Novel ANKRD11 Mutation in KBG Syndrome: A diagnostic triad of hearing loss, radiological macrodontia and artificial intelligence-assisted facial phenotyping

TL;DR

A rare genetic disorder called KBG syndrome was diagnosed in a Moroccan boy using AI facial analysis and confirmed by a new mutation in the ANKRD11 gene.

Contribution

A novel nonsense mutation in the ANKRD11 gene was identified in a patient with KBG syndrome using AI-assisted facial phenotyping.

Findings

AI-assisted facial phenotyping suggested KBG syndrome in a patient with hearing loss and macrodontia.

A novel nonsense mutation (c.1977C>G; p.Tyr659Ter) in the ANKRD11 gene was confirmed as the cause.

Multidisciplinary treatment improved seizure control and developmental progress in the patient.

Abstract

KBG syndrome is a rare autosomal dominant disorder characterised by developmental delay, characteristic facial features, macrodontia and skeletal anomalies, caused by mutations in the ANKRD11 gene. We report a 5.5-year-old Moroccan boy who presented in 2022 to a tertiary military teaching hospital in Rabat, Morocco, with psychomotor delay, autistic traits, epilepsy, bilateral hearing loss with chronic otomastoiditis and radiologically-detected macrodontia before clinical eruption, in whom artificial intelligence-assisted facial phenotyping suggested the diagnosis, subsequently confirmed by identification of a novel nonsense mutation (c.1977C>G; p.Tyr659Ter). Multidisciplinary management including antiepileptic therapy, speech therapy and audiological follow-up resulted in satisfactory seizure control and developmental progress.

Figures

Photographs of the facial phenotype of a 5.5-year-old patient with KBG syndrome showing characteristic dysmorphic features. A: Frontal view demonstrating arched eyebrows, flat nasal bridge and anteverted nostrils. B: Profile view showing micrognathia and ear configuration.

Dental panoramic radiograph revealing macrodontia of the permanent upper central incisor germs (arrows) before clinical eruption.

Artificial intelligence-assisted facial phenotyping using Face2Gene platform (FDNA™, Atlanta, Georgia, USA). The analysis correctly identified KBG syndrome (OMIM: 148050) as the top diagnostic hypothesis with high confidence scores for both gestalt and feature recognition. This digital phenotyping tool proved valuable in guiding genetic testing toward the ANKRD11 gene, where a novel pathogenic mutation was subsequently identified.

Genetic and clinical summary of KBG syndrome. Genetic findings (left panel) showing variant characteristics (c.1977C>G; p.Tyr659Ter), American College of Medical Genetics and Genomics classification (PVS1, PM2, PP5), sequencing methodology and ANKRD11 protein structure with functional domains. The schematic illustrates the location of the truncating variant at position 659, resulting in loss of 75% of the protein including the repression domain, activation domain and C-terminal region. Clinical summary (right panel) consolidating developmental milestones, growth parameters, clinical features across multiple systems, key investigations and management outcomes after 24 months of follow-up. RD = repression domain; AD = activation domain.

Full text

1. Introduction

KBG syndrome is a rare genetic disorder with autosomal dominant transmission, first described by Herrmann et al. in 1975; the acronym comes from the initials of the first 3 identified families.^1^ Its prevalence is estimated at less than 1 case per million births.^2^ Initially characterised by the association of facial dysmorphism, macrodontia of central incisors, psychomotor developmental delay and skeletal anomalies, this syndrome is caused by heterozygous mutations or microdeletions of the ANKRD11 gene (Ankyrin Repeat Domain-containing protein 11) located on chromosome 16q24.3.^13^

The diagnosis of KBG syndrome remains challenging, particularly before the eruption of permanent teeth, with a median age at diagnosis of 8 years.^4^ Expanding the known phenotypic spectrum, especially regarding neurological and audiological manifestations, could allow earlier diagnosis of this condition.^5678^ In parallel, new technologies such as artificial intelligence (AI)-assisted facial phenotyping open promising perspectives for the early identification of rare genetic syndromes.

We report the case of a Moroccan child with KBG syndrome carrying a novel mutation of the ANKRD11 gene, illustrating the interest of a multimodal diagnostic approach combining genetics, imaging and digital phenotyping.

