TAP-GRIN: Interventional Study on Patients With GRIN-related Neurodevelopmental Disorders

L-serine Supplementation in Patients With GRIN-related Neurodevelopmental Disorders: Multicentre Protocol for an Aggregated Series of Randomised, Placebo-controlled N-of-1 Trials

The goal of this clinical study is to find out whether L-serine dietary supplementation helps improve overall clinical functioning in children and young adults (2-30 years) with GRIN-related neurodevelopmental disorders (GRIN-NDD) caused by loss-of-function (LoF) variants in GRIN1, GRIN2A, GRIN2B, or GRIN2D. It will also assess the safety and tolerability of L-serine.

The main questions it aims to answer are:

Does L-serine improve overall clinical status, measured mainly by the Clinical Global Impression-Severity (CGI-S) score?

Does L-serine improve behaviour, cognition, adaptive functioning, motor skills, sleep, and (in those with epilepsy) seizure frequency and EEG findings?

What side effects or medical problems occur during L-serine compared with placebo?

Do neurophysiological measures (including TMS-EMG/TMS-EEG) change with treatment and potentially act as biomarkers of response?

Researchers will compare L-serine to a placebo (maltodextrin powder with similar appearance/texture) using a randomised, double-blind, placebo-controlled "n-of-1" approach, where each participant receives both treatments in alternating periods. Results from multiple single-patient trials will then be combined (aggregated) to estimate the overall treatment effect across the study population.

Participants will:

Complete a 4-week baseline period with assessments (and seizure diary use where applicable)

Receive L-serine and placebo in alternating 3-month periods within each cycle (minimum 2 cycles, up to 4 cycles; each cycle lasts 6 months)

Take the assigned study product by mouth 3 times per day at 500 mg/kg/day (maximum 30 g/day for participants ≥60 kg)

Have the first 7 days of each 3-month period treated as washout, with data from that week not analysed

Attend regular clinic visits for clinical exams, safety labs, and standardized assessments of global status, behaviour/cognition, motor function, and sleep

If they have epilepsy: keep a seizure diary and undergo EEG assessments after each treatment period

In some sites (Italy and France): undergo TMS-based neurophysiology testing

Optionally, a subset may join a cellular biomarker substudy (blood collection to generate iPSC-derived neuronal models and organoids) to explore treatment effects in variant-specific lab models.

GRIN-related neurodevelopmental disorders (GRIN-NDD) are rare genetic conditions caused by pathogenic variants in GRIN1, GRIN2A, GRIN2B, or GRIN2D, which encode subunits of the N-methyl-D-aspartate receptor (NMDAR). NMDARs play a central role in synaptic transmission, plasticity, and neurodevelopment. Variants affecting these genes may result in altered receptor function, most commonly loss of function (LoF), leading to a spectrum of neurodevelopmental phenotypes that include epilepsy, intellectual disability, behavioural abnormalities, sleep disturbances, and motor impairment.

Preclinical and clinical evidence suggests that enhancement of NMDAR co-agonist availability may partially restore receptor function in individuals with LoF GRIN variants. L-serine, a naturally occurring amino acid, serves as a metabolic precursor of D-serine, a major endogenous NMDAR co-agonist. Oral L-serine supplementation has been associated with clinical improvement in small case series and early-phase clinical studies involving individuals with GRIN-NDD due to LoF variants. However, systematic evidence from randomised, blinded studies remains limited, particularly given the rarity and phenotypic heterogeneity of these disorders.

This study is designed to evaluate the clinical effect and safety of L-serine supplementation using an aggregated series of randomised, double-blind, placebo-controlled n-of-1 trials. The n-of-1 approach allows each participant to serve as their own control and is well suited to rare diseases with substantial inter-individual variability. By aggregating data from multiple single-patient trials using Bayesian hierarchical models, the study aims to provide both individual-level and population-level estimates of treatment effect.

Following a baseline observation period, participants will undergo repeated treatment cycles in which L-serine supplementation and placebo are administered in alternating periods. Each cycle consists of two treatment periods of equal duration, with the order of treatments randomised for each cycle. A short washout interval is incorporated at the start of each treatment period to minimise potential carry-over effects. Both participants and investigators remain blinded to treatment allocation throughout the study, except in cases of emergency unblinding for safety reasons.

