A Sequencing Microarray for Mitochondrial Mutations

Johns Hopkins University, Baltimore MD

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Pancreatic cancer is a lethal disease, with the vast majority of patients presenting in an advanced, inoperable stage. There is an urgent need to develop sensitive and specific molecular biomarkers for early detection of pancreatic cancers. Somatic mitochondrial DNA mutations are common in cancers, but their usefulness as a biomarker for early diagnosis has been impeded by low-throughput and lack of adequate sensitivity of direct sequencing technologies. An oligonucleotide-based mitochondrial sequencing microarray ("MitoChip") has been designed that can sequence approximately 29kb of double stranded DNA in a single assay. Both strands of the entire human mitochondrial coding sequence (15,451bp) are arrayed on the MitoChip; both strands of an additional 12,935 bp (84% of coding DNA) are arrayed in duplicate, providing internal validation of sequence data. 1.6 million bp of mitochondrial DNA have been sequenced, with intra- and inter-chip sequence reproducibility of >99.99%. In serial dilution experiments using mixed normal and tumor DNA, the MitoChip was able to detect an aberrant clonal population in as much as 50-fold diluted samples. The eventual goal of this proposal is to enable early diagnosis of pancreas cancer using endoscopically-obtained pancreatic juice samples. The central hypothesis is that array-based sequencing can reliably detect mitochondrial DNA mutations from pancreatic juice samples containing shed neoplastic cells. The R21 phase will therefore (a) identify the frequency and patterns of any confounding tissue-specific coding sequence variations that may occur in non-neoplastic pancreata compared to lymphocyte DNA, (b) determine the frequency of mitochondrial coding sequence mutations in a series of primary pancreatic cancers (using matched lymphocyte DNA as control), and (c) establish the conditions for long PCR amplification and MitoChip assay from pancreatic juice DNA. In the R33 phase, endoscopically obtained pancreatic juice from patients with pancreatic cancer, non-neoplastic pancreatic disease (e.g., chronic pancreatitis), and non-diseased controls (n=40 in each category) will be analyzed for mitochondrial DNA mutations. Juice samples from a specific subset of patients with localized, node-negative pancreatic cancer will also be examined to confirm the utility of the MitoChip in detecting early stage, and hence potentially curable, disease. This study will serve as a model for array-based diagnosis of cancer in clinical samples.