Abstract 5019: Comprehensive pan-cancer analysis of cfDNA methylation marks in tumors reveals complex epigenetic regulatory circuits and diagnostic biomarkers

Abstract Early detection of tumor activity in clinically asymptomatic patients has the potential to improve quality of life and prevent unnecessary exposure to toxic therapies. DNA methylation is one of the most studied, stable, and fundamental epigenetic marks and represents a prominent alteration in most cancers. Analysis of DNA methylation, particularly in liquid biopsies has been introduced in cancer diagnosis and risk stratification. Despite indisputable achievements, translation into clinical practice is lagging behind, primarily due to 1) challenges associated with the analysis of DNA methylation, 2) the fragmented nature of cell-free DNA (cfDNA), aggravated by bisulfite treatment, 3) lack of clinical validation of reported cfDNA methylation marks, and 4) the paucity of knowledge about the functional consequences of cfDNA methylation marks within tumors. Using a sample pooling strategy, complemented with robust whole genome and reduced representation methylation sequencing, we establish a pan-cancer cfDNA methylation resource encompassing 12 cancer entities with diverse clinical phenotypes and treatment modalities. We built this compendium using a robustly performing whole genome and reduced representaion enzymatic methylation sequencing of more than 150 cfDNA and tumor DNA. We complement our data by analyzing DNA methylation data from 15 cancer entities (more than 7, 000 patients) within the TCGA. We developed an enzymatic digital PCR (dPCR) approach for pan-cancer but also entity-specific analysis of cfDNA methylation at single base-resolution. Using this dPCR assay, we analyzed cfDNA and tumor DNA samples from more than 400 patients and compared enzymatic digestion with bisulfite treatment. We demonstrate that this resource can allow for the identification of entity-agnostic and specific markers, and that cfDNA methylation patterns are mirrored in tumor samples. The approach can be applied for methylation profiling of yet unanalyzed entities. Finally, we demonstrate unconventional epigenetic regulation of gene expression by methylated DNA-binding transcription factors whose activities are tissue- and context-specific and dosage-dependent. This work, therefore, provides a reference resource to identify minimally invasive, entity-specific and pan-cancer markers applicable for diagnostics, surveillance and patient stratification. It also sets the stage for further investigation of tissue- and context-specific epigenetic regulation of gene expression in health and disease by methylated DNA-binding transcription factors. Citation Format: Smiths Sengkwawoh Lueong, Martin Metzenmacher, Gregor Zaun, Gina Lauren Mayer, Erik Jan Hemmer, Katharina Lückerath, Kelsey Pomykala, Balazs Hegedüs, Peter A. Horn, Marija Trajkovic-Arsic, Tibor Szarvas, Renáta Váraljai, Corinna Keup, Ingeborg Tinhofer, Stephen George, Sabine Kasimir-Bauer, Samuel Peña-Llopis, Alexander Roesch, Cornelius Kürten, Lukas Boosfeld, Kirsten Bruderek, Christopher Darr, Thomas Hilser, Viktor Grünwald, Sven Brandau, Boris Hadaschik, Hans Neubauer, Irene Irene Esposito Irene Esposito, Tanja Fehm, Csilla Oláh, Anita Csizmarik, Fabinshy Thangarajah, Laura Reetz, Ghanam Jamal, Basant Kumar Thakur, Halime Kalkavan, Alexander Schramm, Martin Schuler, Jens T. Siveke. Comprehensive pan-cancer analysis of cfDNA methylation marks in tumors reveals complex epigenetic regulatory circuits and diagnostic biomarkers 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 (6Suppl): Abstract nr 5019.