Biallelic loss-of-function variants in C19orf44 lead to retinal degeneration

Abstract

Background

Inherited retinal diseases (IRDs) are a group of disorders often resulting in progressive vision loss, ultimately leading to blindness. A significant portion of their genetic causes remain unresolved, partly due to undiscovered disease-associated genes or variants. This study aimed to identify novel genetic links to IRDs.

Methods

All patients underwent comprehensive ophthalmological evaluation, including retinal imaging (fundus autofluorescence and macular optical coherence tomography) and electroretinogram testing. Whole exome sequencing and whole genome sequencing were performed on patients with clinically unsolved IRD, and data were analysed using an in-house pipeline to identify causal variants. Subsequently, Sanger sequencing was performed to confirm identified variants.

Results

Three unrelated patients from Europe, Middle East and East Asia were identified with unique late-onset retinal degeneration (Stargardt-like phenotype) associated with biallelic loss-of-function (LoF) variants in C19orf44 (HGNC: 26141), a gene of unknown function. The homozygous variant NM_032207.2:c.549_550del;p.Ser185Profs*2 was identified in two unrelated patients (European and Middle Eastern). Moreover, an East Asian patient had likely compound heterozygous LoF variants (NM_032207.2:c.1168C>T;p.Gln390*/c.976_977del;p.Leu326Lysfs*15).

Conclusions

Our findings establish C19orf44 as a novel disease-causing gene for IRD with Stargardt-like phenotype, expanding the genetic landscape of retinal degeneration.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Inherited retinal diseases (IRDs) are a genetically diverse group of disorders, and many cases remain unsolved due to unknown disease-associated genes.

WHAT THIS STUDY ADDS

• This study identified C19orf44 as a novel IRD-associated gene through whole exome sequencing and whole genome sequencing of three unrelated patients with a Stargardt-like phenotype. All individuals carried biallelic loss-of-function variants in C19orf44, a gene not previously linked to retinal disease.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This study expands the genetic landscape of IRDs and supports the inclusion of C19orf44 in diagnostic testing. This discovery also lays the groundwork for future functional studies and potential therapeutic development.

Introduction

Inherited retinal diseases (IRDs) are genetic disorders marked by significant diversity in their genetic causes and clinical presentations. IRDs are typically inherited in a Mendelian pattern, following recessive, dominant, X linked or maternal (mitochondrial) patterns, depending on the presence of pathogenic variants in one of the 327 identified genes to date (https://retnet.org/summaries#a-genes, accessed 24 Jan 2025).

ABCA4-related retinopathy, also called Stargardt disease (STGD1; OMIM: 248200), is one of the most common conditions and can primarily affect the macula or in some cases the entire retina. It was first described by Karl Stargardt in 1909 and is marked by a gradual loss of sharpness in central vision and central visual field in both eyes. ABCA4-related retinopathy typically presents with three temporal patterns: childhood onset, early adult onset or late onset. The earlier the disease starts, the more severe the outcome is likely to be.¹ Mutations in several other genes, including ELOVL4, PROM1, PRPH2 and BEST1, can present with phenotypical overlap and have been termed Stargardt-like dystrophies.² Stargardt-like disease refers to a group of rare retinal disorders that closely resemble the clinical characteristics of STGD but are caused by mutations in genes other than ABCA4. Approximately 10–30% of patients clinically diagnosed with STGD1 or Stargardt-like diseases remain genetically unsolved, with variation across different ethnic cohorts.³,¹¹

Despite the advancement in next generation sequencing technologies, causative variants remain undetected in approximately one-third of IRD cases, indicating the presence of unknown genetic variants or undiscovered disease-associated genes.¹² In this study, we identify biallelic loss-of-function (LoF) variants in chromosome 19 open reading frame 44 (C19orf44) gene in three patients who exhibit Stargardt-like phenotype from three different families across distinct geographical regions. Our findings support the notion that the C19orf44 protein is essential for retinal function, and its deficiency results in retinal degeneration.

Materials and methods

Enrolment of cases

All cases enrolled for this study were affected with IRDs as per clinical assessment by expert ophthalmologists and were recruited following the principles of the Declaration of Helsinki. The blood sample of each case was collected after signed informed consent. DNA was extracted using Qiagen blood genomic DNA extraction kits (Qiagen, Hilden, Germany) or XTRACT16 (Autogen, Holliston, Massachusetts).

