Discovery of New DNA Methylation Biomarkers for Predicting the Malignant Outcome of Low-Grade Oral Dysplasia - Oral squamous cell carcinoma (OSCC) is a devastating malignancy that may arise from precursor mucosal
lesions, called oral premalignant lesions (OPLs). OPLs consist of low- and high-grade dysplasia (ODs).
Although most low-grade ODs remain stable for years and some even regress spontaneously, a significant
minority of them (4-11%) rapidly develop OSCC. Currently, there is no biomarker available for screening low-
grade ODs to identify those at high risk of developing OSCC to implement evidence-based cancer prevention
management. Several studies reported DNA methylation changes in oral carcinogenesis, but failed to address
the differences between progressive vs. static low-grade ODs or generate whole genome coverage with
sufficient width and depth for maximal new discoveries. Our long-term goal is to understand the epigenetic
mechanisms that drive the malignant progression of precancerous lesions. The objective of this R21 is to
discover new differentially methylated regions (DMRs) associated with OPL progression using a whole-
genome single-nucleotide-resolution approach, create novel DMR-based models, and validate their power in
predicting the progression of low-grade ODs to OSCC. The central hypothesis of this R21 states that the
progression of OPLs to OSCC is driven by DNA methylation changes that can be discovered by whole-genome
DMR analysis of progressive vs. static OPLs and used to predict the cancer risk of OPLs for management
decision. This hypothesis is formulated based on the scientific premises that: 1) epigenetic modification is a
key disease-driving mechanism that captures early carcinogenic environment-gene interaction events and is
faithfully preserved throughout disease progression, and 2) DNA methylation changes have been shown to
occur in ODs and OSCC compared to normal oral epithelium. The rationale for pursuing the proposed research
is that once the spectrum of DMRs between the progressive and static ODs are identified based on high quality
data and powerful analysis, we will discover key DMR sites that drive and predict the malignant progression of
low-grade ODs, which would otherwise be missed. Aim 1 will discover genomic regions differentially
methylated in progressive vs. static low-grade ODs using whole genome bisulfite sequencing (WGBS) on
samples with longitudinal follow-ups. Aim 2 will create DMR-based models and validate their power in
predicting the risk of low-grade ODs in progression to OSCC. A successful completion is expected to discover
new DMRs capable of differentiating progressive vs. static ODs and create novel risk-prediction DMR models.
The outcome will fill the critical need for a sensitive and reliable screening method to predict the risk of OPLs in
becoming OSCCs at an early stage, when medical interventions are more effective and less damaging, which
will lead to a direct impact on reducing the incidence of OSCC and related health issues, as well as finding
novel DNA methylation mechanisms driving the malignant progression of OSCC that may be druggable. It may
also have a broad impact on advancing research on other tobacco-related HNSCC.
