Studying the Continuum of Autistic Traits with Peripheral Biomarkers: Preliminary Evidence for Vasopressin, Oxytocin, and BDNF

Background: There is still a debate if behavioral symptoms should be considered dimensionally with disability lying at one extreme of the distribution of traits, or whether they should be categorically defined. Some evidence supports the notion that autistic symptoms and behaviors should be regarded as dimensional traits. Recently, autistic traits in healthy controls have been correlated with thinner cortex in the superior temporal gyrus. To date, there are no available data on potential biochemical correlates for the autism spectrum continuum. However, studies have reported altered levels of several biomarkers in Autism: in particular increased levels of brain-derived neurotrophic factor (BDNF) and arginine-vasopressin (AVP) have been observed in individuals with autism spectrum conditions (ASC). On the other hand, a disruption in the oxytocin (OXT) system resulting in lower plasma levels of this peptide has been reported in participants with Autism.  

Objectives:  The present study aimed to investigate the role of AVP, BDNF and OXT as potential biochemical correlates of subclinical autistic traits in a cohort of healthy young adults.

Methods: One hundred and twenty-three subjects (64 males, 59 females) were recruited who were between 20 and 37 years old (M=23.17, SD=3.97). All participants were screened by a senior psychiatrist to exclude the presence of psychiatric disorders. Participants completed the Autism Spectrum Quotient (AQ), a widely used measure for the identification of autistic traits in the general population. Blood samples were obtained from all participants at the same time of the day to allow for circadian variation. All women in the study were taking oral contraceptives and thus we considered their sexual hormone levels as stable. Analyses were conducted using commercially available ELISA kits (R&D Systems, Minneapolis, MN) according to the manufacturer’s instruction.

Results:  We conducted a multiple regression analysis using the AQ score as the dependent variable and sex, vasopressin, oxytocin and BDNF levels as the independent variables. The model was significant (p<0.01), explaining the 46% of the variance of the AQ score. Among the parameters included in the analysis, vasopressin levels (p=0.01) and BDNF levels (p<0.006) were independent predictors of AQ score.

Conclusions:  The association between a gradient of autistic traits and AVP, BDNF and OXT levels in healthy adults and patients with autism will be discussed. These preliminary results support the hypothesis of a continuum underlying autistic traits in the general population.