115 Molecular analysis for ovarian cancer detection in patient-friendly samples assay of the Agilent 4200 TapeStation System (Agilent) for quality control before sequencing. Sequencing libraries of the first pilot series of urine supernatant DNA were prepared using the ThruPLEX Plasma-seq Kit (Takara Bio, Mountain View, CA, USA) for whole-genome sequencing according to manufacturer’s instructions. The remaining samples were prepared using the NEBNext® Enzymatic Methyl-seq (EM-seq) Kit (NEB, Ipswich, MA, USA). EM-seq was performed according to manufacturer’s guidelines for standard insert libraries with 14 PCR cycles. Libraries were quantified and quality checked using the D1000 ScreenTape Analysis Assay (Agilent) before pooling. Pairedend 150 base pair (bp) libraries were pooled in equimolar amounts and sequenced on a NovaSeq6000 (Illumina) (GenomeScan, Leiden). The processing of sequencing data and subsequent analysis of SCNA and cfDNA fragmentation patterns are provided in the Supplemental Methods. Shallow whole-genome sequencing of paired FFPE primary tumor tissue was performed to verify copy number profile concordance and is also described in the Supplemental Methods. Statistical analysis Methylation levels were expressed as 2log-transformed Cq ratios and presented in violin plots. Tissue methylation levels were compared between two groups using the non-parametric Mann-Whitney U test. Methylation levels of each gene in the remaining sample types were compared between healthy controls and patients diagnosed with a benign or malignant ovarian mass using the Kruskal-Wallis test. In case of a significant Kruskal-Wallis test (p<0.05), this was followed by post-hoc testing of 1) healthy controls versus malignant ovarian masses, and 2) benign versus malignant ovarian masses using the Mann-Whitney U test with Bonferroni correction. The correlation between methylation levels of each DNA methylation marker between paired samples of patients diagnosed with ovarian cancer was assessed using Spearman’s rank correlation. Correlation coefficient r was defined as very weak (r = 0.00–0.19), weak (r = 0.20–0.39), moderate (r = 0.40–0.59), strong (r = 0.60–0.79), or very strong (r = 0.80–1.00) and displayed in correlation matrices. Fragment size profiles were visualized by density plots and analyzed by comparing cfDNA reads of healthy controls and ovarian cancer patients with low (<5%) and high (≥5%) tumor fractions. Data was collected using Castor EDC and analyzed using R (version 4.0.3 with packages: cowplot, corrplot, dplyr, ggplot, ggpubr, and rstatix). P-values are two-sided and considered statistically significant when p<0.05. 5
RkJQdWJsaXNoZXIy MjY0ODMw