📋 In This Guide
The short answer
L1-79 is experimental. It has not been proved to improve social communication, language or daily living, and it is not an approved or routine autism treatment in the UK.
There is a reason researchers are interested. A very small uncontrolled case series reported improvements, and two Phase 2 trials have been completed. But the early controlled trial was mainly a four-week safety study, with very small placebo groups. In the newer 58-person trial, the registered main responder outcome did not significantly separate L1-79 from placebo. Some secondary findings favoured L1-79, but only a short conference abstract—not a full peer-reviewed results paper or registry results—was available when this guide was checked.
Company presentations and press releases may sound encouraging. They are not a substitute for complete results that independent clinicians and researchers can examine.
What is L1-79?
L1-79 is the name used for an oral experimental medicine containing a form of alpha-methyl-para-tyrosine. It inhibits tyrosine hydroxylase, an enzyme involved in making brain and stress-related chemicals called catecholamines. These include dopamine, noradrenaline and adrenaline.
The idea is that changing this chemical pathway might affect social interaction or other behaviours. That is a biological theory, not proof that the medicine helps autistic people. Autism is also very varied; one proposed pathway is unlikely to explain every person's strengths, needs or difficulties.
A related medicine called metyrosine is used for a rare hormone-producing tumour. That separate use does not mean L1-79 is licensed, safe or effective for autism.
What studies have been completed?
| Study | What it found | What it cannot tell us |
|---|---|---|
| Eight-person case series Published 2019 | Seven participants improved on some parent or clinician ratings during eight weeks of treatment. | There was no placebo group, participants varied widely in age and the study was too small to establish benefit or uncommon harms. |
| 42-person Phase 2 safety trial NCT02947048 | Four weeks of safety and exploratory outcome data were posted to the registry. | Only small numbers entered each blinded comparison, placebo improvement was substantial on several measures, and the study was not convincing evidence of effectiveness. |
| 58-person Phase 2 crossover trial NCT05067582 | The registered main responder outcome was not significantly different from placebo. A secondary Vineland score and several other measures favoured L1-79 in a smaller first-period analysis. | Problems carrying effects from one treatment period into the next meant the crossover comparison could not be used as planned. Only a conference abstract, not a full paper or registry results, was available. |
The 2019 case series
The only peer-reviewed results paper identified was a prospective case series involving just eight autistic people aged from 2 years 9 months to 24 years. They all received L1-79 for eight weeks. Seven improved on some recorded measures and three mild adverse events were reported.
Without a comparison group, the study cannot separate a medicine effect from expectation, ordinary change over time, changes in other treatment or differences in how behaviour was rated. The lead author was affiliated with the company developing L1-79. That does not make the findings false, but independent confirmation matters.
The early controlled safety trial
The first registered Phase 2 trial enrolled 42 males aged 12 to 21. Some participants received L1-79 openly; others were randomly assigned to L1-79 or placebo for four weeks. The study was designed mainly to assess safety.
The blinded outcome groups were tiny. Depending on the measure, only two to four people contributed data to each placebo group. Results were mixed: on several parent-rated measures, placebo participants improved as much as—or more than—participants receiving L1-79. The registry does not provide a reliable demonstration that L1-79 worked.
The more recent 58-person trial
A later US study enrolled 58 autistic adolescents and young adults aged 12 to 21. It compared two L1-79 doses with placebo in a 12-week crossover design. Participants needed an IQ score around 70 or above and enough spoken language to complete particular assessment modules.
The study was marked completed in June 2024. A 2024 conference abstract reported an important complication: carryover and treatment-sequence effects meant the planned crossover analysis could not be used. The researchers instead compared only the first treatment period, using 44 participants who completed it.
The trial's registered main outcome asked how many participants met a Vineland socialisation “responder” rule. That result did not significantly differ from placebo. A continuous Vineland socialisation score favoured L1-79 by 7.94 points, and several other measures had nominally positive results. But the continuous score was a secondary outcome, several outcomes were tested and the effective comparison was smaller than planned.
As of 30 August 2026, ClinicalTrials.gov still said “No Results Posted”. We did not identify a full peer-reviewed results paper. Claims that the trial was successful therefore need a careful question: where are the complete, independently reviewable numbers?
Does the evidence apply to younger children?
No controlled L1-79 trial listed above studied children under 12. The eight-person case series included younger children, but its uncontrolled design cannot establish either effectiveness or safety for a young child.
Results from selected teenagers and young adults also cannot automatically be applied to a younger child, someone with significant intellectual disability, a minimally speaking child, or a person with epilepsy or other medical conditions. The recent trial's entry criteria excluded many of those groups.
What do we know about side effects and safety?
