Cancer is the leading cause of death worldwide and was responsible for nearly 10 million deaths in 2020. Early diagnosis remains the most effective strategy to improve survival outcomes, yet population-wide screening programs currently exist for only a limited number of cancer types. As a result, many cancers are detected at advanced stages, when therapeutic options are restricted, and survival rates are much lower.
Recent advances in liquid biopsy technologies have demonstrated that cancer-derived signals can be detected in blood. Current non-invasive diagnostic approaches primarily focus on identifying rare tumor-specific mutations or DNA methylation patterns in circulating cell-free DNA (cfDNA). Although promising results have been reported, particularly for methylation-based cancer detection, three major limitations hinder broader clinical implementation. First, tumor-derived cfDNA molecules are typically present at very low abundance compared to cfDNA originating from normal blood cells, leading to limited sensitivity and increased false-positive rates. Second, mutation- or methylation-based assays do not directly provide information about tumor gene expression programs, which are essential for molecular classification and therapeutic stratification. Third, current solutions are costly due to the deep sequencing coverage required for unbiased detection of cancer-associated DNA methylation patterns or cancer mutations above biological noise.
Cell-free DNA fragments in circulation are predominantly associated with histone proteins in the form of nucleosomes rather than existing as ‘naked’ DNA. These circulating cell-free nucleosomes (cf-nucleosomes) carry histone and DNA modifications that reflect the gene regulatory state of their tissue of origin. In the human genome, combinations of histone modifications and DNA methylation patterns define cell-type-specific chromatin states, which are tightly linked to gene activity. During cancer development, extensive reprogramming of gene expression is accompanied by characteristic alterations in these epigenomic patterns.
This project builds on the hypothesis that cf-nucleosomes contain rich regulatory information that can be leveraged to enhance cancer detection and classification. By selectively analyzing chromatin-associated signals in blood, the project aims to increase the proportion of disease-relevant molecular signatures compared to conventional cfDNA approaches. Specifically, EpiCBlood investigates complementary strategies to identify cancer-associated chromatin patterns and to enable tissue-of-origin detection and molecular tumor classification from blood samples.
By shifting the focus from static genetic alterations to functional epigenomic signatures, the project seeks to establish a new conceptual framework for minimally invasive cancer diagnostics. If successful, this approach could improve cancer detection sensitivity, enable biologically informed tumor stratification, and contribute to more precise and cost-effective screening strategies in oncology.
