Saliva-based detection of oral HPV and oral cancer.

Abstract

6057 Background: Head and neck cancers (HNC) consist of a group of biologically and clinically diverse malignancies. Oral cavity squamous cell carcinoma (OCSCC) and oropharyngeal squamous cell carcinoma (OPSCC) are the most common subtypes, together comprising the majority of HNC cases. While these cancers are associated with tobacco use and alcohol consumption, infection with human papilloma virus (HPV) is also etiologic to HNC, with OPSCC known to harbor higher HPV positivity (HPV+) rates relative to OCSCC, which is largely HPV-negative (HPV-). This distinction is clinically relevant as HPV+ tumors are more responsive to therapy and associated with better prognosis. While early-stage OCSCC/OPSCC has 5-year survival rates of >80%, diagnosis typically occurs at advanced stages, where the survival rate drops to 20-40%. Compounding this grim outlook is the growing incidence in patients that do not smoke or drink alcohol, as well as the startling rise in HPV+ OPSCC cases, thus increasing the affected population and burden on the healthcare system. Currently, painful incisional biopsies are the standard method to diagnose HNC, and there are no accepted non-invasive screening options for early disease detection or serial assessment of treatment efficacy. There is an urgent need for non-invasive diagnostic solutions that accurately identify disease at an early stage and quickly discern HPV status to inform effective treatment decisions and improve patient outcomes. Methods: We have recently developed a novel salivary liquid biopsy screening method for early detection of OCSCC, which relies on targeted NGS of 7 commonly mutated genes associated with OCSCC tumorigenesis, and shown that it is able to accurately and reproducibly identify ~93% of patients with OCSCC, including early stage cases. To enhance the capabilities of this screening platform we have incorporated probes targeting high-risk HPV strains (hrHPV16/18) into the sequencing panel. Results: Applying this multi-functional assay to a cohort of 20 primary OCSCC tumors, driving somatic mutations were detected in all cases. Furthermore, using the 5% alignment cutoff, all OCSCC specimens were HPV-. We next sequenced HPV+ OPSCCs using the same criteria, detecting 93% of the cases as HPV positive by our combined assay. We next applied this updated assay to saliva specimens collected from 30 OCSCC patients and 10 healthy individuals. Somatic mutations were detected in all OCSCC saliva specimens, while no actionable aberrations were found in healthy controls. As expected, 27 of 30 OCSCC saliva samples were HPV-, with the remaining 3 samples being inconclusive. Conclusions: While additional validation is needed to accurately assess the performance of this dual panel in saliva specimens, such multi-functional (mutational drivers/HPV) detection assay may facilitate personalized treatment decisions based on tumor biology (mutations) in addition to clinical risk factors (presence of hrHPV).