The European USH2A hotspots were entirely absent from a Korean cohort

Design

retrospective cohort with restricted cubic spline modelling of age-related change

Population

182 Korean patients with biallelic pathogenic USH2A variants; 56 Usher syndrome, 126 non-syndromic retinitis pigmentosa; mean age 44.3 years

Primary outcome

variant spectrum and age-adjusted visual acuity, visual field and ellipsoid zone length

Effect

c.2802T>G in 23.6% of 364 alleles and 45.1% of patients; European hotspots c.2299delG and c.2276G>T absent; missense 53.8% of alleles; worse function with Usher syndrome and greater truncating burden

Gene-directed therapies for USH2A-associated disease are being built around exon 13, which is where the common European pathogenic variants sit. This retrospective cohort assembled 182 Korean patients with biallelic pathogenic or likely pathogenic USH2A variants - 56 with Usher syndrome and 126 with non-syndromic retinitis pigmentosa, mean age 44.3 - and characterised the allele spectrum alongside acuity, Goldmann fields and ellipsoid zone length.

The spectrum is different. Of 364 alleles, c.2802T>G (p.Cys934Trp) was the commonest at 23.6%, followed by c.8559-2A>G at 12.1%, and 45.1% of patients carried at least one c.2802T>G allele, four of them homozygous. The European hotspots c.2299delG and c.2276G>T were not found at all. Missense variants made up 53.8% of alleles, a much higher proportion than in European series. Both Usher syndrome and a greater burden of truncating variants tracked with worse age-adjusted acuity, visual field and ellipsoid zone length.

The relevance to Indian practice is by analogy rather than directly - this is a Korean cohort and the Indian USH2A spectrum is its own question, largely unanswered. But the general point transfers: variant spectra are population-specific, therapies are being designed around the variants found in European cohorts, and a patient tested against a panel or counselled on a prognosis derived from those cohorts may be poorly served. c.2802T>G sits in exon 13, which is the encouraging part of this particular finding, and is the authors' argument for including East Asian patients in exon 13-targeted trials.

• Use a full gene analysis rather than a hotspot panel when testing for USH2A outside European populations
• Counsel prognosis using truncating variant burden and Usher syndrome status, both of which tracked with worse age-adjusted function
• Note that missense variants dominated here, which complicates variant interpretation - expect more variants of uncertain significance
• Ask whether trials of exon 13-targeted therapy are open to patients with non-European variant spectra
• Do not extrapolate these allele frequencies to Indian patients; that data does not exist in this paper

Why it matters

It shows a therapy being designed around one population's variant spectrum, and asks who that leaves out.

Don't overread it

A clinic-based Korean cohort - the allele frequencies describe this referral population and do not transfer to other Asian populations, India included.

The statistics, in plain English

Allele frequencies from a clinic-based cohort reflect who is referred and tested, not the population - a single-centre or single-country series can over-represent families with striking phenotypes. With 182 patients and 364 alleles, the commonest variant's frequency of 23.6% is reasonably stable, while rarer alleles are not. The association between truncating burden and worse function is age-adjusted and modelled with splines, which handles the non-linear course of retinitis pigmentosa better than a simple regression would.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.