Results
Etiology
For understanding AS etiology, we have to delve deeper into the ∼2 Mb genomic parent-specific imprinting region 15q11q13 on the proximal q arm of chromosome 15. The maternal and paternal alleles differ in this region regarding DNA methylation, histone modification, and gene expression. A two-part imprinting center (IC), overlapping with the SNRPN gene, controls the genomic imprinting of this region. Most of the genes in this region, including SNRPN , are only expressed paternally; the maternal allele is methylated in this area. An exception in two ways is UBE3A (ubiquitin-protein ligase E3a): it is unmethylated on both alleles in most tissues and in neurons, it is only expressed on the maternal allele. A paternally derived antisense transcript enables brain-specific silencing of paternal UBE3A .[8 ]
UBE3A plays a role as a ligase in the ubiquitin proteasome pathway as well as a transcriptional coactivator ([Fig. 2 ]).[9 ]
[10 ]
Fig. 2 Localization of the relevant genomic parent-specific imprinting region 15q11q13 on the proximal arm of chromosome 15 for AS. Zoomed in from (male) karyogram to maternal and paternal chromosome 15 and the respective genes in this region with marked breakpoints (bp), the imprinting center (IC), class 1 and 2 deletion, and the UBE3A gene. Illustration of chromosome region 15q1q13 was modified from the AWMF S1 guideline for diagnostic of Angelman syndrome.[10 ] Created in BioRender. Hentschel, J. (2025) https://BioRender.com/r46z766
Loss of function of UBE3A on the maternal allele causes AS. The loss of function can occur in different ways. The most common cause is a 5 to 7 Mb deletion on the maternal chromosome 15. A distinction is made between a smaller deletion 2 (Breakpoint BP2 and BP3), a larger deletion 1 (BP1–BP3), and atypical deletions. Other causes can be a paternal UPD 15, an imprinting defect (paternal imprinting pattern on the maternal allele), or in up to 10% a pathogenic variant in UBE3A on the maternal allele. Approximately 10% of imprinting defects are due to a 5 to 80 kb IC deletion upstream of SNRPN ([Fig. 2 ]).[10 ]
[11 ]
[12 ]
[13 ]
[14 ] Pathogenic variants in the IC have not yet been identified.[15 ]
The recurrence risk of AS is low but depends on the respective genetic cause. Deletions and UPD almost always occur sporadically. A balanced translocation including chromosome 15 in the mother can cause an unbalanced translocation in the child resulting in AS. Imprinting defects without IC deletion or rearrangement appear to be sporadic, as well. Pathogenic variants in UBE3A and IC deletions can occur either sporadically or can be inherited (silent transmission via the mother's paternal allele). If the mother is a carrier, there is a 50% recurrence risk of AS for her offspring. Regarding de novo IC deletions and pathogenic variants in UBE3A , a germ cell mosaicism is possible resulting in an increased recurrence risk.
Genetic Diagnostic Approach
Referring to the “Update of the EMQN best practice guidelines for molecular analysis of Prader–Willi and Angelman syndromes” and the German Guidelines (“AWMF Leitlinien für die molekulare und zytogenetische Diagnostik bei Prader-Willi-Syndrom und Angelman-Syndrom”), the commonly recommended method is methylation-specific (MS)-MLPA (multiplex ligation-dependent probe amplification).[10 ]
[16 ] MS-MLPA detects methylation and gene dosage of the region 15q11q13. In addition to deletion 1 and 2, the MS-MLPA can also detect UPD, imprinting defects, IC deletions, and atypical deletions. For differentiation between UPD and imprinting defects, microsatellite analysis of both parents and the patient needs to be performed. A normal test result of MS-MLPA does not exclude AS. In cases with high suspicion of AS but normal MS-MLPA result, sequencing of UBE3A should be considered. Array analysis or fluorescence in situ hybridization analysis is no longer recommended.
