Rethinking gene therapy for X-linked retinoschisis: from vitreous to subretinal space

Wait 5 sec.

In a recent study published in The New England Journal of Medicine, Liang and colleagues report that subretinal delivery of scAAV8-hRS1 to 12 children with X-linked retinoschisis (XLRS) resulted in closure of macular schisis by 13 weeks in all treated eyes and restoration of outer retinal structure in most participants.1 These findings provide encouraging evidence that RS1 gene augmentation can substantially improve retinal structure in X-linked retinoschisis, while raising the important question of whether anatomical rescue will translate into meaningful and durable improvements in visual function.XLRS (OMIM: 312700) is an X-linked recessive retinal disorder caused by pathogenic variants in RS1, which encodes the 224 amino acid protein retinoschisin. It predominantly affects males and typically presents in childhood with reduced central vision, foveal schisis, and an electronegative full-field electroretinogram (ffERG), indicating predominant inner-retinal dysfunction, and there is considerable variability in disease severity and progression. Retinoschisin is a secreted, disulphide-linked octamer mainly produced by photoreceptors and bipolar cells that supports retinal structure and photoreceptor–bipolar cell signalling. Consequently, the cystic cavities visible on OCT are only one manifestation of XLRS rather than the full cause of the functional disease process.The rationale for RS1 augmentation was provided by studies in Rs1 knockout mice. In these animals, intravitreal AAV-RS1 administration produced improvements in retinal structure and function, establishing an important preclinical basis for human translation.2 However, human clinical trials of intravitreal RS1 gene therapy, while generally safe over 1 year, revealed that inflammatory ocular adverse events were common and there was no overall improvement in best corrected visual acuity (BCVA), visual field, ffERG, or cystic-cavity volume relative to fellow eyes.3,4 These studies may have revealed the key limitation of intravitreal delivery for XLRS in humans. The vector must cross the vitreous and internal limiting membrane, distribute through a structurally abnormal retina, and transduce the bipolar and photoreceptor cells relevant to retinoschisin biology. In humans, this is likely to be less efficient than in the mouse.The recent study of RS1 augmentation in XLRS by Liang et al. provides indication that the delivery route may alter the therapeutic potential of RS1 therapy (Fig. 1). In their study of 12 patients aged 5–11 years, a single subretinal injection of self-complementary AAV8 encoding human RS1 was associated with an average treatment-eye BCVA gain of 10.8 ± 4.9 letters using the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at 1-year post-therapy, compared with 2.4 ± 3.3 letters in untreated fellow eyes. Seven of 12 treated eyes achieved gains of at least 10 letters. Treated eyes showed closure of macular schisis cavities and reduced retinal thickness, but there were no changes in b-wave amplitude or retinal sensitivity.1Fig. 1Full size imageOverview of subretinal RS1 gene augmentation for X-linked retinoschisis including disease biology and therapeutic approach. a Disease biology highlighting schitic cavities present in the disorganized XLRS retina, and the impact of pathogenic variants in RS1 leading to structural and photoreceptor-bipolar signalling defects, with resultant decrease in central vision and b-wave on ffERG. b Therapeutic approach including a representation of the retinoschisin (RS1) protein showing functional domains and the location of pathogenic variants in the treated participants reported by Liang et al. The RS1 protein comprises a signal peptide (residues 1–23), RS1-specific domain (residues 24–62), central discoidin domain (residues 63–219), and a short C-terminal segment (220–224). All missense variants identified in the study map to the discoidin domain. Each symbol represents one participant; multiple symbols at the same residue indicate participants sharing the same variant. Participants were treated with subretinal administration or either 7.5 × 1010 (open circles) or 1 × 1011 (closed circles) vector genomes of the scAAV8-hRS1 vector. One additional participant had a partial RS1 genomic deletion (chrX:g.18,671,528 to 18,675,852) which is not shown on the protein schematic. A self-complementary AAV8 vector encoding human RS1 was administered via subretinal injection with the injection bleb covering the macular region. ffERG full-field electroretinography, ILM internal limiting membrane, OCT optical coherence tomography, PPV pars plana vitrectomy, RPE retinal pigment epitheliumOver 52 weeks, no grade 3 or higher adverse events or ocular inflammation were reported, contrasting with the inflammatory signal observed in previous intravitreal studies. This supports the possibility that localised subretinal administration can offer more effective retinal exposure without adverse safety impact. However, the trial was small, open label, and short in duration. Fellow-eye comparisons are useful in a bilateral inherited disorder and, while they can control for age-related changes and learning effects during visual-acuity testing, they may be limited by eye-specific differences in disease severity or progression, fluctuation of schitic cavities, and treatment responses. The findings should be viewed as an encouraging signal rather than definite evidence of clinical benefit.An important point in considering Liang et al. is the avoidance of equating anatomical response with visual recovery. Macular schisis is an attractive OCT endpoint because it is visible, quantifiable, and often changes after intervention. However, cavity reduction alone may not restore visual acuity if dysfunction occurs because of photoreceptor damage, defective bipolar-cell transmission, synaptic remodelling, or other consequences of RS1 deficiency. This distinction is highlighted by acetazolamide studies in XLRS, which may be associated with reduced schitic lesions, but visual acuity and other outcome measures do not show clearcut improvement. The XLRS outcome-measure literature shows that OCT-derived central macular thickness can vary with cystic changes and correlates imperfectly with function. BCVA may be the outcome measure most likely to demonstrate meaningful change, but more refined structure–function measures are needed.The biology of XLRS helps