Optical Imaging of Head and Neck Cancer

This study examines if certain imaging techniques and devices can aid the surgeon in detecting cancer during the surgical procedure.

The purpose of this study is to determine if optical imaging modalities used at the time of surgical resection for head and neck squamous cell carcinoma can help delineate normal from cancerous mucosa. The High resolution microendoscope, developed by our collaborators at Rice university, can allow for real time visualization of tissue nuclei. The overall aim of this study is to determine if this device can be used to enhance the accuracy of intraoperative margin detection during tumor resection for head and neck cancer. At the time of tumor resection for head and neck squamous cell carcinoma, a wide field imaging device will be used to identify suspicious areas. The High resolution device will then image representative areas from the tumor, the tumor margin, and normal mucosa. A topical dye, proflavin, will be placed on the tissue to enhance the visualization of nuclei prior to imaging with the HRME device. Following imaging, biopsies of the imaged areas will be taken and submitted for pathology diagnosis. The images of the biopsies will then be compared and the device will be evaluated for accuracy of margin detection at the time of tumor resection.

Squamous Cell Carcinoma, Neoplasia, Optical imaging, Head and Neck Cancer, Oropharynx, Larynx, Oral Cavity

High Resolution Microendoscopy (HRME) imaging device that operates as a fluorescence microscope with a fiber optic imaging probe. The probe is placed against the mucosa to obtain images relayed to a tablet computer. 0.01% Proflavine hemisulfate used as a fluorescent contrast agent applied topically to mucosa. HRME is used to capture images of suspicious areas sprayed with proflavine hemisulfate.

High Resolution Microendoscopy imaging device that operates as a fluorescence microscope with a fiber optic imaging probe. The probe is placed against the mucosa to obtain images relayed to a tablet computer.

Accuracy of reviewers in differentiating neoplastic or benign mucosa in comparison to the pathology results

Sensitivity = probability that the HRME correctly classifies as positive those with neoplasia compared to pathology results

Specificity = Probability that the HRME correctly classifies as negative those without neoplasia compared to pathology results

Amount of agreement among the 11 blinded head and neck cancer specialists, determined by the Fleiss Kappa. 33 benign and 65 cancer images were evaluated by the reviewers who were blinded to the anatomical site, tumor subsite, and final histopathologic diagnosis. Each reviewer was asked to classify each image as benign or neoplastic. The reviewers evaluated the images based on nuclear size, nuclear to cytoplasmic ratio, and overall cell architecture. Images were randomized in their presentation to the reviewers as to not establish any pattern. Each reviewer provided their interpretation in isolated settings to avoid influence from other reviewers.

Inclusion Criteria: Biopsy Proven Squamous Cell Carcinoma of the oral cavity, oropharynx, larynx, hypopharynx Must be receiving surgical treatment for their cancer Exclusion Criteria: Presence of medical or psychiatric condition affecting the ability to give informed consent Known allergy to Proflavin Pregnant or nursing Females