Preclinical and Clinical Trials
After the diagnosis of AS in a child, many questions arise, especially about treatment options. Currently, there is no causal therapy established as standard of care for patients with AS. However, there are several clinical trials ongoing and various approaches in preclinical stage, which will be discussed in the following. The therapies focus on three different ways, either to restore UBE3A (e.g., by adeno-associated virus [AAV]-mediated gene replacement), to unsilence the paternal allele (e.g., by antisense oligonucleotide [ASO] therapy), or to target altered downstream mechanisms (e.g., to increase GABA signaling). [Fig. 3 ].[17 ]
[18 ]
[19 ]
[20 ]
[21 ]
[22 ]
[23 ]
Fig. 3 Overview of ongoing clinical and preclinical trials. Currently, three strategies exist to address treatment of AS: to restore UBE3A , using an adeno-associated virus (AAV) vector carrying the UBE3A gene[17 ] or modified hematopoietic stem cells (HSCs) producing UBE3A protein[18 ]; to unsilence UBE3A on the paternal allele, using ASO-therapy,[19 ] Crispr/Cas9,[20 ] or zinc finger proteins[21 ]; to target altered downstream mechanism by administration of Alogabat (a positive allosteric modulator of the α5 subunit containing GABAA receptor[22 ]) or by administration of NNZ-2591 (a synthetic analogue of cyclic glycine proline[23 ]) in clinical trials, and via several small molecules in preclinical development. Created in BioRender. Hentschel, J. (2025) https://BioRender.com/r46z766
Restore UBE3A
Adeno-Associated Virus–Mediated Gene Replacement
In principle, AAV is a small, nonenveloped virus that packages a single-stranded linear DNA genome of the gene that required replacement. It cannot replicate without the aid of host cell machinery or helper viruses. The viral capsid serves as the vehicle for viral gene delivery. Daily et al proofed beneficial effects in AS mouse models by directly injecting an AAV vector carrying the Ube3a gene into the brain of the animals.[17 ] These mice showed improvements in learning and memory. Based on the published knowledge, clinical phases are not yet beyond the preclinical setting.[17 ]
Stem Cell Therapies
Another approach to restore UBE3A levels in neurons is the transplantation of autologous hematopoietic stem cells (HSCs) that have been modified by a lentiviral vector to produce UBE3A.[18 ] These modified stem cells deliver UBE3A via cross-correction to UBE3A -deficient cells, i.e., neurons. After transplantation of these stem cells to immune- and Ube3a -deficient mice, increased levels of Ube3a were detected in the brain and an amelioration of motor and behavioral deficits as well as normalized delta power in the EEG were found.[18 ] Stem cell therapy with gene-corrected HSCs by transfection with a lentiviral vector has been for example successfully used in patients with metachromatic leukodystrophy or cerebral adrenoleukodystrophy.[24 ]
[25 ] Stem cell therapy for patients with AS is in preclinical development. Favorable is the fact that the stem cell therapy could be a permanent treatment.[26 ] On the other hand, a suppression of the immune system is required for the transplantation of HSCs with all its potential side effects.
Unsilence the Paternal Allele
The paternal UBE3A allele of patients with AS is a fully intact gene copy, that is, however, silenced in neurons by an UBE3A antisense transcript. There are several attempts to unsilence this paternal UBE3A allele in neurons.
Antisense Oligonucleotides Therapies
ASOs are chemically modified oligonucleotides, more specifically short, single-stranded synthetic deoxynucleotide or ribonucleotide analogues.[27 ] They do not require integration into the genome and are designed to be complementary to specific messenger RNA (mRNA) molecules. ASOs bind to target mRNA and prevent the mRNA from being translated into a protein by the ribosome. In this way, ASOs modulate the expression of genes and can either enhance or inhibit gene expression. ASO therapy is highly precise because it can be tailored to target-specific genes or gene variants. They are very promising disease-modifying agents and have already received approval from the U.S. Food and Drug Administration and European Medicines Agency for certain neurological disorders (e.g., Nusinersen for spinal muscular atrophy).[28 ]
[29 ]
As mentioned above, in neurons UBE3A is only expressed maternally, the expression of the paternal allele is suppressed by a specific noncoding RNA known as UBE3A antisense transcript (UBE3A -ATS). One promising approach involves reactivating the intact paternal UBE3A allele in neurons by suppressing UBE3A -ATS.[19 ]
[30 ] ASOs have been designed to specifically bind to UBE3A -ATS. By precisely targeting this region, these ASOs efficiently repress the transcription of UBE3A -ATS, effectively “switching on” the paternal UBE3A allele, and restore functional UBE3A expression.