Genetic analysis

Unsolved patients after clinical testing (panel) sequencing underwent whole exome sequencing (WES) or whole genome sequencing (WGS) using genomic DNA of each IRD case at Blueprint Genetics or Baylor College of Medicine, Houston, Texas, USA. As described in our previous studies, variant calling, data alignment and filtration were performed at the Functional Genomics Core at Baylor College of Medicine, USA.^{12 13} Variants passing the filtering steps were evaluated as suggested by the American College of Medical Genetics and Genomics (ACMG) guidelines for their interpretation.¹⁴ Novel variants were evaluated for their potential impact on protein function using multiple in silico tools. Nonsense, frameshift and canonical splice site variants were classified as likely LoF alleles. Missense variants were evaluated based on sequence conservation and in silico predictions. Briefly, sequencing reads were aligned with the Burrows-Wheeler Alignment human genome assembly (hg19),¹⁵ whereas single-nucleotide variants (SNVs) and insertion-deletion variants (INDELs) were identified using GATK4. A 0.5% population-frequency threshold was used to eliminate frequently occurring variants that were not likely to cause IRDs. Coding region SNVs and INDELs were annotated with ANNOVAR and compared with the dbNSFP 3.5a database, while the conservation of remaining variants was estimated following the UCSC Genome Browser’s phastCons.hg19.100way.¹⁶ The effect of missense variants was predicted using REVEL v1.3.¹⁷ WGS-consolidated SNVs of all cases were annotated and filtered using the genomic alteration with a custom pipeline, and the intronic variant effects were predicted with SpliceAI (spliceAI-1.2.1).¹⁸ Sanger sequencing was performed to confirm the identified variants.

Identity-by-descent and homozygosity-by-descent analyses using phased genotype data

We used Beagle (V.5.5) to phase the genotype data for all SNPs associated with patients 1 and 2.¹⁹ The identity-by-descent (IBD) and homozygosity-by-descent (HBD) analyses were subsequently performed using hap-ibd (V.1.0, 15Jun23.92f), which employs linear interpolation to estimate genetic positions between mapped locations. HapMap genetic maps in cM units, formatted for PLINK and generated using 1000 Genomes data, were downloaded from https://bochet.gcc.biostat.washington.edu/beagle/genetic_maps/plink.GRCh37.map.zip.²⁰ The hap-ibd parameters ‘max-gap=1000 min-seed=0.1 min-output=0.1 min-markers=10 min-mac=1’ were used.

Long read sequencing for phasing of C19orf44 variants

A 6.2 kb amplicon was generated from genomic DNA extracted from peripheral blood using PCR with primers 5′-AGGTCCCATCTCAACTGAGAGCATTTAC and 5′-TCACACTTGAAGAGGATTCAGACAAAGC. Following PCR purification, a sequencing library was prepared using the Native Barcoding Kit 24 V14 (Oxford Nanopore Technologies, Cat No SQK-NBD114.24) and subsequently loaded onto an R10.4.1 flow cell (FLO-MIN114) for sequencing on a MinION Mk1C device.

Base calling was performed using Dorado V.0.8.1 in super accuracy mode from POD5 files. The resulting Fastq reads were processed with porechop v0.2.4 and chopper v0.7.0 to retain only reads containing both PCR primers and within ±200 bp of the expected amplicon size. High-quality reads were then aligned to the GRCh38 reference genome using minimap2 v2.26. Variant calling was conducted using deepvariant v1.6.0, followed by haplotype phasing with whatshap v2.3.

Results

Clinical findings

Patient 1: A European male in his 60s presented with a late-onset macular dystrophy. The patient’s medical history included hypertension and gout. There was no reported family history of vision loss or macular degeneration, although his paternal grandparents and father were affected by glaucoma (figure 1A). Central vision loss began in the early 40s, with a clinical diagnosis of Stargardt-like disease in the early 50s. Vision loss was more pronounced in the left eye, which led to legal blindness and cessation of driving. Despite early progressive decline, visual function was relatively stabilised in subsequent years.

