Nonverbal Cognitive Skills in Children With Aicardi Goutières Syndrome

Abstract

Background and objectives: Aicardi Goutières syndrome (AGS) is type I interferonopathy characterized by severe neurologic impairment. Although many children with AGS demonstrate motor and expressive language deficits, the magnitude of receptive language impairment is uncharacterized. We sought to characterize cognitive function in AGS-affected children using assessment tools with reduced dependence on motor abilities and compare cognitive testing outcomes with overall severity and parental assessment of adaptive behavior.

Methods: We performed a cross-sectional study. Children were recruited as part of the Myelin Disorders Biorepository Project at the Children's Hospital of Philadelphia. We included individuals with a confirmed diagnosis of AGS. We administered the Leiter International Performance Scale, third edition (Leiter-3), and the Vineland Adaptive Behavior Scale, third edition (VABS-3), in the context of research encounters. Motor skills were categorized by AGS Severity Scale mobility levels. Descriptive statistics and Spearman's rank correlation were used to compare assessments. Mann-Whitney and Kruskal-Wallis tests with correction with Dunn's multiple comparison test were used to compare test performance between mobility groups.

Results: Cognitive and adaptive behavior performance was captured in 57 children. The mean age at encounters was 8.51 (SD 5.15) years. The median (IQR) Leiter-3 score was 51 (interquartile range [IQR] 60), with administration failure in 20 of 57 (35%) individuals. On the VABS-3, the Motor Domain (median 29, IQR 36.25) was more impacted than the Communication (median 50, IQR 52), Daily Living Skills (median 52, IQR 31), and Socialization (median 54, IQR 40) Domains (p < 0.0001). The AGS Scale correlated with VABS-3 (r = 0.86, p < 0.0001) and Leiter-3 (r = 0.87, p < 0.0001). There was correlation between VABS-3 Domains and Leiter-3 (r-range 0.83-0.97). Gross motor and fine motor categories, respectively, correlated with VABS-3 (H = 39.37, p < 0.0001; U = 63, p < 0.0001) and Leiter-3 (H = 40.43, p < 0.0001; U = 66, p < 0.0001). Within each gross motor and fine motor category of the AGS Scale, a subset of children scored within normal IQ range.

Discussion: Parental assessment of function by the VABS-3 correlated with directly assessed performance measures. Our data underscore the potential value of VABS-3 and Leiter-3 as tools to assess psychometric function in AGS. With a deeper understanding of our patients' abilities, we can better guide clinicians and families to provide appropriate support and personalized interventions to empower children with leukodystrophies to maximize their communication and educational potential.

Figure 1. Performance of the AGS Cohort at the VABS-3

(A) By VABS-3 domain performance, Motor Skills were the lowest compared with Communication, Daily Living Skills, and Social Domains (mixed effects with multiple comparisons with Bonferroni correction, all pairwise comparisons, p < 0.0001). (B) Within VABS-3 subdomains, the performance within the expressive language and writing abilities was lower than the performance of receptive language, whereas gross motor skills were noted by families to be more impacted than fine motor skills (analysis details included in eTable 1). ABC = Adaptive Behavior Composite; AGS = Aicardi Goutières syndrome; VABS-3 = Vineland Adaptive Behavior Scale, third edition.

Figure 2. Performance of the AGS Cohort at the Leiter-3

Mean Leiter-3 nonverbal IQ was 60.6 standard score (range 30–117, IQR 60). Of the 57 children to whom the Leiter-3 was administered, 20 children were unable to complete any items. Excluding the failed administration cohort, the median recorded nonverbal IQ was 77 (IQR 54–117; n = 37). AGS = Aicardi Goutières syndrome; IQR = interquartile range; Leiter-3 = Leiter International Performance Scale, third edition.

Figure 3. Internal Validity Between VABS-3 and Leiter-3

There was strong internal validity between domains of the VABS-3 with performance on the Leiter-3 (correlation matrix range 0.83–0.97). Leiter-3 = Leiter International Performance Scale, third edition; VABS-3 = Vineland Adaptive Behavior Scale, third edition.

Figure 4. Correlation Between VABS-3, Letier-3, and AGS Severity Scale

Performance on the AGS scale was strongly correlated with VABS-3 (Spearman's rank correlation coefficient r = 0.86) and Leiter-3 (r = 0.87) scores. AGS = Aicardi Goutières syndrome; Leiter-3 = Leiter International Performance Scale, third edition; VABS-3 = Vineland Adaptive Behavior Scale, third edition.

Figure 5. Correlation Between VABS-3 and Leiter 3 and Mobility Categorization

(A, B) Gross motor categorization correlated with VABS-3 and Leiter-3 performance (Kruskal-Wallis with Dunn correction, p < 0.0001). Within each mobility category, a subset of children scored within an IQ normal range: 7% (2/27) nonmobile, 67% (10/15) rolling/crawling, 80% (4/5) walking with assistance, and 100% (10/10) of children who were independently ambulatory. (C, D) Fine motor categorization correlated with VABS-3 and Leiter-3 performance (Kruskal-Wallis with Dunn correction, p < 0.0001). Overall, 12% (3/26) of children unable to self-feed or achieve pincer grasp and 74% (23/31) scored within an IQ normal range. Leiter-3 = Leiter International Performance Scale, third edition; VABS-3 = Vineland Adaptive Behavior Scale, third edition.