In laboratory settings using induced pluripotent stem cells derived from both individuals with and those without AS, ASOs successfully reactivated the expression of paternal UBE3A allele in neurons.[31 ]
[32 ]
Also in animal studies, ASOs were shown to be effective. Following ASO treatment, UBE3A expression was observed throughout the central nervous system in treated mice.[33 ]
[34 ]
[35 ]
[36 ]
[37 ]
Behavioral testing in these mice demonstrated significant improvements. Sensitivity to audiogenic seizures was prevented. Additional improvements were observed in anxiety-related behaviors, forced swim tests, and motor coordination (rotarod performance).[34 ] Milazzo et al showed also a trend toward reducing microcephaly and evidence for enhanced hippocampal synaptic plasticity.[34 ] They also highlighted the importance of the timing of ASO treatment and emphasized a critical window for ASO intervention. Administering ASOs around birth appeared to be more effective than treatment at later stages.
In this regard, Lee et al recently showed that ASO therapy in juvenile or adult mice can still rescue abnormal EEG patterns and sleep disturbances, but the impact in adult mice was less pronounced.[33 ]
Currently, ASO therapy for AS is in early clinical trial stages. Two ongoing clinical trials are indicated by its registration on ClinicalTrials.gov. One trial (NCT04259281) is a Phase 1/2, open-label, multiple-dose study to evaluate the safety, tolerability, and plasma and cerebrospinal fluid concentrations of GTX-102 in pediatric participants with AS.[38 ] A second trial (NCT05127226) evaluates safety, tolerability, pharmacokinetics, and pharmacodynamics of intrathecally administered ION582 in patients with AS aged 2 to 50 years.[39 ] A third clinical trial was stopped by the sponsor in 2023 (NCT04428281) as efficacy data of the study drug (RO7248824, Rugonersen) have so far fallen short of criteria that would be needed to justify continued investment in the therapy.[40 ] The negative charge, high molecular weight, and hydrophilic nature of ASOs inhibit their diffusion across the blood–brain barrier and limit the efficacy after systemic administration. For this reason, intrathecal injection is necessary in AS. Potential side effects include type I hypersensitivity, radiculopathy, fever, thrombocytopenia, arthralgia, nephrotoxicity, and hepatotoxicity.
CRISPR/Cas9
Unsilencing of the paternal allele via CRISPR/Cas9 has been shown in preclinical studies. The used guide RNAs differ in efficiency when tested in mice models. Especially Spjw33 and ATS-GE showed good results in reducing the transcription of targeted Ube3a -ATS areas and resulted in positive effects in motor and behavior phenotypes.[20 ]
[41 ] Both approaches needed intracerebroventricular injections with gene editing in only a subset of neurons (up to 20%).[41 ]
Zinc Finger Artificial Transcription Factor
Another therapeutic approach to unsilence the paternal allele is the use of artificial zinc finger–based transcription factors. Zinc finger proteins are artificially produced zinc finger nucleases that enable gene-specific suppression. They interact specifically with DNA or RNA and can be programmed to target specific sequences in the genome. For the AS mouse model, an artificial transcription factor ATF-S1K26 was developed that specifically binds to the mouse Snurf/Snrpn transcription start site to downregulate transcription in a target-specific manner.[21 ]
[42 ]
[43 ] A positive effect was shown in the mouse model and only a single injection appears to be necessary.[21 ] These artificial transcription factors have already been tested in clinical trials for HIV-1 and β-thalassaemia.[21 ] However, there have been no clinical studies for AS to date.
Downstream Therapies
Beside the intention to activate the silenced paternal allele, there are several attempts to target molecular downstream mechanisms.
