A comprehensive systematic review and meta-analysis of pharmacological and dietary supplement interventions in paediatric autism: moderators of treatment response and recommendations for future research.

BACKGROUND: Autism spectrum disorders (ASDs) are pervasive and multifactorial neurodevelopmental conditions, characterized by impairments in social communication and interaction, and restricted, repetitive patterns of behaviour, interests or activities. Treatment options to ameliorate symptoms of ASDs are limited. Heterogeneity complicates the quest for personalized medicine in this population. Our aim was to investigate if there are baseline characteristics of patients that moderate response or trial design features that impede the identification of efficacious interventions for ASDs.

METHOD: Literature searches of EMBASE, MEDLINE and PsycINFO identified 43 studies for qualitative assessment of baseline characterization of participants and 37 studies for quantitative analysis of moderators of treatment response. Criteria included blinded randomized controlled trials (RCTs) in paediatric ASD, with at least 10 participants per arm or 20 overall, of oral treatments, including pharmacological interventions and dietary supplements.

RESULTS: Random-effects meta-analysis of 1997 participants (81% male) identified three moderators associated with an increase in treatment response: trials located in Europe and the Middle-East; outcome measures designated primary status; and the type of outcome measure. Inconsistent reporting of baseline symptom severity and intellectual functioning prevented analysis of these variables. Qualitative synthesis of baseline characteristics identified at least 31 variables, with only age and gender reported in all trials. Biological markers were included in six RCTs.

CONCLUSIONS: Few trials reported adequate baseline characteristics to permit detailed analysis of response to treatment. Consideration of geographical location, baseline severity and intellectual function is required to ensure generalizability of results. The use of biological markers and correlates in ASD trials remains in its infancy. There is great need to improve the application of baseline characterization and incorporation of biological markers and correlates to permit selection of participants into homogeneous subgroups and to inform response to treatment in ASD.