Photographs of the facial phenotype of a 5.5-year-old patient with KBG syndrome showing characteristic dysmorphic features. A: Frontal view demonstrating arched eyebrows, flat nasal bridge and anteverted nostrils. B: Profile view showing micrognathia and ear configuration.

2. Case report

We report a 5.5-year-old Moroccan boy, referred in 2022 to the neuropediatric consultation at a tertiary military teaching hospital in Rabat, Morocco, for evaluation of psychomotor developmental delay, behavioural disorders and epileptic seizures. The anamnesis revealed perinatal history marked by neonatal hypotrophy (birth weight: 2,200 g), obstetric dystocia with transverse presentation and neonatal asphyxia with delayed cry. Growth has remained insufficient since birth.

Psychomotor development showed significant delay with acquisition of sitting position at 13 months, crawling at 15 months and autonomous walking at 2 years. Language development was particularly affected, with the appearance of the first word only at the age of 4 years, requiring twice-weekly speech therapy. Behaviourally, the child exhibited difficulties in social interaction, poor eye contact, stereotypies and intolerance to changes suggestive of autistic traits. The patient had also experienced generalized tonic-clonic epileptic seizures since the age of 3 years, currently well controlled by sodium valproate, with no recurrence since January 2023.

Clinical examination revealed a child with manifest growth retardation (weight = 15 kg, Z-score = –3; height = 104 cm, Z-score = –1.5) and characteristic facial dysmorphism combining microcephaly (head circumference = 48 cm, Z-score = –3), arched eyebrows, micrognathia and flat nose [Fig. 1]. Somatic examination revealed poor dental condition with multiple caries, a systolic heart murmur on auscultation and an inflamed tympanic membrane on ears, nose and throat (ENT) examination.

The initial paraclinical workup included brain magnetic resonance imaging (MRI) and electroencephalogram performed in February 2022, both normal. An audiological evaluation conducted in November 2023 revealed asymmetric bilateral hearing loss with pure-tone auditory thresholds at 20 dB on the left and 45 dB on the right on auditory evoked potentials. Impedancemetry showed measured tympanic volumes of 0.45 mL on the right and 0.46 mL on the left. A temporal bone computed tomography (CT) scan performed in October 2023 documented bilateral chronic otomastoiditis with opacification of mastoid air cells and middle ear cavity involvement.

A dental panoramic radiograph performed in December 2023 revealed macrodontia of the permanent upper central incisor germs, a characteristic sign but not yet clinically visible at this age [Fig. 2]. A second brain MRI from January 2024 remained normal, as did the Doppler echocardiography performed the same month. The standard karyotype (46,XY) performed in April 2022 was normal.

Dental panoramic radiograph revealing macrodontia of the permanent upper central incisor germs (arrows) before clinical eruption.

Artificial intelligence-assisted facial phenotyping using Face2Gene platform (FDNA™, Atlanta, Georgia, USA). The analysis correctly identified KBG syndrome (OMIM: 148050) as the top diagnostic hypothesis with high confidence scores for both gestalt and feature recognition. This digital phenotyping tool proved valuable in guiding genetic testing toward the ANKRD11 gene, where a novel pathogenic mutation was subsequently identified.

Genetic and clinical summary of KBG syndrome. Genetic findings (left panel) showing variant characteristics (c.1977C>G; p.Tyr659Ter), American College of Medical Genetics and Genomics classification (PVS1, PM2, PP5), sequencing methodology and ANKRD11 protein structure with functional domains. The schematic illustrates the location of the truncating variant at position 659, resulting in loss of 75% of the protein including the repression domain, activation domain and C-terminal region. Clinical summary (right panel) consolidating developmental milestones, growth parameters, clinical features across multiple systems, key investigations and management outcomes after 24 months of follow-up. RD = repression domain; AD = activation domain.

Given the constellation of clinical signs including neurodevelopmental delay, autistic traits, facial dysmorphism, epilepsy and hearing impairment, a facial morphometric analysis assisted by AI was used as a complementary diagnostic tool in December 2023. The frontal facial image of the patient, submitted to the Face2Gene platform (FDNA™, Atlanta, Georgia, USA), identified KBG syndrome among the 3 main diagnostic hypotheses with a high phenotypic similarity score [Fig. 3].