Targeting GABA Receptors
AS children with deletion genotype usually lack a copy of UBE3A as well as of other genes, i.e., GABRB3 , GABRA5 , and GABRG3 encoding for GABAA receptor subunits β3, α5, and γ3.[35 ] α5-containing GABAA receptors are known to play a key role in learning, memory, and development.[44 ] Furthermore, they are involved in seizure control, and pathogenic variants in GABRA5 receptors are associated with developmental and epileptic encephalopathies.[45 ]
[46 ] Several studies found that children with a deletion subtype show more severe phenotypes than nondeletion patients.[47 ]
[48 ] Postmortem examination of cortex of patients with AS showed decreased GABA α5 subunit protein expression indicating a contribution of the hemizygous GABR gene cluster to the more affected phenotype of children with a deletion.[49 ]
Alogabat is a selective, positive allosteric modulator of the α5 subunit containing GABAA receptor.[22 ] In July 2023, a phase IIa, multicenter, open-label study (Aldebaran, NCT05630066) with children with AS started and will enroll up to 56 participants with deletion genotype. Completion is estimated for March 2025.[50 ]
Effect of Insulin-Like Growth Factors, Their Metabolite Cyclic Glycine Proline and Analogues
Insulin-like growth factors (IGFs) IGF-1 and IGF-2 and their metabolites like cyclic glycine proline (cGP) have effects on neuroprotection and neuronal recovery.[51 ] The subcutaneous administration of IGF-2, which binds to the CIM6P/IGF-2- receptor, in an AS mouse model ameliorated motor and cognitive deficits and reduced acoustically induced seizure.[52 ] NNZ-2591, a synthetic analogue of cGP, was tested in an AS mouse model and showed an improvement in experiments of behavior, motor, and intellectual skills.[23 ] In July 2022, an open-label phase II study of the safety, tolerability, and pharmacokinetics of oral NNZ-2591 in AS (Neu-2591-AS-001) started, the study is ongoing, and study completion is estimated for the end of June 2024.[53 ]
Exogenous Ketones as Treatment Option
A high prevalence of epilepsy in children with AS, especially with deletion genotype, is found in natural history studies.[54 ] Beside antiseizure medication, dietetic therapies like low-glycemic index diet and ketogenic diet are recommended, especially for children difficult to treat.[55 ]
[56 ] As ketogenic and low glycemic index diets demand good compliance of patients, the administration of exogenous ketones via a fat-based nutritional formulation was proposed.[57 ] In an AS mouse model, the supplementation of ketone esters ameliorated the phenotype of AS mice.[58 ] In a double-blind, placebo-controlled study of exogenous ketones (FANS, NCT03644693) with 15 children, a feasible safety and tolerability profile as the primary endpoint was found.[59 ] Furthermore, delta power in the EEG decreased during the treatment period and improved fine motor function was shown in five of six patients with nutritional formulation.[57 ] Due to several limitation factors of the clinical trial (i.e., small cohort, short observation period), further research is recommended.[57 ]
Downstream Pathways in Preclinical Status
Taurine
Taurine, a sulfur-containing amino acid, is known to be neuroprotective and plays a critical role in normal brain development.[60 ] Taurine functions as a weak, partial GABAA receptor agonist.[61 ] When administered in the AS mouse model, motor and learning deficits of AS mice improved.[62 ] The positive effect of taurine on neurological disorders, i.e., stroke, epilepsy, and memory dysfunction, has been investigated in numerous mouse models.[63 ] Currently, there is a clinical trial to evaluate the effect of administration of taurine on core symptoms of children with autism spectrum disorders (NCT05980520), but so far none in children with AS.[64 ]
LB-100
LB-100 is a potent inhibitor of the protein phosphatase 2 (PP2A), a serine/threonine phosphatase involved in numerous cell regulation processes.[65 ] In an AS mouse model, elevated concentrations of PP2A and decreased levels of its activator phosphotyrosyl phosphatase activator, which is a substrate of UBE3A, were found. The administration of LB-100 improved motor functions of AS mice.[66 ] Due to its antitumor effect, LB-100 is currently studied in clinical trials for several solid tumor entities, including tumors of the central nervous system (NCT03027388).[67 ]
NSI-189
Another small molecule in preclinical development for AS is the neuroprotective agent NSI-189 phosphate.[68 ] Treatment of AS mice with NSI-189 phosphate ameliorated motor and cognitive functions.[69 ] Patients with major depressive disorder treated with NSI-189 in a double-blind, placebo-controlled clinical trial (NCT02695472) benefited compared with placebo.[70 ]
[71 ]
[72 ]
CN2097
An altered signaling of tropomyosin receptor kinase B (TrkB) and its ligand brain-derived neurotrophic factor (BDNF) are supposed to be involved in several neurological, psychiatric, and proliferative disorders, as TrkB/BDNF play a key role in development and synaptic plasticity of neurons.[73 ] Also, in AS mice an impaired signaling of TrkB due to an attenuated association with the postsynaptic density protein-95 (PSD-95) was shown.[74 ] The administration of CN2097, a bridged cyclic peptide that binds to the PDZ1 domain of PSD-95, raised concentration of synaptic proteins in the hippocampus and ameliorated cognitive and motor function of AS mice.[75 ]
