Cell-free DNA (cfDNA) in peripheral blood contain tumor-derived DNA and are promising non- invasive bi- omarkers for cancer therapy selection and recurrence monitoring. However, the high cost of current commercial next-generation sequencing (NGS) panels for profiling cancer mutations in cfDNA prevents them from being routinely used. Furthermore, the high sequencing depth and reads required means that cfDNA profiling is es- sentially only possible in specialized centralized reference laboratories and inaccessible to hospitals with lower- throughput sequencing instruments. Here, we propose to use allele-selective enrichment sequencing (ASES) to develop NGS panels that are capable of profiling rare cancer mutations in cfDNA with less than 0.1% variant allele frequency (VAF) using low-depth sequencing. We achieve this using allele-specific toehold probes that deplete wildtype DNA sequences and/or enrich for known mutations, increasing the VAF of rare cancer mutations by roughly 100-fold. Unlike other hy- bridization- based probes, toehold probes discriminate single nucleotide variants based on molecular competition rather than melting temperature optimization, and thus is uniquely scalable to high multiplexing. We have ob- tained preliminary data on a NGS panel that includes 118 amplicons, and observed similar mutation sensitivity using 98% fewer NGS reads. In this Phase I SBIR application, we will first develop an ASES NGS panel for cfDNA profiling covering 7 genes bearing 150 frequent mutations in non-small cell lung cancer patients. Subsequently, we will scale up to a 150 kb hypothesis-free panel covering 50 genes.
Noninvasive profiling of cancer mutations in cell-free DNA can inform optimal target therapy selection and pro- vide timely early warning of cancer recurrence, but is currently not economically feasible due to the high sequencing depths required. We have develop Allele-Specific Enrichment Sequencing (ASES), in which rare cancer mutations can be reliably and economically analyzed using low-depth sequencing. In this SBIR Phase I proposal, we will first show feasibility on a 7-gene lung cancer ASES panel targeting 150 recurrent mutations, and then demonstrate a hypothesis-free ASES panel covering the full exon regions of 50 genes (150kb).