Figure 1. The family trees of the patients affected with C19orf44-related retinopathy, Sanger sequencing and predicted C19orf44 protein structure. (A) Pedigree of the affected patient 1. (B) Pedigree of the affected patient 2. (C) Pedigree of the affected patient 3. (D) Sanger sequencing of patient 1 and his sister, patient 2 and for chr19-16612150-AAC-A:c.549_550del (p.Ser185Profs*2). (E) Phasing analysis from gnomAD v2 of identified variants in patient 3 (chr19-16620328-C-T:c.1168C>T (p.Gln390*)/chr19-16614090-CTT-C:c.976_977del (p.Leu326Lysfs*15)). Black-filled symbols with arrows represent probands diagnosed with C19orf44-related retinopathy. M1, chr19-16612150-AAC-A:c.549_550del:p.Ser185Profs*2; M2, chr19-16620328-C-T:c.1168C>T:p.Gln390*; M3, chr19-16614090-CTT-C:c.976_977del:p.Leu326Lysfs*15.(F) Predicted protein structure of C19orf44 and position of identified variants.

During his 60s, tonometry revealed intraocular pressures of 10 mm Hg in the right eye and 11 mm Hg in the left eye. Refraction values were −3.5/+1.5×75° dioptre (D) for the right eye and −3.0/+1×114° D for the left eye, with best-corrected visual acuity (BCVA) of 20/50−1 in the right eye and counting fingers (CF) in the left eye. Extraocular movement was normal. Colour vision testing showed significant impairment, with both eyes scoring 0/20 on Hardy-Rand-Rittler colour vision test. In the early 60s, the Humphrey Visual Field (HVF) test, followed by kinetic visual field testing with the Octopus 900 perimeter in the mid-60s and early 70s (online supplemental figure 1A–C), demonstrated bilateral central scotomas, while peripheral fields remained relatively preserved. These findings revealed progressive and characteristic visual field changes consistent with disease progression.

Slit-lamp examination identified nuclear sclerotic cataracts graded at +2 in both lenses, with cortical spoking also graded at +2. Fundus examination revealed atrophy nasal to the optic disc, pigmentary changes in the temporal macula and prominent retinal pigment epithelium (RPE) flecks and mottling bilaterally (figure 2). The cup-to-disc ratio was 0.5, and retinal vessels appeared normal. Optical coherence tomography (OCT) indicated severe retinal and choroidal atrophy throughout the macula, while fundus autofluorescence (FAF) imaging showed a central area of absolute hypoautofluorescence, complete ring of peripapillary atrophy and nasal sparing (figure 2). The progressive retinal degeneration over time was evident based on assessments using OCT and AF imaging shown in online supplemental figure 2.

Figure 2. Top: Pseudocolour fundus images of the left and right eyes from patient 1, ‘multicolor’ images from the left and right eyes from patient 2 and pseudocolour image from the right eye from patient 3. Bottom: fundus autofluorescence and optical coherence tomography photographs of the left and right eyes from patient 1, the left and right eyes from patient 2 and the right eye from patient 3.

In the early 60s, full-field electroretinography (ffERG) revealed significantly abnormal response parameters across all stimuli with prolonged timing. Cone responses were barely detectable, with an average reduction in response amplitudes of 25–30% compared with the previous assessment conducted 2 years ago, indicating progressive retinal dysfunction (online supplemental figure 3A). By the mid-60s, ffERG revealed severely decreased amplitudes that were undetectable bilaterally. Scotopic (DA 0.01) ffERG recordings were limited by blink artefact, but photopic (LA 6.0) cone-dependent responses were markedly reduced in amplitude with normal implicit times (online supplemental figures 3B and 4A). The ffERG results indicate a significant generalised cone dysfunction with no detectable macular cone function. Overall, the earlier ERG showed limited but preserved retinal activity, with distinguishable photopic and scotopic responses, suggesting the patient still had functional photoreceptors and bipolar cells. In contrast, the later ERG demonstrated a marked loss of both rod and cone activity, with minimal or no recordable responses detected (online supplemental figures 3A,B and 4A).

At a follow-up visit in the early 70s, retinal degeneration had progressed, accompanied by systemic organ dysfunction. Notable clinical findings included generalised weakness, tophaceous gout, stage 4 chronic kidney disease, eosinophilia, acquired ptosis, hypertension, peripheral neuropathy, acute gastric ulcer, urinary retention and vitamin D deficiency. The patient’s overall health deteriorated, and he subsequently passed away.

Patient 2: A Middle Eastern male in his early 60s, born to second-degree cousin parents, was clinically diagnosed with Stargardt-like disease in his early 50s (figure 1B). The patient reported no family history of retinal disease. He is the father of teenage triplets, all of whom are unaffected. Additionally, the patient had a history of renal cell carcinoma, which was successfully managed with surgical intervention, with no evidence of recurrence.

