the strongest killing in monolayer and spheroid models, with antigen-dependent proliferation and a moderated acute cytokine profile. All efficacy evidence remains in vitro.
Key findings
- GPC3 was positive in 82.6% of biopsies. At an effector-to-target ratio of 10:1, CAR4 cytolysis reached 52.55 ± 10.04% against HepG2 (p=0.003) and 64.45 ± 7.81% against HeLa-GPC3 (p=0.0002), exceeding GPC3-negative controls, and it also led in spheroids. CAR4 retained antigen-dependent proliferation, with lower PD-1 but higher TIM-3 than CAR2 after acute exposure.
Why this matters globally
The high GPC3 prevalence and stronger CAR4 activity support further development for a solid tumour that remains challenging for CAR T therapy, where trafficking and the in-vivo microenvironment are major barriers.
Thai researcher contribution
Siriraj's cancer-immunotherapy, molecular-medicine, pathology and hepatopancreatobiliary-surgery teams collaborated from patient tissue profiling through engineered-cell evaluation.
Limitations to consider
Monolayers and spheroids do not model trafficking, persistence, cytokine toxicity, on-target/off-tumour effects or systemic immunosuppression. Tissue prevalence does not predict response, and higher TIM-3 may have complex implications under chronic exposure.