To the Editor:

We read with interest the series of patients described by Ba-Abbad et al. [1]. Whilst biallelic truncating CDHR1 variants result in cone-rod or rod-cone dystrophies—consistent with the Cdhr1−/− knockout mouse [2]—it is intriguing that certain biallelic splice or missense variants in CDHR1 may result in late-onset macular dystrophy.

Variants in CDHR1 described in association with late-onset macular dystrophy [1] appear to affect the ectodomains of CDHR1. Although the six extracellular cadherin repeats form the longest part of the protein, they are the most conserved, suggesting an important biological function [3]. Ultrastructural analysis with immunogold labelling of CDHR1 in murine retinae identified that evaginating, nascent discs at the base of the photoreceptor outer segments form CDHR1-based connections with the inner segments [4]. Formed from the ectodomains of CDHR1, the contacts are lost as outer segment discs mature. This physical separation occurs through proteolytic cleavage of the CDHR1 ectodomain [3]. Non-truncating CDHR1 variants affecting the ectodomain may interfere with its proteolytic cleavage, or attachments to the inner segment, thereby resulting in dystrophic outer segments without widespread photoreceptor degeneration seen with biallelic truncating variants.

c.783G>ACDHR1 appears to be the most common CDHR1 variant associated with late-onset macular dystrophy, based on mean allele frequency [1, 5]. Indeed, when hypomorphic variants are included, CDHR1 was amongst the more common causes of macular and cone- or cone-rod dystrophy in one large series [6]. Although isolated macular dystrophy appears to be the most common phenotype in c.783G>ACDHR1 homozygotes [5], peripheral retinal degeneration has been reported [7], suggesting the influence of unknown modifiers which may result in rod photoreceptor degeneration instead of only macular involvement. The identification the inner segment binding partner of CDHR1 and the enzyme catalysing the cleavage of the CDHR1 ectodomain may help to explain the macular-predominant dystrophy in this cohort. Variants in these unknown genes may themselves act as potential disease modifiers, or even be disease causing in their own right.

Similar phenotypes are found in association with CDHR1, PROM1 and PRPH2 variants—all of which encode proteins supporting photoreceptor outer segment structure. It remains to be determined why cones, or macular photoreceptors in general, appear to be more susceptible to the effects of specific variants in these genes.