MicroRNA-based prognostic model for early-stage oral cancer patients

PROJECT SUMMARY / ABSTRACT Each year an estimated 400,000 people globally, among them 30,000 Americans, are newly diagnosed with oral squamous cell carcinoma (OSCC). Of the newly diagnosed oral cancer cases, ~80% are in the tumor- node-metastasis (TNM) Stage I/II without regional lymph node involvement or distant metastasis. For these early-stage oral cancer patients, the five-year survival rate is estimated to be 60%. Currently, there is no means to distinguish 60% of early-stage patients who will most likely survive from the 40% who are at high-risk for cancer-specific death. As 80% of oral cancer patients are in early stage at the time of diagnosis, a window of opportunity exists in which accurate prognostication and subsequent decisions for proper treatment will dramatically improve 5-year survival of patients with this deadly disease. We have previously discovered microRNA (miRNA) signatures, miRNA-375 and 214-3p, using next generation sequencing that have high prognostic power to identify individuals at increased risk for cancer-specific death. In the current proposal, we aim to develop, refine and validate the miRNA-based prognostic model of 5-year cancer survival and to identify optimal therapeutic strategy for the high and low risk strata. In Aim 1, using a retrospective study design, we will discover prognostic miRNA signatures using next-generation sequencing and validated in the independent internal validation cohort from Columbia University Medical Center using qRT-PCR in early-stage oral cancer patients treated with surgery only. We will develop a miRNA-based prognostic model consisting of miRNA signatures and prognostically significant clinico-demographic covariates. We will then externally validate the miRNA-based prognostic model in a second, independent cohort from the NCI-sponsored Cooperative Human Tissue Network (CHTN). Moreover, we will construct and validate a cancer-specific mortality risk score formula, which is personalized formula consisting of the patient?s miRNA expression levels and the clinico- demographic covariates, weighted by their respective regression coefficient. In Aim 2, we will identify treatment regimens (elective neck dissection, radiation therapy, or elective neck dissection with radiotherapy following surgery of primary tumor) associated with survival benefit in high and low risk groups, stratified based on the miRNA-based prognostic model developed in Aim 1. This R01 proposal describes a 4-year research project aimed at developing clinically applicable strategy in personalized molecular management of early-stage OSCC patients.