Table 5.

Clinical studies on melatonin in autism spectrum disorder.

Authors (Ref.) country	Study design	Sample size	ASD assessment	Sleep–wake cycle assessment	Melatonin (MT)	Results	Major limitations
Nir et al. (138) Jerusalem, Israel	Case–control study	10 males with ASD (16–30 years); 5 TD subjects.	DSM-III diagnostic criteria.		Blood melatonin level every 4 h for 24 h.	Abnormal melatonin circadian rhythm in autistic; amplitude of melatonin peak lower in children with autism than in controls; serum melatonin higher during day and lower during night than in controls.	Small size sample. No evaluation of sleep parameters.
Kulman et al. (139) Italy	Case–control study	14 children with autism; 20 age-matched TD children.	DSM-III diagnostic criteria.		Blood melatonin level every 4 h for 24 h.	Significantly low melatonin level in autistic; abnormal melatonin circadian rhythm in all 14 autistic children compared with controls: children with ASD did not demonstrate physiological increase in melatonin during the night.	Small size sample; no objective sleep measures. No evaluation of sleep parameters.
Tordjman et al. (140)	Case–control study	49 children and adolescents with autistic disorder; 88 TD children matched on age, sex, and Tanner stage of puberty (6–15 years).	DSM-IV, ICD-10, and CFTME diagnostic criteria; ADI-R; ADOS-G; Wechsler intelligence scales; Kaufman-Assessment Battery for Children (K-ABC)		Urinary 6-SM 12 h collection from 8 pm to 8 am.	Mean 6-SM lower than in controls; 63% of children with ASD had low 6-SM, low 6-SM level was significantly more common in males and prepubertal children.	No diurnal melatonin evaluation. No evaluation of sleep parameters.
Melke J et al. (141) Paris Autism Research International Sipair study	Case–control study; multicentric study	250 autistic patients and their parents; 255 TD subjects.	DSM-IV diagnostic criteria; ADI-R; Social and Communication Disorders (DISCO-10); AS Diagnostic Interview.		Sequencing ASMT exons and promoters; biochemical analyses performed on blood platelets and/or cultured cells.	Non-conservative variations of ASMT including a splicing mutation present in two families with ASD, but not in controls. Two polymorphisms located in the promoter (rs4446909 and rs5989681) were more frequent in ASD compared to control and were associated with a decrease in ASMT transcripts in blood cell lines. Highly significant decrease in ASMT activity and melatonin level in individuals with ASD.	No evaluation of sleep parameters.
Chaste P et al. (142) Paris Autism Research International Sib-pair study	Case–control study; multicentric study	295 patients with ASD (AD = 222; AS = 61; PDD-NOS = 12); 362 TD subjects; 284 individuals from different ethnic backgrounds.	DSM-IV diagnostic criteria; ADI-R; AS Diagnostic Interview.	Actigraphy; analysis of MT1-I49N mutation.	Sequenced MTNR1A, MTNR1B, and GPR50 genes (coding for the orphan melatonin-related receptor GPR50) in patients and controls.	6 non-synonymous mutations for MTNR1A and 10 for MTNR1B. Most of these variations altered receptor function. Mutants are MT1-I49N, which is devoid of any melatonin binding and cell surface expression, and MT1-G166E and MT1-I212T, which showed severely impaired cell surface expression. The prevalence of these deleterious mutations in cases and controls indicates that they do not represent a major risk factor for ASD (MTNR1A case 3.6% vs. controls 4.4%; MTNR1B case 4.7% vs. 3% controls). They detected a significant association between ASD and two variations OF GPR50, D502–505 and T532A, in affected males.
Jonsson et al. (143) Sweden	Case–control study	109 patients with ASD; 188 TD subjects.	DSM-IV diagnostic criteria; ADI-R.		They have investigated all the genes involved in the melatonin pathway by mutation screening of AA-NAT, ASMT, MTNR1A, MTNR1B, and GPR50.	Several rare variants were identified in patients with ASD, including splice site mutation in ASMT (IVS5+2T > C). However, mutations were found in upstream regulatory regions in three of the genes investigated, ASMT, MTNR1A, and MTNR1B.	No evaluation of sleep parameters.
Mulder et al. (144) Netherlands	Case–control study	10 normoserotonemic and 10 hyperserotonemic age-matched autistic individuals.	DSM-IV-TR diagnostic criteria: ADI-R; ADOS.		Urinary excretion of 5-hydroxyindoleacetic acid (5-HIAA) and serotonin (5-HT).	In the hyperserotonemic group, significant increases at trend level in urinary excretion of 5-HIAA and 5-HT and a significant decrease for 6-SM were found. The urinary 5-HIAA:5-HT ratio was similar in the normoserotonemic versus the hyperserotonemic groups.	No characterization of daytime and nighttime melatonin production in autism. No evaluation of sleep parameters.
