PROJECT SUMMARY/ABSTRACT
Epigenetic modifications play a crucial role in mammalian development, cellular differentiation, and disease, with
5-methylcytosine (5mC) being the most well-studied epigenetic modification. 5mC globally represses gene
expression and has significant gene-regulatory importance in various diseases. The ten-eleven translocation
(TET) family of enzymes can oxidize 5mC to 5-hydroxymethylcytosine (5hmC), which is a stable marker of global
transcriptional activation. While previous studies have characterized the role of 5hmC modifications in healthy
human tissues and evaluated plasma cell-free DNA (cfDNA) for its biomarker potential in various malignancies,
urine cfDNA has not yet been studied. Urine represents a non-invasive, easily collectible, and proximal biofluid
to a number of GU diseases. At present, it is unclear how urine cell-free DNA (cfDNA) 5hmC profiles reflect
normal physiological conditions and mirror the composition of neighboring tissues. Researchers wonder whether
urine cfDNA 5hmC profiles might offer a superior alternative to plasma cfDNA, the current gold standard in liquid
biopsy, in the diagnosis of GU cancers like bladder cancer. To address these questions, this proposal seeks to
characterize the genome-wide 5hmC landscape of healthy urine and develop biomarkers in the setting of
localized bladder cancer. The proposal aims to achieve these objectives through two specific aims. The first aim
focuses on identifying sources of unwanted variation, characterizing the genomic and epigenomic 5hmC
landscape, and inferring the tissue-of-origin of cell-free DNA (cfDNA) derived from healthy urine samples. To
accomplish this aim, a bioinformatic pipeline will be optimized for processing 5hmC profiles from urine cfDNA.
The 5hmC regulation in urine cfDNA will be characterized by analyzing its activity on gene bodies, promoters,
enhancers, and various regulatory epigenomic regions. This aim also proposes the creation of tools to determine
the tissue-of-origin of urine cfDNA and confirm its enrichment for GU tissues compared to plasma cfDNA,
highlighting the applicability of urine cfDNA in GU contexts. The insights gained from Aim 1 will provide a
reference for evaluating the deviations in 5hmC distribution in urine cfDNA in various cancer and non-cancer
disease states. The second aim focuses on developing robust diagnostic and predictive biomarkers for localized
bladder cancer. The study will demonstrate the utility of urine cfDNA 5hmC-profiles in bladder cancer diagnosis
and treatment surveillance and compare its performance to plasma cfDNA 5hmC profiles. Additionally, a multi-
omics classifier model incorporating 5hmC, copy number, and targeted mutational data will be created and
validated for predicting minimal residual disease in bladder cancer patients who underwent curative-intent
bladder removal surgery. The results from Aim 2 will generate 5hmC signatures that can be used to predict the
onset of localized bladder cancer and to predict minimal residual disease. Ultimately, the data from Aim 2 could
deliver a set of new gene and enhancer targets that maybe serve as exciting therapeutic targets for prevenation
and treatment of bladder cancer.
