Long-term intranasal oxytocin (16 IU/day for 8 weeks) was well-tolerated in 29 males with autism and intellectual disabilities, with no improvement on standard rating scales but increased social interactions observed in daily life during early treatment.
Safe over 8 weeks in ASD+IDThe first study demonstrating tolerability of long-term intranasal oxytocin in the high-risk ASD population with intellectual disabilities and elevated seizure risk
What the researchers found
In this randomized, double-blind, placebo-controlled crossover pilot study, 29 males aged 15–40 with ASD and intellectual disabilities received intranasal oxytocin (16 IU/day) or placebo for 8 weeks each. No serious adverse events occurred except one seizure in one participant — an important safety finding given that this population has elevated seizure risk.
Primary outcome (Childhood Autism Rating Scale) and secondary standard scale measures showed no significant difference between oxytocin and placebo. However, exploratory analysis revealed significantly more frequent social interactions during play sessions and daily life in the initial half of the oxytocin-first arm. Plasma oxytocin concentrations significantly correlated with irritability subscale scores on the Aberrant Behavior Checklist, suggesting a biological relationship between oxytocin levels and behavioral symptoms.
Why it matters
People with autism and intellectual disabilities represent the most underserved population in autism research — they're typically excluded from clinical trials due to the difficulty of obtaining consent and measuring outcomes. This study demonstrates that long-term oxytocin treatment is feasible and safe in this group, even with their elevated seizure risk. The discrepancy between null results on rating scales and positive signals in real-world social observations raises important questions about how we measure treatment outcomes in this population.
How the study worked
Randomized, double-blind, placebo-controlled crossover pilot study (UMIN000007250). Twenty-nine males aged 15–40 with ASD and comorbid intellectual disabilities participated. Each received 8 weeks of intranasal oxytocin (16 IU/day) and 8 weeks of placebo in crossover fashion. Primary outcome was the Childhood Autism Rating Scale. Secondary measures included multiple standardized behavioral assessments. Exploratory measures included direct observation of social interactions during play sessions and daily life. Plasma oxytocin concentrations were measured.
What this study cannot tell us
This is a small pilot study (29 participants) not powered for definitive efficacy conclusions. The positive social interaction findings were from exploratory analyses, not pre-specified primary outcomes, increasing the risk of false positives. The crossover design with 8-week periods may have been complicated by carryover effects. The seizure event, while isolated, is concerning in a seizure-prone population and warrants monitoring in larger trials. Only males were included, and results may not generalize to females with ASD.
How to read the evidence
This is a randomized, double-blind, placebo-controlled crossover pilot study — a rigorous design — but the small sample size (29), null primary endpoint, and exploratory nature of the positive findings limit the evidence to hypothesis-generating. The safety data is the strongest contribution, demonstrating feasibility for larger trials.
When this study was published
Published in 2016, this study has informed subsequent larger oxytocin trials in autism. The field has continued to show inconsistent results, with the measurement challenge highlighted here remaining unresolved.
The bigger picture
Intranasal oxytocin has been widely studied for autism, with inconsistent results across trials. This study adds a unique perspective by including the most severely affected subgroup (ASD + ID) who have been largely ignored in the literature. The finding that standard rating scales missed improvements visible in daily life observations highlights a fundamental challenge in autism treatment research: our measurement tools may not capture the real-world social changes that matter most to patients and families.
Questions still open
- Should future oxytocin autism trials use naturalistic observation rather than rating scales as primary endpoints, especially in intellectually disabled populations?
- Why did the social interaction improvement appear mainly in the first half of treatment — does tolerance develop with chronic oxytocin administration?
- What is the seizure risk profile of long-term intranasal oxytocin in populations with elevated baseline seizure susceptibility?
Common questions
Why didn't oxytocin work on the autism rating scales but did improve real-world social behavior?
Is intranasal oxytocin approved for treating autism?
Read the original research
Oxytocin for Male Subjects with Autism Spectrum Disorder and Comorbid Intellectual Disabilities: A Randomized Pilot Study.
Frontiers in psychiatry, 7, 2
Citation
Munesue, Toshio; Nakamura, Hiroyuki; Kikuchi, Mitsuru; Miura, Yui; Takeuchi, Noriyuki; Anme, Tokie; Nanba, Eiji; Adachi, Kaori; Tsubouchi, Kiyotaka; Sai, Yoshimichi; Miyamoto, Ken-Ichi; Horike, Shin-Ichi; Yokoyama, Shigeru; Nakatani, Hideo; Niida, Yo; Kosaka, Hirotaka; Minabe, Yoshio; Higashida, Haruhiro. (2016). Oxytocin for Male Subjects with Autism Spectrum Disorder and Comorbid Intellectual Disabilities: A Randomized Pilot Study.. Frontiers in psychiatry, 7, 2. https://doi.org/10.3389/fpsyt.2016.00002