TMIC-30. TREM1-TARGETED PET IMAGING OF TUMOR-ASSOCIATED MACROPHAGES IN AN ORTHOTOPIC GLIOBLASTOMA MOUSE MODEL

Abstract

Tumor-associated macrophages (TAMs) are a key component of the tumor microenvironment and play an integral role in cancer pathogenesis and metastasis. Triggering receptor expressed on myeloid cells 1 (TREM1) is a membrane receptor that acts as a potent amplifier of proinflammatory responses and immune signaling. TREM1 is highly expressed on TAMs, but not cancer cells [1,2], and is intimately associated with cancer recurrence, invasion, and poor survival rates. Since TREM1 levels are strongly linked to maladaptive inflammatory processes, imaging this target could provide insights into the functional status of myeloid cells in the context of cancer biology. Here, we aimed to synthesize a TREM1-targeted PET tracer and evaluate its ability to detect TAMs in an orthotopic glioblastoma (GBM) mouse model. Anti-TREM1 monoclonal antibody was conjugated to DOTA and radiolabeled with 64Cu (t1/2=12.7h), affording [64 Cu]TREM1-mAb with high specific radioactivity (>75GBq/μmol), radiochemical purity >99%, and labeling efficiency of 70-95%. Cell uptake studies revealed 24-fold higher binding of [64Cu]TREM1-mAb in TREM1-transfected vs. untransfected HEK293T cells (0.056±0.002 vs. 0.002±0.0002, p<0.001, n=4), confirming its high specificity for TREM1 in vitro. Blocking studies, using 100-fold unlabeled TREM1-mAb, led to a significant reduction of [64Cu]TREM1-mAb binding in TREM1-transfected cells (p<0.001, n=4). Next, [64Cu]TREM1-mAb was evaluated in vivo using an orthotopic GBM model. 1million murine-derived GL26 cells in HBSS (3mL) were stereotactically implanted supratentorially in C57Bl/6 albino mice. Sham mice underwent the same procedure but were only injected with HBSS. 7days after implantation, mice were injected with [64Cu]TREM1- mAb (∼90μCi) and imaged 1day later using MR and PET/CT. After imaging, mice were perfused to remove possible unbound intravascular [64Cu]TREM1-mAb, and radioactivity in dissected tissues was measured using a gammacounter. Brain sections from selected GBM and sham mice were further analyzed via ex vivo digital autoradiography and stained with H&E. PET signal from brain tumors, identified by co-registering brain PET and T2-weighted MR images, was significantly higher than that observed in healthy brain regions for GBM and sham mice (tumor: 4.10±0.67; GBM mouse normal brain: 1.74±0.57; sham brain: 1.07±0.17%ID/g, p<0.01, n=4). Similarly, biodistribution results revealed significantly higher uptake of [64Cu]TREM1-mAb in the whole brain of GBM mice compared to shams (0.35±0.12 vs. 0.14±0.02 %ID/g, p<0.05, n=4). Moreover, autoradiography showed markedly higher [64Cu]TREM1-mAb binding within the tumor compared to normal brain tissue of GBM mice and sham brain slices. To our knowledge, this is the first report of a PET tracer targeting TREM1. The ability to image TREM1 levels in vivo and home in on maladaptive immune responses in the tumor microenvironment could profoundly impact the way we understand the neuroinflammatory contribution to GBM development and progression, as well our ability to select and monitor GBM therapies. (Figure Presented).