Abstract 7003: Microfluidic platform for DNA methylation profiling towards early detection of ovarian cancer

Abstract

Abstract We present a microfluidic platform for molecule-by-molecule detection of heterogeneous epigenetic patterns of rare tumor-derived DNA by highly parallelized digital melt. The platform digitizes DNA molecules in a 4-module, 40,000 1-nL array microfluidic device and observes sequence-dependent fluorescence changes during temperature ramping. The platform is applied to quantify DNA methylation heterogeneity of a biomarker panel within Pap specimens in a small cohort of healthy and ovarian cancer patients, and demonstrates an accuracy of 0.9. DNA hypermethylation of tumor-suppressor genes is an epigenetic phenomena that occurs early in tumorigenesis. Variable methylation patterns are observed in ovarian cancer precursor tissue, and trace amounts of ovarian-tumor-derived DNA can be found in Pap specimens, albeit at low frequencies (<0.01%). This suggests that methylation profiles of DNA extracted from routinely-obtained Pap specimens could contain early cancer biomarkers. However, due to the scarcity of these molecules, current techniques, such as sequencing and digital PCR (dPCR), have technical limitations precluding their ability to profile methylation patterns reliably. Sequencing, although comprehensive, has limited sensitivity, undermining its utility for routine detection of fractions below 0.1%. dPCR achieves high sensitivity, but is limited to known sequences, thus cannot provide comprehensive analysis of molecular heterogeneity. To overcome these limitations, we developed a technique called digital high resolution melt (dHRM), a thermodynamics-based approach to detecting molecular variability by observing the sequence-specific release of a DNA-intercalating dye under a thermal ramp. dHRM can discriminate methylation patterns on CpG-by-CpG basis and detect methylated epialleles at frequencies as low as 0.00005%. Here, DNA from Pap specimens was loaded on a microfluidic chip that digitizes rare target molecules into individual reaction chambers. A thermal-optical platform was developed to perform parallelized dPCR and dHRM. Methylation patterns of a biomarker panel were analyzed to produce a probability score that the Pap specimen contained tumor-derived DNA. 12 healthy and 12 ovarian cancer Pap specimens were assessed for methylation heterogeneity of 9 genes. Optimal methylation density thresholds were determined, and data above each threshold was combined into a single score for each possible combination of biomarkers. In this small cohort, methylation heterogeneity analysis of just 2 loci produced an area under the receiver-operator characteristics curve (AUC) of 0.9. This highly sensitive methylation profiling technology demonstrates promising utility towards early cancer detection. Future work aims to expand the testing cohort for ovarian cancer detection and permit further studies on the impact of methylation heterogeneity on early cancer development. Citation Format: Chrissy O'Hersey, Yang Zhao, Thomas R. Pisanic, Tian-Li Wang, Ie-Ming Shih, Tza-Huei Jeff Wang. Microfluidic platform for DNA methylation profiling towards early detection of ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7003.