Poster Presentation NSW State Cancer Conference 2023

Development of a shortlist of tumour-specific methylated DNA regions with potential as a blood-based test for early ovarian cancer diagnosis (#171)

Kristina Warton 1 , Teagan Fisher 1 , Elyse Powell 1 , Bonnita Werner 1 , Nicole Yuwono 1 , Jennifer Duggan 2 , Jason Abbott 1 2 , Ramanand Athavale 1 2 , Caroline Ford 1
  1. University of New South Wales, Kensington, NSW, Australia
  2. Royal Hospital for Women, Sydney, NSW, Australia

Ovarian cancer (OC) has no screening test, thus, is typically diagnosed late. An early detection test has potential to improve outcomes by allowing surgical removal of the tumour before it spreads. Detection of tumour-derived cell-free DNA (cfDNA) in blood presents an opportunity for OC screening. Previously, we identified 73 regions of DNA that are differentially methylated in OC and are promising biomarkers for an early detection test.

This study aimed to develop methylation-specific qPCR (MS-qPCR) assays for the 73 biomarkers and narrow these down based on their sensitivity and specificity by testing the assays on healthy and OC samples to determine the best performing biomarkers specific to OC.

MS-qPCR assays were designed for the biomarker DNA regions. Specificity was optimised using fully-methylated and fully-unmethylated DNA. Next, assays were tested on buffy coat from healthy women (n=30), and women with benign gynaecological conditions (n=16) to eliminate regions that were methylated in cancer-free samples (more than 0.5% methylated DNA). Assays that passed this initial screening phase were tested in OC ascites (n=16), which is known to contain a high proportion of tumour DNA. Assays were considered positive if more than 25% of the target was methylated in the cancer biospecimens. The 8 biomarkers that were positive in the highest proportion of cancer samples were tested in matched plasma and buffy coat from women without OC (n=50) to eliminate false positives.

Primers for 138 MS-qPCR assays were designed, covering 52 of the differentially methylated biomarker regions. After optimisation, 58 assays covering 35 regions were found to be methylation specific on fully-methylated DNA and progressed to testing on buffy coat from women without OC. Here, 10/58 assays were removed due to false positive signal. The remaining 48 assays covering 29 regions were tested on OC ascites samples. Of these, 11 regions had assays that were positive in at least 50% of samples. Combining the cancer-free buffy coat and the OC results, the best 8 regions were selected and tested on matched healthy plasma cfDNA and buffy coat DNA. Two of the regions were methylated in 80% and 84% of healthy plasma samples and were removed from the biomarker shortlist.

We have identified 6 promising biomarkers of OC based on methylation patterns in cancer-free blood and OC ascites. Next, these assays will be optimised for sensitivity, and tested in samples from an expanded cohort of women with OC to develop an early detection test.