Abstract 1960: Evaluation of mono- and di-nucleosomal cell-free DNA in fragmentome analyses

Abstract INTRODUCTION: Liquid biopsies are promising, noninvasive tools for cancer detection. Genome-wide fragmentomes reflect the nucleosomal origin of cell-free DNA (cfDNA) in the blood. Variations in cfDNA concentrations and characteristics of mono-nucleosomal (mono) compared to di-nucleosomal (di) cfDNA have not been well described. METHODS: We evaluated liquid biopsies from 356 individuals in the LUCAS cohort (Mathios et al, 2021) including 146 without cancer, 85 with benign lung lesions, and 125 with lung cancer. Plasma (2.2-4 mL) was isolated from whole blood, cfDNA was extracted, and genomic libraries for whole-genome sequencing (WGS) were prepared. WGS was performed on the Illumina NovaSeq6000 S4 at 10X coverage. Mono and di cfDNA concentrations were quantified post-extraction and post-library preparation using the Agilent Bioanalyzer at each peak, and post-sequencing using fragment sizes of 0-250bp (mono) or 251-500 bp (di). Chromosomal representation and fragmentation patterns in cfDNA were analyzed using z-scores and short-to-long fragment ratios, respectively. The DNA Evaluation of Fragments for early Interception (DELFI) approach for lung cancer detection in high-risk individuals was used to evaluate total extracted cfDNA concentration (ng/mL) as an additional feature. RESULTS: The fraction of mono to total cfDNA per mL of plasma increased significantly from cfDNA extraction (a), to library preparation (b), to sequencing (c) in individuals without cancer (median=0.77(a), 0.87(b), 0.92(c); IQR=0.71-0.83(a), 0.85-0.90(b), 0.91- 0.94(c); p<0.0001 (a vs b and a vs c)), in those with benign lung nodules (median=0.78(a), 0.88(b), 0.93(c); IQR=0.72-0.85(a), 0.84-0.90(b), 0.92-0.94(c); p<0.0001 (a vs b and a vs c)), or in patients with lung cancer (median=0.76(a), 0.87(b), 0.92(c); IQR=0.71-0.82(a), 0.85-0.91(b), 0.91-0.94(c); p<0.0001 (a vs b and a vs c)). Changes appeared to be driven by decreases in di cfDNA across experimental steps in all individuals. Mono fragment sequences more accurately reflected copy number changes in individuals with or without cancer compared to di fragments. Inclusion of cfDNA concentration (ng/mL), reflective of both mono and di cfDNA, as an additional DELFI feature did not improve lung cancer detection (AUC=0.93, 95% CI: 0.88-0.97) compared to fragmentation and copy number features alone (AUC=0.93, 95% CI: 0.89-0.97). CONCLUSION: We observed that mono cfDNA was retained, while di cfDNA was largely lost throughout experimental steps. Copy number changes were more apparent from sequenced mono cfDNA compared to di cfDNA. Including total cfDNA concentration as an additional feature to DELFI did not enhance performance. Citation Format: Sarah Short, Shashikant Koul, Akshaya V. Annapragada, Dimitrios Mathios, Jamie E. Medina, Zachariah H. Foda, Carolyn Hruban, Elaine J. Chiao, Vilmos Adleff, Robert B. Scharpf, Victor E. Velculescu, Jillian Phallen. Sarah Short, Shashikant Koul, Akshaya V. Annapragada, Dimitrios Mathios, Jamie E. Medina, Zachariah H. Foda, Carolyn Hruban, Elaine J. Chiao, Vilmos Adleff, Robert B. Scharpf, Victor E. Velculescu, Jillian Phallen [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1960.