Short double-stranded cfDNA: A novel precision biomarker class for cancer diagnostics

Background: Cancer diagnostics relies on highly specific and sensitive biomarkers to detect early stages in cancer progression as well as recurrence of disease (MRD). However, current approaches often fail to fulfill these needs, however, liquid biopsy based on cell-free DNA (cfDNA) and in particular short double-stranded cell-free DNA (footprint DNA) holds great promise for improvements. Derived from interaction of regulatory proteins including transcription factors at DNA-binding motifs, footprint DNA comprises genome-wide DNA footprints from plasma samples. Selective enrichment and quantitative analyses of footprint DNA reveal significantly different signatures for complex diseases diagnostics including various cancer types. Methods: Footprint DNA of more than 200 plasma samples from various tumor types, including colorectal cancer and pancreatic ductal adenocarcinoma, as well as non-cancer controls were analyzed by high-throughput sequencing followed by bioinformatic peak identifications to generate a comprehensive reference annotation for genome-wide liquid footprint sites. Additionally, quantitative evaluation of footprint DNA from individual patient samples was used to differentiate between clinical conditions and benchmarking against standard of care diagnostics. Results: Peak calling revealed approximately 8.6 million different peaks in about 6.5 million footprint regions. 3.3 million peaks overlap with annotated transcription factor bindings sites (TFBS). Moreover, 33,675 promotor regions, 55,406 CTCF sites and more than 100,000 enhancer binding regions are covered by footprint DNA signals. We also found that footprint DNA at defined genomic loci semi-quantitatively correlated with physiological markers like ALT or urea from major organ systems including liver or kidney. In a proof-of-concept cohort, differential footprint DNA biomarkers distinguished between clinical conditions, notably also closely related cancer types of colorectal cancer and pancreatic ductal adenocarcinoma (PDAC). Even same organ diseases (SOD) like pancreatitis and PDAC could be specifically discriminated using footprint DNA biomarkers. Conclusion: Footprint DNA bears strong potential for cancer diagnostics, demonstrated by disease and cancer discrimination as well as differentiation between same organ disease. Footprint DNA enrichment sets the groundwork as identification platform of discriminatory footprint DNA biomarkers, which allows panel-based cancer detection for clinical applications. This opens new possibilities in early cancer detection and monitoring of minimal residual disease (MRD). Comprehensive footprint annotation will therefore provide a strong basis for the diagnostic of cancer and other complex diseases.