Linoleic Acids
In sensory neurons of Ube3a- deficient mice, the activity of PIEZO2, a mechanosensitive ion channel, is reduced.[76 ] There are no known agonists of PIEZO2, but response of PIEZO1 to mechanical stimuli can be modulated by fatty acids.[77 ] When Ube3a -deficient mice received a diet enriched of linoleic acid, an essential ω-6 fatty acid and a structural component of plasma membranes, AS mice showed an improvement of gait ataxia.[76 ]
Bumetanide
Another target for treatments of various neurological disorders is the imbalanced regulation of intracellular chloride concentrations by the Cl− importer Na+ –K+ –Cl– transporter isoform 1 (NKCC1) and the Cl− exporter K+ –Cl– transporter isoform 2 (KCC2) leading to an altered GABA signaling.[78 ]
[79 ] In hippocampi of Ube3a -deficient mice, an elevated expression of NKCC1 and a decreased expression of KCC2 were found.[80 ] The administration of an inhibitor of NKCC1, bumetanide, ameliorated cognitive deficits in AS mice, but not motor deficits and EEG abnormalities.[80 ] The effect of bumetanide on neurological disorders was tested in several preclinical and clinical studies, for example, for neonatal seizures and autism, partially with promising results (reviewed in Savardi et al[81 ]). However, two randomized, placebo controlled phase III studies of children with autism spectrum disorders failed to demonstrate effectiveness.[78 ]
Clinical Interventional Trials That Missed Primary and/or Secondary Endpoints
Targeting GABA Receptors: Gaboxadol
The ubiquitin E3 ligase UBE3A binds and regulates the concentration of presynaptic GABA transporter 1 (GAT1) by degradation.[82 ] In Ube3a -deficient AS mice, elevated concentration of presynaptic GAT1 was found, causing an increased reuptake of GABA. In consequence, extra-synaptic GABAA receptors containing a δ-subunit are less activated leading to decreased tonic inhibition.[82 ]
[83 ] The administration of a δ-selective agonist of GABAA receptors (4,5,6,7-tetrahydroisothiazole-[5,4]-pyridine-3-ol, THIP, also known as Gaboxadol) balanced tonic inhibition and improved motor dysfunction of Ube3a -deficient mice.[82 ]
Gaboxadol (OV101) was administered in a randomized placebo-controlled phase II study in adolescents and adults with AS (STARS, NCT02996305) for 12 weeks leading to an improvement in the CGI-I-AS scores, especially regarding the sleep symptoms with a reduced sleep-onset latency.[84 ]
[85 ] In children, a double-blind, randomized placebo-controlled phase III study (NEPTUNE, NCT04106557) did not demonstrate a significant difference in the CGI-I-AS scale between children treated with Gaboxadol versus placebo.[86 ]
[87 ] Due to these results, the open-label phase III study to evaluate the long-term safety, tolerability, and efficacy of OV101 in individuals with AS (ELARA, NCT03882918), where patients received a treatment with OV101 for 52 weeks, was stopped.[88 ]
Targeting Different Pathways: Methyl Donors, Minocycline, and Levodopa
The attempt to increase the expression of UBE3A and DNA methylation levels by the administration of the methyl donors betaine and folic acid failed in clinical trials, the clinical trials did not show differences between treated and placebo groups.[89 ]
[90 ]
[91 ]
Minocycline is a tetracycline antibiotic that is also known for its anti-inflammatory and antioxidative effect in neurons.[92 ] A randomized placebo controlled trial for minocycline in patients with AS (A-Manece study, NCT02056665) failed to improve developmental parameters.[93 ]
[94 ]
Another altered downstream pathway in the pathophysiology in AS is a decreased activity of calcium/calmodulin-dependent kinase II (CaMKII) in the hippocampus of AS mice associated with an altered phosphorylation status of threonine residues.[95 ] Levodopa was shown to normalize phosphorylation level of CaMKII.[96 ] In a randomized placebo controlled clinical trial with Levodopa (Levodopa trial register, NCT01281475) no statistically significant change was seen in the outcome parameters neurodevelopment and behavior.[96 ]
[97 ]
Current Clinical Trials
Currently, there are four clinical trials taking place worldwide, two phase I/II studies about ASO therapies (GTX-102, NCT04259281 and ION582, NCT05127226) and two phase II clinical trials about downstream mechanisms (Aldebaran, NCT05630066 and NNZ-2591, NCT05011851).[38 ]
[39 ]
[50 ]
[53 ] ASO therapies have to be administered regularly intrathecally requiring a short anesthesia targeting the underlying genetic cause of AS. In contrast, Alogabat and NNZ-2591 are taken orally and aim to ameliorate deficits by targeting different altered downstream mechanisms. In Germany in 2024, the Alogabat study in Munich and the GTX-102 study in Hamburg and Leipzig are ongoing. An overview of current clinical trials worldwide is provided in [Table 1 ].
