Development study on noninvasively tracking CAR T cells with PSMA reporter gene PET/CT

Introduction: For solid tumors, chimeric antigen receptor T (CAR-T) cells therapy confronts obstacles of improving the infiltration and overcoming the immunosuppressive tumor microenvironment. Furthermore, the cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome are related to the abnormal activation and localization into the central nervous system of CAR-T cells. However, current clinical examinations have difficulties in tracking the real-time distribution of CAR-T cells in vivo. Our group previous study verified the longitudinal [⁶⁸Ga]Ga-PSMA-617 PET/CT for noninvasively tracking CAR T cells targeting transferrin receptor (TfR) in breast cancer by using truncated prostate specific membrane antigen (ΔPSMA) reporter gene. Furthermore, other previous researches also have proved the reporter gene imaging is one promising method for noninvasively real-time tracking CAR-T cells in vivo. Mesothelin (MSLN) is one promising target for CAR-T cells therapy in pancreas cancer, which lacks effective treatment and has poor prognosis. In this study, we constructed one novel CAR-T cells targeting MSLN with ΔPSMA reporter gene (denoted as MSLN CAR-ΔPSMA T cells). To gain a deeper understanding of the dynamics of CAR-T cells distribution in vivo, we investigated the sensitivity and specificity of [⁶⁸Ga]Ga-PSMA-617 PET/CT for tracking MSLN CAR-ΔPSMA T cells.

Methods: On the basis of constructing MSLN CAR-ΔPSMA T cells, we launched the proof of concept studies by using the [⁶⁸Ga]Ga-PSMA-617 to visualize and quantify MSLN CAR-ΔPSMA T cells in vitro and in vivo to explore the minimum detection threshold. The procedure of in vivo experiments was as Fig. 1A exhibited. MSLN CAR-ΔPSMA T cells or blank T cells (non-transduced) with different gradient numbers (0, 10 K, 50 K, 100 K, 500 K and 1 M; 1 K = 10³, 1 M = 10⁶) in 100 µL 50% mixed matrix gel were innoculated on both shoulders of Balb/c nude mice. Mice were injected with 5.55 MBq of [⁶⁸Ga]Ga-PSMA-617 via the tail vein and PET/CT scans were conducted at 30 min after injection. The regions of interests (ROIs) of cell mass and the blood pool were drawn by three times and calculated the ratio (Mass/blood). For in vitro experiment, MSLN CAR-ΔPSMA T cells or blank T cells were incubated with 74 MBq [⁶⁸Ga]Ga-PSMA-617 for 60 min respectively. Next, we resuspended the cells and added 20 µL of the cell suspension per well to evaluate the binding specificity (n = 5) after two washes to exclude the non-specific combination. The final mean cell gradients ranged from 8 × 10³ to 1 × 10⁶ cells per well prepared for PET/CT scans.

Results: The in vivo experiment exhibited that the threshold was 100 cells/mm³ (Fig. 1B). The more MSLN CAR-ΔPSMA T cells possessed significantly higher uptake of cell mass and the ratio in each gradient group (Fig. 1C-E). Similarly, MSLN CAR-ΔPSMA T cells showed significantly higher uptake than blank T cells in vivo (Fig. 1F). The in vitro study showed that the detection minimum number was 400 cells/mm³ (Fig. 1G), and there were linear relationship between MSLN CAR-ΔPSMA T cells with [⁶⁸Ga]Ga-PSMA-617 PET signal (Bq/mL) (R² = 0.9992) (Fig. 1I). Furthermore, compared with the MSLN CAR-ΔPSMA T cells, the uptake of blank T cells was near the background signal (Fig. 1H, J).

Conclusions: Above results confirmed that [⁶⁸Ga]Ga-PSMA-617 PET/CT could sensitively and specifically detect MSLN CAR-ΔPSMA T cells in vitro and in vivo. In the future, we will explore the utility of [⁶⁸Ga]Ga-PSMA-617 PET/CT for tracking CAR-T cells in the pancreas cancer model.

Acknowledgement: National Natural Science Foundation of China (grants 82030052), Hubei Province Science and Technology Innovation Team ([2022] No. 72) and Key Project of Hubei Province Natural Science Foundation (No. 2021CFA008).

• Download figure
• Open in new tab
• Download powerpoint