Abstract This project addresses the current lack of quantifiable and clinically relevant imaging endpoints for use in patient-derived organoid models. Microphysiological tumor models (μPTMs) are tissue-engineered 3D tumors that can be grown inside microfluidic devices to form multicellular tissue-like constructs that retain the biological and functional characteristics of the tissue of origin. These μPTMs provide a powerful model of individual patients’ tumor and are used in drug discovery, cancer research, and personalized medicine. However, a critical hurdle remains that, unlike xenotransplanted tumors, μPTMs are incompatible with positron emission tomography (PET) and other diagnostic imaging tools used in oncology. There is a dearth of quantitative imaging methods that can be applied seamlessly across physical scales, ranging from in vitro cell cultures to animal models and cancer patients. Drawing from extensive preliminary work, we will bridge this gap by harnessing the ability of radioluminescence microscopy (RLM) to image clinical radionuclides in organoids with ultra-high spatial resolution. Upon completion, this project will enable routine imaging of in vitro tumor models using the growing array of diagnostic and therapeutic radiopharmaceuticals, many of which are used as clinical standard of care. This goal will be achieved by pursuing three specific aims. First, we will demonstrate that quantitative image-based metrics can be acquired using PET tracers in patient-derived organoids. Validation will be conducted for three PET tracers against mouse xenograft models derived from the same set of cancer patients. Second, we will refine the μPTMs by incorporating functional (perfusable) human microvascular networks within the 3D matrix and, using imaging and other assays, determine the effect of the vasculature on image-based endpoints. Third, as a pilot translational study, we will develop patient-specific μPTMs (n=10) and compare fluorodeoxyglucose (FDG) metabolic activity in these organoids against biomarkers derived from clinical FDG-PET. In sum, this project will enhance the ability of researchers to run clinical trials “on a chip”, using the patient’s own tumor and clinically approved radiopharmaceuticals. Ultimately, these advances could be translated to predict the efficacy of new drugs, test biological hypotheses, and individualize patient therapy.

Narrative Microphysiological tumor systems, which are small human tumors grown outside the body, are useful models for cancer research, drug discovery, and personalized therapy, but they lack clinically relevant imaging endpoints. This project will develop a quantitative and high-resolution approach to image these microtumors using the existing array of clinical PET radiotracers, thus bridging the gap in physical scale that exist between clinical tumor imaging and microphysiological tumor models. The approach will be demonstrated in a pilot study in a cohort of head-and-neck cancer patients.