Oxidative Stress and Immune Cytokines in Plasma of Young Children with Autism Spectrum Disorder and Recent Language and/or Social Regression: A Prospective Case-Control Study

Background:  Regression occurs in 22%-41% of children with Autism Spectrum Disorder (ASD) at the average age of 20-23 months. At present, no clear contributors to regression have been identified. Increased levels of several immune cytokines have been reported in the subgroup of regressive autism, but no studies have examined the role of oxidative stress in this process.

Objectives:  To compare biomarkers of oxidative stress and immune cytokines in children with ASD, with and without regression.

Methods:   Three sites in the Autism Speaks Autism Treatment Network: Toronto, ON; Baltimore, MD; and Little Rock, AS, recruited children, aged 18-42 months with ASD, with regression (ASD-R) (n=25), and without regression (ASD-NR) (n=24). The two groups were compared with regards to oxidative stress markers (free reduced glutathione, oxidized glutathione, homocysteine, cysteine, 3-nitrotyrosine and 3-chlorotyrosine) and immune cytokines (interleukin 1, macrophage chemoattractant protein 1, macrophage inflammatory protein 1A, interferon gamma, TNF alpha, and CD 40L). Regression was defined as having established a skill for at least one month, followed by a significant loss of that skill (more than 90% for language, and 75% for social) for more than 1 month. Regression was operationalized using a modified Regression Validation Interview and inter rater reliability for the assessment of “definite” or “no” regression was 89%. The ASD-R group completed a standardized medical investigation including screening for metabolic and genetic disorders. Children met criteria for an ASD according to the Autism Diagnostic Observation Schedule and the DSM-IV.

Results:  The mean ages of the ASD-R and ASD-NR groups were similar, 2.5 and 2.6 years respectively. The mean age of regression in the ASD-R group was 1.9 ± 0.4 years and the duration from when regression was definitively noted to plasma analysis was 8.0 ± 4.5 months.  The ASD-R group had a significantly lower mean free reduced glutathione level (1.71 ± 0.13) compared to the ASD-NR group (1.86 ± 0.20) (p=0.006).  There was also an increase in a marker for protein oxidative damage, 3-nitrotyrosine (p=0.04) in the ASD-R (74.19 ± 17.14) compared to the ASD-NR (64.78 ± 12.30) group. There were no differences in other oxidative stress markers or immune cytokines between groups.

Conclusions:  In this pilot study, ASD and recent regression in young children was associated with decreased free glutathione and increased 3-nitrotyrosine, supporting evidence of decreased antioxidant potential and increased oxidative damage in plasma, compared to young children with ASD and no regression. There were no group differences in other biomarkers of oxidative stress. Lower free glutathione levels may be a risk factor for regression in a subgroup of children with ASD and regression.  This study was limited by a small sample size and the analysis of oxidative stress biomarkers and cytokine levels was limited to plasma. Future studies should examine oxidative stress biomarkers in the plasma and cerebrospinal fluid in young children with ASD and recent regression to assess for possible differences in oxidative stress and damage in the central nervous system.