Materials and Methods
The present double-blind, placebo-controlled, multicentre, crossover trial with two administration periods was conducted at seven university hospitals in Japan. The recruitment process and screening, primary outcome scoring and data and drug management were performed at the main site (Hamamatsu University School of Medicine Hospital). Testing eligibility, confirmation of diagnosis, registration, oxytocin/placebo treatments and clinical assessments were performed at each site (further details and the and prespecified full protocol are provided in the Supplementary material). The study protocol was approved by the institutional review boards for each site and was registered (ClinicalTrials.gov Identifiers: NCT03466671/UMIN000031412). Written informed consent was obtained from all participants at each site.
Eligibility Criteria for Participants
Inclusion criteria were as follows: (i) males aged 18–55; (ii) ASD diagnosis based on the Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5)[36] with a score exceeding the cut-off value of 10 for qualitative abnormalities in social reciprocity on the Autism Diagnostic Interview Revised;[37] and (iii) Full Scale Intelligence Quotient above 80, measured using the Wechsler Adult Intelligence Scale-III.[38]
Exclusion criteria were as follows: (ii) diagnosis of bipolar disorder or schizophrenia spectrum disorder; (ii) primary diagnosis of psychiatric disorders other than ASD; (iii) instability in symptoms of comorbid mental disorders; (iv) history of changes in medication or doses of psychotropics within 1 month before registration; (v) current treatment with more than one psychotropic; (vi) history of hypersensitivity to oxytocin; (vii) history of seizures or traumatic brain injury with loss of consciousness for longer than 5 min; (viii) history of alcohol-related disorders, substance abuse or addiction; (ix) family history of male breast cancer; and (x) severe complications or known hypersensitivity to some drugs and foods.
Interventions
The interventions with crossover design included the following eight groups [i.e. two doses (TTA-121 3/10U) × two orders (TTA-121→placebo/placebo→TTA-121) × two frequencies (once/twice per day); Figure 1]. Groups 1–4: four weeks of TTA-121 at 3 U or 10 U once per day in the morning and placebo once per day in the evening in the first/second period (3 or 10 U/day). After 4 weeks washout, 4 weeks of placebo twice per day in the second/first period. Groups 5–8: four weeks of TTA-121 at 3 or 10 U twice per day in the morning and evening in the first/second period (6 or 20 U/day). After 4 weeks washout, 4 weeks of placebo twice per day in the second/first period.
Figure 1.
Participant flow diagram of the clinical trial. AE = adverse event; AQ = autism-spectrum quotient; E = evening M = morning OXT = oxytocin; Pla = placebo; SRS = Social Responsiveness Scale.
To avoid any subjective effects of the substances other than oxytocin, the placebo contained all of the inactive ingredients from the oxytocin spray. All participants underwent training regarding intranasal administration. Identical instructions and training materials (Supplementary material) were used at each site before the trial initiation, and the effectiveness of training was confirmed at each 2-week assessment point. Treatment adherence was assessed using a self-reported daily record (Supplementary material).
Randomization and Masking of Drug Administration
Each participant was randomized to one of eight interventions. Randomization was performed by an unblinded randomization manager who was not a research team member using a machine-generated treatment schedule, which allocated each participant to an intervention using a randomly permuted block (block size 8). This protocol allowed the participants, clinicians and research team members to remain blinded throughout the trial's duration.
Outcomes
The primary outcome measure was changes in the social core symptom of ASD assessed using the social reciprocity score of Autism Diagnostic Observation Schedule (ADOS)[39] module 4 (range: 0–14, higher values represent worse outcomes) between the baseline and end point of each administration period. ADOS is a standard diagnosis tool for ASD, but has also repeatedly been adopted as a primary outcome in ASD-related trials.[17,34,40–43] The baseline was just before first drug administration on the same day, and the end point was from ~15 min after the drug last administration. The ADOS administration were conducted by trained psychiatrists or psychologists who had completed a training course regarding the research use of ADOS and whose credentials had been validated by another certified administrator (Mi.K.). The four ADOS evaluations for each participant were conducted by the same administrator. To minimize interrater variability, all final ADOS scores were rated by one of seven certified administrators (Mi.K., A.Y., C.K., N.I., H.K., T.H., or K. M.) via video recordings. The four ADOS evaluations for each participant were finally scored by the same certified administrator, blinded to the intervention and order of administration (i.e. baseline or end point). The interrater reliability between two certified administrators who rated each case independently was >90%.
The secondary outcomes were items other than reciprocity in ADOS module 4 and general clinical assessments including Clinical Global Impression[44] and Global Assessment of Functioning (GAF).[45] Quantitative measurements of social behaviours that had significant improvements in our previous trial of intranasal oxytocin (Syntocinon®),[19,34] including fixation time of eye gaze on social region and facial expressions during social interactions in an ADOS activity, were also included as secondary outcomes. Exploratory outcomes included items in quantitative measurements of social behaviours that did not have significant improvements in our previous trial of Syntocinon®,[19,34] including fixation time of eye gaze on social regions and facial expressions during social interactions. Quantitative measurements of speech prosody that characterize ASD compared with typically developed individuals[46] were also included as exploratory outcomes (details in the Supplementary material).
