NEW YORK – The European Society of Human Genetics recently updated its guidelines for next-generation sequencing to reflect the increased adoption of whole-genome sequencing in the diagnostic setting.
But the recommendations, published online in the European Journal of Human Genetics last month, differ in several aspects from similar guidelines for clinical WGS in the US, such as those of the Medical Genome Initiative.
Previous ESHG guidelines, published in 2016, focused mainly on the clinical use of whole-exome sequencing. EuroGenTest, the working group within ESHG that prepared both sets of recommendations, said that an update and reassessment was needed given the increased uptake of whole-genome sequencing in the clinic. Solve-RD, a Horizon 2020-funded project, also helped to develop the guidelines.
According to Janneke Weiss, corresponding author on the updated recommendations, the popularity of whole-genome sequencing in diagnostics has been increasing in Europe for multiple reasons, making the guidelines update necessary. Weiss is a laboratory specialist at Radboud University Medical Center in Nijmegen, Netherlands.
"Whole-genome sequencing is able to replace more techniques, such as karyotyping, multiplex ligation-dependent probe amplification, and arrays, than whole-exome sequencing is currently able to do," she said, adding that the field might "move to a one-test-fits-all principle."
Weiss drew attention to the new European In Vitro Diagnostic Regulation (IVDR) covering molecular tests in Europe, which mandates that manufacturers obtain regulatory clearance for most of their assays, as well as that laboratories follow a variety of new rules, as another potential reason for clinicians tending to favor a single, catch-all test rather than multiple approaches.
Additionally, whole-genome sequencing has become a go-to technology in the clinic simply because of its high diagnostic yield. Weiss also noted that costs for whole-genome sequencing continue to drop, meaning that going forward, it will only become more affordable and feasible for labs to offer it in routine diagnostics.
With that added breadth of coverage, however, come challenges regarding data interpretation, particularly variants of unknown significance. In Weiss' opinion, the most important additions to the guidelines therefore cover bioinformatics and interpretation.
"We do not know how to interpret the sequence beyond the exome very well, and we therefore have to invest in this field in upcoming years," she said. "When we learn how to interpret the variants beyond the exome, we may be able to increase the diagnostic yield " she said.
But rather than draw up strict protocols for European labs to follow, the authors instead provided more general suggestions of what they should aim for. There are 44 separate recommendations included in the guidelines, covering general implementation of whole-genome sequencing, diagnostic routing, bioinformatics, quality assessment, ethical considerations, and reporting.
The authors wrote that they recommend adopting whole-genome sequencing when it is necessary to improve diagnostic quality, efficiency, and yield, and to do so in accredited laboratories only. Tests should be validated and findings confirmed via retesting of independent samples. The authors also provide guidance around interpreting variants of unknown significance, stating they should only be reported when follow-up studies can provide more insight into pathogenicity.
Diagnostic labs are expected to maintain structured databases for classified variants, and to report their variants to databases accessible by other geneticists. Other recommendations cover implementing and maintaining the bioinformatics pipeline, as well as data storage solutions.
Weiss said that the authors purposefully provided generalizable recommendations, while avoiding specifics on how to validate tests, for instance, because of different local policies regarding unsolicited findings and "different setups and infrastructure" in European labs.
Helger IJntema, head of genomic diagnostics at Radboud and a coauthor on the paper, said that while whole-genome sequencing is growing in popularity, it has not yet eclipsed other methods, such as array comparative genomic hybridization or whole-exome sequencing, as it is still considered expensive for most labs. However, it is used more often for unsolved cases, she said, and she expects this to continue as the price of whole-genome sequencing falls in the future.
With regards to the recent updates, IJntema said that the most significant changes were around interpreting and reporting different variants, as well as what kinds of validation are necessary. Like Weiss, she said that interpreting the recommendations will depend on national policies, as well as required quality standards in different jurisdictions.
IJntema added that Radboud is already following protocols like those described in the guidelines, meaning that the new publication won't necessarily impact its operations. However, she said there would likely be updates going forward, as more labs adopt whole-genome sequencing for diagnostics.
The Medical Genome Initiative
The publication of the new European guidelines came just weeks after the Medical Genome Initiative (MGI), a consortium of North American research and healthcare organizations put together by San Diego vendor Illumina, published its own set of practices and standards to guide the adoption of whole-genome sequencing in diagnostics.
According to the authors of both papers, there are similarities and differences between the two. Weiss pointed out that the most recent MGI paper, published April 8, focused on best practices for the interpretation and reporting of clinical whole-genome sequencing, while a previous October 2020 paper discussed the technical aspects of implementing the approach.
"We decided to put the whole process into a single and short publication and hope this may be a helpful tool for labs to start with [introducing] whole-genome sequencing into diagnostics," Weiss said. "All of the mentioned publications are very good to read when starting with whole-genome sequencing."
According to IJntema, another difference between the two is that the recent MGI paper discussed genotype-driven analyses, while the current European recommendations are more focused on phenotype-based analysis, suggesting, for example, that clinicians analyze one or more in silico gene panels and use filtering strategies, such as parent-child trios for disorders frequently caused by de novo variants.
"While the MGI recommendations do advise labs to utilize phenotype-driven approaches such as gene panels, we also advocate for genotype-driven approaches that prioritize highly suspicious variants across the genome," said Christina Austin-Tse, analysis team lead in the rare disease group at the Broad Institute of MIT and Harvard and lead author on the recent MGI paper. Such an approach could reveal phenotype expansions or unexpected diagnoses relevant to the primary indication for testing, she said in an email.
Heidi Rehm, codirector of the program in medical and population genetics at the Broad and a member of the MGI steering committee, noted in a separate email that there are numerous differences between the two sets of guidelines.
The authors of the European guidelines, for instance, recommend that only genes for which a clear association with disease has been confirmed should be reported, she pointed out, and that variants in genes of unknown function should be listed in an independent research report. In the US, in contrast, labs are encouraged to report genes of uncertain significance if there is a "strong possibility of causality," said Rehm, and clinicians are encouraged to use Matchmaker Exchange, a project that aims to build evidence for genetic causes for patients with rare diseases.
"In general, US labs do not include a research report and generally keep research and clinical [findings] separate," said Rehm.
Rehm also echoed Austin-Tse in noting that the European guidelines recommend that testing be directed toward answering a clinical question, whereas MGI "strongly encourages genotype-first strategies" in addition to phenotype-driven approaches. She also drew attention to the European recommendation that variants of unknown significance should be reported only if the phenotype associated with the respective disease gene matches the clinical features of the patient, and when follow-up studies can be performed to gain more information about pathogenicity of the variant.
"The key difference is that [in the US,] variants of unknown significance would be reported if they are in genes that match the phenotype even if there is not a specific follow-up study that can be done," remarked Rehm. "This is a difference in guidance for all US practices and guidelines."
There are efforts underway to harmonize guidelines between Europe, including the UK, and North America, Rehm added. She noted that Sian Ellard, professor of genomic medicine at the University of Exeter Medical School in the UK, was recently invited to join the steering committee of the American College of Medical Genetics and Genomics, ClinGen, and the Association for Molecular Pathology, to help develop standards for variant classification.
Rehm also mentioned the Global Alliance for Genomics and Health (GA4GH), an international standards organization that focuses mostly on research but has been addressing issues around returning significant findings to patients and consent policy via its regulatory and ethics working group.