The diagnosis was genetically confirmed in January 2025 by whole exome sequencing. Genomic DNA was extracted from peripheral blood collected in EDTA tubes using the QIAmp DNA Blood Mini Kit (Qiagen, Venlo, Netherlands). DNA quantity and quality were assessed by spectrophotometry (Nanodrop 2000, Thermofisher Scientific, Waltham, Massachusetts, USA) and fluorometry (Qubit 3.0). Exome enrichment targeting coding regions and intronic flanking sequences up to 10 base pairs was performed using a hybridization-based protocol with the KAPA Hypercapture kit (Roche Diagnostics, Basel, Switzerland). Massive parallel sequencing was conducted on the DNBSEQ-G400 platform (MGI). Bioinformatic analysis was performed using GenoSystem Variant Analysis software, including quality control, filtering, alignment to the hg19 reference genome and annotation of variants in exonic and splicing regions with allelic frequency greater than 30%. Variants were annotated using ClinVar, LOVD, InSIGHT, UMD, ExAc and BIC databases. In silico pathogenicity prediction was performed using SIFT, PolyPhen-2, Mutation Taster, Provean and Human Splicing Finder. Variant classification followed American College of Medical Genetics and Genomics (ACMG) guidelines.

Sanger sequencing confirmation was not performed due to limited financial resources in the current setting. For the same reason, whole exome sequencing was performed as a singleton (proband-only) analysis, without parental samples for trio analysis or segregation study. Nevertheless, the variant was detected with high sequencing quality (allelic frequency 67.74%) and was independently classified as pathogenic by 3 laboratories in ClinVar (rs749201074). The concordance between the clinical phenotype and the expected KBG syndrome manifestations, including characteristic facial features, radiological macrodontia, hearing loss, epilepsy and neurodevelopmental delay, strongly supports the pathogenicity of this variant which identified a pathogenic heterozygous mutation in the ANKRD11 gene (NM_013275.5; c.1977C>G; p.Tyr659Ter). This nonsense variant, causing the substitution of a tyrosine by a premature STOP codon at position 659, is predicted to generate a truncated protein of 658 amino acids instead of the normal 2663 based on in silico analysis. The mutation was classified as pathogenic according to ACMG criteria (PVS1, PM2, PP5), definitively establishing the diagnosis of KBG syndrome [Fig. 4].

The patient currently benefits from a multidisciplinary management including neuropaediatric follow-up for his epilepsy, speech therapy for his language delay, behavioural intervention for his autistic traits and ENT follow-up for his hearing loss. After 24 months of follow-up, the clinical evolution has been favourable with significant neurodevelopmental improvement. The patient now demonstrates coherent language with effective verbal communication. Notably, the autistic traits initially observed have resolved, likely attributable to the successful management of recurrent ENT infections and subsequent improvement in hearing function. Epilepsy remains well controlled under valproate monotherapy with no seizure recurrence.

3. Discussion

The current observations revolve around 3 key diagnostic elements of KBG syndrome that allowed early identification of this rare syndrome in the patient: (1) a characteristic neurological and audiological presentation, (2) radiologically detectable incisor macrodontia before its clinical expression and (3) the contribution of digital facial phenotyping.

The patient presents several classic manifestations of KBG syndrome initially described by Herrmann et al.^1^ The psychomotor and language developmental delay, characteristic facial dysmorphism including microcephaly, arched eyebrows and micrognathia constitute cardinal elements of the syndrome. These clinical signs are consistent with recent descriptions of the KBG phenotype in the literature.^2^ As reported in the current case, low birth weight and insufficient growth are frequently found, reinforcing the need for specific nutritional monitoring. The presence of autistic traits in the patient is also consistent with the literature. The observed behavioural disorders (difficulties in social interaction, poor eye contact, stereotypies) are part of the neurodevelopmental manifestations described by Sirmaci et al. and Li et al.^34^ This association between KBG syndrome and autism spectrum disorder underscores the importance of systematic behavioural assessment in these patients.

