PROJECT SUMMARY/ABSTRACT DNA shed from tumors into the blood stream, termed circulating tumor DNA (ctDNA), is an easily obtained source of tumor material. As most ctDNA is identical to normal DNA, some distinguishing feature is needed to demark a cancer origin. In this regard, a fifth or more of all human cancers harbor a cancer-causing (oncogenic) point mutation in the gene KRAS. This raises the exciting possibility that sequencing for the presence of KRAS-mutant ctDNA could be used to detect many types of cancers from a simple blood draw. Indeed, the Guardant360® ctDNA-detection assay is used for just this purpose in the clinical care of cancer patients. The challenge to detecting ctDNA is that this form of DNA is found at vanishingly low levels in the blood. This limitation is borne out in our own clinical experience at Duke, where we find that the Guardant360® assay successfully detected KRAS-mutant ctDNA in only half the cases in which the patient's cancer was documented to be KRAS mutation- positive by direct sequencing of resected or biopsy tumor tissue. Thus, while Guardant360® is real-world proof that ctDNA can be used as a `liquid biopsy' in the clinic, there is clearly much room for improvement. In this regard, we adopted the Maximum Depth Sequencing (MDS) technology, originally developed in the microbiology field to detect rare antibiotic-resistance mutations in bacteria populations, to detect oncogenic mutations in KRAS. By barcoding the original KRAS template and making multiple first-strand replicates thereof, coupled with ultra-deep sequencing of these targeted DNA products, we were able to detect mutations engineered into KRAS templates at a limit of 5x10-7, which is 2,500 to 50,000 times more sensitive than the detection limit of 1x10-3 to 2x10-4 reported for the Guardant360® assay. Given this, we will combine the basic research of Dr. Counter into this KRAS-specific MDS (K-MDS) assay with the clinical and translational expertise of oncologist Dr. Abbruzzese to optimize the K-MDS assay for blood samples (aim 1) and then evaluate K-MDS to Guardant360® a prospective clinical comparison (aim 2). Completion of this study will thus provide an new technology to screen for KRAS-mutant ctDNA in the blood of cancer patients at a sensitivity orders of magnitude greater than current clinical assays, initially to monitor either recurrence of KRAS-mutant cancers or detect such cancers in high-risk patients, but more long term, in combination with screening for other hotspot mutations and using different sources of tissue, for the early detection of multiple cancer types.

PROJECT NARRATIVE Early detection of cancer holds great promise as a way to identify cancer at a curative stage or recurrent disease before radiologically visible to initiate treatment at a more vulnerable stage of disease. As such, our approach to develop an ultra-sensitive method to detect cancer DNA in the bloodstream has significant clinical implications.