Preview

PDF, English Download (7MB) | Terms of use

Abstract

Compared to tissue biopsies, ctDNA provides a more comprehensive tumor landscape, allowing for repeated non-invasive sampling. Therefore, ctDNA detection has a broad application prospect in tumor diagnosis, treatment and monitoring. However, due to the low content of ctDNA in cfDNA, ctDNA detection requires a high sequencing depth to achieve high sensitivity. Currently, the commonly used ctDNA testing scheme is panel sequencing with high coverage, so that genes of interest can be tested at a lower cost. However, panel sequencing has limited ability to detect other variants such as SV and CNV. Another option is lcWGS. lcWGS is able to identify CNVs, providing valuable insights into genomic alterations. Based on the data of HIPO-K34 and INFORM, this study explored the ability of ctDNA detection by the two schemes. The HIPO-K34 project focused on patients with ALK gene-fused non-small cell lung cancer (NSCLC) and contains multi-time point sequencing data from lcWGS and panel sequencing. The INFORM project consists of liquid biopsy samples and tissue samples taken from the same patient at the same time point. In addition, various detection tools were benchmarked using simulated data with known tumor DNA fractions and CNVs. In order to improve the detection performance, the tools were optimized and the tool with the best performance was selected. Finally, a pipeline combining the panel analysis process with the optimized lcWGS analysis process was established for the accurate analysis of liquid biopsy samples.