The heterozygous nonsense mutation identified in the ANKRD11 gene (c.1977C>G; p.Tyr659Ter) in the current patient has not been previously reported. This variant results in the production of a severely truncated ANKRD11 protein (658/2,663 amino acids), adding to the pathogenic variants described by Murphy et al.^5^ The probable functional impact of this truncation is an alteration of ANKRD11's transcriptional regulation capabilities, consistent with its role in neuronal development highlighted by Gallagher et al.^6^ It is interesting to note that this truncated protein retains the N-terminal ankyrin repeat domain but loses the crucial repression and activation functional domains. This molecular configuration could explain certain phenotypic particularities observed in the current patient, notably the severity of language delay and autistic traits.

Comparison of the p.Tyr659Ter variant with previously reported truncating mutations provides insights into genotype-phenotype correlations in KBG syndrome. Li et al. demonstrated that intellectual disability severity correlates with variant position within the ANKRD11 gene: patients carrying truncating variants located between the first repression domain (RD1, aa 318–611) and the activation domain (AD, aa 1851–2145) showed significantly milder intellectual disability compared to those with variants disrupting RD2 alone or all functional domains.^4^ The p.Tyr659Ter mutation identified in the current patient, located at amino acid position 659, falls precisely within this intermediate region, which may explain the favourable neurodevelopmental outcome observed after 24 months of follow-up, including resolution of autistic traits and acquisition of coherent language. Martinez-Cayuelas et al. reported in the largest KBG cohort to date (340 patients) that sequence variants were associated with a higher phenotypic score compared to copy number variations, with specific variants showing variable neurodevelopmental outcomes.^2^ The variant generates a truncated protein retaining only 24.7% of the full-length sequence (658/2,663 amino acids), comparable to other early truncating variants such as c.2305delT (p.Ser769GlnfsX8) initially reported by Sirmaci et al.^3^ This genotype-phenotype correlation supports the hypothesis that early diagnosis and comprehensive management of associated comorbidities can positively influence developmental trajectory in patients with truncating variants in the RD1-AD region.

Audiological manifestations constitute an important diagnostic element in the atuhors' observation. The asymmetric bilateral hearing loss (20 dB on the left and 45 dB on the right) associated with chronic otomastoiditis perfectly matches the audiological phenotype recently characterized by Rhamati et al.^7^ Their multicentre study demonstrated that hearing loss associated with KBG syndrome is typically conductive (71%), bilateral (81%) and mild to moderate (84%), with frequent abnormalities of the ossicular chain on imaging. The recurrent otitis observed in the current patient also fits into this clinical picture, as recurrent middle ear infections are reported in nearly 70% of KBG patients with hearing loss.^7^ The temporal bone CT findings in the current patient, showing bilateral chronic otomastoiditis with mastoid and middle ear involvement, are consistent with the chronic otological complications frequently observed in this syndrome. This phenotypic correlation reinforces the importance of a complete audiological evaluation in any patient suspected of KBG syndrome, especially since hearing loss may precede the appearance of other characteristic signs such as macrodontia.

Epilepsy constitutes an important neurological manifestation in the current patient. The generalised tonic-clonic seizures that began at age 3 correspond to the data reported by Buijsse et al., who established that 26.9% of patients with KBG syndrome present with epilepsy, with a median age of seizure onset at 4 years.^8^ As in the current case, the majority of patients (77.8%) achieve satisfactory seizure control with one or two antiepileptic medications. It is important to note that epilepsy in KBG syndrome is associated with a less favourable neurodevelopmental prognosis according to Buijsse et al., emphasising the importance of close neurological monitoring.^8^ This particularity perfectly illustrates the interest of early diagnosis, which allowed appropriate management of epilepsy in the current patient, potentially beneficial for his neurodevelopmental prognosis.

In the current patient, the favourable neurodevelopmental outcome observed after 24 months of follow-up, including resolution of autistic traits and acquisition of coherent language, supports the hypothesis that early diagnosis and comprehensive management of associated comorbidities, particularly chronic otological infections affecting hearing, can positively influence the developmental trajectory in KBG syndrome.

