Anandamide As a Blood-Based Biomarker in Children with Autism Spectrum Disorder

Background: Autism spectrum disorder (ASD) is a brain disorder that emerges in early childhood and is characterized by social functioning impairments and repetitive, and stereotyped behaviors. Optimal outcomes rely on early identification and intervention, but progress has been hindered by a poor understanding of ASD pathophysiology. Identification of a diagnostic or phenotypic ASD biomarker would enhance therapeutic development by identifying novel intervention targets. Recent data have shown that endogenous cannabinoids (endocannabinoids, or eCBs), a class of lipid neuromodulators that regulate neurotransmitter release, also contribute to social functioning. Moreover, preclinical ASD research implicates the major eCBs, anandamide (AEA) and 2-arachidonoylglycerol (2-AG), in ASD pathophysiology.

Objectives: We sought to develop a reproducible liquid chromatography with tandem mass spectrometry (LC-MS/MS) method to detect and quantify endocannabinoid (eCB) levels in small volumes of peripherally sampled human blood. LC-MS/MS is the most comprehensive method to quantitatively characterize lipids in children with ASD. The methodology was used to compare plasma eCB levels between children with and without ASD and to test whether eCB levels predicted behavioral measures. We hypothesized that children with ASD would have altered eCB tone compared to neurotypical controls, and that plasma eCB concentrations would be related to variation in ASD behavioral functioning.

Methods: A conventional toluene liquid-liquid extraction was compared to a modified salting-out assisted liquid-liquid extraction (SALLE) with acetonitrile on reproducibility, efficiency, and matrix effects. Following lipid extraction, an optimized LC-MS/MS method was used to quantify AEA (350Da) levels in banked plasma samples collected from phenotypically well-characterized children with ASD (N=57) and matched neurotypical controls (N=54). A TSQ Vantage triple quadrupole mass spectrometer coupled with an Accela 1250 HPLC system was operated in positive mode using heated electrospray ionization. Data acquisition was performed in selected reaction monitoring mode and processed with Xcalibur software. We used JMP 12.1 to analyze data with general linear models that corrected for age, gender, ethnicity, and LC-MS/MS sample run.

Results: SALLE reduced extraction time, increased recovery, and enhanced inter-sample consistency. LC-MS/MS methodology was highly sensitive with lower limits of quantitation for AEA at 50fg. Blood AEA levels were similar in children with ASD and neurotypical controls (mean ± SE: 334±14 vs. 343±14 pg/mL; F_1,102=0.73; p=0.395). Among children with ASD, IQ was both lower than in controls (83.6±3.7 vs. 115.5±1.3; F_1,102=59.46; p<0.0001) and negatively correlated with AEA concentrations (F_1,57=8.05, p=.007). Thus, when controlling for IQ, blood AEA concentrations were significantly lower in children with ASD than in neurotypical controls (F_1,101=6.63; p=0.012). In children with ASD, blood AEA levels, controlling for IQ, negatively predicted total scores on the Vineland Adaptive Behavior Scale (VABS; F_1,46=5.22; p=0.027), and more specifically, scores on the Socialization (F_1,47=4.28, p=0.044) and Motor Skills (F_1,46=7.17, p=0.010) subscales.

Conclusions: These preliminary findings demonstrate that children with ASD show abnormalities in blood eCB levels compared to neurotypical control children, and that blood eCB levels may predict behavioral measures in children with ASD. Results implicate a meaningful role of AEA tone in ASD pathophysiology and suggest that AEA may hold promise as a peripheral marker of ASD pathophysiology.