Table 1
Overview of current clinical trials worldwide
Trial title
Trial ID
Phase
Sites
No
Age (y)
Administration
Study completion (estimated)
Ref.
A Study of the Safety and Tolerability of GTX-102 in Children With Angelman
Syndrome
NCT04259281
I/II
United States
Australia
Canada
France
Germany
Spain
Israel
United Kingdom
74
4–17
Intrathecal
12/2025
[38 ]
HALOS: A Safety, Tolerability, Pharmacokinetics and Pharmacodynamics Study of Multiple Ascending Doses of ION582 in Participants with Angelman Syndrome
NCT05127226
I/II
United States
Australia
France
Israel
Italy
United Kingdom
51
2–50
Intrathecal
3/2029
[39 ]
Study to Investigate the Pharmacokinetics and Safety and to Provide Proof of Mechanism of Alogabat in Children and Adolescents Aged 5–17 Years with Angelman Syndrome (AS) with Deletion Genotype (Aldebaran)
NCT05630066
II
United States
Australia
France
Germany
Italy
Spain
56
5–17
Oral
3/2025
[50 ]
An Open-Label Study of the Safety, Tolerability, and Pharmacokinetics of Oral NNZ-2591 in Angelman Syndrome (AS-001)
NCT05011851
II
Australia
20
3–17
Oral
6/2024
[53 ]
Neuropediatrics Evaluation Parameter
To assess the effectiveness of new therapeutic approaches, evaluation criteria should be used that capture symptoms that are typical for AS. According to Tjeertes et al, measurements should be “noninvasive and minimally burdensome.”[98 ] The parameters used in current studies are, on the one hand, assessments by caregivers and, on the other hand, data collected by clinical experts. In addition to clinical outcome measures such as EEG, laboratory parameters, developmental tests, semi-structured interviews, and questionnaires, the use of digital health technologies is of great importance.[98 ] These include, for example, video-recorded sleep EEGs and actigraphs.
EEG as a Biomarker
The EEG in children with AS is generally conspicuous regardless of epilepsy, and diagnosis of AS can often be made on the basis of EEG alone. High-amplitude rhythmic delta waves are described over the frontal and posterior regions, sometimes with embedded spikes as “notched delta” and generalized rhythmic theta activity without visual blockage.[99 ] Fewer and shorter sleep spindles are described during sleep.[100 ] Hypsarrhythmia-like patterns are rarely found.[101 ]
There are some attempts to establish the EEG as an objective biomarker. The proportion of delta waves was measured for the first time in 2017 using spectral analysis (delta power). Increased delta power was detected in the AS mouse model and in children with AS.[102 ] Theta power was also established as a biomarker and differences between the phenotypes and the severity of the disease were identified.[103 ] Deletion and nondeletion children were compared and the increased theta power was attributed to the hemizygosity for the GABAR genes in the deletion group. The changes in the delta band were attributed to UBE3A .[103 ] Delta power has been shown to correlate with cognitive function measured by the Bayley III score.[104 ] In the mouse model, effects of ASO therapy on delta power were shown as a biomarker.[105 ]
Clinical Global Impression Scales for Angelman Syndrome CGI-S (Severity) and CGI-I (Improvement)
The CGI has long been established as an evaluation parameter for clinical studies, especially in the psychiatric field.[106 ]
In the phase II STARS study (NCT02996305) with the GABA analog gaboxadol, these scales were used for the first time in a form adapted for AS and modified for the subsequent phase III NEPTUNE study (NCT04106557). The CGI-S-AS scales consist of six domains, behavior, gross and fine motor skills, expressive and receptive language, and sleep, special anchors were developed with seven subclasses. The CGI-I-AS scales focus on changes in the individual domains, with seven subclasses ranging from 1 (very much improved) to 7 (very much worse). Advantages are the relatively simple implementation, whereby it should always be the same rater, who must be specially trained. Disadvantages are the subjectivity, especially with the CGI-I scales.[106 ] Sleep and behavior are more difficult to capture in the scales than motor skills and communication. After evaluating the NEPTUNE study (NCT04106557), these scales were once again critically scrutinized and compared with other parameters of the study.[107 ] It was found that the CGI scales S and I were consistent in the domains of behavior and sleep and inconsistent in the domains of motor skills and communication. In comparison with VABS-III, the results were different, particularly in the area of communication. A comparison with the sleep diaries showed a good correlation, but not with the actigraphs used in the study.
