Six New Gene Variants Found Behind Rare Inherited Retinal Dystrophies

Inherited retinal diseases (IRDs) are genetically heterogeneous conditions — over 330 causative genes are known — and represent a leading cause of working-age blindness. IMPG1 and IMPG2 encode glycoproteins (SPACR and SPACRCAN) integral to the interphotoreceptor matrix (IPM), a carbohydrate-rich scaffold that supports photoreceptor homeostasis, nutrient transport, and retinal attachment. Because IMPG1 localization depends on IMPG2 and vice versa, variants in either gene can produce overlapping phenotypes ranging from mild maculopathy to severe rod-cone dystrophy.

This retrospective Oxford cohort study identified 13 unrelated patients with IMPG1 or IMPG2 variants via next-generation sequencing (NGS) through the Oxford University Hospitals Medical Genetics Laboratory. Patients underwent comprehensive clinical evaluation including ultra-widefield fundus imaging, fundus autofluorescence (FAF), and optical coherence tomography (OCT). In silico tools (SIFT, PolyPhen-2, MutationTaster, SpliceAI) assessed variant pathogenicity, and evolutionary conservation analysis was applied to novel missense variants.

Two IMPG1 patients were monoallelic: one carried a missense variant and presented with AVMD and drusenoid changes at age 81; the other carried an exon deletion and presented with pattern dystrophy at age 72. Both had relatively mild phenotypes consistent with autosomal dominant disease. Among 11 IMPG2 patients, five monoallelic cases presented with maculopathy (4 AVMD, 1 PD) at a mean age of ~54 years and retained good visual acuity (mean BCVA ~0.15 logMAR). In contrast, six biallelic IMPG2 patients had autosomal recessive retinitis pigmentosa with severe early maculopathy, symptom onset at ~18 years, and near-total vision loss by a mean age of ~69 years (mean BCVA ~1.9 logMAR). Eleven distinct IMPG2 variants were found (4 missense, 7 truncating); one missense variant (c.871C>A, p.Arg291Ser) was predicted by SpliceAI to cause aberrant splicing despite appearing missense. Six variants across both genes were novel, expanding the known genetic landscape.

The presence of truncating variants in both monoallelic and biallelic IMPG2 cases supports haploinsufficiency as the likely mechanism driving maculopathy in heterozygous carriers. OCT findings in monoallelic IMPG2 cases included characteristic hyper-reflective material above the RPE/Bruch's membrane complex — the hallmark vitelliform lesion — with some cases showing lesion resorption, cavity formation, and eventual central macular atrophy over time, consistent with prior longitudinal reports.

These findings have direct clinical implications: accurate genotyping enables correct inheritance pattern counseling, informs prognosis (monoallelic = slower, milder course; biallelic = severe, early onset), and could guide patient selection for emerging gene therapy trials targeting IMPG1/IMPG2 pathways. Limitations include the small sample size and retrospective design, which preclude robust genotype-phenotype statistical analysis.