Cortisol Levels In Adolescents with ASD and Typical Development

Background: Research suggests that anxiety, depression and behavioral difficulties may increase during adolescence for some individuals with ASD.  Hypothalamo-pituitary-adrenal (HPA) system dysregulation is a general risk factor for psychopathology in typically developing populations. Patterns of cortisol secretion are one way to assess HPA function.  Alterations in expected circadian rhythm of cortisol secretion have been observed in children with ASD and an abnormal suppression of the typical pattern of cortisol release after waking has been reported in one study of adolescents with ASD.  However, additional studies are needed to understand the potential for cortisol measurement to inform psychopathology risk or presentation in this population.

Objectives: We aim (1) to compare levels of cortisol in adolescents with ASD vs. typical development (TYP), (2) to compare variability in cortisol levels across two days in ASD vs. TYP groups, (3) to investigate the relationship of ASD symptom severity, associated symptoms, and behavioral functioning with cortisol levels.

Methods: In the context of a longitudinal study, salivary cortisol samples were collected via Salimetrics home collection kits from adolescents with ASD (n=16 males, 4 females) and those with a history of typical development (TYP; n=13 males, 5 females). Cortisol was measured as nanograms per milliliter (ng/mL) three times; upon awakening (wake), 30-minutes after awakening (30-min), and 12 hours after awakening (12-hrs) across two days. The groups did not differ significantly by age (ASD M=14.72, SD=1.13; TYP M=14.68, SD=1.49), or gender.  The TYP group had higher general cognitive ability scores on the DAS-II (ASD M=96.95; range=30-127; TYP M=123.72, range=104-150, F(1,39)=14.33, p<.001).  Autism symptom severity was assessed with the ADOS and associated symptoms and behavioral functioning with the CBCL through parent report.

Results: (1) Cortisol levels did not significantly differ between the ASD and TYP groups at any point of the day (ASD M(SD) wake=3.10(1.47), 30-min=3.70(1.09), 12-hrs=0.62(0.39); TYP M(SD) wake=3.19(1.48), 30-min=4.27(1.87), 12-hrs=0.66(0.35); all F tests n.s.). (2) Cortisol levels at 30-min showed significantly less day-to-day stability in the ASD (r=0.04) compared with the TYP (r=0.70) group via multiple regression analysis (t=2.47, p=0.019). (3) The ASD group showed a significant correlation between parent-reported internalizing problems and the 30-min cortisol level (r=0.52, p=0.49).  Although larger, this relationship was not significantly different from the TYP group (r=0.29, p=0.25) group (t=-1.47, p=0.15).  ASD symptoms did not correlate significantly with cortisol level in the ASD group.

Conclusions: This study found that adolescents with ASD demonstrated patterns of cortisol secretion consistent with typically developing adolescents.  However, they demonstrated more variability in cortisol levels 30 minutes after awakening compared with typically developing adolescents across two days.  Inconsistent cortisol levels 30 minutes after awakening may indicate HPA dysregulation under stress. Evidence was also found for an association between 30-min cortisol and parent report of internalizing problems in the ASD and TYP groups.  Investigating cortisol in adolescents with ASD may offer insight into the development of associated psychiatric conditions in ASD.