Sample Size
Our previous placebo-controlled, double-blind, crossover trial in which 24 U of intranasal oxytocin in Syntocinon® spray were administered twice per day for 6 weeks in 18 participants with ASD showed changes in the reciprocity scores of ADOS.[17] The mean of amount of change was −0.5 [standard deviation (SD): 1.42] in the oxytocin group, 0.78 (SD: 1.63) in the placebo group and the difference between groups was 1.28 (SD: 1.64), which was a significant result (P = 0.034). On the other hand, we examined the data collected from 104 participants with ASD in another previous clinical trial[34] to test the relationship between the amount of change in the ADOS reciprocity score and clinical function in daily life. The clinically significant amount of change in the GAF score that reflects clinical function in daily life was reported to be 10 points. Based on the data of the above clinical trial, as the GAF score and the ADOS reciprocity score were significantly correlated (correlation coefficient = −0.35, P < 0.0001), it was shown that the amount of change in the GAF score of 10 points corresponds to the change in the ADOS reciprocity score of 1.2 points. Based on the above results, the difference between oxytocin and placebo treatment at the dose at which the maximum effect observed was set to 1.3. In addition, the SD of each treatment group was expected to be 1.5, and the multiple dose–response relationships listed in Supplementary Table 1 were assumed for the ADOS reciprocity score.
The contrasts corresponding to the above dose-response relationships were considered as follows: (i) [2, 1, 0, −1, −2]; (ii) [9, 4, −1, −6, −6]; (iii) [8, 3, −2, −7, −2]; (iv) [7, 2, −3, −3, −3]; (v) [4, −1, −6, −1, 4]; (vi) [1, 0, −1, 0, 0]; (vii) [4, −1, −1, −1, −1]; and (viii) [1, −1, 0, 0, 0]. If a one-way ANOVA was used with a two-sided significance level of 5%, the powers at which the respective comparisons were significant under the setting of 120 cases were as shown in Supplementary Table 2. Using the corresponding contrasts, a power of greater than ~80% was ensured in all dose–response relationships (diagonal contrasts correspond to true dose–response relationships).
Because this clinical trial was a crossover trial, it was expected that the SD of the difference between the groups would be smaller in the pairwise comparison with placebo treatment. Therefore, 120 cases would ensure sufficient power. Based on the above rationale, the required number of participants was 120 in this clinical trial. However, assuming that 20% of the cases were dropped during the trial, the target number of cases was set to 144 cases.
Statistical Analysis
Further details of statistical analyses are provided in the Supplementary material.
Primary Outcome. The change in ADOS reciprocity score during 4 weeks of administration was calculated for each subject and summary statistics were calculated for each administered group. As the fixed effects on the change in each administration period, the content of administration (TTA-121/placebo), order of crossover administration (TTA-121→placebo/placebo→TTA-121) and interaction between content and order were included in a mixed-effects model in which the individual effects were included as random effects. The Kenward–Roger method was used to calculate the degrees of freedom. In addition, for the dose contents [3 U once daily (3 U), 3 U twice daily (6 U), 10 U once daily (10 U), and 10 U twice daily (20 U)], contrasts were set to test dose–response relationships,[47] and the P-value for each contrast was calculated. Contrasts for placebo, 3 U, 6 U, 10 U and 20 U were as follows: [2, 1, 0, −1, −2], [9, 4, −1, −6, −6], [8, 3, −2, −7, −2], [7, 2, −3, −3, −3], [4, −1, −6, −1, 4], [1, 0, −1, 0, 0], [4, −1, −1, −1, −1], [1, −1, 0, 0, 0]. Without correcting for multiple testing, we considered the contrast with the minimum P-value as the dose–response relationship.
The dose that was considered to be most effective based on the identified dose–response relationship was compared with the placebo. The differences between groups and their 95% confidence intervals (CIs) and P-values were calculated using paired t-tests. Similarly, for other doses, pairwise comparisons with placebo were performed.
Secondary and Exploratory Outcomes. The secondary and exploratory outcomes were used to calculate changes compared with baseline measurements and summary statistics for these outcomes were calculated for each treatment group. The mean differences between treatment groups and their 95% CIs were estimated. The statistical model was similar to that of the primary outcome, and the effect on the outcomes of each dose compared with the placebo was evaluated.
Data Availability
Individual participant data after de-identification underlying the results reported in this article are available on request from investigators providing a methodologically sound proposal and whose proposed use of the data has been approved by an independent review committee identified for this purpose. Proposals should be directed to yamasue@hama-med.ac.jp. Maintenance of the dataset in the participants of clinical trials will be ended 5 years following article publication. The data are not publicly available due to them containing information that could compromise research participant privacy or consent.
Brain. 2022;145(2):490-499. © 2022 Oxford University Press
