The present study examines the safety profile of intravenous onasemnogene abeparvovec gene therapy in a real-world setting, both alone or in combination with intrathecal antisense oligonucleotide nusinersen therapy in two cohorts of patients with spinal muscular atrophy (SMA). The first cohort included eight presymptomatic infants treated solely with onasemnogene abeparvovec, while the second cohort comprised six symptomatic infants receiving onasemnogene abeparvovec and nusinersen co-therapy. All patients received the corticosteroid prednisolone coincident with gene therapy. Circulating alanine aminotransferase (ALT) and aspartate transaminase (AST) levels were measured to determine potential hepatoxicity, the primary focus of this study. Elevated ALT and AST levels were observed in one pre-symptomatic and three symptomatic patients post-treatment. However, all values returned to normal levels within 3 months of onasemnogene abeparvovec injection. Nusinersen treatment received previously or coincident with gene therapy did not impact the transient elevation of liver transaminases. This study highlights the importance of early intervention with molecular treatments for SMA and indicates that prior or coincident treatment with nusinersen is unlikely to impact safety of onasemnogene apoparvovec and could theoretically improve clinical outcomes in symptomatic infants or in those with gene therapy delayed beyond the immediate neonatal period.
Advancements in our understanding of genetic disease and adeno-associated virus has prompted great excitement into the field of AAV-mediated gene therapy, particularly for genetic diseases of the central nervous system, including retinal disorders. Despite significant progress, exemplified by the approval of therapies such as Luxturna® and Zolgensma®, a substantial number of therapies remain in pre-clinical or early clinical stages, with many failing to advance to later phases. Whilst the use of animal models to test safety and delivery route efficacy of AAV treatments is imperative, differences in tissue structure and physiology between humans and animal models has restricted precise disease modelling and gene therapy development for many CNS disorders. Alongside the FDA push for non-animal alternative models, researchers are increasingly turning to human-based models, including stem cell-derived organoids, which can offer a more accurate representation of human cellular microenvironments and niches. As such, this review explores the advantages and limitations of brain and retinal organoids as pre-clinical models of disease, with a primary focus on their utility in identifying novel AAV capsids, cell-specific promoters, and their role in recent pre-clinical AAV gene therapy studies.

