Marc Bissonnette

M.D.

University of Chicago

Associate Professor

PLCO (Learn more about this study)

2020-1003

Jan 15, 2021

Development of cell-free DNA (cfDNA) 5-hydroxymethylcytosine (5hmC) biomarkers for early detection of pancreatic ductal adenocarcinoma (PDAC)

Pancreatic ductal adenocarcinoma (PDAC) and the importance of early detection: PDAC is a highly aggressive malignancy, estimated to become the 2nd leading cause of cancer-related deaths by 2030. Prognosis is dismal, with a 5-year survival rate of only ~8% (the lowest of any major cancer). PDAC is rarely curable, because it is rarely diagnosed early. Detection of early-stage PDAC and monitoring of pre-cancerous pancreatic neoplasms are the most desirable approaches for early PDAC diagnosis. However, unlike many other cancers, there is no effective methodology for detecting PDAC early. There is thus a strong, unmet need to develop sensitive and robust biomarkers for the early detection of localized PDAC.
cfDNA as a liquid biopsy analyte for epigenetic mapping: The discovery that circulating cell-free DNA (cfDNA) originates from multiple tissues in the body suggests a revolutionary potential for liquid biopsies. To date, however, biological variables measured during liquid biopsy development are typically useful only for very specific diseases. Similarly, although tumor-related mutations can be detected in cfDNA via liquid biopsy, many mutations are private and are also not tissue-specific. Further, genetic mutations often do not provide information on disease stage or progression. By contrast, genome-wide epigenetic alterations in DNA provide a rich dataset for developing diagnostic tests for many conditions. Disease-specific epigenomic patterns detectable in circulating blood have already been shown to serve as unique biomarkers for disease diagnosis and prognosis.
DNA cytosine 5-hydroxymethylation as a potential epigenetic biomarker for PDAC: Considered the “sixth DNA base,” 5-hydroxymethylcytosine (5hmC) is an epigenetic DNA modification of cytosine and an oxidized form of 5-methylcytosine (5mC). There are several advantages to measuring genome-wide 5hmC epigenetic changes for the development of biomarkers, including a) aberrant alterations in DNA methylation state occur early in cancer development; b) changes in the genomic distribution of 5hmC occur in a tissue-specific and cancer-specific manner and track with disease progression. Furthermore, when used as a surrogate for gene expression, cancer-associated 5hmC-modified loci reveal information about underlying biological mechanisms and microenvironmental influences promoting cancer progression.
The Investigators here have extensive experience developing blood–based biomarkers for gastrointestinal cancers. We have recently developed a highly sensitive and selective analytic approach, called nano-hmC-Seal, to capture 5hmC-containing DNA fragments in genomic DNA and cfDNA, followed by next-generation sequencing, to map genome-wide 5hmC distributions. The nano-hmC-Seal technique can be applied to small amounts of input DNA (e.g. <5 ng cfDNA from <500l plasma), making this state-of-the-art technology valuable for development of liquid biopsy assays. In preliminary studies we have already discovered exciting 5hmC profile differences in cfDNA that appear to distinguish patients with PDAC from controls (see Available Preliminary Data section 3.2).
Informed by the 5hmC profiles we will derive from tumor tissue, we propose to identify diagnostic 5hmC biomarkers from circulating cfDNA to create a sensitive, specific, and clinically feasible blood test for PDAC detection. We will assess the sensitivity of this test and its ability to detect a cancer signature prior to clinical diagnosis using PLCO blood samples from patients who later developed PDAC.

Aim 1: Discover and validate genome-wide 5hmC biomarkers diagnostic for PDAC.

a: Perform genome-wide 5hmC profiling in patient-matched plasma-derived cfDNA and pancreatic cancer tissue from individuals with PDAC versus healthy controls, followed by use of rigorous bioinformatic and machine-learning approaches to discover 5hmC genomic signatures diagnostic for PDAC.

b: Using cfDNA from an independent cohort, validate 5hmC biomarker signatures discovered in aim 1a that distinguish patients with PDAC from healthy individuals.

Aim 2: Determine if 5hmC profiles obtained from circulating cfDNA in PLCO samples from asymptomatic individuals can detect occult PDAC prior to clinical cancer presentation.

Perform genome-wide 5hmC profiling on plasma-derived cfDNA from asymptomatic PLCO Trial participants who subsequently developed pancreatic cancer versus cancer-free matched control participants.