Info
Background:
iPSC-derived chimeric antigen receptor macrophages (CAR-iMacs) are emerging as scalable off-the-shelf cell therapies for solid tumors, but translation requires quantitative tools to define in vivo exposure, biodistribution, persistence, and dosing interval. Magnetic particle imaging (MPI) directly detects superparamagnetic iron oxide nanoparticle (SPION)-labeled cells with positive contrast, low background, and linear quantification, making it suitable for longitudinal cell-therapy PK/PD assessment.
Hypothesis/Objective:
We hypothesized that MPI could quantify the in vivo fate of engineered CAR-iMacs in orthotopic hepatocellular carcinoma (HCC) and guide dose optimization. The objective was to use MPI/CT to link CAR-iMac exposure and decay kinetics with tumor control and macrophage polarization.
Methods:
Umbilical cord blood-derived iPSCs were engineered to generate glypican-3-targeted CAR-iMacs incorporating ID3 endowment and TGFBR2 deletion to enhance phagocytic activity and reduce TGF-β-mediated immunosuppression. Terminally differentiated TGFBR2-deficient CAR-iMacs were labeled with SPIONs and evaluated for iron loading, viability, phenotype, and MPI calibration. Orthotopic HCC-bearing NSG mice received intravenous SPION-labeled CAR-iMacs at 1 × 10⁶ cells, 2.5 × 10⁶ cells, 2.5 × 10⁶ cells plus a pro-inflammatory supplement, or 5 × 10⁶ cells. Longitudinal MPI/CT quantified cell-associated signal kinetics, while bioluminescence imaging monitored tumor burden. Ex vivo organ MPI, Perls’ Prussian blue staining, human CD68 immunostaining, and multiplex M1/M2 phenotyping validated localization and linked persistence to functional state. MPI-guided dosing was further tested in humanized HCC models.
Results:
SPION labeling achieved efficient intracellular iron loading while preserving macrophage marker profiles and viability. In vitro MPI calibration showed a strong linear relationship between MPI signal and labeled-cell number. After infusion, MPI/CT demonstrated predominant hepatic accumulation with dose-dependent persistence. The 1 × 10⁶-cell dose approached baseline by approximately day 7, whereas 5 × 10⁶ cells remained detectable through day 30. The 2.5 × 10⁶-cell dose showed low residual signal at day 30, while 2.5 × 10⁶ cells plus pro-inflammatory stimulation approached background by approximately day 15. Importantly, longer persistence was not equivalent to greater antitumor efficacy. Bioluminescence imaging showed stronger early-to-mid tumor suppression with the 2.5 × 10⁶-cell plus pro-inflammatory supplement regimen, whereas 5 × 10⁶ cells did not provide consistent incremental benefit and was associated with a greater M2-associated bias. Ex vivo MPI confirmed liver-dominant biodistribution with lower signals in lung and spleen. Histology showed co-localization of Perls blue-positive iron deposits with human CD68-positive macrophages within tumor regions, and MPI signals were detectable in metastatic lung lesions. Guided by MPI, split dosing with 2.5 × 10⁶ cells on days 0 and 15 improved tolerability and sustained antitumor efficacy in humanized models.
Conclusion:
MPI enabled high-sensitivity, noninvasive, quantitative tracking of iPSC-derived CAR macrophages in orthotopic HCC. By linking cell exposure, biodistribution, decay kinetics, tumor response, and macrophage polarization, MPI revealed that maximal persistence is not necessarily optimal efficacy. MPI therefore provides a practical platform for cell-therapy PK/PD assessment, dose selection, and dosing-interval optimization in solid-tumor immunotherapy.
Author
TUO SHAO
Soochow University