HVF test revealed bilateral central scotomas, and peripheral fields remained preserved (online supplemental figure 1D). BCVA was 20/30−2 in the right eye and CF in the left eye. Refraction values were +0.50/+0.50×166° D for the right eye and plano (0 D) for the left eye. Fundus examination, supported by AF imaging and OCT, revealed classical features consistent with Stargardt-like disease. Over the past decade, there was significant progression of paracentral geographic atrophy, as documented through serial imaging (figure 2, online supplemental figure 3D). ffERG revealed severely reduced scotopic and photopic responses, with diminished a-wave and b-wave amplitudes and attenuated photopic flicker responses, indicating progressive rod and cone dysfunction consistent with advanced Stargardt-like retinal degeneration (online supplemental figure 3C).

Patient 3: An East Asian woman in her late 80s presented with a history of retinal degeneration, having first noticed visual symptoms in her late 60s, reporting central blurriness and difficulty seeing in dim light (figure 1C). Her medical history includes hypertension and anaemia.

BCVA was 20/25 in the right eye and 20/50 in the left eye, with refractive errors of −0.25/+1.00×180° D and −0.75/+1.25×175° D, respectively. The patient was pseudophakic in both eyes. Fundus examination revealed macular geographic atrophy with granular RPE changes extending to the periphery in both eyes (figure 2). Additionally, a round vitreous condensation was observed in the right eye.

Colour vision testing showed tritan discrimination deficits bilaterally on the D15 test. Goldmann perimetry indicated a constricted visual field. OCT demonstrated retinal thinning and loss of the inner segment/outer segment junction, with a preserved outer retinal structure centrally (figure 2). ffERG revealed severely reduced scotopic and photopic responses, indicating widespread retinal dysfunction.

The demographic and clinical data of patients are provided in table 1.

Table 1. Demographic and clinical information of patients affected with C19orf44-related retinopathy.

	Patient 1		Patient 2	Patient 3
Sex	Male		Male	Female
Average age at examination (years)	71.75
Average age of onset (years)	51.67
Ethnicity	Europe		Middle East	East Asia
Genomic variant (GRCh37)	chr19-16612150-AAC-A		chr19-16612150-AAC-A	chr19-16620328-C-T/chr19-16614090-CTT-C
HGVS annotation	c.549_550del;p.Ser185Profs*2		c.549_550del;p.Ser185Profs*2	c.1168C>T;p.Gln390*/c.976_977del;p.Leu326Lysfs*15
gnomAD v4.1.0 AF	0.0001425		0.0001425	0.0001172/0.0001096
gnomAD v4.1.0 subpopulation AF	European (non-Finnish) 0.00009152, European (Finnish) 0.00001562; Middle East 0		European (non-Finnish) 0.00009152, European (Finnish) 0.00001562; Middle East 0	East Asians 0.001714/0.001554
Zygosity	Homozygous		Homozygous	Compound heterozygous
Clinical diagnosis	C19orf44-related retinopathy		C19orf44-related retinopathy	C19orf44-related retinopathy
Consanguinity	No		Yes	No
BCVA	NA	RE: 20/50-1 LE: CF	RE: 20/30−2 LE: CF	RE: 20/25 LE: 20/50
Visual acuity	Central scotoma and intact peripheral fields	>J10 both eyes	NA	Constricted
Refraction	RE: −3.25/+1.25×80° D LE: NI	NI both eyes	RE: +0.50/+0.50×166° D LE: plano	RE: −0.25/+1.00×180° D LE: −0.75/+1.25×175° D
Fundus findings	Central chorioretinal atrophy with flecks and RPE mottling	Central chorioretinal atrophy with flecks and RPE mottling	Notable progression of the paracentral geographic atrophy	Macular geographic atrophy and granular RPE changes out to the periphery
Autofluorescence	Hypoautofluorescence patches, complete ring of peripapillary atrophy and nasal sparing	Large central area and patches of atrophy nasal to disc and nasal sparing with mild progression	Significant progression of the paracentral geographic atrophy	NA
Optical coherence tomography	Severe atrophy of both retina and choroid through macula	Severe atrophy of both retina and choroid through macula which appears relatively stable	Progressive paracentral geographic atrophy	Retinal thinning
Multifocal ERG	Severely abnormal, amplitude of both eyes	NA	NA	NA
Full-field ERG	Moderate reduction of rod-driven responses as well as cone responses	NA	NA	Severely reduced responses in both scotopic and photopic waveforms
Comorbidities	Hypertension and gout	Tophaceous gout, CKD at stage 4, eosinophilia, acquired ptosis, hypertension, peripheral neuropathy, acute gastric ulcer, urinary retention, vitamin D deficiency	Renal cell cancer	Pseudophakia, hypertension and anaemia