Observer Reported Communication Ability Measure
The ORCA (Observer Reported Communication Ability) measurement was originally developed to assess communication skills in AS. The main purpose of the instrument is to record the individual baseline communication level and possible changes in competences within the framework of studies. The ORCA takes into account the different types of communication and differentiates between gestures/signs, words, assistive technology, and combinations of these types of communication.[108 ]
[109 ]
Pediatric Quality-of-Life Inventory
The Pediatric Quality of Life Inventory was developed by Varni's working group to measure the health-related quality of life of children and adolescents.[110 ] Subsequently, a series of disease-specific additional modules and the module “Family Impact” were added.[111 ] The disease-specific additional modules “Cerebral Palsy” and “Multidimensional Fatigue” as well as the module “Family Impact” are used in studies.[110 ]
[111 ]
[112 ] To date, there is no disease-specific module for AS.
Aberrant Behavior Checklist
The items of the Aberrant Behavior Checklist are distributed across the scales “irritability/agitation,” “lethargy/withdrawal,” “stereotypic behavior,” “hyperactivity/noncompliance,” and “inappropriate speech.”[113 ] Studies, which recorded the behavioral abnormalities of individuals with AS and with other genetic disorders, usually found lower scores for the group of individuals with AS.[114 ] Possibly, the aberrant behavior checklist does not adequately reflect characteristics of AS, such as hyperactivity, restlessness, and fascination with water in combination with impaired language development.
Bayley Scales of Infant and Toddler Development
The Bayley Scales of Infant and Toddler Development is a developmental test for examining the developmental status of infants and toddlers aged between 1 and 42 months.[115 ] The Bayley scales are also frequently used for older children with severe developmental retardation.[116 ]
In an AS natural history study, it was shown that 6-year-old children with AS showed a cognitive developmental level of 14 to 27 months when using the age-equivalent scores.[117 ] Depending on the genetic subtype, the developmental progress was 1 to 2 months per year of life. The authors emphasize that age-equivalent scores are not sufficiently sensitive to reflect small developmental progress.[117 ]
Vineland Adaptive Behavior Scales
The Vineland Adaptive Behavior Scales are a multidimensional measurement instrument for assessing the adaptive behavior of children, adolescents, and young adults aged between 3.0 and 21.11 years.[118 ] It is a semi-structured interview that is conducted with a primary caregiver. Gwaltney et al examined adaptive skills and trajectories in 257 individuals with AS using the second edition of the Vineland Adaptive Behavior Scales. The authors showed that individuals with deletion subtypes demonstrate a lower level of adaptive skills than nondeletion subtypes.[119 ]
The NEPTUNE study (NCT04106557) showed that changes in the values from baseline to week 6 of the interpersonal relationship and coping skill subscales corresponded with the CGI-I-AS global improvement scores.[107 ]
Children's Sleep Habit Questionnaire
The Children's Sleep Habit Questionnaire is a questionnaire in which the caregiver is asked to retrospectively assess the sleep behavior of younger children.[120 ]
The analysis of the NEPTUNE study (NCT04106557) data showed some validity between these scales and the CGI, but not in all subscales.[107 ]
Development of a Newborn Screening
The feasibility of newborn screening for AS, Prader–Willi syndrome (PWS), and Dup15q was tested using SNRPN methylation.[121 ] The test was performed on 109 children with PWS, 48 children with AS, 9 children with Dup15q, and 1,190 healthy newborns. A high sensitivity was found, but only a positive predictivity for AS of 67%. However, this could be improved by changing the threshold values. UBE3A mutations show a normal methylation pattern and are not recognized. The main problem at present is the lack of a causal therapy as a basis for newborn screening.
Summary and Outlook
The genetic background of AS is well understood, as it is caused by loss of UBE3A on the maternal allele in neurons. It is known that UBE3A plays a key role in synaptic development and is involved in GABA metabolism, but detailed information about pathophysiology and the effect of altered downstream pathways on the phenotype of patients with AS remain unclear.[35 ] Depending on the genetic variant, there is considerable phenotypic variation.[47 ] And still, even among individuals with the same type of genetic alteration, phenotype severity can vary, leading to a complex clinical picture with various individual problems of each patient. Overall, the burden of families of children with AS and the need for causal therapies is enormous.
