Using Nucleosome Levels, Copy Number Alterations, and DNA Methylation to Profile Disease State Through Liquid Biopsy.

e15014 Background: Nucleosomes are the repeating unit of chromatin that contain important signals for proper genomic function and transcriptional activity. Upon cell death, chromatin is fragmented, and nucleosomes are released into circulation and thus, detectable in liquid biopsy. Cell death naturally occurs over time and increases in a variety of diseases resulting in elevated levels of circulating nucleosomes and the genetic and epigenetic signatures they carry. These signatures contain important information about the cells from which they were derived, as well as the reason for them being in circulation. Methods: We measured circulating nucleosome levels, using Volition’s Nu.Q H3.1 sandwich ELISA across a healthy cohort spanning 40-85 years of age as well as patients whom had been diagnosed with Non-Hodgkin's Lymphoma (NHL), some of which were untreated and others that had undergone treatment. Additionally, to demonstrate that the correlation between nucleosome (Nu.Q) levels and copy number alterations is specific to cancer, we assessed the genomic integrity of an NHL sample with a Nu.Q level of 744 ng/ml and a sepsis sample with Nu.Q level of 3,281 ng/ml using low coverage Oxford Nanopore whole genome sequencing to identify genomic amplifications and deletions. We further used cell type specific methylation patterns to identify the cell of origin of cfDNA in each sample. Results: Importantly, we found no significant correlation between nucleosome level and age in our healthy cohort (R2= 0.0004; 40-49: N = 10, mean = 39 ng/ml; 50-59: N = 10, mean = 18 ng/ml; 60-69: N = 10, mean = 35 ng/ml; 70-79: N = 15, mean = 34 ng/ml; 80-85: N = 5, mean = 32 ng/ml). We did however find elevated nucleosome levels in untreated NHL patients (N = 4, mean 646 ng/ml) and that nucleosome levels in NHL patients that had undergone treatment were reduced (N = 29, mean = 180 ng/ml). Using copy number changes, we found that the fraction of cfDNA estimated to be derived from the tumor in the NHL sample was 54% and were able to detect amplifications in chromosomes 10, 11 and 18 and deletions in chromosomes 1, 8, 17, 18 and 23, whereas we found no copy number alterations present in the cfDNA sample derived from a sepsis patient and a tumor fraction of 0. Furthermore, using cell type specific methylation patterns to deconvolute the data we found that a large fraction of the cfDNA in the NHL sample was derived from B-cells consistent with the NHL sample being of B-cell origin whereas we determined that the cell of origin of the cfDNA in the sepsis sample was Neutrophils. Conclusions: This data shows that nucleosome levels, as measured by Nu.Q, can be used in conjunction with shallow sequencing and DNA methylation profiles to define tumor fraction and cell of origin in liquid biopsy.