Next-generation sequencing (NGS) as a diagnostic tool for retinal degeneration reveals a much higher detection rate in early-onset disease

Abstract

Inherited retinal degeneration (IRD) is a common cause of visual impairment (prevalence ∼1/3500). There is considerable phenotype and genotype heterogeneity, making a specific diagnosis very difficult without molecular testing. We investigated targeted capture combined with next-generation sequencing using Nimblegen 12plex arrays and the Roche 454 sequencing platform to explore its potential for clinical diagnostics in two common types of IRD, retinitis pigmentosa and cone-rod dystrophy. 50 patients (36 unknowns and 14 positive controls) were screened, and pathogenic mutations were identified in 25% of patients in the unknown, with 53% in the early-onset cases. All patients with new mutations detected had an age of onset <21 years and 44% had a family history. Thirty-one percent of mutations detected were novel. A de novo mutation in rhodopsin was identified in one early-onset case without a family history. Bioinformatic pipelines were developed to identify likely pathogenic mutations and stringent criteria were used for assignment of pathogenicity. Analysis of sequencing metrics revealed significant variability in capture efficiency and depth of coverage. We conclude that targeted capture and next-generation sequencing are likely to be very useful in a diagnostic setting, but patients with earlier onset of disease are more likely to benefit from using this strategy. The mutation-detection rate suggests that many patients are likely to have mutations in novel genes.

Figure 1

Data-analysis pipeline; the approximate number of variants at each analysis stage is shown.

Figure 2

Age range and family history status of patients. (a) Age range of patients with pathogenic mutation(s) identified. (b) Age range of patients without pathogenic mutation(s) identified. (c) Family history in patients with pathogenic mutations identified (d) Family history in patients without a pathogenic mutation identified (e) Distribution of phenotypes in patients with a pathogenic mutation (red) compared with the total (blue).

Figure 3

Sample coverage. The percentage of bases covered at 10 × , 20 × and 30 × for each patient. Patients with mutation(s) identified (blue), patients with no identified mutations (black), control patients (green) and patients with