Over four years, the Deciphering Developmental Disorders (DDD) study, a UK-wide collaborative, has collected samples and clinical information of 12,000 undiagnosed children with developmental disorders in the UK as well data from parents. The aim of the project is to better understand the genetic biology of developmental disorders as well as assist in bringing new sequencing technologies into clinical practice. The project began in 2011 and is slated to close in 2021. Over this time-period, the clinical teams, scientists, and bioinformaticians have been and will continue analyzing the data to find diagnoses for the patients. The initial results from the study published in 2015 includes data from 1133 complete family trios (undiagnosed children)1.
A recent paper on the project shared results from a re-analysis in 2017 of the initial 1133 children using new knowledge and findings that have accumulated since the initial analysis2. The pace at which improvements in genomic data technologies, analysis, and knowledge are moving, the group hypothesized that it is likely that diagnostic yield would increase via the re-analysis. They sought to determine how much of an improvement can be made over time.
An automated filtering pipeline was used to narrow down the candidate diagnostic variants using rules (e.g. based on allele frequency, predicted consequence, inheritance) for the family trios. Candidate variants identified via additional variant detection algorithms (including UPD, de novo non-essential splice sites…) were analyzed outside the filtering pipeline described above. Those variants passing the automated workflow were then evaluated by the DDD study’s internal clinical review team. Likely diagnostic variants were then confirmed in an accredited diagnostic laboratory. Variant interpretation included using guidelines from both the ACMG (American College of Medical Genetics and Genomics) and the UK Association for Clinical Genetic Science.
In the initial reporting in 2014, a diagnostic yield of 27% was achieved via whole-exome sequencing. With the re-analysis of the same patients using improved methodologies, variant calling, annotations, etc., the diagnostic yield increased to 40% and an additional 182 patients diagnosed. Two-thirds of these novel diagnoses were attributed to new disease-associated genes identified over the last three years (about half of these were in 30 new disease-associated genes discovered by the DDD study and the rest were in additional published disease genes found via literature searches). Twenty-three percent of the novel diagnoses resulted from improved analyses and 8% from additional analytical methods. 272 previously diagnosed probands remained diagnosed and 39 probands had their previous diagnoses clinically reclassified as uncertain or likely benign and six probands received diagnoses from an independent test as they were missed by the DDD workflow due to low-depth sequencing in at least one member of the trio. All variants with associated phenotypes are available via DECIPHER and all are in published development disorder genes with sufficient evidence to merit inclusion in their clinician curated gene-to-phenotype database.
Based on previous estimates of a diagnostic yield of 10% from micaroarray testing and a small additional yield from single gene testing (cases excluded from the DDD study), the group estimates greater than 50% diagnostic yield from whole-exome sequencing if implemented as a first-line test for DD. In addition, the group values trio sequencing as a first-line strategy in sporadic cases due to pathogenic classification of approx. 80% of reported de novo mutations in a known-dominant DD. The group suggests adopting reiterative reinterpretation of reported clinical sequencing data as routine but does acknowledge major changes would be required, one has to also consider how all this information would affect families, and appropriate systems would need to be established for this huge undertaking.
There is a lot of great information and points for discussion in this paper so please do take a look at the entire paper. Also note, that the DECIPHER and DDD curated databases mentioned above are integrated with NxClinical software for analysis and interpretation of variants - take a look at the recent updates to the reference tracks available in NxClinical.
References
1. Deciphering Developmental Disorders Study. et al. Large-scale discovery of novel genetic causes of developmental disorders. Nature. 2015 Mar 12;519(7542):223-8.
2. Wright CF et al. Making new genetic diagnoses with old data: iterative reanalysis and reporting from genome-wide data in 1,133 families with developmental disorders. Genet Med. 2018 Jan 11. [Epub ahead of print]