Leu et al. (145), Nashville, USA	Cross-sectional study	23 children with ASD (4–10 years)	ADOS; ADI-R; Peabody Picture Vocabulary Test-III (PPVT-III); the Parental Concerns Questionnaire (PCQ).	CSHQ; PSG.	Overnight Urinary 6-SM.	Urinary 6-SM excretion rates are low in autistic subjects. Higher urinary 6-SM excretion rates were associated with increased N3 phase of sleep, decreased N2 phase of sleep, and daytime sleepiness.	Small sample size; lack of a control group; did not obtain information on the specific segments of sleep; no repeated blood specimens of melatonin.
Tordjman et al. (146) France	Case–control study	Post-pubertal individuals with autism (N = 43) and TD controls (N = 26)	DSM-IV-TR, ICD-10 diagnostic criteria; ADI-R; Wechsler intelligence scale; Kaufman K-ABC		Daytime and nighttime urinary excretion of 6-SM during a 24-h period.	Low 6-SM excretion in autism; nocturnal excretion of 6-SM was negatively correlated with autism severity in the overall level of verbal language and repetitive use of objects.	No evaluation of sleep parameters.
Wang et al. (147) Beijing, Republic of China	Case–control study	398 patients with autism (2–17 years); 437 healthy controls.	DSM-IV diagnostic criteria; Autism Behavior Checklist (ABC); CARS.		They sequenced all ASMT exons and their neighboring region.	The study did not detect significant differences of genotypic distribution and allele frequencies of the common SNPs in ASMT between patients with autism and healthy controls; new rare coding mutations of ASMT.	Small sample size; lack of clinical information; not sequence other genes in melatonin pathway.
Jonsson et al. (148) Sweden	Cross-sectional study	1747 subjects (357 monozygotic (MZ) twin pairs, 500 dizygotic (DZ) twin pairs, and 33 subjects without their co-twin).	Telephonic interview: The Autism—Tics, ADHD and other Comorbidities is a sensitive tool for screening the general population for child ASDs and associated conditions.		Analysis of the SNPs and the duplication of exons 2–8 in ASMT. A panel of 47 SNPs to determine twin zygosity.	Significant association, in girls, between an intronic SNP and social interaction impairment and restricted and repetitive behavior where the C-allele carriers were shown to have higher scores. They also investigated a microduplication of exons 2–8 in the ASMT gene, which was found in 27 individuals (1.7%). All these individuals had one extra copy of the region investigated, except for one MZ twin pair, who had two extra copies. This duplication was analyzed with respect to the total ASD scores, although no significant associations could be shown.	Small sample size; no evaluation of sleep parameters.
Goldman et al. (149) Nashville, USA	Cross-sectional study	9 children (3–10 years) and took at least 30 min or longer at baseline to fall asleep on three or more nights (by parent report and actigraphy), free of psychotropic medication.	DSM-IV-TR diagnostic criteria; ADOS.	Comprehensive sleep interview; video-EEG-PSG; actigraphy; CSHQ.	Endogenous plasma melatonin dim light melatonin onset (DLMO) and supplemental melatonin.	In endogenous samples, maximal melatonin concentration and time to peak concentration were comparable to the literature results for TD children. DLMO were captured in the majority of children. Children with ASD and insomnia responsive to low dose melatonin treatment have relatively normal profiles of endogenous and supplemental melatonin.	Small sample size, lack of a control group; variability in the specific start time of the serial blood sampling.
Pagan et al. (150) Paris Autism Research International Sib-pair study	Case–control study; multicentric study	278 patients with ASD, their 506 first-degree relatives (129 unaffected siblings, 199 mothers and 178 fathers); 416 sex- and age-matched controls.	Social Responsiveness Scale (SRS), in first-degree relatives and in controls; RBS (Repetitive Behavior Scale) for probands and their relatives; ADI-R; Diagnostic Interview for Genetic Studies for adults and Kiddie-Schedule for Affective Disorders and Schizophrenia for children; Wechsler scales or Raven’s progressive matrices for nonverbal individuals.	Sleep self-report and/or parent questionnaire; CSHQ; actigraphy; Pittsburgh Sleep Quality Index; a self-assessment questionnaire to determine morningness–eveningness in human circadian rhythms; Epworth sleepiness scale.	Whole-blood serotonin was measured by high-performance liquid chromatography. Plasma melatonin was measured using a radioimmunoassay. NAS and 14-3-3 were determined in platelet pellets; urine samples were collected overnight (2000–0800 hours) from 16 adult patients with HFA and 10 adult controls; 6-SM was measured by a radio immunological method.	Patients showed higher serotonin and NAS levels and lower melatonin levels than healthy controls. Impairments of melatonin synthesis in ASD may be linked with decreased 14-3-3 proteins. The melatonin deficit was only significantly associated with insomnia.	The assessment of melatonin only from plasma sampled in the morning; the association finding was not fully replicated in an independent study.