explain why restoration of retinoschisin may have variable effects across patients. Retinoschisin contributes to retinal structural integrity but also appears integral to signalling across photoreceptor and bipolar-cell layers. The characteristic reduction in the ffERG b-wave is therefore not simply a secondary consequence of schisis but is also likely due to functional inner-retinal abnormality. Notably, the conventional definition of an electronegative ffERG in XLRS with a b-wave:a-wave (b:a) ratio below 1.0 is relatively insensitive for genetically confirmed XLRS, with cases often with a b:a ratio of 1.0 or higher. This supports use of more nuanced electrophysiological measures rather than a binary ffERG classification in therapeutic trials.A further issue in XLRS clinical trial interpretation is genotype. Most preclinical RS1 work has used Rs1 knockout models, which model absent protein rather than the missense variants that constitute more than 50% of patient alleles (ClinVar, https://www.ncbi.nlm.nih.gov/clinvar/) and comprise most patients included in clinical trials to date, including the cohort reported by Liang et al. Missense variants may produce misfolded, intracellularly retained, poorly secreted, or dysfunctional retinoschisin, with potentially distinct disease mechanisms and responses to gene augmentation. Augmentation therapy has been tested in an Rs1 R213W missense mouse model. Intravitreal delivery using a novel rationally designed capsid and an optimised RS1 expression cassette reduced retinoschisis, normalised the ffERG b:a ratio, and restored b-wave amplitude to wild-type levels by 16 weeks.5 These results support the plausibility of gene augmentation in missense disease, but they do not remove the need for human genotype-stratified analyses.Liang et al. suggest that subretinal delivery can overcome some barriers encountered by intravitreal vectors, particularly in obtaining adequate relevant retinal cell-type exposure. There may be downsides to this, including that subretinal injection may be technically challenging in an already fragile, schitic macula with required retinal detachment at the injection site. This makes evolving vector technology highly relevant. There has been a description of AAV.SPR, a laterally spreading capsid that, after peripheral subretinal injection in non-human primates, produced central retinal expression beyond the bleb margin, reported to be up to 7.2 mm from the injection edge, without requiring a central retinal detachment. Such vectors could combine the transduction advantage of subretinal administration with a potentially safer surgical approach for XLRS. This approach is currently being evaluated for XLRS with ATSN-201 in the LIGHTHOUSE trial (NCT05878860; Atsena Therapeutics).Future studies should consider disease stage and RS1 genotype and incorporate a comprehensive assessment of visual function such as low luminance visual acuity as well as BCVA, microperimetry and electrophysiology in addition to structural measures and patient-reported outcomes. Cognisance of natural-history data in selection of the most suitable outcome measures is especially important because XLRS visual acuity often declines slowly, making short-term changes difficult to interpret.The findings described in Liang et al. represent an important step in XLRS augmentation therapy. They provide evidence that subretinal RS1 augmentation can achieve anatomical rescue and suggestion of an initial visual-acuity signal in children with XLRS. The next challenge is to establish whether this benefit is durable, generalisable across genotypes, linked to restored retinal function, and achievable with delivery systems that minimise risk.ReferencesLiang, L. et al. Subretinal gene therapy for X-linked retinoschisis. N. Engl. J. Med. 394, 2223–2234 (2026).Article  CAS  PubMed  Google Scholar Bush, R. A. et al. Preclinical dose-escalation study of intravitreal AAV-RS1 gene therapy in a mouse model of X-linked retinoschisis: dose-dependent expression and improved retinal structure and function. Hum. Gene Ther. 27, 376–389 (2016).Article  CAS  PubMed  PubMed Central  Google Scholar Cukras, C. et al. Retinal AAV8-RS1 gene therapy for X-linked retinoschisis: initial findings from a phase I/IIa trial by intravitreal delivery. Mol. Ther. 26, 2282–2294 (2018).Article  CAS  PubMed  PubMed Central  Google Scholar Pennesi, M. E. et al. Intravitreal delivery of rAAV2tYF-CB-hRS1 vector for gene augmentation therapy in patients with X-linked retinoschisis. Ophthalmol. Retina 6, 1130–1144 (2022).Article  PubMed  Google Scholar Wei, X. et al. A novel AAV capsid-mediated RS1 gene therapy restored retinal function to wild-type levels in Rs1R213W mouse model. Investig. Ophthalmol. Vis. Sci. 66, 37 (2025).Article  Google Scholar Download referencesAcknowledgementsThe figure was created using Adobe Illustrator with components from BioRender.com.Author informationAuthors and AffiliationsEye Genetics Research Unit, Children’s Medical Research Institute, Sydney Children’s Hospitals Network, Save Sight Institute, The University of Sydney, Sydney, NSW, AustraliaRobyn V. Jamieson & John R. GriggDepartment of Clinical Genetics, Western Sydney Genetics Program, Sydney Children’s Hospitals Network, Sydney, NSW, AustraliaRobyn V. JamiesonSpecialty of Genomic Medicine, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, AustraliaRobyn V. JamiesonSydney Eye Hospital, Sydney, NSW, AustraliaJohn R. GriggSpecialty of Ophthalmology, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, AustraliaJohn R. GriggGene Therapy Research Unit, Children’s Medical Research Institute, Faculty of Medicine and Health, The University of Sydney and Sydney Children’s Hospitals Network, Sydney, NSW, AustraliaSamantha L. GinnAuthorsRobyn V. JamiesonView author publicationsSearch author on:PubMed Google ScholarJohn R. GriggView author publicationsSearch author on:PubMed Google ScholarSamantha L. GinnView author publicationsSearch author on:PubMed Google ScholarContributionsJ.R.G. supplied the spectral domain OCT scan from a deidentified patient used in Fig. 1a. All authors wrote the article and created the figure. All authors have read and approved the article.Corresponding authorCorrespondence to Samantha L. Ginn.Ethics declarationsCompeting interestsThe authors declare no competing interests.Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Rights and permissionsOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.Reprints and permissionsAbout this article