Genome sequencing with gene panel-based analysis for rare inherited conditions in a publicly funded healthcare system: implications for future testing

Abstract

NHS genetics centres in Scotland sought to investigate the Genomics England 100,000 Genomes Project diagnostic utility to evaluate genome sequencing for in rare, inherited conditions. Four regional services recruited 999 individuals from 394 families in 200 rare phenotype categories, with negative historic genetic testing. Genome sequencing was performed at Edinburgh Genomics, and phenotype and sequence data were transferred to Genomics England for variant calling, gene-based filtering and variant prioritisation. NHS Scotland genetics laboratories performed interpretation, validation and reporting. New diagnoses were made in 23% cases - 19% in genes implicated in disease at the time of variant prioritisation, and 4% from later review of additional genes. Diagnostic yield varied considerably between phenotype categories and was minimal in cases with prior exome testing. Genome sequencing with gene panel filtering and reporting achieved improved diagnostic yield over previous historic testing but not over now routine trio-exome sequence tests. Re-interpretation of genomic data with updated gene panels modestly improved diagnostic yield at minimal cost. However, to justify the additional costs of genome vs exome sequencing, efficient methods for analysis of structural variation will be required and / or cost of genome analysis and storage will need to decrease.

Fig. 1. Distribution of recruitment categories among 394 probands recruited to the SGP Study.

Disease categories are as those set out at Level 2 within the Genomics England Rare Disease List. Where a proband was entered with conditions in more than one category, the proband is included within counts for all relevant categories.

Fig. 2. Distribution of variant numbers by family structure for primary and secondary assessment.

Circles and corresponding numbers indicate median number of variants, bars and vertical lines indicate range.

Fig. 3. Summary of the number of cases reviewed at each stage of variant interpretation and associated outcomes.

Bold-outlined boxes indicate cases where a confirmed genetic diagnosis fully explained the condition; the associated additional diagnostic yield is given alongside.

Fig. 4. Diagnostic yield following primary (black) and secondary (grey) assessment, for all cases and by Genomics England rare disease category.

Diagnostic yield is also included for probands with any ID (a subset of the Neurology and Neurodevelopmental Disorders category). The number of probands in each category is shown in brackets. Some probands are in more than one category and are included for both categories where the variant fully explained their phenotype. Bars indicate 95% confidence intervals for the overall diagnostic yield after primary and secondary assessment.