ABSTRACT Pre-clinical imaging provides wonderful structural features, but is lacking in the spatial resolution for molecular features which are deep into the animal body. This is due to fundamental physical limits on optical scattering & absorption, and is especially problematic for orthotopic tumors, such as pancreas or glioma, which are grown in the middle of the body. The most relevant molecular tracers of tumor metabolism and immunology are often imaged well through the skin in subcutaneous tumors, but these images are highly superficial or achieved with microscopic imaging. There is no method to image 1-3 cm into tissue with molecular sensitivity in the microMolar to nanoMolar range. A new high-resolution, deep-tissue, imaging approach has been invented, and in this application will be further developed for whole body scanning of concentrations in the sub-microMolar range. The new approach uses thin sheets of MegaVolt (MV) x-ray from a linear accelerator (LINAC) shaped by a multileaf collimator, to induce Cherenkov excitation of luminescence for scanned imaging (CELSI). These sheets are swept over the animal to localize the excitation via Cherenkov within the animal, allowing precise knowledge of where the detected light came from. The emission is captured with time-gated low-light detector array, using an approach similar to sheet illumination microscopy. The key benefit is that the spatial resolution is determined by the LINAC beam size and location in an otherwise optically turbid sample. The design implicitly allows high precision spatial localization, and we hypothesize and test the functionality of linear correction algorithms such as spatial deconvolution and depth-dependent attenuation correction, as compared to non-linear diffusion based reconstruction. The proposed project develops the basic science of a working prototype system, as well as a collaboration to develop a commercial prototype system. The Cerenkov emission excites either phosphorescent or fluorescence molecules, which are used to directly measure metabolism or to tag molecular reporters. Initial animal studies showed CELSI could be achieved either i) at therapeutic doses at a very low molecular probe concentration (2Gy with nanoMolar probe) or ii) low radiation doses for moderately higher probe doses (0.1 Gy with microMolar probe). Recovery of images with spatial resolution less than 300 microns is readily achieved, throughout the entire body of an animal. Three parameters directly influence image quality, including 1) sheet depth, 2) delivered dose, and 3) probe concentration, and the reciprocity between these will be quantitatively examined to define acceptable and biologically relevant modes of operation. In this work, the system to image multiple rodents is developed with detection sensitivity being optimized for luminescence in a clinical LINAC. A commercial partner will provide a custom short pulsed LINAC for superior signal-to-noise and production of a prototype commercial system. Metabolic and immune sensing probes will be optimized for orthotopic pancreas cancer imaging, which is critical to understand responses of tumors that effectively recapitulate the growth and pathophysiology of human disease within the pancreas. This full-body high-resolution molecular optical imaging has particular relevance to advancing research into orthotopic cancer models and internal organ diseases, which are not resolved well with any current molecular imaging tools.

Narrative A new high-resolution deep tissue imaging system has been invented, using thin sheets of high energy gamma rays to excite molecular probes. In this application will be further developed into a system which allows whole body scanning to image concentrations in the sub-micromolar range, and can be implemented in cancer treatment hospitals with high animal throughput. It allows molecular imaging of orthotopic cancer models such as pancreas, which are within the interior of rodents, and so not easily possible to image with current optical molecular systems.