Development of an effective, low-cost, point-of-care early detection test for oral cancer in low resource settings

ABSTRACT Oral cavity and oropharyngeal squamous cell carcinoma (OOPSCC) affects 537,000 people worldwide each year resulting in 313,000 reported deaths. These cancers continue to increase in incidence. The main risk factors for OOPSCC include use of tobacco, alcohol, and betel nut in Southeast Asia, and oncogenic human papillomavirus (HPV). Lower socio-demographic index regions exhibit the highest mortality rates for this disease. OOPSCC usually presents with advanced local or regional disease when 5-year survival rates only reach 30-40%. Thus, there is a significant, unmet need for improved detection of OOPSCC. as cure rates could reach 90% if detected earlier. A low-cost, noninvasive, point-of-care (POC) test would improve time to OOPSCC diagnosis and reduce the morbidity and mortality of this horrible disease. Our group invented patented technology combining three biomarkers for OOPSCC including the stem cancer cell marker CD44, total protein (TP), and p16 to develop an oral rinse early detection test. The test has been improved further by adding p16INK4a (p16). Our preliminary data on 149 subjects using a POC prototype for the 3 markers and a machine learning algorithm incorporating risk and demographic variables yields a sensitivity of 80% and specificity of 89%. To advance this technology further in Specific Aim 1 we will develop CD44, TP and p16 point of care prototypes into commercializable test strips. Vigilant Laboratories will work with DCN Dx, an engineering and diagnostics contract manufacturer, to optimize the test strips for readability and reproducibility and use with DCN's reader system which will provide quantitative outputs for each marker. Oral rinses collected from OOPSCC cases and controls obtained from the University of Miami (UM) and University of California, San Diego (UCSD) will be used to fine tune individual test strips. In Specific Aim 2 we will develop an OOPSCC risk model using the triple biomarker test combined with risk factor and demographic data. We will use archived and prospective samples from 250 cases and 500 controls (50 cases and 100 controls from UCSD). Risk factors such as age, gender, tobacco, alcohol and oral health are recorded. In Specific Aim 3 we will test the comprehensive OOPSCC risk assessment model in cases and matched controls. The model in Aim 2 will be frozen and the algorithm from Aim 2 will be programmed into the reader which will display the final risk level for each tested sample. Data will be collected prospectively from 75 cases and 175 controls (25 cases and 50 controls from UCSD) in clinical settings at UM and UCSD using DCN's customized reader with incorporated algorithm. Successful completion of these aims will result in a risk stratification system for patients at high risk for OOPSCC with the next step being a Phase IIB multi-institutional trial for FDA PMA regulatory approval.