A particularly original aspect of the authors' observation concerns the early radiological detection of macrodontia of the upper central incisors. Although the current patient had not yet developed clinically evident macrodontia, the panoramic radiographic examination revealed a proportionally increased size of the germs of these teeth. This finding is significant because macrodontia is classically considered a cardinal sign of KBG syndrome, but generally clinically visible only after the complete eruption of permanent teeth at 7–8 years of age. The authors' observation joins that of Rodrigues Alves Barbosa et al. who also emphasised the importance of dental imaging in the early identification of characteristic signs of KBG syndrome.^9^ The systematic use of dental panoramic radiography could thus contribute to reducing the frequent diagnostic wandering in this syndrome.

Although the current patient currently shows a good response to antiepileptic treatment, vigilance regarding long-term evolution is warranted. Babunovska et al. reported a case of KBG syndrome associated with drug-resistant epilepsy, suggesting a possible worsening of epilepsy over time.^10^ This variability of the epileptic phenotype underscores the need for prolonged neurological follow-up. Additionally, recent discoveries about the role of ANKRD11 in cardiac development reported by Kibalnyk et al. justify the cardiac monitoring established in the current patient, who presented with a systolic murmur on auscultation.^11^ The demonstration of ANKRD11's role in outflow tract remodelling by cardiac neural crest cells explains the frequent association of cardiac anomalies in this syndrome. The neurological manifestations of KBG syndrome continue to expand, as illustrated by Stehr et al.'s observation describing a movement disorder with predominant tremors.^12^ This phenotypic diversity reinforces the importance of a multidisciplinary approach in following these patients.

The use of facial morphometric analysis assisted by AI represents an innovative aspect of the authors' diagnostic approach. Face2Gene (FDNA™) correctly identified KBG syndrome among the main diagnostic hypotheses despite the absence of clinically visible macrodontia, demonstrating the utility of this technology for early detection of rare genetic syndromes. This digital facial phenotyping method proves particularly valuable in cases like the current one where certain cardinal signs are not yet apparent, but where other subtle facial features can guide toward the correct diagnosis. Carrer et al. recently demonstrated the effectiveness of this tool in an Italian cohort, with a diagnostic accuracy of 90% in the first 3 suggestions for all syndromes and 75% for ultra-rare syndromes such as KBG syndrome.^13^ This technological approach complements traditional clinical evaluation and can significantly contribute to reducing diagnostic wandering in rare genetic syndromes.

4. Conclusion

This report enriches the mutational spectrum of KBG syndrome by describing a novel pathogenic mutation in the ANKRD11 gene (c.1977C>G; p.Tyr659Ter) in a Moroccan child. The diagnostic triad of characteristic audio-neurological manifestations, radiological incisor macrodontia detectable before clinical eruption and digital facial phenotyping enabled early diagnosis at 5.5 years, before the typical age of 8 years. Early recognition through this multimodal approach combining clinical evaluation, audiological analysis, dental imaging and AI-assisted phenotyping allows for timely multidisciplinary management that could improve neurodevelopmental outcomes. Additional studies are needed to validate dental panoramic radiography and digital facial phenotyping as systematic screening tools for neurodevelopmental disorders of genetic origin.

Authors' contribution

Azzeddine Laaraje: Conceptualization, Methodology, Investigation, Data curation, Writing – original draft, Project administration. Khadija Belkadi Abassi: Investigation, Validation, Writing – review and editing. Mouna Lemaamer: Investigation, Resources, Writing – review and editing. Abdelilah Radi: Investigation, Validation, Writing – review and editing. Amale Hassani: Investigation, Validation, Writing – review and editing. Rachid Abilkassem: Supervision, Validation, Resources, Writing – review and editing.

Acknowledgement

The authors express their sincere gratitude to the patient's family for their cooperation and consent to publish this case report. We acknowledge the genetics laboratory team for performing the whole exome sequencing analysis and thank FDNA Inc. for providing access to the Face2Gene platform for AI-assisted facial phenotyping.

Ethics statement

Written informed consent was obtained from the patient's parents for publication of this case report and accompanying images.

Conflict of interest

The authors declare no conflict of interest.

Funding statement

No funding was received for this study.

Data availability

Data is available upon reasonable request from the corresponding author.