Development of a Method to Implement Whole-Genome Bisulfite Sequencing of cfDNA from Cancer Patients and a Mouse Tumor Model.

Elaine C Maggi,Jan Vijg,Cristina Montagna,Steven K Libutti,Silvia Gravina,Haiying Cheng,Bilal Piperdi,Ziqiang Yuan,Xiao Dong

doi:10.3389/fgene.2018.00006

Abstract

The goal of this study was to develop a method for whole genome cell-free DNA (cfDNA) methylation analysis in humans and mice with the ultimate goal to facilitate the identification of tumor derived DNA methylation changes in the blood. Plasma or serum from patients with pancreatic neuroendocrine tumors or lung cancer, and plasma from a murine model of pancreatic adenocarcinoma was used to develop a protocol for cfDNA isolation, library preparation and whole-genome bisulfite sequencing of ultra low quantities of cfDNA, including tumor-specific DNA. The protocol developed produced high quality libraries consistently generating a conversion rate >98% that will be applicable for the analysis of human and mouse plasma or serum to detect tumor-derived changes in DNA methylation.

Highlights

Non-invasive blood based screening is emerging as a promising alterative to traditional tissue biopsies in the management of cancer patients
We have described a new method for cell-free DNA (cfDNA) isolation and library generation for use in genome wide DNA methylation profiling of both murine and human samples
That DNA methylation profiling of cfDNA holds the potential of a sensitive biomarker is suggested by whole-genome bisulfite sequencing (WGBS) of cfDNA of pregnant women where placenta specific DNA methylation regions could be detected in the blood of the mothers (Jensen et al, 2015)

Summary

Introduction

Non-invasive blood based screening is emerging as a promising alterative to traditional tissue biopsies in the management of cancer patients. Non-invasive blood based screenings can be performed when traditional tissue biopsies are not feasible or when the collected tissue is not sufficient for diagnostic analysis and testing of biomarkers of interest. This is a common scenario in lung cancer diagnosis and clinical follow up (Ilie et al, 2014)

Objectives

Methods

Conclusion