504 Background: Current approaches based on invasive biopsy genetic analysis can fail to capture an accurate picture of the real-time cancer profile due to limited spatial window of biopsy into residual disease throughout the body. Moreover, tumor rebiopsy has significant challenges to be widely used in practice for serial monitoring of disease progression or acquired resistance. Analysis of circulating tumor nucleic acids (ctDNA), on the other hand, presents a new tool for the monitoring and treatment of cancer. However, due to high-quality false positives in current NGS assays, the majority of studies on ctDNA have been limited to hotspot analyses, and typically only involve patients where ctDNA fractions are high (>1-5%). Methods: We have developed a differentiated sequencing assay, Digital Sequencing Technology (DST) that enables detection of rare genomic abnormalities with ultra high-specificity and sensitivity. Our assay is able to eliminate the error and distortion created by sample-prep and sequencing processes in standard NGS workflows and produce near-perfect representations of all rare variants. Results: We have shown that in sequencing a comprehensive cancer panel of 80kbp in 0.1% cancer cell line titration samples, standard Illumina SBS generates many high-quality false positive variant calls in the range of 0.05-5%, while our assay resulted in highly sensitive and completely error-free variant calls across the entire panel. We then profiled ctDNA and matched primary tumor samples in more than 50 plasma samples collected from metastatic colorectal cancer patients. We found mutations in ctDNA with allele frequencies in the range of 0.05-40%. We established the concordance of ctDNA mutations with different clones within primary tumors. We further investigated the potential clinical usefulness of discordant mutants based on the treatment history of each patient. Conclusions: This work indicates the remarkable potential of using our assay in deep analysis of ctDNA, thereby allowing researchers and clinicians to comprehensively and non-invasively monitor the genetic dimension of cancer throughout the body.
KRAS mutation analysis: a comparison between primary tumours and matched liver metastases in 305 colorectal cancer patients