The Road towards Gene Therapy for X-Linked Juvenile Retinoschisis: A Systematic Review of Preclinical Gene Therapy in Cell-Based and Rodent Models of XLRS

Abstract

X-linked juvenile retinoschisis (XLRS) is an early-onset progressive inherited retinopathy affecting males. It is characterized by abnormalities in the macula, with formation of cystoid retinal cavities, frequently accompanied by splitting of the retinal layers, impaired synaptic transmission of visual signals, and associated loss of visual acuity. XLRS is caused by loss-of-function mutations in the retinoschisin gene located on the X chromosome (RS1, MIM 30083). While proof-of-concept studies for gene augmentation therapy have been promising in in vitro and rodent models, clinical trials in XLRS patients have not been successful thus far. We performed a systematic literature investigation using search strings related to XLRS and gene therapy in in vivo and in vitro models. Three rounds of screening (title/abstract, full text and qualitative) were performed by two independent reviewers until consensus was reached. Characteristics related to study design and intervention were extracted from all studies. Results were divided into studies using (1) viral and (2) non-viral therapies. All in vivo rodent studies that used viral vectors were assessed for quality and risk of bias using the SYRCLE’s risk-of-bias tool. Studies using alternative and non-viral delivery techniques, either in vivo or in vitro, were extracted and reviewed qualitatively, given the diverse and dispersed nature of the information. For in-depth analysis of in vivo studies using viral vectors, outcome data for optical coherence tomography (OCT), immunohistopathology and electroretinography (ERG) were extracted. Meta-analyses were performed on the effect of recombinant adeno-associated viral vector (AAV)-mediated gene augmentation therapies on a- and b-wave amplitude as well as the ratio between b- and a-wave amplitudes (b/a-ratio) extracted from ERG data. Subgroup analyses and meta-regression were performed for model, dose, age at injection, follow-up time point and delivery method. Between-study heterogeneity was assessed with a Chi-square test of homogeneity (I2). We identified 25 studies that target RS1 and met our search string. A total of 19 of these studies reported rodent viral methods in vivo. Six of the 25 studies used non-viral or alternative delivery methods, either in vitro or in vivo. Of these, five studies described non-viral methods and one study described an alternative delivery method. The 19 aforementioned in vivo studies were assessed for risk of bias and quality assessments and showed inconsistency in reporting. This resulted in an unclear risk of bias in most included studies. All 19 studies used AAVs to deliver intact human or murine RS1 in rodent models for XLRS. Meta-analyses of a-wave amplitude, b-wave amplitude, and b/a-ratio showed that, overall, AAV-mediated gene augmentation therapy significantly ameliorated the disease phenotype on these parameters. Subgroup analyses and meta-regression showed significant correlations between b-wave amplitude effect size and dose, although between-study heterogeneity was high. This systematic review reiterates the high potential for gene therapy in XLRS, while highlighting the importance of careful preclinical study design and reporting. The establishment of a systematic approach in these studies is essential to effectively translate this knowledge into novel and improved treatment alternatives.