First demonstration of in vivo tracking of single cells using PET

Objectives: With the progress in cell-based therapies, molecular imaging methods are increasingly being used for in vivo cell-tracking applications. They are also being applied to study various physiological and pathological processes, such as stem cell homing and cancer metastasis. Due to its high sensitivity and clinical applicability, positron emission tomography (PET) is the ideal modality for tracking cells. The purpose of this research is to demonstrate that, contrary to common knowledge, PET has sufficient sensitivity to track the migration of single cells in mice, in real-time and at the whole-body level.

Methods: To enable PET tracking, human breast cancer cells (MDA-MB-231) were passively labeled with ⁶⁸Ga-labeled mesoporous silica nanoparticles (MSNs). Using a fluorescent dye label (DiO) and fluorescence microscopy, a small number of single cells were isolated in a 96-well plate through a statistical process of limiting dilution, and their radioactivity was quantified by radioluminescence microscopy (RLM) and gamma counting. Finally, single cells were tracked both in phantoms and healthy mice using a small-animal PET/CT scanner (Inveon) . A helical phantom was constructed by coiling a length of tubing around a 3D-printed cylinder (51 mm diameter, 27 mm helical pitch). List-mode data were acquired while a single cell was circulated through the phantom at a speed of 1.36 mm/s. Healthy mice were injected both intravenously (tail-vein) and in the foot pad, then they were moved to the PET scanner for real-time tracking. Finally, list-mode data were reconstructed using both conventional OSEM and a custom spatiotemporal trajectory reconstruction algorithm. The 3D trajectory of single cells was analyzed in terms of velocity and accuracy.

Results: Radioactivity was 30 Bq/cell on average, with a range of 0-110 Bq/cell. RLM imaging confirmed uptake of 68Ga-MSNs by single cells but suggested heterogeneous distribution of the 68Ga-MSN label. The specificity of the radiolabeling was confirmed by verifying that the radioactivity of a single cell cannot be split into multiple vials. Higher-radioactivity cells were then tracked using PET, both in phantom and in mice. A single cell (67 Bq) moving a helical trajectory at 1.36 mm/s was tracked in real-time with accuracy better than 5 mm (root-mean-square error). Its velocity was estimated and found to be consistent with the flow rate in the tubing. Another single cell (30 Bq) injected IV was located in the mouse lung but no motion was detected over a 10 min acquisition. Migration of MDA-MB-231 cells labeled with DiO to the lungs was later confirmed in ex-vivo lung tissue by confocal microscopy. Another single cell (27 Bq) injected in the footpad remained at the injection site, with no detectable motion. For static cells, focal radiotracer signal was also detectable in the reconstructed 2D OSEM image, and it matched the position obtained through the trajectory reconstruction algorithm.

Conclusions: Single breast cancer cells were successfully labeled with 68Ga-MSNs and tracked using PET, both in a phantom and in vivo in mice. This is the first demonstration of PET imaging and real-time tracking of radiolabeled single cells in vivo. This new platform for single-cell tracking could provide valuable information for further application to cell-based therapy and basic science research.