Clinical relevance of blood-based ctDNA analysis: mutation detection and beyond

Abstract

Cell-free DNA (cfDNA) derived from tumours is present in the plasma of cancer patients. The majority of currently available studies on the use of this circulating tumour DNA (ctDNA) deal with the detection of mutations. The analysis of cfDNA is often discussed in the context of the noninvasive detection of mutations that lead to resistance mechanisms and therapeutic and disease monitoring in cancer patients. Indeed, substantial advances have been made in this area, with the development of methods that reach high sensitivity and can interrogate a large number of genes. Interestingly, however, cfDNA can also be used to analyse different features of DNA, such as methylation status, size fragment patterns, transcriptomics and viral load, which open new avenues for the analysis of liquid biopsy samples from cancer patients. This review will focus on the new perspectives and challenges of cfDNA analysis from mutation detection in patients with solid malignancies.

Fig. 1. Different features of ctDNA and potential clinical implications.

This figure summarises the tumour-relevant clinical information that can be gained from the study of different features of cfDNA. Somatic genomic aberrations detectable on ctDNA include mutations, chromosomal rearrangements and copy number aberrations. Additional features characteristic for ctDNA are specific epigenetic aberrations like methylation patterns or different DNA fragment lengths. Information on tumour-specific transcription can also be obtained from ctDNA analysis by reading the inter-nucleosome depth coverage. In virus-induced tumours (e.g. EBV-related nasopharyngeal carcinomas or HPV-related head and neck tumours), the quantitative assessment of virus sequences have diagnostic validity. TSS transcription starting sites.

Fig. 2. Clinical applications of genome-wide fragmentation analysis of cfDNA in cancer patients.

Analysis of length distribution of cfDNA has revealed that cancer patients present a more fragmented pattern (and consequently shorter fragments) than healthy donors. This feature can be leveraged to detect cancer without previous knowledge of genomic aberration but also to enhance sensitivity of mutation detection when monitoring tumour evolution. cfDNA coverage signal around TSS correlates with gene expression. Actively transcribed promoters at TSS display low nucleosome occupancy (that is translated in very low read numbers of cfDNA fragments) flanked by well-phased nucleosomes (translated in relatively high and well-phased read numbers of cfDNA fragments). Nonetheless, the region around an active TSS exhibits an overall lower coverage in comparison to inactive TSS promoters, which exhibit an increased coverage signal indicative of denser nucleosome packaging. Therefore, unravelling nucleosome occupancy at promoters from plasma DNA sequencing might help inferring expression levels of genes in the contributing cell types. cfDNA fragment ends pattern reflects nucleosome-depleted region and well-phased nucleosome arrays around the tissue-specific open chromatin region. This analytical approach allows by comparison of nuclear DNA from tissues for the determination of the relative contributions of various tissues in plasma DNA. For the design of our Figure, we were inspired by the figures in the publications of Van der Pol et Mouliere and Murtaza et Caldas.