Abstract
Circulating tumor DNA (ctDNA) is a biomarker that could potentially improve the survival rate of ovarian cancer (OC), e.g., by monitoring treatment response and early relapse detection. However, an optimal method for ctDNA analysis in OC remains to be established. We developed a method for tumor-informed single-nucleotide variant detection of ctDNA in OC using whole-genome sequencing. Tumor and plasma samples obtained at the time of diagnosis from 10 patients with OC were included. The tested method involved applying basic filters with different cut-offs of read depth, allelic depth, and variant allele frequency of tumor and normal DNA. In addition, we applied a new filtering approach using plasma samples from the other included OC patients (the plasma pool) for specific removal of artefacts. The basic filters with varying cut-offs showed minor improvement in signal-to-noise ratio (S2N). However, the addition of the plasma pool filter resulted in a considerable ctDNA signal improvement, indicated by both S2N and z-score. This study demonstrates a promising method for ctDNA detection in OC patients using a tumor-informed approach for whole-genome sequencing. Despite the limited number of patients involved, the results suggest a significant potential of the method for ctDNA signal detection in patients with OC.