0730

19F-perfluorocarbon-labeled human peripheral blood mononuclear cells can be detected in vivo using clinical MRI parameters in a therapeutic cell setting

Corby Fink¹, Jeffrey Gaudet², Matthew S Fox³, Shashank Bhatt⁴, Sowmya Viswanathan⁴, Michael Smith¹, Joseph Chin⁵, Paula Foster², and Gregory Dekaban¹

¹Microbiology and Immunology, Robarts Research Institute, London, ON, Canada, ²Medical Biophysics, Robarts Research Institute, London, ON, Canada, ³Medical Biophysics, Lawson Health Sciences Centre, London, ON, Canada, ⁴Cell Therapy Program, Krembil Research Institute, Toronto, ON, Canada, ⁵Urology, University of Western Ontario, London, ON, Canada

Synopsis

A major hurdle in advancing cell-based cancer vaccines is the inability to track where therapeutic cells migrate post-injection. We present the first study conducted in which primary Good Manufacturing Practice-grade ¹⁹Fluorine (¹⁹F) labeled all peripheral blood mononuclear cell lineages (PBMC) and ¹⁹F cellular MRI was used to track and quantify in vivo migration in a mouse model. Secondly, we present the highest sensitivity for ¹⁹F detection reported in the literature and the first time a 1.2cm deep injection of ¹⁹F-labeled PBMC using a small dual-tuned surface coil has been detected using a clinical scanner and MR protocol.

Purpose

With respect to antigen presenting cell-based (APC) cancer vaccines, the quantity of tumor antigen-specific APC reaching a secondary lymphoid organ post-injection is directly proportional to the magnitude of the ensuing anti-cancer immune response. We performed a pre-clinical validation study in order to progress towards a clinical trial. The main purpose, as a proof of principle study, was to determine if we could label human PBMC under GMP conditions with a ¹⁹F labeling agent and detect these cells in vivo at an immunotherapeutic-relevant depth using a clinical MR scanner.

Methods

For manufacturing validation studies, human PBMC were obtained from 5 subjects and labeled with ¹⁹F (5mg/mL) overnight under GMP conditions. Phenotyping and viability assessment (using 7-AAD) took place both pre- and post-transport to Robarts Research Institute. Prior to in vivo imaging studies, NMR was performed to determine average ¹⁹F uptake per cell. Nude mice were then injected with ¹⁹F-PBMC and subjected to MR imaging.

Pre-clinical mouse MRI was conducted at 9.4T using a balanced steady state free precession (bSSFP) sequence with a resolution of 1x1x1mm³ for ¹⁹F scans and 200x200x200μm³ for ¹H scans. ¹⁹F-human PBMC detection at both the subcutaneous flank injection site and draining lymph nodes was measured and quantified by using the in vivo ¹⁹F signal and comparing it to a ¹⁹F reference tube concentration.

Clinical in vivo MRI was performed on a 3T system using a dual-tuned surface coil (4.3cmx4.3cm) approved for human use. Intradermal mock injections of ¹⁹F-human PBMC ranging from 1x10⁶ to 20x10⁶ were performed in a ham shank as well as a subcutaneous injection of 4.5x10⁶ cells 1.2 cm below the surface. A resolution of 0.5x0.5x1cm³ and scan time of 15min was used to acquire ¹⁹F-bSSFP images.

During this pre-clinical study, the question arose as to which main cell lineages (B cells, T cells and monocytes) within the heterogeneous PBMC formulation were producing in vivo signal in the lymph node. To address this important question, PBMC were negatively selected to purify out T cells, B cells and monocytes. Each lineage was then doubly labeled with ¹⁹F and a unique intracellular fluorophore. Prior to footpad injection of 3x10⁶ PBMC in mice, lineages were recombined at physiologic proportions. Two days later, draining popliteal lymph nodes were removed, cryopreserved and sectioned for fluorescence microscopy to determine the degree of migration by each lineage.

Results/Discussion

Our laboratory demonstrated near 100% ¹⁹F labeling of human PBMC and NMR revealed loading at 6.17x10¹⁰±1.80x10¹⁰ ¹⁹F/cell (Figure 1A/B). GMP-labeling and transport of human PBMC had no significant effect on viability (Figure 2) and all sterility and quality assurance requirements were met. The functionality of APC within the PBMC mixture, as assessed by a mixed lymphocyte reaction, is unaffected by ¹⁹F-labeling. Together, this indicated that the current manufacturing process is suitable for an upcoming clinical trial. Human ¹⁹F-PBMC were detected and quantified in vivo in mice at both the subcutaneous flank injection site (Figure 3A/C) and draining lymph node (Figure 3B).