Veatch et al. (151) Nashville, USA	Cross-sectional	15 ASD children (3–9 years) with sleep disturbances of which 11 in treatment with melatonin.	DSM-IV-TR diagnostic criteria; ADOS.	Sleep interview followed by structured parent education to provide instructions on daytime and evening habits to promote sleep. Children were confirmed to have sleep onset delay of at least 30 min at baseline on C3 nights per week, and none had sleep disturbance limited to specific seasons.	They evaluated variation in two melatonin pathway genes, ASMT and cytochrome P450 1A2 (CYP1A2).	Higher frequencies than currently reported for variants evidenced to decrease ASMT expression and related to decreased CYP1A2 enzyme activity relationship between genotypes in ASMT and CYP1A2	Lack of a control group; small size sample; unable to assess potential differences in ASMT and CYP1A2 between responders and non-responder to melatonin.
Abdulamir et al. (152) Baghdad, Iraq	Case–control study	60 males with ASD (3–13 years) divided into 3 subgroups: mild, moderate, and severe; 26 TD subjects age-/gender- matched.	DSM-5 diagnostic criteria.	76% of autistic subjects showed sleep problems. The severe autistic patients showed the highest number of sleep problems (18 patients) in comparison with moderate (15 patients), and mild autistic patients (13 patients).	Serum levels of melatonin and oxytocin.	Levels of oxytocin and melatonin in patients were significantly lower than that of age-matched and gender-matched controls and were associated with the severity of the disease that was indicated by the significant decrease in the levels of oxytocin and melatonin in moderate patients.	Small sample size; only single area; no systematic assessment of sleep–wake rhythm.
Pagan et al. (153) France	Cross-sectional study	Melatonin: 9 patients and 22 controls; gut samples for serotonin: 11 patients and 13 controls; blood platelets: 239 individuals with ASD and their first-degree relatives and 278 controls.	DSM-IV-TR diagnostic criteria.		Melatonin in plasma and tissues was measured using a radioimmunoassay; serotonin was measured by high-performance liquid chromatography; NAS, AANAT, and ASMT were determined by radio enzymology; the amount of 14-3-3 proteins was determined using the commercial 14-3-3 Pro ELISA kit.	Melatonin deficit in ASD, reduction of AANAT and ASMT observed in the pineal gland as well as in gut and platelets of patients. Reduced levels of 14-3-3 proteins that regulate AANAT and ASMT activities and increased levels of miR-451.	Small samples size; no evaluation of sleep parameters.
Baker et al. (154) Australia	Case–control study	16 adults with ASD (ASD-Only); 12 adults with ASD medicated for comorbid diagnoses of anxiety and/or depression (ASD-Med); 32 TD subjects.	ADOS-2; Autism Quotient (AQ); Wechsler Abbreviated Scale of Intelligence (WASI); Wechsler Adult Intelligence Scale-Fourth Ed (WAIS IV): 3 ASD-Only.	14-day sleep/wake diary and actigraphy assessment; State–Trait Anxiety Inventory (STAI); Patient Health Questionnaire-8 (PHQ-8); Sleep Anticipatory Anxiety Questionnaire (SAAQ).	Sit in dim light 1-h prior to their first saliva sample with saliva sampling commencing 3 h prior to their habitual sleep time and ceasing 1 h past their habitual sleep time, with an hourly sampling rate; salivary melatonin concentrations were determined by a commercially available Melatonin EIA kit.	The timing of DLMO did not differ between the two groups, advances and delays of the melatonin rhythm were observed in individual profiles. Overall mean melatonin levels were lower in the ASD-Med group compared to the two other groups; greater increases in melatonin in the hour prior to sleep were associated with greater sleep efficiency in the ASD groups.	Small sample sizes; use of the individual saliva collection protocols to assess DLMO in adults with ASD; inability to measure and control participants’ exposure to blue light.

DSM, Diagnostic and Statistical Manual; ICD-10, International Classification of Diseases-10; ASD, autism spectrum disorder; AD, autism disorder; AS, Asperger disorder; ADHD, attention-deficit hyperactivity disorder; PDD-NOS, pervasive developmental disorders–not otherwise specified; TD, typically development; CSHQ, Children’s Sleep Habits Questionnaire; ADI_R, Autism Diagnostic Interview-Revised; ADOS, Autism Diagnostic Observation Schedules; CARS, Childhood Autism Rating Scale; PSG, polysomnography; CBCL, child behavior checklist; ASMT, acetylserotonin O-methyltransferase; MTNR1A, MTNR1B, melatonin receptor 1A and 1B; GPR50, G protein-coupled receptor 50; AA-NAT, arylalkylamine N-acetyltransferase; 6-SM, 6-sulfatoxymelatonin.