← Back
NeuroTrials.ai
Neurology Clinical Trial Database

Levacetylleucine for AT

Safety and efficacy of levacetylleucine in ataxia-telangiectasia: a phase 3, randomised, double-blind, placebo-controlled crossover trial

Year of Publication: 2026

Authors: Martakis K, Bremova-Ertl T, Bolton C, et al.

Journal: Lancet Neurology

Citation: Lancet Neurol. 2026 Jul;25(7):633-644

Link: https://doi.org/10.1016/S1474-4422(26)00158-4

Bottom Line

In this first positive phase 3 trial in ataxia-telangiectasia, levacetylleucine produced a statistically significant and clinically meaningful improvement in SARA (treatment effect −1·88 points, 95% CI −2·70 to −1·06; p<0·0001) over 12 weeks with a favourable safety profile and no treatment-related serious adverse events.

Major Points

  • First completed positive phase 3 RCT for any therapy in ataxia-telangiectasia, a rare autosomal-recessive neurodegenerative disorder with no approved treatments.
  • 73 patients (47 paediatric, 26 adult) across 10 sites in 6 countries received 12 weeks of oral levacetylleucine and 12 weeks of matching placebo in a crossover design; 70 completed both periods.
  • Primary SARA outcome: mean change −1·92 with levacetylleucine vs −0·14 with placebo (linear mixed model treatment effect −1·88, 95% CI −2·70 to −1·06; p<0·0001), exceeding the 1–1·5 point clinically meaningful threshold.
  • Supportive secondary results: FDA f-SARA −0·57 (p=0·001), ICARS −2·84 (p=0·003), and investigator CGI-I −0·4 (p=0·02); subgroup benefit was consistent across age, disease-onset group, weight-based dose, and gender.
  • Patients randomised to levacetylleucine→placebo deteriorated by 1·81 SARA points during period 2 washout, mirroring the drug's expected pharmacokinetic effect.
  • 81% of exit interviews (57/70) reported neurological improvement on levacetylleucine; 66% described the changes as meaningful for everyday life.
  • Safety: 54 AEs in 29 patients (40%) on levacetylleucine vs 75 AEs in 25 patients (35%) on placebo; only 3 drug-related AEs (mild diarrhoea, mild eczema, moderate insomnia), no treatment-related serious adverse events and no deaths.
  • Weight-tiered oral dosing (2 g/day for 15–<25 kg, 3 g/day for 25–<35 kg, 4 g/day for ≥35 kg) as granules for suspension in water, orange juice, or almond milk was well tolerated.

Design

Study Type: Randomised Controlled Trial (Phase 3, crossover)

Randomization: 1

Blinding: Double-blind (participants, investigators, assessors, sponsor, families)

Enrollment Period: March 18, 2025 – June 30, 2025

Follow-up Duration: Two consecutive 12-week treatment periods (24 weeks total); open-label extension ongoing

Centers: 10

Countries: Germany, Slovakia, Spain, Switzerland, UK, USA

Sample Size: 73

Analysis: Linear mixed-effects model in all patients who received at least one dose (n=73); permuted 4-patient block randomisation via centralised interactive response technology (Medpace)


Inclusion Criteria

  • Age ≥4 years
  • Genetically confirmed diagnosis of ataxia-telangiectasia
  • SARA total score 7–34 at screening
  • Gait subtest of the SARA scale within the 2–7 range, or able to perform the 9-hole peg test with the dominant hand in 20–150 seconds
  • Weight ≥15 kg at screening
  • Written informed consent by patient, parent, caregiver, or legal representative

Exclusion Criteria

  • Not meeting the SARA range or functional test criteria above
  • Advanced disease state precluding reliable completion of functional assessments
  • Asymptomatic ataxia-telangiectasia (not eligible under this symptomatic protocol)
  • Use of prohibited concomitant medication (monitored by regular urine analyses to confirm adherence)
  • Detailed inclusion/exclusion list in trial appendix (pp 2–3)

Baseline Characteristics

CharacteristicLevacetylleucine → Placebo (n=36)Placebo → Levacetylleucine (n=37)
Paediatric (<18 years)23 (64%)24 (65%)
Adult (≥18 years)13 (36%)13 (35%)
Female17 (47%)21 (57%)
Male19 (53%)16 (43%)
Hispanic or Latino2 (6%)2 (5%)
White28 (78%)27 (73%)
Asian2 (6%)2 (5%)
Age at diagnosis - Early-infantile (<2 years)24 (67%)27 (73%)
Age at diagnosis - Late-infantile (2 to <6 years)8 (22%)8 (22%)
Age at diagnosis - Juvenile (6 to <15 years)2 (6%)1 (3%)
Age at diagnosis - Adolescent/Adult (≥15 years)2 (6%)1 (3%)
Weight 15 to <25 kg (2 g/day)8 (22%)10 (27%)
Weight 25 to <35 kg (3 g/day)9 (25%)8 (22%)
Weight ≥35 kg (4 g/day)19 (53%)19 (51%)
Baseline SARA total score (mean, SD)19·81 (6·43)19·87 (6·41)
Baseline FDA functional SARA (mean, SD)7·83 (2·60)7·86 (2·60)
Baseline ICARS (mean, SD)51·20 (17·05)51·54 (16·97)

