To our knowledge, this is the first study to employ an acute treatment trial of daily administration of IN-OXT in ASD. Although no significant improvements were noted in the primary outcome measures when controlling for baseline differences, 6-week use of IN-OXT versus placebo resulted in improvements in aspects of social cognition, repetitive behaviors and emotional well-being in some, although not all, measures used – suggesting potential therapeutic benefit for IN-OXT in this population, which needs to be explored in follow-up larger studies. Further, our study did not report any serious adverse effects and IN-OXT was well tolerated. Our data are consistent with previous single-dose studies in individuals with ASD. Our data are also consistent with recent data from schizophrenia and social anxiety. Feifel and colleagues published an augmentation study of oxytocin in schizophrenia. In 19 participants with residual symptoms treated with 40 IU oxytocin versus placebo, the authors reported that oxytocin significantly reduced scores on the Negative Symptom Scale in addition to the Positive Symptom Scale and CGI – improvement scale compared with placebo at the 3-week end point. Oxytocin was well tolerated and no adverse effects were reported. In a randomized, double-blind, placebo-controlled trial, Guastella and colleagues administered 24 IU oxytocin or a placebo in combination with exposure therapy to 25 participants who met primary diagnosis for social anxiety disorder. Participants administered with oxytocin showed improved positive evaluations of appearance and speech performance, suggesting that following exposure therapy the administration of oxytocin improved the mental representations of self.
In this study the documented effects are on social perception (RMET), lower-order repetitive behaviors (RBS-R) and quality of life as it relates to emotion (WHOQOL – emotion, patients described an effect of well-being). There is large debate in the literature about the nature of central/peripheral effects of oxytocin on cognition/perception and behavior. Both amygdala and the nucleus accumbens are rich in OXT-R. The question remains whether the effects of oxytocin on social function are related to anxiety, social reward, social perception (for example, emotion, identity detection) or social cognition (for example, theory of mind). Evidence exists for each of these hypotheses. The anxiolytic properties of oxytocin have been well documented for both exogenous and endogenous release and are mediated by both central and peripheral mechanisms. However, effects on trust, theory of mind, emotion detection and positive and negative symptoms of schizophrenia have been repeatedly documented. In addition, in studies where the investigators controlled for anxiety using self-report measures (for example), oxytocin doses that were adequate to produce social cognition effects did not show effects of anxiety. Still, given that the relations between anxiety, affiliative behaviors and social cognition/perception are not yet well delineated, this remains an area of active research.
The effects on repetitive behaviors were interesting, with no effects observed for higher-order behaviors (for example, compulsive-like) but improvements noted for low order behaviors (for example, stereotypy)[32, 33], whereas no effects were noted in a measure that combined both types of behaviors in a single severity score (YBOCS). The possibility that the lower-order behavior score may be preferentially capturing pleasure-seeking repetitive behaviors and differentially responding to oxytocin compared with the higher-order domain, traditionally thought to be similar to the egodystonic behaviors of obsessive–compulsive disorder, is intriguing and needs to be examined further. This may be of particular interest given the paucity of available interventions for lower-order repetitive behaviors.
The effect sizes in our study for social cognition and function ranged from small/medium to very large. This was a short-duration (6 weeks), small-sample study and all such effect sizes need to be viewed with caution.
We acknowledge that our relatively small sample size resulted in baseline differences, which although not significant were large in some of the measures. However, the effects remained significant for both social cognition as measured by the RMET and the WHOQOL – emotional subscale after adding baseline scores as a covariate and there was a strong trend noted for improvements in repetitive behaviors as measured by the lower-order subscale of the RBS-R. In addition, the duration of the study was only 6 weeks and as such we may have underestimated the potential impact of oxytocin on core symptom domains. Furthermore, the sample size precludes us from examining the impact of other characteristics such as IQ and concurrent medications on treatment response and safety (Additional file1). The three female participants were all premenopausal and on no hormonal contraception, but we did not collect data on the time of menstrual cycle that may potentially interact with oxytocin. In addition, no information on relationship status was collected in this study.
A concern raised about trials of IN-OXT is related to the very short half-life of IN-OXT in the blood. However, in the context of documented behavioral effects lasting several hours to days, it seems that plasma levels of oxytocin are not related to such effects. Although much clarity is needed about the mechanism of the behavioral effects noted with IN-OXT administration, it is worth noting that levels of vasopressin in the cerebral spinal fluid were still rising 4 hours post administration of a single intranasal dose of the peptide in a cerebral spinal fluid pharmacokinetic study.
Placebo in this study was normal saline. Although the participants detected no taste/smell for the active compound, normal saline is not the vehicle for intranasal Syntocinon and future studies should use placebo with identical inactive ingredients.
Lastly, the study did not follow-up the participants once the medication had been stopped. Given the animal literature suggesting that oxytocin promotes neural plasticity changes, we would predict that any effects should outlast the period of actual administration. Follow-up studies need to include such a design.