Detection of Copy Number Aberration and Tumor Fraction in Archival Cervical Specimens from Ovarian Cancer Patients using Shallow Whole Genome Sequencing.

Ovarian cancer, often called the silent killer due to its diffuse symptoms at early stage, poor prognosis after treatments and high mortality, is also a heterogeneous disease consisting of different histological subtypes with potentially different origins. About 90% of all cases derive from epithelial cells and high-grade serous ovarian carcinoma (HGSOC) is the most common and aggressive form of epithelial ovarian cancer. Recent data indicate the p53 signature lesions and serous tubal intraepithelial carcinomas (STICs) in the fallopian tube are likely to be the common origin of HGSOC, and neoplastic cells containing TP53 somatic mutations could be detected in the cervical specimens collected from 20 months to 6 years before the diagnosis. Our ongoing project is to validate pre-diagnostic cervical specimens from HGSOC by using shallow whole genome sequencing (sWGS), which can detect copy number aberrations (CNAs) even in preserved tumor DNA samples with advantages of low cost, high multiplex and easy data handling. The sWGS will be performed on Illumina sequencing platforms and the sequencing data will be processed with BWA (alignment), SAMtools (cleanup), Picardtools (duplicate) and gatk (BQSR). QDNAseq will be used for the downstream copy number analysis, and GISTIC2.0 for identifying focal gain and loss region. It will be a challenge to estimate the tumor purity and ploidy on the scarce amount of ovarian tumorigenic precursors in the cervical specimens, and we will need to use a probabilistic graphical model without a priori information from normal fallopian tissue to estimate tumor fraction, corrected CNAs as well as tumorigenic copy number signatures. We hope the sWGS approach will allow us to study and detect the early onset of ovarian cancers on large population based cervical screening in a non-invasive and cost-efficient manner. (Less)