"We've created the first pan-cancer DNA sequencing panel that we know to be optimised for cell free DNA (cfDNA) in paediatric solid tumours" - Reda Stankunaite, Institute of Cancer Research
Paediatric cancers are the leading cause of disease-related deaths among children who survive past infancy in the Western world. A substantial proportion (43-87%) of these tumors have potentially actionable alterations that if detected could guide treatment. Whilst solid tumor profiling is well advanced for children, liquid biopsy profiling, lags behind. To address this, a group at the Institute of Cancer Research recently published the validation of a pan-cancer sequencing panel for liquid biopsy molecular diagnostics which could be used to profile tumor DNA in the blood of paediatric cancer patients. This opens up the possibility to routinely sample patients, monitoring their response to treatment and following their disease progression to detect relapse as early as possible.
Nonacus spoke to Reda Stankuniate from the Institute for more information:
Dr Mike Hubank, Reda Stankunaite and Professor Louis Chesler from the Institute of Cancer Research
Where does your interest in minimally invasive liquid biopsy tests derive from?
Unlike tissue sampling, which can only really be done once or twice on any individual, blood-based testing offers the possibility of serial sampling of a patient throughout their treatment. This allows us to monitor the response of a patient to treatment, and potentially spot resistance and relapse earlier than we can at the moment. It also offers the possibility for us to profile tumors where a tissue biopsy is difficult or undesirable.
Why did you decide to design a panel specifically for children?
There are differences between the landscape of genomic alterations in adult and childhood cancers and a number of variants that we know are enriched in childhood cancers are not present in commercially available adult panels. We wanted to create a panel that would specifically target genes that contained the most clinically relevant variants in most common paediatric solid tumors.
You already had a panel for paediatric solid tumor sequencing, why did you need a different panel for liquid biopsy?
We were focusing our liquid biopsy efforts on molecular profiling of cell-free DNA (cfDNA) circulating in the blood. Some of this DNA will contain DNA shed by the tumor ,known as circulating tumor DNA (ctDNA)-this is what we are trying to detect. However, in many cases ctDNA is present in very small quantities often less than 1% of the total cfDNA fraction. Consequently, in order to detect the tumor mutations we need to sequence a cfDNA sample at much greater depth (20,000x) than the 500x depth we would sequence a solid tumor. This has implications on the cost of sequencing, so we wanted to make the panel as small as possible whilst still detecting the most clinically relevant genetic alterations.
How did you decide on the final panel design?
We started with our 91-gene Paediatric Solid tumor panel (currently in routine diagnostic use in the UK) and focused on reducing the size of the panel while maintaining the coverage of the genes most commonly reported in solid paediatric tumors. We focused on known hotspot regions of the oncogenes, with full coverage for important tumor suppressors with predictive, prognostic, and diagnostic SNVs and CNVs. And of course, we gathered opinions from paediatric oncologists! We also used the expertise of scientists at Nonacus to optimise our design ensuring we had sufficient and even coverage of our targets. This in turn allowed us to minimise the amount of sequencing needed per sample keeping our sequencing costs as low as possible. In the end our panel covers a total of 67 genes.
Ultra-deep sequencing can make variant calling challenging due to the background of errors from PCR and sequencing. How did you achieve the sensitivity you needed to detect tumor mutations in cfDNA?
We knew we would need to be able to detect very low frequency variants-perhaps even as low as<0.3%. We used the Nonacus Cell3 Target technology as this not only provides a highly efficient library preparation and capture but uses unique molecular identifiers (UMIs)which combined with background noise suppression and our bioinformatic pipeline allowed us to push the limit of detection for known variants to 0.125% and 0.3% for cfDNA-unique variants.
As well as targeted sequencing, you used low pass whole genome sequencing (lpWGS) for copy number analysis, why?
In a lot of paediatric cancer patients the amount of DNA we can receive from a patient is limited, therefore we wanted to get as much information from each sample as possible. We used the same sequencing library preparation to detect large scale CNVs by lcWGS as well as SNV and indels from the panel sequencing. lcWGS analysis also allowed us to estimate the fraction of ctDNA in a cfDNA sample which helped us with clinical interpretation of the panel results.
What is the most exciting thing in your mind, to come out of this research?
We were able to demonstrate that our ctDNA panel meets clinical diagnostic standards and with more validation could help bring liquid biopsies into the clinical setting. This is really exciting as it has the possibility to enable doctors to make more informed treatment selections and monitor the treatment response of their patients far more closely than they can today. At the end of the day, we want to provide patients with the best treatment for their cancer and give them the best possible outcome.
What are your future plans?
We are working on collecting more evidence and evaluating the concordance between tissue and blood based cfDNA profiling in different paediatric cancer types to be able to bring the test into the clinical setting. Like others before, we were unable to detect variants in the plasma of brain tumor patients, perhaps not surprising as cfDNA doesn't seem to cross the blood-brain barrier. As a third of paediatric solid tumors are brain tumors we are currently collecting CSF samples and hope to find it provides a better source of cfDNA so we can help improve the outcomes for these patients too.
This work was supported by the National Institute for Health Research (NIHR) Biomedical Research Centre at The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research, London. Funding was also provided by CRUK, Christopher's Smile, the Royal Marsden Cancer Charity (RMCC) Royal Marsden CYP Unit and Children's Unit Fund and research nurse funding by an Experimental Cancer Medicines Centre (ECMC) grant to the Paediatric ECMC network.
To read the full paper click here.
To find out more about how molecular barcoding with UMIs can improve sensitivity - read our blog post here.
To find out more about Nonacus custom panels for oncology - go here.
Efficient DNA extraction is critical to successful cfDNA sequencing. To find out how to maximise cfDNA yield (from blood or CSF) - click here.
