Cell-free DNA sequencing and methylation detection — promising potential for non-invasive cancer monitoring

In 2022, there were approximately 9.7 million cancer-related deaths worldwide¹, and it is estimated that this will increase to 18.6 million by 2050². Early diagnosis is critical, but diagnostic methods currently require invasive and time-consuming tissue biopsies³.

Liquid biopsy could enable non-invasive cancer detection and surveillance through the analysis of cell-free DNA (cfDNA), including circulating tumour DNA (ctDNA) shed from tumours. However, detecting low-abundance ctDNA is challenging due to the low fraction of tumour-derived DNA in blood⁴.

In this case study, we highlight how researchers have used Oxford Nanopore sequencing to extract multiple layers of biological information from a single liquid biopsy research sample. By analysing tumour-derived cfDNA, including methylation and fragmentation simultaneously, these researchers are uncovering new insights into tumour biology and exploring novel approaches for cancer detection and monitoring.

Revealing epigenetic signals in cfDNA

Building on early demonstrations that nanopore sequencing could detect copy number alterations in cfDNA⁵, Katsman, Orlanski, and Martignano et al. showed that shallow nanopore whole-genome sequencing (WGS) could simultaneously detect methylation, fragmentation patterns, and cell-of-origin signals from liquid biopsy samples⁶.

Methylation is critical to capture from cfDNA, as it offers multiple insights. Aberrant methylation is associated with cancer, and methylation patterns can be used to detect the turnover of damaged cells⁶ and their tissue of origin⁴. However, research in this area has previously been limited because traditional methylation detection methods require chemical conversion steps that can affect sample quality, strip methylation, and obscure fragmentation patterns^3,6.

Katsman, Orlanski, and Martignano et al. also performed downsampling experiments to better understand the impact of low sequencing depth on Oxford Nanopore and Illumina sequencing and found that results were consistent between sequencing methods, even at a low sequencing depth of 0.2x coverage⁶.

Overall, DNA methylation, fragmentation, and copy number aberration profiles were broadly concordant between Illumina WGS, whole-genome bisulfite sequencing datasets, and Oxford Nanopore data, highlighting the utility of nanopore sequencing for liquid biopsy applications⁶.

Tracking tumour evolution through liquid biopsy

To demonstrate that cfDNA methylation analysis has the ‘potential to impact liquid biopsy diagnostics for cancer detection and characterisation’, Lau et al. developed a single-molecule methylation classifier to distinguish between tumour- and immune-derived cfDNA. They used Oxford Nanopore cfDNA sequencing to detect cancer-associated methylation profiles from liquid biopsies at scale and applied the classifier to longitudinal cancer monitoring as a proof of concept⁷.

By analysing blood research samples from patients with cancer over three years, the team found cancer-specific cfDNA molecules correlated with clinical events, such as tumour response and recurrence. For example, they observed a significant increase in reads with tumour-specific methylation changes when new metastases emerged⁷ (Figure 1), highlighting the potential clinical utility of monitoring cfDNA during cancer treatment.

Extending methylation analysis beyond blood-derived biopsies, Chen et al. expanded cfDNA analysis to cerebrospinal fluid (CSF) research samples to unravel methylation patterns specific to brain metastasis in patients with non-small-cell lung cancer (NSCLC)⁸. Brain metastases are associated with poor prognosis and are common in patients with NSCLC, meaning there is considerable interest in identifying biomarkers that could support earlier detection⁸.

In this first study to characterise NSCLC-specific fragmentation and hydroxymethylation profiles from CSF-derived liquid biopsies, Chen et al. discovered distinct methylation and hydroxymethylation patterns between research samples from patients with brain metastases compared with healthy controls⁸. These findings suggest that cfDNA-derived epigenetic signatures may represent promising biomarkers for early brain metastasis detection and disease monitoring.

Uncovering hidden fragmentation biomarkers

Fragmentation patterns in cfDNA are increasingly recognised as a valuable source of cancer biomarkers, with fragmentation hotspots revealing cell type of origin based on cell-type-specific positioning of nucleosomes. However, the cause of altered patterns in patients with cancer is poorly understood⁹.

Fragmentation can occur by both cellular and circulating endonuclease activity, creating fragmented cfDNA of approximately 167 bp with specific fragment end motifs^10,11. However, fragmentomics has previously only been investigated with short-read sequencing, limiting analysis to fragments of less than 250–500 bp^8,10, reflecting fragmentation patterns of only mono-nucleosomes, which fragment cfDNA to their characteristic 167 bp length. With nanopore reads of unrestricted length, researchers can capture longer cfDNA fragments, revealing previously unknown distinct fragmentation profiles caused by di- and tri-nucleosomes⁸.

Using nanopore sequencing, Berman et al.¹⁰ provided a systematic survey of fragmentomic alterations, identifying four major classes using nanopore data alone (Figure 2). The team characterised the different classes of fragmentomic patterns, such as hyperfragmentation, which potentially reflects an inflammatory response rather than a cancer-specific fragmentation mechanism. By improving our understanding of fragmentomic variation, this work could help researchers develop more informative biomarkers while uncovering previously hidden aspects of tumour biology.

Together, these studies demonstrate how nanopore sequencing provides multiomic insights in one go from a single liquid biopsy sample. By simultaneously analysing tumour-derived cfDNA molecules, their methylation patterns, and fragmentation, researchers are gaining more insights into tumour biology and advancing new approaches for earlier detection and disease monitoring to tackle cancer effectively.

Oxford Nanopore Technologies products are not intended for use for health assessment or to diagnose, treat, mitigate, cure, or prevent any disease or condition.