A PET/CT scanner for guiding treatment of head and neck cancer - ABSTRACT/SUMMARY Head and neck carcinomas (HNC) are some of the most challenging cancers to effectively treat (five-year survival rates for some HNCs are as low as 25%); recurrence rates range from 8% to 43%. The number of new cases is rising due to the current, near epidemic of HPV-associated HNCs. Unfortunately, most treatments are associated with significant morbidity due to damage of sensitive structures in the region (spinal cord, salivary glands, parotid gland, esophagus, carotid arteries and thyroid gland). HNCs are treated with surgical excision, radiation therapy, or a combination of methods. Selection of therapy is based upon assessment of tumor location, size, proximity to bone, amount of infiltration into surrounding tissues and spread to regional lymph nodes performed with advanced imaging methods. Perhaps the most promising of these methods is metabolic-based imaging, specifically PET, often utilizing FDG. The non-optimal spatial resolution and fixed geometry of whole body scanners, however, limits PET in fulfilling its promise in this role by inhibiting the ability to accurately detect small tumor masses in lymph nodes, and in quantifying the size and nature of the primary tumor. Thus, there is an unmet need for improved PET/CT scanner technology to enhance treatment planning of HNCs. To address this opportunity, we propose the creation and testing of a lower-cost, flexible geometry, high-resolution PET/CT system (approaching the spatial resolution of pre-clinical PET scanners), called HNPET/CT, designed specifically for the imaging of the head and neck region. It will consist of a novel pair of large area, immersion-cooled, PET detectors and a cone beam CT (CBCT) scanner mounted on a rotating gantry whose geometry can be tailored to patient size and anatomy to be scanned. To capitalize on HNPET/CT’s high-resolution images, an image segmentation method that utilizes both CBCT and PET images will be developed and tested. We plan to explore the potential utility of HNPET/CT for enhancing therapy planning (surgical and radiation). This assessment will first be performed with anthropomorphic phantoms and then in a limited human trial. In addition to the novel detector design and adjustable geometry, HNPET/CT will introduce a new capability to the treatment planning of HNCs not yet broadly employed by end users in the clinical. It could enhance current planning techniques (reduce treatment margin size, for example), enable the effective application of advanced methods (dose painting, for example), and perhaps inspire development of new, more effective personalized treatment strategies that require high resolution, multi-modality imaging. The lower cost of the system promises the introduction of cutting-edge, image-guided treatments to under-served populations in areas treated by small or private clinics that often cannot afford such technology. This project will be performed by a multi-disciplinary academic industrial partnership joining the Departments of Radiology, Head and Neck Surgery and Radiation Oncology at West Virginia University, and Xoran Technologies, LLC.