The research on AS is facing various challenges. Results obtained in a Ube3a- deficient mouse model cannot easily be transmitted to patients with AS, as the majority of AS is caused by a large maternal deletion including more genes then UBE3A . There are several attempts to establish a deletion mouse model, one for example was presented on FAST summit gala 2023.[122 ] Characterization and the effect of drugs in an AS mouse model are based on different tests like forced swim test, rotarod test, marble burying, and open fields.[35 ] These tests try to reflect the complexity of the phenotype of AS, but are mainly focused on motor deficits and show difficulties to measure seizure susceptibility, speech, and behavioral problems. However, the research on mouse models can identify involved target molecules and helps to find the critical time window for the application of drugs.[35 ] The understanding of the essential role of UBE3A in the pathophysiology of AS enabled researchers to search for specific therapies.
The application of AAV gene replacement and of stem cell therapy to patients with AS is in preclinical status. The successful restoration of UBE3A in UBE3A -deficient neurons by autologous genetically modified stem cells has been shown in an immunodeficient AS mouse model.[18 ] In humans, neurodegenerative diseases like metachromatic leukodystrophy have been treated with stem cell therapy.[25 ] But in contrast to these diseases, patients with AS have a normal life expectancy without progressive neurodegeneration leading to death. Stepping forward to clinical evaluation will be an ethical challenge.
To activate the paternal allele by the administration of artificial zinc finger-based transcription factors or by ASOs represents an opportunity for precision medicine in patients with AS. Whereas the artificial transcription factors are in preclinical research, ASO therapy in AS has evolved to clinical development, there are currently two clinical trials (phase I/II) ongoing. It is a promising therapy for addressing AS at its genetic root, but potential long-term effects and safety concerns need further investigation. Also, responses to ASO therapy may vary among patients. Moreover, it remains unclear when in human development it becomes too late to reactivate a silenced gene in the nervous system and whether neurogenetic diseases can be effectively treated in adulthood.
Further developments in ASO therapy for AS may include continued refinement of ASO design for most precise targeting, optimizing dosing regimens, and delivery methods to ease the ASO treatment. In future, it might be combined with other treatments, such as small molecules addressing altered downstream mechanisms.
An advantage of these small molecules is the oral administration and the research of tolerability and safety in a broad field of neurological disorders as many of these substances are known to be in general neuroprotective. However, they do not target the underlying genetic cause of AS which will presumably limit their clinical efficacy in clinical trials. Alogabat is focused on balancing the GABA deficit of patients with deletion genotype, but not on restoration of UBE3A and its downstream pathways.
For the successful establishment of new therapies in patients with AS, evaluation parameters of high quality are essential. In clinical trials, a great number of varying assessment parameters are used to quantify and describe change in behavior, sleep, development, communication, and seizure. But as many of them are not adapted to the characteristics of patients with AS, they cannot fully reflect the complexity of this disease and are used due to missing alternatives. Research on reliable, objective, reproducible, and nonrater-dependent parameters is ongoing, for example, an AS Biomarker and Outcome Consortium (ABOM) is supported by the FAST organization to establish AS-specific outcome parameters. Promising is the measurement of the delta and theta power in the EEG as an evaluation parameter.[103 ]
[105 ] Also, the use of other digital health systems, e.g., actigraphs and home polysomnography, might be helpful.[98 ] Overall, an enormous challenge for every single parameter is the quantification of change. The observation period in clinical trials is often short and the change in one single parameter does not automatically mean a clinically significant improvement for the patient.
As for now we are lacking a disease-modifying treatment, patients are treated on a supportive and symptomatic level. To improve care of patients with AS, Angelman clinics were founded worldwide. Patients treated in an Angelman center can benefit from the experience of clinicians and therapists specialized on AS, as they have a profound knowledge about the natural progression of the disease combined with expertise about latest therapeutic approaches. Medical and psychological problems can be challenging in a patient with AS and require a qualified and proficient interdisciplinary team.
Here we are now, and where are we going? There are many potential therapeutic agents in preclinical development. In the next few years, some will probably take the next step to clinical evaluation. A newborn screening for AS might be established when a therapy for children with AS is found, that should be administered within the first weeks of life to reach the best outcome of treated children. Natural history studies are ongoing in many different countries worldwide, the data obtained will help us to understand the needs, problems, and hopes of families with a patient with AS. Our work is supported by national and international family organizations, and we would like to thank them for their great commitment.