Clinical 3T imaging was first optimized using cell pellets in which as little as 4x10¹⁷ ¹⁹F/spins were detected (data not shown). This to date is the highest sensitivity reported using clinical 3T parameters¹, despite human PBMC labeling with low ¹⁹F spins per cell (6.17x10¹⁰). Next, MRI of a ham shank after surface intradermal and deeper subcutaneous (1.2cm, a depth similar to inguinal lymph nodes) injections ranging from 1-20x10⁶ ¹⁹F-PBMC permitted a minimum detection of 4.11x10⁶ and 3.76x10⁶ ¹⁹F-labeled PBMC, respectively (Figure 4).

Immunofluorescent microscopy confirmed that the ¹⁹F lymph node signal was the result of originally injected PBMC and that all 3 of the main lineages within PBMC (B cells, T cells and monocytes) are present at physiologic proportions (Figure 5).

Conclusions

This proof of principle pre-clinical study provides a GMP-labeling procedure and experimental layout that yields near 100% ¹⁹F human PBMC labeling without affecting viability or compromising sterility and other quality assurance criteria that must be met for injection into humans. We outline a clinical 3T MR protocol approved for human use that is able to detect cell numbers in vivo an order of magnitude lower than what will be injected into prostate cancer patients upon clinical trial approval. Furthermore, we report the lowest sensitivity for ¹⁹F detection using a clinical set-up reported thus far in the literature and report the detection of a ¹⁹F signal deeper in vivo than what has ever been reported using a surface coil². Further optimization is being performed while awaiting clinical trial approval to image ¹⁹F-labeled PBMC in humans.

Acknowledgements

This pre-clinical study was supported by both a Movember Discovery Grant from Prostate Cancer Canada and the Smarter Imaging Program from the Ontario Institute for Cancer Research (OICR). JG is funded by the Sam Ciccolini Graduate Studentship from Prostate Cancer Canada. CF and JG are both members of the Western Molecular Imaging collaborative program and have received funding from the London Cancer Research and Technology Transfer (CaRTT) training program. CF has received funding from the Ontario Graduate Scholarship (OGS) Program.

References

1) Ahrens ET, Helfer BM, O’Hanlon CF, Schirda C. Clinical cell therapy imaging using a perfluorocarbon tracer and fluorine-19 MRI. Magn Reson Med. 2014; 72(6):1696-701.

2) Rose LC, Kadayakkara DK, Wang G, Bar-Shir A, Helfer BM, O’Hanlon CF, Kraitchman DL, Rodriguez RL, Bulte JW. Fluorine-19 labeling of stromal vascular fraction cells for clinical imaging applications. Stem Cells Transl Med. 2015; 4(12):1472-81.

Figures

Figure 1. Human PBMC can be efficiently labeled with ¹⁹F cell labeling agent. PBMC labeled with a red fluorescent version of ¹⁹F cell labeling agent (5 mg/mL) for 24 hours resulted in 99.8% labeling of PBMC (black histogram) (unlabeled PBMC blue histogram) (A). The number of ¹⁹F atoms per cell for five prostate cancer patient PBMC preparations.

Figure 2. The viability of human PBMC is not affected by transport or ¹⁹F-labelling. PBMC from 3 prostate cancer patients were gated on CD45 to examine the viability of CD45⁺ PBMC following 7-AAD staining. Viability was determined on day 0 and day 1 post-labeling before and after transport from Princess Margaret Hospital (Toronto, Canada) to Robarts Research Institute (London, Canada).

Figure 3. ¹⁹F MRI of human PBMC after flank administration in murine model. Overlay produced from ¹⁹F and ¹H MR images following a 6x10⁶ human PBMC subcutaneous flank injection. On day 0, a ¹⁹F signal was observed in both the flank (blue arrow) (A) and in the popliteal lymph node (yellow arrow) (B). Forty-eight hours after administration, the ¹⁹F signal was decreased but visible in the flank (C) but not detectable in the lymph node (D). The reference tube used to quantify the number of PBMC is marked with R in (C).

Figure 4. ¹⁹F-labeled human cells can be detected using a 3T clinical MRI scanner. A dual-tuned surface coil is required to detect ¹⁹F-labeled cells (A). Realistic mock infusions were performed in a ham shank. Following a 4.5x10⁶ intradermal (yellow arrow) and subcutaneous (blue arrow) cell injection,¹⁹F-labeled cells were detected using a clinical 3T MRI protocol (B). ¹⁹F-cell detection at a depth of 1.2cm is important to note as this is similar to the depth of inguinal lymph nodes in humans (C).

Figure 5. Popliteal lymph node ¹⁹F signal is a result of originally injected human PBMC. Prior to injection, B cells, T cells and monocytes were separated via negative selection and differentially labeled with a fluorescent cytoplasmic dye. Cell lineages were then recombined back at physiological proportion and footpad injections of 3x10⁶ human PBMC were performed. Two days later, popliteal lymph nodes were removed and sectioned to identify fluorescent cells. Scale bar represents 100µm.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)

0730