Researchers from the University of California Los Angeles have developed a novel assay that can overcome the challenges of detecting cell-free (cf)DNA as an early indicator of cancer in a cost-effective way.
Cell-free DNA describes small, freely floating DNA fragments in the bloodstream that, in the context of cancer, have been cleaved from a tumor during apoptosis or necrosis. Normally, the fragments are removed by macrophages, but the overproduction of cells in cancer leaves more of the cfDNA behind, making it a good, non-invasive marker of disease.
Liquid biopsy detection of cfDNA is becoming increasingly relevant in clinical oncology with potential applications including cancer screening, early diagnosis, therapeutic evaluation, monitoring disease progression, and determining disease prognosis.
It is not without challenges, however, particularly for the detection of early cancer where the amount of cfDNA might be low and the molecular heterogeneity of cancer makes it difficult for researchers to find what they are looking for.
To address these challenges, Xianghong Jasmine Zhou, professor of pathology and laboratory medicine at UCLA, and colleagues have developed a methylome-based test that can cover the broad landscape of different cancer methylation markers and exploit as many cfDNA fragments in the blood as possible.
Methylome-based tests look for DNA fragments that have had a methyl group added to a cytosine or adenine nucleotide. This regulatory biologic process can change the activity of a DNA sequence without affecting the genetic code.
“The genome-wide methylation information of cfDNA would be very helpful for cancer detection, but the cost is very high, which is why we developed a new cost-effective method to sequence the cfDNA methylome,” Zhou explained.
Focusing on this small area makes the test “dramatically cheaper” than traditional whole genome bisulfite sequencing while still capturing more than 90% of CpG islands in the whole genome, wrote Zhou and co-authors in Nature Communications.
The test, which is combined with a computational method to extract information from the DNA sequencing to aid early detection and diagnosis, was validated on blood samples from 217 people with colon, liver, lung, or stomach cancer and 191 cancer-free controls.
The researchers report that the model had a sensitivity and specificity of 80.7% and 97.9%, respectively, for detecting cancers of any stage and a sensitivity of 74.5% for early-stage cancers.
The tissue-of-origin was correctly identified in 89.1% of all-stage cancers and 85.0% of early-stage cancer, and Zhou said that these results “compare favorably with other state-of-art methods.”
She added that the team are now looking to conduct a large clinical trial of the test and seek regulatory approval, ultimately aiming to “provide highly accurate, non-invasive, and cost-effective test to facilitate the early cancer detection for the broad population.”