AF, allele frequency; BCVA, best-corrected visual acuity; CF, counting fingers; CKD, chronic kidney disease; ERG, electroretinography; HGVS, Human Genome Variation Society; LE, left eye; NA, not available; NI, no improvement; RE, right eye; RPE, retinal pigment epithelium.

Genetic findings

A large number of patients with unsolved IRD underwent WES and WGS in multiple centres in the USA after screening for mutations in known IRD-associated genes. Following the application of stringent variant filtering criteria, as detailed in the Materials and methods section, LoF variants in the novel candidate gene C19orf44 were identified in three unrelated patients with IRD (table 1).

Two patients from distinct geographical regions were found to share a homozygous LoF variant, NM_032207.2:c.549_550del;p.Ser185Profs*2 (figure 1F, online supplemental figure 5A). Sanger sequencing confirmed that the identified variant was homozygous in patient 1, heterozygous in his sister and homozygous in patient 2 (figure 1D). This variant is predicted to introduce a premature stop codon at amino acid position 186, likely resulting in complete loss of protein function. The overall allele frequency (AF) of this variant in gnomAD v4.1.0 is 0.0001425. Subpopulation-specific analysis revealed an AF of 0.00001562 in the European Finnish population and AF of 0.00009152 in the non-Finnish population, while the variant is absent in the Middle Eastern population. This geographical variation in AF highlights the potential for population-specific enrichment or genetic drift influencing the variant’s distribution.

The homozygous deletion (NM_032207.2:c.549_550del;p.Ser185Profs*2) was located within an IBD segment (chr19:16415688-16850140, hg19) shared by both patients (patient 1 and patient 2), suggesting inheritance from a common ancestor. Additionally, deletion was found within regions of HBD in both patients: patient 1 (chr19:16341056-17491472, hg19) and patient 2 (chr19:16413062-18210938, hg19). These findings strongly indicate a founder effect, where the mutation originated from a shared ancestor and has been inherited through subsequent generations, resulting in its homozygosity in both patients.

Patient 3, from East Asia, was found to harbour two heterozygous LoF variants in the C19orf44 gene (NM_032207.2: c.1168C>T, p.Gln390*; c.976_977del, p.Leu326Lysfs*15). These LoF variants are predicted to result in complete loss of gene function, potentially contributing to the observed phenotype. Population AF from gnomAD v4.1.0 revealed a global AF of 0.0001172 for the c.1168C>T (p.Gln390*) variant and AF of 0.0001096 for the c.976_977del (p.Leu326Lysfs*15) variant. For the East Asian subpopulation, these frequencies were higher, with an AF of 0.001714 and 0.001554, respectively.

Phasing analysis using gnomAD v2 indicated that the two variants occurred on different haplotypes within the East Asian subpopulation (figure 1E), consistent with a probable compound heterozygosity in patient 3. Interestingly, long-read DNA sequencing from patient 3 confirmed that the identified variants are present on different haplotypes (online supplemental figure 4B).

Discussion

In this study, we used WGS and WES to identify disease-causing variants in C19orf44 in three unrelated patients, each clinically diagnosed with C19orf44-related retinopathy. These patients were evaluated at distinct medical centres in the USA, including the Casey Eye Institute (Oregon), Jules Stein Eye Institute (UCLA) and the National Eye Institute (Maryland). The identified variants introduce premature stop codons in both transcripts NM_032207.2 and NM_001288834.2, which are predicted to result in nonsense-mediated mRNA decay, thereby leading to a complete loss of protein function (figure 1F, online supplemental figure 5B). In instances where translation proceeds, the truncated protein is expected to lack structural and functional integrity. Our findings provide evidence that LoF variants in C19orf44 are associated with a late-onset, progressive IRD phenotype, specifically resembling STGD classified as C19orf44-related retinopathy. The phenotype observed in these patients is consistent with prior reports linking LoF in novel genes to macular degeneration, underscoring the importance of C19orf44 in retinal health.²¹,²³ Detailed phenotyping is also essential for accurately correlating genetic variants with clinical manifestations, enabling precise diagnosis, understanding disease mechanisms and guiding personalised therapeutic strategies. Based on the 2015 ACMG/AMP criteria, the identified variants meet the criteria for likely pathogenicity, further supporting their role in disease causation.¹⁴

