Circulating fragments of cell-free DNA in the blood could signal the presence of many diseases and provide an indication of clinical outcomes, research suggests.
The findings, in Science Translational Medicine, provide hitherto unrecognized possibilities for genome-wide, cell-free (cfDNA) fragmentation in liquid biopsies.
This “fragmentome” of genetic information, found in the circulation after being released from dying cells, has already shown promise for flagging cancer.
Now researchers have found that it can noninvasively detect liver disease, and they suggest it could help with the unmet clinical need for accessible biomarkers to enhance early detection across a wide spectrum of diseases.
“The disease specificity of the approach and the ability to extract genome-wide fragmentome information related to multiple diseases from a single blood sample could one day enable a facile and accessible liquid biopsy test for noninvasive, multi-disease screening,” according to researcher Victor Velculescu, PhD, from Johns Hopkins University School of Medicine, and colleagues.
Plasma biomarkers based on cfDNA have shown promise for cancer detection, identifying response to therapy and tumor recurrence, prenatal testing, and monitoring of organ transplantation.
The genome-wide size, coverage distribution, and genomic content of cfDNA fragments reflect the underlying genomic, chromatin, transcriptomic and epigenomic states of the cells from which they originate.
The researchers therefore reasoned that cfDNA fragmentomes could serve as noninvasive or minimally invasive biomarkers for complex illnesses that lack effective screening, including common diseases of inflammatory or fibrotic origins that become more common with age.
To investigate further, Velculescu and team applied whole-genome sequencing to cfDNA fragmentomes from 1576 people, including those with liver disease, autoimmune disorders, or neurodegenerative conditions.
A machine learning algorithm was able to identify markers in the fragmentome that distinguished early liver disease, advanced fibrosis, and cirrhosis, they reported.
The markers displayed high sensitivity when applied in a separate discovery group of 423 individuals and a validation cohort of 221 individuals.
The authors note that these changes reflected both target organ contributions to the cfDNA as well as immune-mediated changes in white blood cell content in cfDNA.
Changes in the fragmentome were also linked with heart disease, diabetes, and other chronic disorders, and the machine learning model could predict overall survival in a discovery cohort of 571 people as well as a validation cohort of 231 individuals.
The researchers therefore suggested that changes in the cfDNA fragmentome appear to be present across additional conditions beyond liver disease and can serve as a measure of clinical morbidity.
They added: “The genomic location, extent, and type of these alterations vary by disease state and may be used for noninvasive screening of multiple pathologies with high performance and limited cross-reactivity.”
