In Youth with Autism Spectrum Disorder, Memantine Improves Social Impairment

October 27, 2025

Ruta Nonacs, MD PhD

In this randomized controlled trial in youth with ASD, memantine significantly improved social functioning, particularly in those with evidence of glutamate dysregulation.

For individuals with autism spectrum disorder (ASD), treatment options remain limited. Available pharmacologic treatments may alleviate comorbid symptoms, such as depression, anxiety, and attention deficits, but have minimal impact on core ASD features, including deficits in social functioning. A recent randomized-controlled trial conducted by Gagan Joshi, MD, and colleagues from the Alan and Lorraine Bressler Clinical and Research Program for Autism Spectrum Disorder evaluated the safety and efficacy of memantine in treating social impairments in youth with ASD.

Why Study Memantine?

Beginning in the late 1990s, autism researchers started to focus on the role of glutamate in the pathophysiology and symptomatology of ASD. Glutamate is the brain’s primary excitatory neurotransmitter and plays a critical role in neuronal development and synaptic plasticity through activation of N-methyl-D-aspartate (NMDA) receptors. Multiple lines of investigation have demonstrated evidence of glutamate dysregulation in individuals with ASD, a finding that is particularly relevant to understanding social deficits, as glutamate plays an important role in brain regions governing social cognition, emotional awareness, and self-reflection—processes often impaired in individuals with ASD.

In a previous study of youth with ASD, Joshi and his team used proton magnetic resonance spectroscopy (¹H-MRS) and documented elevated glutamate levels in the pregenual anterior cingulate cortex (pgACC), a brain region rich in glutamatergic activity that supports social and emotional functioning.

In terms of treatment, previous studies have examined the effects of glutamate-modulating agents, including lamotrigine, amantadine, and N-acetylcysteine, as potential treatments for core ASD symptoms. However, these agents have shown only modest efficacy. In contrast, preliminary data from small retrospective and open-label trials of memantine have suggested greater promise. Memantine, a moderate-affinity, noncompetitive NMDA receptor antagonist, has demonstrated an acceptable safety profile and potential benefits for improving social behaviors in ASD populations.

Study Design

In this 12-week, randomized, double-blind, placebo-controlled trial, youth with ASD (ages 8–17 years) without intellectual disability were recruited from outpatient psychiatry clinics. Participants were randomly assigned (1:1) to receive either memantine or placebo. The medication was titrated to a maximum daily dose of 20 mg over the first four weeks and maintained at the highest tolerated dose for the remainder of the trial. Dosing adjustments were guided by tolerability and clinical judgment.

Core social and behavioral symptoms were measured using the Social Responsiveness Scale–Second Edition (SRS-2), including subscales assessing social communication and interaction (SRS-SCI) and restricted and repetitive behaviors (SRS-RRB). Additional measures included the Aberrant Behavior Checklist–Social Withdrawal subscale (ABC-SW) and the Clinical Global Impression–Improvement (CGI-I) scale. Response was defined as a greater than 25% reduction in SRS-2 total scores and a clinician-rated CGI-I subscale score of 2 or less (anchored for ASD).

Glutamate levels in the pgACC were assessed using ¹H-MRS imaging prior to treatment and repeated at 12-week intervals. A separate group of age- and sex-matched healthy controls underwent imaging for comparison.

Memantine and Social Functioning

Treatment with memantine was superior to placebo in improving measures of social functioning. Youth in the memantine group had a significantly higher treatment response rate than those receiving placebo (56.2% vs. 21.0%). Participants who received memantine were nearly five times more likely to respond positively compared with those who received placebo (odds ratio, 4.8 [95% CI, 1.1–21.2]; P = .03).

Although individuals in the memantine group showed greater improvements in other ASD domains—such as social cognition (SRS-SCI), restricted and repetitive behaviors (SRS-RRB), and adaptive behavior (ABC-SW)—these differences did not reach statistical significance.

Memantine was very well tolerated. In general, adverse events were transient and considered mild or moderate. The most frequently reported AEs in the memantine group were headaches, insomnia, upper respiratory conditions, and increased appetite.

Impact of Pre-Treatment Glutamate Levels on Response

¹H-MRS data from 37 participants with ASD and 16 healthy controls revealed significantly higher pgACC glutamate levels in participants with ASD. Of note, elevated glutamate was not universal, occurring in approximately half (54%, n = 20 of 37) of the ASD participants.

Treatment response differed based on pre-treatment pgACC glutamate levels. Among participants with ASD and high glutamate levels, 8 of 10 individuals (80.0%) responded positively to treatment with memantine; however, there were no memantine responders in those with ASD and medium glutamate levels. All memantine responders exhibited elevated pretreatment pgACC glutamate levels.

Other MGH researchers involved in this project include Maura DiSalvo, Steven Faraone, T. Atilla Ceranoglu, Mai Uchida, Christopher McDougle, and Janet Wozniak.