Development of blood-based methylation biomarkers for CRC risk prediction Colorectal cancer (CRC) incidence and mortality rates are disproportionately higher in African Americans (AA) compared to Caucasian Americans (CA). Current non-invasive screening tools like fecal occult blood test (FOBT) or Cologuard detect cancer after it occurs, and more effective tools for prevention and treatment of higher risk individuals, such as colonoscopy or endoscopy, are invasive, less popular and subjective. Therefore, identification of early biomarkers that distinguish normal colon mucosa of cancer patients from normal colon mucosa of patients without cancer might decrease racial disparities in CRC. We have identified a subgroup of patients as having “Outlier Methylation Phenotype” (OMP) using normal tissue methylome. OMPs are highly epigenetically disrupted and display abnormal DNA methylation patterns throughout their epigenome. We have been able to significantly associate this phenotype with CRC patients over healthy controls. Furthermore, AA CRC patients appear more than twice as likely to have OMP than CA. In our current grant application, we propose to develop OMP as a less- invasive CRC screening and prognostic tool by evaluating the consistency of OMP status in a less-invasive (whole blood) tissue with an invasive tissue (normal colorectal mucosa). In Specific Aim 1A, we will test whether OMPs identified in colorectal tissues can also be identified in whole blood samples in 200 CRC patients (100 AA, 100CA) and age, sex, racial ancestry and location (for colorectal tissues) matched 400 healthy controls (200 with history of adenomas and 200 without history of adenomas) using epigenome-wide data from >850K CpGs. In Specific Aim 1B, we will analyze the association of OMP in CRC patients with known CRC molecular subtypes or mutations like CpG island methylation phenotype (CIMP), KRAS, BRAF, MLH1 to estimate whether or not OMP is a surrogate marker of any known CRC molecular subtype. In Specific Aim 1C, we will follow-up the controls (especially OMPs) after 3-4years of screening colonoscopy and compare the clinical outcomes to evaluate the relevance of this phenotype in screening or CRC prevention. We will also study the association of genetic (Specific Aim 2) and environmental (Specific Aim 3) factors with OMPs. In Specific Aim 2, we will genotype the blood DNA to confirm the self-identified racial ancestry of our samples and to study the association of genetic variants with abnormal methylation in OMPs. In Specific Aim 3, we will evaluate the effect of diet and social determinants of health on OMP. Overall, the proposed study aims to identify and characterize molecular markers (OMP) in a less- invasive tissue (whole blood) that can be used both as a diagnostic and a prognostic tool, especially in the underprivileged AA population who have the lowest CRC screening rates, highest CRC incidence and highest CRC mortality rates.

Narrative We have identified a subgroup of colorectal cancer (CRC) patients as having “Outlier Methylation Phenotype” (OMP) using normal tissue methylome and OMPs are significantly associated with CRC patients over healthy controls and African American CRC patients are more than twice as likely to have OMP than Caucasians. In our current grant application, we propose to develop OMP as a less- invasive CRC screening and prognostic tool by evaluating the consistency of OMP status in a less-invasive (whole blood) tissue with an invasive tissue (normal colorectal mucosa). We will test whether OMPs identified in colorectal tissues can also be identified in whole blood samples using epigenome-wide data from >850K CpGs and analyze the association of OMP in CRC patients with known CRC molecular subtypes and with future clinical outcomes in controls; and study the association of genetic (single nucleotide polymorphisms including ancestry informative markers) and environmental (diet and social determinants of health) factors with OMPs.