ABSTRACT The goal of this project is to develop a novel bimodal intraoral imaging device for assessment of the oral and oropharynx mucosa for detection of neoplasia, aiding in early detection. Oral and oropharyngeal cancers account for 640,000 new cases of cancer worldwide each year. The number of cases increases each year, in part due to the spread of human papillomavirus. The 5-year survival rate has remained relatively unchanged over several decades, despite accessibility of the oral cavity and oropharynx for examination. With early detection, the 5-year survival rate increases to 80%, providing hope of improved patient outcomes given development of effective early detection approaches. The current clinical approach of detection by visual examination with palpation to guide biopsy acquisition, is insufficient to reliably detect neoplasia at the earliest, most treatable stages, including oral epithelial dysplasia (OED) and early oral/oropharyngeal squamous cell carcinoma (OSCC). Our preclinical and pilot clinical sample studies indicate that label-free nonlinear optical microscopy (NLOM), consisting of multiphoton autofluorescence microscopy and second harmonic generation, is highly promising as an optical biopsy method. It provides deep mucosal imaging at subcellular resolution and detects parameters associated with neoplasia that parallel histology. In addition to optical histology, NLOM provides additional spectroscopic metrics based on the metabolic and biochemical microenvironment, which may add to optical histology metrics. Our results assessing NLOM for this application show high sensitivity and specificity for detection of neoplasia, and a high degree of agreement with histopathology. Despite these advantages, a limitation of microscopy is the inherently limited image field of view on the order of hundreds of microns. On the other hand, clinical large area screening methods comprised of widefield autofluorescence (WF) imaging provide the fields needed (centimeters) with high sensitivity for detection, but lack specificity and cannot image below the surface or assess microscopic features. In this project, we will combine the benefits of both approaches into a single handheld intraoral imaging device that will provide large area screening for regions of suspicion for neoplasia using WF coupled with detailed, depth-resolved subsurface imaging by NLOM for highly sensitive and specific detection of high risk lesions (OED) and early OSCC. To accomplish this goal we will 1) construct a biomodal WF-NLOM intraoral imaging device and optimize performance and suitability for noninvasive evaluation of neoplasia in human oral/oropharyngeal mucosa; 2) refine and translate methodology and metrics defined in preclinical studies, to the intraoral device for use in human OSCC; and 3) evaluate the bimodal intraoral imaging device performance and conduct a pilot study to evaluate the device in patients in vivo, obtaining valuable feedback regarding potential for detection of neoplasia and adaptability for the clinic.

PUBLIC HEALTH RELEVANCE This project will develop a new imaging tool to promote early detection of oral cancer, which accounts for nearly 640,000 new cases of cancer worldwide each year. We will develop a handheld wand that fits in the mouth and will illuminate the surface with light to highlight areas suspicious for early cancer and then image below the surface to see if cellular changes associated with high-risk precancer and cancer are found in those sites. This capability is aimed to improve biopsy guidance over the current practice of direct observation by the eye and palpation, ultimately improving patient survival.