To characterize Aicardi Goutières Syndrome neurodevelopment

A mimic of congenital infections and rare genetic cause of interferon overproduction, Aicardi Goutières Syndrome (AGS) results in significant neurologic disability. AGS results from pathogenic changes in the intracellular nucleic acid sensing machinery (TREX1, RNASEH2A, RNASEH2B, RNASEH2C, SAMHD1, ADAR1, and IFIH1). All individuals exhibit neurologic impairment: from mild spastic paraparesis to severe tetraparesis and global developmental delay. We hypothesize that genotype influences the heterogeneous developmental trajectory found in AGS.

To characterize this spectrum, age and symptoms at presentation and longitudinal developmental skill acquisition was collected from an international cohort of children (n=88) with genetically confirmed AGS.

We found that individuals present at variable ages, with the largest range in SAMHD1, ADAR, and IFIH1. There are 3 clusters of symptoms at presentation: altered mental status (irritability or lethargy), systemic inflammatory symptoms, and acute neurologic symptoms, with variability across all genotypes. By creating Kaplan-Meier curves for developmental milestones, we were able to create genotype-based developmental trajectories for the children affected by the 5 most common genotypes: TREX1, IFIH1, SAMHD1, ADAR, and RNASEH2B. Individuals with TREX1-related AGS were the most severely affected, significantly less likely to reach milestones compared to the other genotypes, including head control, sitting, and nonspecific mama/dada (p-value <0.005). Individuals affected by SAMHD1, IFIH1, and ADAR collectively attained the most advanced milestones, with 44% of the population achieving a minimum of a single word and 31% able to walk independently. Three retrospective scales were also applied: Gross Motor Function Classification System, Manual Ability Classification Scale, and Communication Function Classification System. Within each genotypic cohort, there was pronounced heterogeneity.

Our results demonstrate the influence of genotype on early development, but also suggest the importance of other unidentified variables. These results underscore the need for deep phenotyping to better characterize subcohorts within the AGS population.