Genome Sequencing Shows Clinical Promise in AML and ALL

Liam Davenport

April 02, 2019

GLASGOW — Whole genome sequencing (WGS) could eventually replace the myriad of cytogenetic panels and tests used to detect single and copy number variants in acute myeloid leukaemia (AML) and acute lymphoblastic leukaemia (ALL), potentially improving clinical practice and benefiting patients, results from a UK study suggest.

Shirley Henderson, PhD, lead for cancer molecular diagnostics, Genomics England, compared WGS and standard of care cytogenetic tests in over 30 patients, in whom a total of more than 250 fluorescence in situ hybridisation (FISH) had been performed.

Presenting the findings at the British Society for Haematology Annual Scientific Meeting in Glasgow, she showed that WGS and FISH had percentage agreements of more than 90%, with false positive rates of less than 5%.

Moreover, WGS was able to identify a single variant associated with paediatric ALL that had been missed by FISH.

Henderson said: "We really hope that these early results show that this will work in clinical practices and provide results that are useful clinically."

She added that the hope is that, "as we go on, patients will be able to benefit from the therapeutic and diagnostic opportunities afforded" by WGS.

Henderson underlined that AML and ALL "are fairly fast moving" diseases, with the continual identification of novel biomarkers.

Greater Genomic Stratification

As, by its nature, WGS is able to test for new variants as they are identified, she said this "future-proofing" aspect "could be particularly useful in those conditions".

Henderson added that the extra information generated by WGS will also allow for greater genomic stratification of patients, which, in the context of the "very poor" survival rates in AML and ALL, could offer potential benefits.

Dr Adele Fielding, University College London Cancer Institute, who co-chaired the session, commented: "It's vitally important that the clinical community engage with this and understand both the power and the limitations of this technique, and how this work is going to be interpreted for the benefit of patients.”

Henderson began her presentation by noting that genomic testing is widely used in haematological malignancies, including for diagnosis, risk group stratification, choosing therapy, minimal residual disease and stem cell transplant monitoring, and prognostication.

However, the National Genomic Test Directory lists a huge number of genes that can be tested for in both AML and ALL, covering small variants and structural and copy number variants, as well as, in ALL, complex variants.

She said that there are a number of techniques currently available for detecting small and copy number variants.

These include karyotyping, FISH, various forms of polymerase chain reaction (PCR) testing, and arrays for comparative genomic hybridisation (CGH) and single nucleotide polymorphisms (SNPs).

While these all have a number of advantages, Henderson pointed out that FISH and PCR are only able to detect the mutations that were sought.

Karyotyping and CGH/SNP arrays cover the entire genome, but they have their own limitations, including issues with automation for the former and low sensitivity to low level clones in the latter.

WGS, in contrast, has the advantage of being able to detect variants across the whole genome with a high resolution and a low failure rate, and without the need for live cells.

Moreover, Henderson emphasised that WGC is 'future resistant', in that it detects all mutations, so it will not be necessary to develop additional testing or redesign wet panels as novel biomarkers are discovered.


Emerging biomarkers, such as mutational nursery and signatures not easily detectable with other technologies, can also be picked up with the technique.

It is not without its challenges, however.

Cost remains an issue, although the increasing number of biomarkers that need to be tested for and the falling price of WSG means that a tipping point, after which it becomes cost-effective, will be reached.

For the technique to move from being a research tool to being adopted clinically, turnaround times will also have to be comparable to current methodologies, and it will require validation and accreditation.

To go some way to addressing those issues, the researchers examined WGS results in AML and ALL samples from the 100,000 Genomes Project and compared them with results from three clinical diagnostic laboratories performing standard of care tests on haematological tumours.

They gathered data on a total of 252 FISH tests performed in 34 patients, including 138 single variants, of which 18 were positive and 120 negative, and 114 copy number variants, of which 85 were positive and 29 negative.

For the copy number variants, there was agreement between FISH and WGS positive results in 78 (92%) cases, with seven false negative variants detected on WGS.

These included the sample having a tumour content of less than 20% in two cases, low level mutations in the FISH test in another two cases, and low quality samples in a further two.

In contrast, there was agreement between the two techniques in 28 (97%) of negative cases, giving a positive predictive value for WGS of 99%, and a false positive rate of 3%.

For the single variants, there was 100% agreement between FISH and WGS in positive cases, while WGS additionally identified four structural variants.

The team therefore calculated that, in terms of single variants, WGS had a positive predictive value of 82% and a false positive rate of 3%.

Interestingly, when they looked more closely at the additional structural variants  identified with WGS, they found three T cell receptor delta locus arrangements, and an ETV6-RUNX1 fusion.

The latter turned out to be a variant breakpoint associated with paediatric ALL that was not detected by FISH due to limitations with the probe.

Next Steps

Next, Henderson and colleagues would like to increase the number of samples and further correlate the WGS with standard of care data, as well as look at other variant types, such as those identified on karyotyping.

In the post-presentation discussion, Henderson said that the transition from cytogenetics to WGS for AML and ALL will be "gradual".

Fielding pointed out that this could have consequences for multidisciplinary meetings (MDTs), as clinicians are "used to" having the individual who performed the tests in the room.

Henderson replied that there will be a "whole shift" in molecular testing, with the "wet work" of performing WGS on a sample more centralised nationally but with an increased need for local so-called dry labs to interpret the data.

"At the MDT, there will be somebody that will hopefully be able to look at the genome and interrogate it, so I think there'll still be a huge level of expertise," she said.

"In fact, much greater expertise is going to be required," Henderson added, noting it's just that the nature of that expertise will change.

No conflicts of interest or funding declared.

BSH 2019: Abstract BSH19-OR-002. Presented April 1.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.