Memantine, a drug approved for Alzheimer’s disease, was associated with significant improvement in social functioning in some children and adolescents with autism spectrum disorder (ASD) without intellectual disabilities, results from a small randomized clinical trial showed.

More than half of youth treated with memantine showed an improvement in social competence and a reduction in autism symptom severity compared with 21% receiving a placebo. The benefits were most pronounced in those with abnormally elevated brain glutamate levels.

“This is the first controlled study to show significant benefit on the core social impairments of autism with memantine. These results suggest the drug could be a viable option for some patients and point to the promise of biomarker-based treatment in ASD,” lead investigator Gagan Joshi, MD, director of the Alan and Lorraine Bressler Clinical and Research Program for Autism Spectrum Disorder at Massachusetts General Hospital and associate professor of psychiatry at Harvard Medical School, both in Boston, told Medscape Medical News.

However, Joshi and other autism experts urged caution when interpreting the findings from the small study, noting that more research is needed.

The findings were published online on October 1 in JAMA Network Open.

Few Options for Social Symptoms

ASD affects more than 2% of children in the US and is characterized by persistent challenges with social interaction and communication. Despite the prevalence of the condition, no pharmacologic treatment has reliably improved core social deficits.

Approved by the FDA in 2003 for Alzheimer’s-related dementia, memantine is an N-methyl-D-aspartate (NMDA) receptor antagonist that blocks glutamate, the brain’s primary excitatory neurotransmitter.

Researchers have expressed interest in memantine as an ASD treatment due to the drug’s distinctive action on glutamate, which is dysregulated in a subset of individuals with autism. Still, prior trials of glutamate-modulating agents in ASD, including memantine, have produced mixed and often inconclusive results.

“Early reports hinted that memantine could help children with autism, but the only previous large study, done mostly in children with intellectual disability, did not show clear benefits,” Joshi said. “This may have been because the dose was too low and didn’t consider key brain chemistry differences.”

The current study enrolled 42 youths aged 8-17 years with ASD and an intelligence quotient of at least 85 (mean age, 13 years; 74% men). All had moderately severe symptoms and were randomly assigned to receive memantine titrated up to 20 mg/d or placebo.

Of the total cohort, 33 participants completed the trial, with 16 receiving memantine and 17 receiving a placebo.

By the end of the study, 56% of those on memantine showed meaningful gains on both parent and clinician measures of social functioning, compared with 21% on placebo. Participants taking memantine were 4.8 times more likely to respond to treatment than the placebo group (odds ratio, 4.8; P = .03), with most improvements evident by week 6.

Glutamate Levels Tied to Response

A distinctive element of the trial was the use of proton magnetic resonance spectroscopy to measure glutamate in the pregenual anterior cingulate cortex (pgACC), a brain region central to social cognition.

Slightly more than half of the participants showed elevated glutamate levels, and they were the ones most likely to respond to treatment.

Among children with high pgACC glutamate, 80% of those receiving memantine improved compared with 20% of those on placebo. Those with normal or moderate glutamate levels showed little added benefit.

pgACC glutamate concentration may serve as a potential biomarker to identify patients most likely to respond to memantine and other glutamate-modulating therapies, investigators concluded.

Memantine was well tolerated, with most participants reaching the maximum dose of 20 mg/d. The most common adverse events were mild to moderate, including headaches and transient changes in sleep or appetite.

Promising but Preliminary

While promising, the researchers acknowledged that the findings are at an early stage and do not establish memantine as a standard therapy for autism. Larger, multisite studies are needed to confirm efficacy, assess the durability of benefit, and determine whether pgACC glutamate levels can predict response to other glutamate-modulating agents, they said.

“Although these results are promising, definitive conclusions regarding the utility of pgACC glutamate levels as a therapeutic biomarker await larger controlled trials that prospectively examine memantine response in individuals with autism based on glutamate concentration,” the authors wrote.

Commenting for Medscape Medical News, Amanda Brignell, PhD, senior research fellow at Monash University, based in Melbourne, Australia , agreed with that caution, calling the results “interesting but preliminary.”

Brignell, who was not part of the study, noted that the trial’s small size and relatively homogenous sample limit generalizability.

She also cited a 2022 Cochrane review of memantine in people with autism, which she coauthored with Tamara May, PhD, senior research fellow at Monash University. That study identified only three small trials with 204 participants and concluded that the certainty of evidence was low to very low across all outcomes.

“In this new study, while social communication abilities were reported to improve, there were no significant group differences in measures assessing autistic and related traits. This study only reports on outcomes immediately after treatment, so the maintenance of these effects is unknown,” she said.

Brignell also noted the high rate of co-occurring conditions such as attention-deficit/hyperactivity disorder among participants, which may limit the specificity of the findings.

“While families often seek medications like memantine when conventional therapies have been ineffective, it is essential to carefully design studies that reflect the priorities of the autism community,” she said. “At this point, the evidence base for memantine remains limited, and further well-designed trials with autistic people’s involvement are needed.”

Daniel Felsky, PhD, of the Centre for Addiction and Mental Health and the University of Toronto, both in Toronto, Ontario, Canada, agreed.

“Promising as they are, the findings of the study by Joshi et al. must be taken into context so that further work can yield the most benefit,” Felsky wrote in an accompanying editorial.

Felsky cited an earlier study in youth with autism aged 6-12 years that showed no difference in social functioning between those who received memantine and a placebo group. Joshi and colleagues referenced this trial in the new study but suggested that different outcomes may be due to low dosing of memantine and the inclusion of youth with intellectual disabilities.

However, that suggestion “remains untested,” Felsky wrote. “Trials of other NMDA antagonists in ASD have also shown mixed results for efficacy. As such, the current trial by Joshi et al. is not definitive and should be replicated in larger groups.”

The National Institute of Mental Health funded the study. Joshi and colleagues reported having multiple research and consulting relationships, detailed in the original article. Brignell reported having no conflicts of interest and no funding related to any drug trials.