open-label trial of an allogeneic dual-target CAR-NK product directed against GPC3 and B7-H3 for adults with advanced hepatocellular carcinoma. The design intentionally uses GPC3 as the primary target anchor because GPC3 is the dominant HCC cell-therapy antigen in current clinical development, while adding B7-H3 to reduce antigen escape and to broaden coverage across tumor and tumor-microenvironment compartments. The study first evaluates safety and dose-limiting toxicities, then expands at the recommended phase 2 dose.
Hepatocellular carcinoma (HCC) remains a high-mortality solid tumor, and many patients progress after or are not candidates for surgery, transplant, locoregional therapy, immune-checkpoint inhibitors, or tyrosine-kinase inhibitors. Among HCC-associated cell-surface targets, GPC3 has the strongest disease-specific cell-therapy track record and is widely used as the anchor antigen in HCC CAR-T programs. However, not all HCCs express GPC3 uniformly, and loss or low-density expression may contribute to resistance and escape. B7-H3 is an attractive secondary co-target because it is frequently expressed in HCC tumor and stromal/vascular compartments, is associated with immune suppression and aggressive biology, and may complement GPC3 when antigen heterogeneity is present. The investigational product in this draft is an allogeneic donor-derived peripheral blood NK-cell product genetically engineered to express a dual-target CAR recognizing GPC3 and B7-H3. To keep this example realistic and conservative, the core registration fields below assume a standard dual-target CAR-NK product plus fludarabine/cyclophosphamide lymphodepletion. If the sponsor later prefers an armored platform , those features can be added in a subsequent protocol version or IND-enabling package. The study is structured as a phase 1/2 single-group trial. Phase 1 uses dose-escalation to determine safety, the recommended phase 2 dose (RP2D), and feasibility. Phase 2 expands at the RP2D to estimate preliminary anti-tumor activity. Participants receive lymphodepleting chemotherapy before CAR-NK infusion, then undergo protocol-defined safety monitoring, serial imaging, and translational assessments including CAR-NK persistence, serum AFP, cytokine profiling, and ctDNA dynamics. Because B7-H3 can also be detected at low levels in some normal tissues, the trial is intentionally conservative: it uses central biomarker confirmation, stepwise dose escalation, strict liver function eligibility, and close monitoring for infusion reactions, cytokine-release syndrome, immune effector cell-associated neurotoxicity, hepatotoxicity, cytopenias, infection, and any evidence of off-tumor toxicity.
Study Type
INTERVENTIONAL
Allocation
NA
Purpose
TREATMENT
Masking
NONE
Enrollment
30
Allogeneic donor-derived NK cells genetically modified to express a dual-target CAR recognizing GPC3 and B7-H3/CD276. Administered intravenously after lymphodepletion.
Lymphodepleting chemotherapy given before CAR-NK infusion.
Lymphodepleting chemotherapy given before CAR-NK infusion.
Peking University Shenzhen Hospital
Shenzhen, Guangdong, China
RECRUITINGIncidence of dose-limiting toxicities (DLTs)
Time frame: 28 Days
Incidence, type, and severity of treatment-emergent adverse events (TEAEs) and serious adverse events (SAEs)
Time frame: 12 months
response rate (ORR) by RECIST 1.1 and mRECIST
Time frame: 6 months
Disease control rate (DCR)
Time frame: 6 months
Duration of response (DoR)
Time frame: 12 months
Progression-free survival (PFS)
Time frame: 12 months
Overall survival (OS)
Time frame: 24 months
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.