Notably, the three probands originated from diverse geographical regions: Europe, the Middle East and East Asia. This broad distribution across multiple ethnic groups suggests that the identified variants in C19orf44 are relevant in answering genetically unresolved cases in the global IRD genetic landscape. The apparent ethnic diversity of these findings emphasises the importance of including under-represented populations in genetic studies to uncover rare pathogenic variants and expand our understanding of the genetic aetiology of IRDs.

Interestingly, we identified a homozygous deletion (NM_032207.2:c.549_550del;p.Ser185Profs*2) in two patients within a shared IBD segment (chr19:16415688-16850140, hg19) and overlapping HBD regions. While this article was under preparation, the same variant was detected in nine patients affected with retinal dystrophies from Israel and Iran.²⁴ These findings strongly suggest a founder effect with the mutation originating in a common ancestor and being inherited through generations, resulting in homozygosity in both patients. Moreover, these findings suggest that c.549_550del;p.Ser185Profs*2 may represent a mutation hotspot within this gene for similar phenotypes and should be considered in future genetic testing.

The C19orf44 gene spans nine exons and encodes a 657 amino acid long protein of yet-to-be-characterised function. Expression analyses indicate that C19orf44 is ubiquitously expressed across diverse tissues, as reported in publicly available database (https://www.proteinatlas.org/ENSG00000105072-C19orf44/tissue). Although RNA sequencing data suggest that C19orf44 expression levels are low, the gene is detectable in both human fetal and adult retina, as well as across major retinal cell types (online supplemental figure 5B). Additionally, single-cell and single-nuclei expression profiles of C19orf44 across major retinal cell classes show similar results (online supplemental figure 5D,E). These findings imply a potential role for C19orf44 in retinal physiology, despite its relatively modest expression levels. A comparative sequence alignment of C19orf44 across species showed high conservation of amino acids in the DUF and C-terminus, suggesting these regions are functionally important and that truncating mutations may have harmful effects (figure 1F and online supplemental figure 6). Usually, the C-terminus is known to be critical for various cellular functions, including protein-protein interactions, post-translational modifications and intracellular trafficking.²⁵

To date, C19orf44 has not been implicated in any human disease in published literature, nor have its precise biological functions been elucidated. However, the identification of a SUMOylation domain within the protein provides insights into its possible functional roles. SUMOylation, a post-translational modification process, involves the covalent attachment of small ubiquitin-like modifier (SUMO) proteins to specific lysine residues on target proteins (figure 1F). This modification influences various cellular processes, including transcription regulation, cell cycle progression, cell signalling, and DNA synthesis and repair.²⁶ The presence of a SUMOylation domain in C19orf44 suggests that it may participate in these critical cellular pathways, potentially impacting retinal cell function. Further functional studies are required to delineate the specific roles of C19orf44, particularly in the context of retinal biology. Elucidating these roles may uncover previously unrecognised mechanisms of retinal maintenance and provide insights into its potential contributions to retinal diseases.²⁶

Future research should also assess the prevalence and penetrance of C19orf44 variants in IRD across populations and investigate the impact of LoF variants. Genetic and clinical evidence implicates LoF variants in C19orf44 as the cause of late-onset progressive retinal degeneration, expanding the genetic landscape of IRDs and highlighting the need to investigate the cellular roles of C19orf44 in vitro and in vivo. Furthermore, including C19orf44 in IRD gene panels can enhance diagnosis, genetic counselling and targeted therapies, advancing the understanding and management of C19orf44-related retinopathy.

In summary, this study provides compelling clinical and genetic evidence supporting the association of biallelic LoF variants in C19orf44 with autosomal recessive retinal dystrophies, specifically C19orf44-related retinopathy. The phenotype is characterised by late-adult-onset retinal degeneration, with progressive macular atrophy and significant visual field loss over time. The findings underscore the pathogenic role of C19orf44 LoF variants in this distinct retinal dystrophy and highlight the importance of considering C19orf44 in the genetic diagnosis of unexplained macular degenerations manifesting in later life.

The funding agencies do not have any involvement in this study.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.