The safety evidence is incomplete. Because L1-79 changes catecholamine production, the early trial monitored blood pressure when standing, heart rhythm and blood tests as well as reported symptoms.
In the posted four-week trial record:
- diarrhoea was recorded in three participants in one active-treatment group;
- one seizure was recorded in an active-treatment participant;
- one fainting episode was recorded in another active-treatment participant; and
- other events included fatigue, anxiety and laboratory changes.
A registry list does not prove that the medicine caused each event. It does show why “well tolerated” should not be treated as “risk free”. The latest 58-person trial has not posted detailed safety results, so parents do not yet have a complete picture of short-term harms, interactions or longer-term safety.
Is L1-79 available in the UK?
L1-79 remains an investigational medicine. It is not included in NICE-recommended autism care and is not a routine NHS autism treatment. The registered studies above took place in the United States, not the UK.
No Phase 3 L1-79 trial appeared in ClinicalTrials.gov when this guide was checked. A company's intention to run another trial is not the same as a registered, recruiting or successful Phase 3 study.
A future positive trial would still be only one step. Researchers would need complete evidence on benefit and harm, followed by the relevant regulatory and NHS decisions. “Phase 2 completed” does not mean “approved”, “available” or “suitable for my child”.
Questions to ask when you see a new L1-79 claim
- Is this a peer-reviewed full paper or a company announcement?
- Did the study meet its pre-declared primary outcome against placebo?
- How many people completed each comparison?
- Were the participants similar in age, language and support needs to my child?
- Were all adverse events and dropouts reported?
- Has an independent research group repeated the finding?
For a wider explanation of these questions, read why autism trials can look promising before larger tests. Our new autism treatment research guide compares other experimental medicines and devices without treating a press release as proof.
The bottom line
L1-79 is a research candidate, not a proven autism treatment. The peer-reviewed human evidence is an uncontrolled series of eight people. The early controlled study was short and mainly about safety, with very small comparison groups. In the newer 58-person trial, the registered main responder outcome was negative; some secondary measures favoured L1-79 after the planned crossover analysis could not be used. Complete results and adverse-event tables have not been made available through the registry or a full peer-reviewed paper.
That does not mean L1-79 can never work. It means the honest answer today is we do not know. Families should not be asked to turn that uncertainty into hope, payment or off-trial use.
Frequently asked questions
Is L1-79 approved for autism in the UK?
No. L1-79 remains an investigational medicine. It is not a NICE-recommended or routine NHS autism treatment.
Has L1-79 been proved to improve social communication?
No. An eight-person uncontrolled case series and a small safety-focused trial cannot establish benefit. In a later 58-person Phase 2 study, the registered main responder outcome did not significantly separate L1-79 from placebo. Some secondary findings were positive, but only a conference abstract—not a full peer-reviewed paper or registry results—was available when checked.
Has L1-79 been properly studied in young children?
No. The two controlled trial records enrolled people aged 12 to 21. A small uncontrolled case series included younger participants, but it cannot show whether the medicine caused any change or establish safety for young children.
Is L1-79 the same as metyrosine?
They are related but not interchangeable treatments. L1-79 is a racemic form of alpha-methyl-para-tyrosine; metyrosine is one form used for a rare tumour-related indication. Metyrosine's existing medical use does not establish L1-79 as safe or effective for autism.
Are the reported Phase 2 results encouraging?
The sponsor has described favourable findings, but parents cannot properly judge them without the complete pre-declared outcomes, statistical analysis and safety results. “Encouraging” is an opinion; it is not an evidence grade.
Sources and further reading
- Rothman, Bartky and Halas (2019) — L1-79 prospective case series of eight autistic participants
- ClinicalTrials.gov NCT02947048 — four-week Phase 2 safety study; results posted
- ClinicalTrials.gov NCT05067582 — 12-week Phase 2 crossover study; completed with no results posted when checked
- Megerian et al. (2024) — conference abstract reporting the later Phase 2 analysis (abstract 343)
- NICE QS51 — medication should not be used for the core features of autism
Disclaimer: Information only, not medical advice. Evidence and source status were checked on 30 August 2026. Research can change. Speak to an appropriately qualified clinician before starting, stopping or changing any medicine or treatment.
Related Evidence Guides
Calm explanations of experimental autism-treatment claims, written for families.
New Autism Treatment Research in 2026
What was positive, what failed and what remains experimental.
Read guide →AST-001 and Autism
The small Phase 2 signal, the Phase 3 miss and why L-serine is not a substitute.
Read guide →Arbaclofen and Autism
What a negative primary outcome means when some secondary measures look positive.
Read guide →Leucovorin and Autism
The small language study, the retraction and the narrow rare-disease approval.
Read guide →