L-Serine Improves Core Autism Symptoms: Phase II Trial Extended

L-serine held its benefits for a year in children who responded to it

STUDY: Kim JI et al, Front Psychiatry 2026
STUDY TYPE: Long-term open-label follow-up of a randomized controlled trial
FUNDING: Astrogen Inc. (manufacturer of AST-001). One author is employed by Astrogen and holds a patent on the study medication.

Background

L-serine is a non-essential amino acid that the body can produce, but it is also widely consumed in the diet (soy, seaweed, sweet potatoes, eggs, and meat) and taken as a supplement. A patented version is under development (AST-001) for autism. It showed promise in an earlier phase II placebo-controlled trial (n=151, although many p values were borderline at 0.04). This study gives us the one-year follow up on those responders, without a placebo comparison.

L-serine is involved in:

• Neuroprotection: building block for phosphatidylserine and sphingolipids, which are essential fats that make up brain cell membranes and nerve coverings
• Immune function: was enhanced in a trial of autism
• Neurotransmitters: restores dopamine tranmission; involved in glycine, glutamate, and GABA synthesis
• Sleep: It is converted into D-serine, which may improve sleep quality

The Study

• 61 children aged 2–11 with autism spectrum disorder (85% also had intellectual disability) who had responded to l-serine (AST-001) in the phase II trial; most were male (79%)
• They continued at their prior dose (low or high) for 52 weeks

Results

In the original trial, l-serine brought significant improvement in overall symptoms, communication, motor skills, and ratings of parental distress compared to placebo. In this follow up study, the global improvement held steady for a full year, though without a placebo comparison.

The mean severity score (CGI-S) dropped from 4.25 at baseline to 4.10 at week 52, a small but consistent improvement.

Among the 48 children who had already responded by the end of the phase II trial, 45 (94%) still met the responder threshold at week 52. Over 90% of those responders maintained that status at every time point throughout the follow-up.

There was no difference in outcome between the low-dose and high-dose groups at any assessment point.

Side Effects

l-serine was well tolerated. Only two children had adverse events attributed to l-serine: one had decreased appetite (low-dose), one had enteritis (high-dose); both were mild and resolved. .

Limitations

• Small sample, no placebo, only prior responders enrolled, industry-funded
• The CGI scales are subjective and not norm-referenced, making it hard to distinguish treatment gains from normal developmental progress in young children

Practice Implications

1. The data is limited to one randomized trial (n=151) and this follow up, but few alternatives are available in autism and l-serine is safe. No major risks showed up in other trials and it is part of the diet.
2. While AST-001 is not available, l-serine is, and is typically dosed at 3-25 grams daily in psychiatric studies. One study of autism dosed it at 400 mg/kg/day.
3. I am not aware of lab-tested brands of l-serine, but NOW makes a version and this manufacturer’s products consistently pass in independent tests.

—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report

What’s Your Take? Share in Comments