L-serine and placebo are provided as oral powders for solution and are administered at a fixed weight-based dose divided into three daily administrations. No dose titration is planned. Concomitant treatments, including antiseizure medications and other standard therapies, are permitted and remain stable whenever clinically feasible. Participants are monitored longitudinally through scheduled visits that include clinical evaluations, safety laboratory testing, and protocol-specified assessments.

The primary focus of the study is the assessment of global clinical change associated with L-serine supplementation compared with placebo. Secondary evaluations examine treatment effects across multiple functional domains, including behaviour, cognition, adaptive functioning, motor abilities, sleep quality, and, in participants with epilepsy, seizure frequency and electroencephalographic (EEG) features. These assessments are conducted repeatedly at the end of each treatment period, enabling within-subject comparisons across treatment conditions.

In addition to clinical outcomes, the study incorporates neurophysiological assessments to explore objective biomarkers of treatment response. Transcranial magnetic stimulation (TMS) combined with electromyography (EMG) and electroencephalography (EEG) is used at selected sites to characterise cortical excitability, inhibitory and excitatory balance, and network-level responses to stimulation. These measures are intended to provide mechanistic insights into NMDAR-related cortical function and to identify potential biomarkers that correlate with clinical response to L-serine.

An optional preclinical biomarker substudy is included for a subset of participants who provide additional consent. In this substudy, peripheral blood samples are used to generate induced pluripotent stem cells (iPSCs), which are subsequently differentiated into neuronal cell models and, in selected cases, three-dimensional brain organoids. These patient-derived models are used to examine neuronal development, synaptic function, network activity, and plasticity in vitro, and to assess the effects of L-serine exposure at the cellular and molecular levels. Data from this substudy are exploratory and are intended to support translational interpretation of clinical findings.

Safety monitoring is conducted throughout the study in accordance with Good Clinical Practice principles. Adverse events are systematically collected and evaluated for severity and potential relationship to the study product. Predefined criteria are in place for treatment discontinuation and early withdrawal in the event of clinical worsening or safety concerns. Given the non-pharmacological classification of L-serine as a food for special medical purposes and its prior clinical use, no specific toxicity signals are anticipated; however, ongoing monitoring is implemented to ensure participant safety.

The study is conducted across multiple international centres with expertise in GRIN-NDD. The innovative design of aggregated n-of-1 trials is intended to maximise the interpretability of results in a rare disease context, allowing identification of both average treatment effects and inter-individual variability in response. The findings are expected to inform clinical decision-making, contribute to evidence-based use of L-serine in GRIN-NDD, and support the broader application of n-of-1 trial methodologies in rare neurological disorders.

Inclusion Criteria:

• Clinical diagnosis of a GRIN-related neurodevelopmental disorder (GRIN-NDD)
• Presence of a pathogenic or likely pathogenic loss-of-function (LoF) variant in GRIN1, GRIN2A, GRIN2B, or GRIN2D
• Parent(s), caregiver(s), or legally authorised representative(s) have been informed of the nature of the study and have provided written informed consent.
• Participants who are able to do so have provided written informed consent or assent, according to local regulations and cognitive capacity.
• Parent(s)/caregiver(s) are willing and able to comply with study procedures and visits, in the opinion of the investigator.
• Participants who have previously received L-serine supplementation are willing to discontinue L-serine for at least one week prior to the baseline observation period.

Exclusion Criteria:

• Age younger than 2 years at screening.
• Known hypersensitivity or intolerance to L-serine, placebo, or any excipients used in the study formulations.
• Presence of a clinically significant unstable medical condition (other than epilepsy) that, in the investigator's judgement, may place the participant at increased risk or interfere with study participation.
• Any other significant disease or disorder that may compromise participant safety, affect study outcomes, or impair the participant's ability to complete the study procedures.
• Inadequate supervision by parent(s) or caregiver(s), as judged by the investigator.
• Participation in another clinical trial involving an investigational medicinal product within the previous 6 months.
• Female participants who are pregnant or breastfeeding.
• Presence of a GRIN1, GRIN2A, GRIN2B, or GRIN2D variant for which a clear loss-of-function effect cannot be demonstrated.