Arms

FieldLevacetylleucine → PlaceboControl
InterventionOral levacetylleucine (N-acetyl-L-leucine) granules for suspension: 4 g/day divided TID (≥35 kg) or weight-tiered ≈0·1 g/kg/day BID–TID (<35 kg) for 12 weeks (period 1), then matching placebo for 12 weeks (period 2)Matching placebo granules (same colour, taste, appearance, solubility) for 12 weeks, then crossover to levacetylleucine 4 g/day TID or weight-tiered ≈0·1 g/kg/day for 12 weeks
Duration24 weeks (two 12-week periods, back-to-back)24 weeks (two 12-week periods, back-to-back)

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Mean change from baseline in SARA (Scale for the Assessment and Rating of Ataxia) total score after each 12-week treatment periodPrimary−0·14 (SD 2·38)−1·92 (SD 2·81)N/A (linear mixed-effects model)<0·0001
FDA functional SARA (f-SARA, axial 4-item)Secondary−0·10 (SD 1·02)−0·65 (SD 1·16)0·001
ICARS (International Cooperative Ataxia Rating Scale)Secondary−1·69 (SD 6·09)−4·22 (SD 7·77)0·003
Investigator Clinical Global Impression of Improvement (CGI-I)Secondary−0·2 (SD 0·6)−0·6 (SD 0·8)0·02
SCAFI composite (8MWT, 9HPT-D, 9HPT-ND, PATA)Secondary0·04 (SD 0·24)0·06 (SD 0·28)Not significant
Caregiver CGI-ISecondary−0·2 (SD 0·8)−0·2 (SD 0·8)Not tested
EQ-Visual Analogue Scale (quality of life)Secondary−0·3 (SD 19·2)1·3 (SD 15·5)
Exit interviews reporting neurological improvementSecondary57 (81%) of 70 responders reported improvement on levacetylleucine; 46 (66%) called changes meaningful to daily life; 29 (41%) noted safety-related functional gains
Any treatment-emergent AEAdverse29 (40%) on levacetylleucine vs 25 (35%) on placebo (54 vs 75 events)
Infections and infestationsAdverse15 (21%) vs 13 (18%)
CoughAdverse6 (8%) vs 2 (3%)
FallAdverse3 (4%) vs 2 (3%)
Upper respiratory tract infectionAdverse3 (4%) vs 3 (4%)
PyrexiaAdverse3 (4%) vs 2 (3%)
DiarrhoeaAdverse2 (3%) vs 4 (6%)
VomitingAdverse1 (1%) vs 4 (6%)
Abdominal painAdverse1 (1%) vs 3 (4%)
Gastrointestinal disorders (SOC)Adverse5 (7%) vs 10 (14%)
Injury, poisoning, procedural complications (SOC)Adverse7 (10%) vs 5 (7%)
Drug-related AEs (all mild or moderate, transient)AdverseDiarrhoea (mild), eczema (mild), insomnia (moderate) — 3 events in 2 patients on levacetylleucine
Treatment-related serious AEsAdverse0
DeathsAdverse0
Discontinuations due to AEAdverse0 (three unrelated SAE withdrawals: T-cell acute leukaemia, diffuse large B-cell lymphoma, lower respiratory tract infection)

Subgroup Analysis

Prespecified subgroup analyses of the primary SARA endpoint showed consistent benefit across paediatric vs adult, age at disease onset, disease severity, weight-based dose group, and gender, with no evidence of heterogeneity of treatment effect.


Criticisms

  • Small sample (n=73) reflecting a rare disease population; crossover design increases power but cannot fully exclude carry-over effects despite the observed washout deterioration.
  • Short 12-week per-period exposure; effects on ataxia-telangiectasia hallmarks such as immunological defects and cancer risk cannot be assessed in this timeframe.
  • Symptomatic entry criteria excluded children younger than 4 years, asymptomatic patients, and those with advanced disease unable to complete functional assessments — limiting generalisability at both extremes of the disease.
  • Predominantly White North American and European population (75% White); ethnic/racial diversity was scarce.
  • No validated biomarker or surrogate endpoint for ataxia-telangiectasia to corroborate the clinical improvement.
  • Secondary endpoints reported without multiplicity adjustment, so p-values are hypothesis-generating rather than confirmatory.
  • Not all functional scales are formally validated in paediatric populations.
  • Industry-sponsored (IntraBio); the sponsor collaborated on design but had no role in data collection, analysis, interpretation, or writing per the disclosures.

Funding

IntraBio (industry sponsor); the funder collaborated on study design only, with no role in data collection, analysis, interpretation, or manuscript writing.

Based on: Levacetylleucine for AT (Lancet Neurology, 2026)

Authors: Martakis K, Bremova-Ertl T, Bolton C, et al.

Citation: Lancet Neurol. 2026 Jul;25(7):633-644

Content summarized and formatted by NeuroTrials.ai.