Development and clinical utility of a novel diagnostic nystagmus gene panel using targeted next-generation sequencing

Abstract

Infantile nystagmus (IN) is a genetically heterogeneous disorder arising from variants of genes expressed within the developing retina and brain. IN presents a diagnostic challenge and patients often undergo numerous investigations. We aimed to develop and assess the utility of a next-generation sequencing (NGS) panel to enhance the diagnosis of IN. We identified 336 genes associated with IN from the literature and OMIM. NimbleGen Human custom array was used to enrich the target genes and sequencing was performed using HiSeq2000. Using reference genome material (NA12878), we show the sensitivity (98.5%) and specificity (99.9%) of the panel. Fifteen patients with familial IN were sequenced using the panel. Two authors were masked to the clinical diagnosis. We identified variants in 12/15 patients in the following genes: FRMD7 (n=3), CACNA1F (n=2), TYR (n=5), CRYBA1 (n=1) and TYRP1 (n=1). In 9/12 patients, the clinical diagnosis was consistent with the genetic diagnosis. In 3/12 patients, the results from the genetic diagnoses (TYR, CRYBA1 and TYRP1 variants) enabled revision of clinical diagnoses. In 3/15 patients, we were unable to determine a genetic diagnosis. In one patient, copy number variation analysis revealed a FRMD7 deletion. This is the first study establishing the clinical utility of a diagnostic NGS panel for IN. We show that the panel has high sensitivity and specificity. The genetic information from the panel will lead to personalised diagnosis and management of IN and enable accurate genetic counselling. This will allow development of a new clinical care pathway for IN.

Figure 1

Multichannel VEPs to investigate intracranial visual pathway dysfunction (a and b). Monocular stimulation is achieved by occluding one eye and presenting a pattern stimulus (checkerboard stimulus). Four lateral electrodes (1 and 2 on left occiput, 4 and 5 on the right occiput) and one midline electrode (electrode 3) are used. Five raw traces corresponding to the polarity of each electrode are obtained during monocular stimulation. In order to improve signal to noise, VEP traces shown are based on the polarity differences between the electrodes placed on the left and right occiput and obtained after averaging. Individuals with a normal visual pathway would not have any significant difference in the polarity between the corresponding lateral electrodes (1 vs 5 or 2 vs 4). Thus a subtracted waveform (1–5 and 2–4) would show no significant polarity differences as seen in the IIN patient (NYS-007) (a). However, in a patient with chiasmal misrouting as seen in albinism (NYS-012) asymmetric responses are seen (blue arrow) (b). Different degrees of iris transillumination defects seen with albinism (c). Pale fundus associated with albinism (d).

Figure 2

Optical coherence tomograms showing the retinal structure in patients with IN. Figure (a) shows an illustration of normal foveal morphology with the features that indicate foveal specialisation which include: (1) deep foveal pit, (2) displacement of inner retinal layers, (3) ONL widening, and (4) OS lengthening. In family NYS-005, the mother (II:2) had a normal foveal structure (b) with all four features present (labelled 1–4); however, her son (III:1) did not have a foveal pit and only had ONL widening (c). Similarly, in family NYS-010 we observe phenotypical heterogeneity with regards to the foveal structure. Both sisters had compound heterozygous variants of the TYR gene; however, in subject II:1 there was only ONL widening, whereas in subject II:2 there was both ONL widening and a rudimentary foveal pit (denoted by 1*). In family NYS-007, OCT from subject III:1 with FRMD7 deletions has normal foveal morphology. Subject I:1 in family NYS-011 with albinism has no foveal pit but has ONL widening and OS lengthening. In family NYS-012 with albinism, we observe foveal hypoplasia with a rudimentary foveal pit (1*) in subject IV:3, as well as ONL widening and OS lengthening. In family NYS-013, we were unable to identify any variants that affect function within the known nystagmus genes; however, subject II:1 was noted to have subtle foveal hypoplasia with a shallow foveal pit (1*) and persistence of inner retinal layers posterior to the fovea. The numbers (1–4) within the tomograms represents the presence of the foveal specialisation, with 1* representing a shallow pit in foveal hypoplasia. Abbreviations: NFL, nerve fibre layer; GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; ELM, external limiting membrane; ISe, inner segment ellipsoid; OS, outer segment; RPE, retinal pigment epithelium; BM/CC, Bruchs membrane/choriocapillaries.

Figure 3

Congenital cataracts associated with CRYBA1 variant in NYS-008. The two images from subjects III:2 and III:3 show (monozygotic twins) sutural cataracts, seen as the opacification of the Y-suture of the foetal nucleus of the lens. Subject II:1 was pseudophakic.

Figure 4

Diagnostic workflow for patients with nystagmus. In the current pathway, patients would undergo a number of different investigations prior to being able to identify the underlying aetiology. The order of performing investigations vary between centres, depends on the clinical context, patient cooperation and availability. Some of the limitations associated with each investigative modality is shown in the box. Not all patients would undergo MRI scans; however, it is one of the investigations of choice for specific forms of nystagmus (for eg, vertical nystagmus) or the presence of additional neurological features such as ataxia were noted. The new pathway that involves using NGS as a frontline diagnostic tool would help establish a genetic diagnosis and thus guide further investigations and targeted treatment.