Behavioral effects of multiple-dose oxytocin treatment in autism: a randomized, placebo-controlled trial with long-term follow-up

Abstract

Background: Intranasal administration of the "prosocial" neuropeptide oxytocin is increasingly explored as a potential treatment for targeting the core characteristics of autism spectrum disorder (ASD). However, long-term follow-up studies, evaluating the possibility of long-lasting retention effects, are currently lacking.

Methods: Using a double-blind, randomized, placebo-controlled, parallel design, this pilot clinical trial explored the possibility of long-lasting behavioral effects of 4 weeks of intranasal oxytocin treatment (24 International Units once daily in the morning) in 40 adult men with ASD. To do so, self-report and informant-based questionnaires assessing core autism symptoms and characterizations of attachment were administered at baseline, immediately after 4 weeks of treatment (approximately 24 h after the last nasal spray administration), and at two follow-up sessions, 4 weeks and 1 year post-treatment.

Results: No treatment-specific effects were identified in the primary outcome assessing social symptoms (Social Responsiveness Scale, self- and informant-rated). In particular, with respect to self-reported social responsiveness, improvements were evident both in the oxytocin and in the placebo group, yielding no significant between-group difference (p = .37). Also informant-rated improvements in social responsiveness were not significantly larger in the oxytocin, compared to the placebo group (between-group difference: p = .19). Among the secondary outcome measures, treatment-specific improvements were identified in the Repetitive Behavior Scale and State Adult Attachment Measure, indicating reductions in self-reported repetitive behaviors (p = .04) and reduced feelings of avoidance toward others (p = .03) in the oxytocin group compared to the placebo group, up to 1 month and even 1 year post-treatment. Treatment-specific effects were also revealed in screenings of mood states (Profile of Mood States), indicating higher reports of "vigor" (feeling energetic, active, lively) in the oxytocin, compared to the placebo group (p = .03).

Conclusions: While no treatment-specific improvements were evident in terms of core social symptoms, the current observations of long-term beneficial effects on repetitive behaviors and feelings of avoidance are promising and suggestive of a therapeutic potential of oxytocin treatment for ASD. However, given the exploratory nature of this pilot study, future studies are warranted to evaluate the long-term effects of OT administration further.

Trial registration: The trial was registered with the European Clinical Trial Registry (Eudract 2014-000586-45) on January 22, 2014 (https://www.clinicaltrialsregister.eu/ctr-search/trial/2014-000586-45/BE).

Keywords: Attachment; Autism spectrum disorder; Oxytocin; Repetitive and restricted behavior; Social responsiveness.

Fig. 1

CONSORT Flow diagram. Data were analyzed using an intention-to-treat format with last-observations-carried-forward to replace missing data. For participants with missing baseline data, analysis for that measure was excluded list-wise. SRS-A Social Responsiveness Scale adult version, RBS-R Repetitive Behavior Scale Revised, SAAM: State Adult Attachment Measure, IPPA Inventory of Parent and Peer Attachment, WHO-QOL World Health Organization Quality of Life

Fig. 2

Effects of oxytocin treatment on autism symptoms and attachment. Mean pre-to-post changes (change from baseline) on self-report and informant-based questionnaires are visualized for the oxytocin (OT) and placebo (PL) treatment groups at assessment session “T1” (immediately after the four-week treatment), “T2” (at follow-up, one month post-treatment), and “T3” (at follow-up, 1 year post-treatment). Mean changes from baseline are visualized separately for a Social Responsiveness Scale (SRS-A) self-report version, b SRS-A informant-based version, c Repetitive Behavior Scale-Revised (RBS-R), d State Adult Attachment Measure (SAAM) Avoidance subscale, e SAAM Security subscale, and f SAAM Anxiety subscale. Lower scores indicate improvement for the SRS-A, RBS-R, SAAM Avoidance, and SAAM Anxiety questionnaires. For the SAAM Security questionnaire, higher scores indicate improvement. Vertical bars denote ± standard errors. Asterisks (*) indicate Cohen’s d ≥ .50 (medium-sized effect). Circles (°) indicate Cohen’s d ≥ .80 (large-sized effect)